Compare commits

..

13 Commits

Author SHA1 Message Date
Dustin L. Howett
b39ba27612 Migrate spelling-0.0.21 changes from main 2020-12-17 11:15:57 -06:00
Dustin L. Howett
d148c536c7 Migrate spelling-0.0.19 changes from main 2020-12-17 11:15:57 -06:00
Mike Griese
a3faed6b7d finish renaming this test 2020-12-17 11:15:57 -06:00
Mike Griese
0579b2417b LocalTests_Remoting -> UnitTests_Remoting 2020-12-17 11:10:09 -06:00
Mike Griese
590b9ff0c7 this macro makes me feel dirty 2020-12-17 11:01:52 -06:00
Mike Griese
03bfc6e8a9 This works as a unittest, but not a local test. That's batty 2020-12-17 10:50:48 -06:00
Mike Griese
5cabcfb452 add a note to future me 2020-12-17 07:43:46 -06:00
Mike Griese
9a41647ffe HOLY SHIT I GOT THE COMANDLINE TO EXECUTE IN THE CURRENT WINDOW!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 2020-12-17 07:24:16 -06:00
Mike Griese
27ace16652 whoop, we pass the commandline from the peasant, to the monarch, and then back! 2020-12-16 16:31:51 -06:00
Mike Griese
36539cfa47 This won't work, but I'm committing this becaus I finally got it to compile a String[] 2020-12-16 16:03:23 -06:00
Mike Griese
5a9cdc8b0b Shockingly, this works, it works elevated, and it works unpackaged 2020-12-16 09:51:24 -06:00
Mike Griese
1f52d35833 Yank all the M/P files, this builds?! 2020-12-16 08:08:55 -06:00
Mike Griese
3bef7bbb38 Get all the projects created and hooked up to the sln 2020-12-16 07:59:52 -06:00
1203 changed files with 28176 additions and 100579 deletions

View File

@@ -1,12 +0,0 @@
{
"version": 1,
"isRoot": true,
"tools": {
"XamlStyler.Console": {
"version": "3.2008.4",
"commands": [
"xstyler"
]
}
}
}

View File

@@ -1,7 +1,6 @@
root = true
[*]
charset = utf-8
trim_trailing_whitespace = true
insert_final_newline = true

2
.gitattributes vendored
View File

@@ -3,8 +3,6 @@
###############################################################################
* -text
*.inc linguist-language=cpp
###############################################################################
# Set default behavior for command prompt diff.
#

54
.github/ISSUE_TEMPLATE/Bug_Report.md vendored Normal file
View File

@@ -0,0 +1,54 @@
---
name: "Bug report 🐛"
about: Report errors or unexpected behavior
title: ''
labels: ''
assignees: ''
---
<!--
🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
I ACKNOWLEDGE THE FOLLOWING BEFORE PROCEEDING:
1. If I delete this entire template and go my own path, the core team may close my issue without further explanation or engagement.
2. If I list multiple bugs/concerns in this one issue, the core team may close my issue without further explanation or engagement.
3. If I write an issue that has many duplicates, the core team may close my issue without further explanation or engagement (and without necessarily spending time to find the exact duplicate ID number).
4. If I leave the title incomplete when filing the issue, the core team may close my issue without further explanation or engagement.
5. If I file something completely blank in the body, the core team may close my issue without further explanation or engagement.
All good? Then proceed!
-->
<!--
This bug tracker is monitored by Windows Terminal development team and other technical folks.
**Important: When reporting BSODs or security issues, DO NOT attach memory dumps, logs, or traces to Github issues**.
Instead, send dumps/traces to secure@microsoft.com, referencing this GitHub issue.
If this is an application crash, please also provide a Feedback Hub submission link so we can find your diagnostic data on the backend. Use the category "Apps > Windows Terminal (Preview)" and choose "Share My Feedback" after submission to get the link.
Please use this form and describe your issue, concisely but precisely, with as much detail as possible.
-->
# Environment
```none
Windows build number: [run `[Environment]::OSVersion` for powershell, or `ver` for cmd]
Windows Terminal version (if applicable):
Any other software?
```
# Steps to reproduce
<!-- A description of how to trigger this bug. -->
# Expected behavior
<!-- A description of what you're expecting, possibly containing screenshots or reference material. -->
# Actual behavior
<!-- What's actually happening? -->

View File

@@ -1,53 +0,0 @@
name: "Bug report 🐛"
description: Report errors or unexpected behavior
body:
- type: markdown
attributes:
value: |
Please make sure to [search for existing issues](https://github.com/microsoft/terminal/issues) and [check the FAQ](https://github.com/microsoft/terminal/wiki/Frequently-Asked-Questions-(FAQ)) before filing a new one!
If this is an application crash, please also provide a [Feedback Hub](https://aka.ms/terminal-feedback-hub) submission link so we can find your diagnostic data on the backend. Use the category "Apps > Windows Terminal" and choose "Share My Feedback" after submission to get the link.
- type: input
attributes:
label: Windows Terminal version (or Windows build number)
placeholder: "10.0.19042.0, 1.7.3651.0"
description: |
If you are reporting an issue in Windows Terminal, you can find the version in the about dialog.
If you are reporting an issue with the Windows Console, please run `ver` or `[Environment]::OSVersion`.
validations:
required: true
- type: textarea
attributes:
label: Other Software
description: If you're reporting a bug about our interaction with other software, what software? What versions?
placeholder: |
vim 8.2 (inside WSL)
OpenSSH_for_Windows_8.1p1
My Cool Application v0.3 (include a code snippet if it would help!)
validations:
required: false
- type: textarea
attributes:
label: Steps to reproduce
placeholder: Tell us the steps required to trigger your bug.
validations:
required: true
- type: textarea
attributes:
label: Expected Behavior
description: If you want to include screenshots, paste them into the markdown editor below.
placeholder: What were you expecting?
validations:
required: false
- type: textarea
attributes:
label: Actual Behavior
placeholder: What happened instead?
validations:
required: true

View File

@@ -1,4 +1,4 @@
<?xml version="1.0" encoding="utf-8"?>
<packages>
<package id="vswhere" version="2.6.7" />
</packages>
</packages>

View File

@@ -1,13 +0,0 @@
{
// See https://go.microsoft.com/fwlink/?LinkId=827846 to learn about workspace recommendations.
// Extension identifier format: ${publisher}.${name}. Example: vscode.csharp
// List of extensions which should be recommended for users of this workspace.
"recommendations": [
"ms-vscode.cpptools"
],
// List of extensions recommended by VS Code that should not be recommended for users of this workspace.
"unwantedRecommendations": [
]
}

24
.vscode/launch.json vendored
View File

@@ -1,24 +0,0 @@
{
// Use IntelliSense to learn about possible attributes.
// Hover to view descriptions of existing attributes.
// For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
"version": "0.2.0",
"configurations": [
{
"name": "Debug OpenConsole by Launching (x64, debug)",
"type": "cppvsdbg",
"request": "launch",
"program": "${workspaceFolder}\\bin\\x64\\debug\\openconsole.exe",
"args": [],
"stopAtEntry": false,
"cwd": "${workspaceFolder}",
"environment": [],
},
{
"name": "Debug Terminal by Attaching (You go build/register/launch it first.)",
"type": "cppvsdbg",
"request": "attach",
"processId": "${command:pickProcess}"
}
]
}

105
.vscode/settings.json vendored
View File

@@ -1,105 +0,0 @@
{
"C_Cpp.default.browse.databaseFilename": "${workspaceFolder}\\.vscode\\.BROWSE.VC.DB",
"C_Cpp.default.browse.path": [
"${workspaceFolder}"
],
"C_Cpp.loggingLevel": "None",
"files.associations": {
"xstring": "cpp",
"*.idl": "cpp",
"array": "cpp",
"future": "cpp",
"istream": "cpp",
"memory": "cpp",
"tuple": "cpp",
"type_traits": "cpp",
"utility": "cpp",
"variant": "cpp",
"xlocmes": "cpp",
"xlocmon": "cpp",
"xlocnum": "cpp",
"xloctime": "cpp",
"multi_span": "cpp",
"pointers": "cpp",
"vector": "cpp",
"bitset": "cpp",
"deque": "cpp",
"initializer_list": "cpp",
"list": "cpp",
"queue": "cpp",
"random": "cpp",
"regex": "cpp",
"stack": "cpp",
"xhash": "cpp",
"xtree": "cpp",
"xutility": "cpp",
"span": "cpp",
"string_span": "cpp",
"algorithm": "cpp",
"atomic": "cpp",
"bit": "cpp",
"cctype": "cpp",
"charconv": "cpp",
"chrono": "cpp",
"clocale": "cpp",
"cmath": "cpp",
"compare": "cpp",
"complex": "cpp",
"concepts": "cpp",
"condition_variable": "cpp",
"cstdarg": "cpp",
"cstddef": "cpp",
"cstdint": "cpp",
"cstdio": "cpp",
"cstdlib": "cpp",
"cstring": "cpp",
"ctime": "cpp",
"cwchar": "cpp",
"cwctype": "cpp",
"exception": "cpp",
"filesystem": "cpp",
"fstream": "cpp",
"functional": "cpp",
"iomanip": "cpp",
"ios": "cpp",
"iosfwd": "cpp",
"iostream": "cpp",
"iterator": "cpp",
"limits": "cpp",
"locale": "cpp",
"map": "cpp",
"memory_resource": "cpp",
"mutex": "cpp",
"new": "cpp",
"numeric": "cpp",
"optional": "cpp",
"ostream": "cpp",
"ratio": "cpp",
"set": "cpp",
"shared_mutex": "cpp",
"sstream": "cpp",
"stdexcept": "cpp",
"stop_token": "cpp",
"streambuf": "cpp",
"string": "cpp",
"system_error": "cpp",
"thread": "cpp",
"typeinfo": "cpp",
"unordered_map": "cpp",
"unordered_set": "cpp",
"xfacet": "cpp",
"xiosbase": "cpp",
"xlocale": "cpp",
"xlocbuf": "cpp",
"xlocinfo": "cpp",
"xmemory": "cpp",
"xstddef": "cpp",
"xtr1common": "cpp"
},
"files.exclude": {
"**/bin/**": true,
"**/obj/**": true,
"**/packages/**": true,
"**/Generated Files/**": true
}
}

121
.vscode/tasks.json vendored
View File

@@ -1,121 +0,0 @@
{
"version": "2.0.0",
"tasks": [
{
"type": "process",
"label": "Build Terminal/Console",
"command": "powershell.exe",
"args": [
"-Command",
"Import-Module ${workspaceFolder}\\tools\\OpenConsole.psm1;",
"Set-MsBuildDevEnvironment;",
"$project = switch(\"${input:buildProjectChoice}\"){OpenConsole{\"Conhost\\Host_EXE\"} Terminal{\"Terminal\\CascadiaPackage\"} TermControl{\"Terminal\\TerminalControl\"}};",
"$target = switch(\"${input:buildModeChoice}\"){Build{\"\"} Rebuild{\":Rebuild\"} Clean{\":Clean\"}};",
"$target = $project + $target;",
"msbuild",
"${workspaceFolder}\\OpenConsole.sln",
"/p:Configuration=${input:configChoice}",
"/p:Platform=${input:platformChoice}",
"/p:AppxSymbolPackageEnabled=false", // This takes a long time, so false if we don't really need it.
"/t:$target",
"/m", // Parallel builds
"/verbosity:minimal"
],
"problemMatcher": ["$msCompile"],
"group": {
"kind": "build",
"isDefault": true
},
"runOptions": {
"reevaluateOnRerun": false,
"instanceLimit": 1,
"runOn": "default"
}
},
{
"type": "process",
"label": "Register Windows Terminal x64 Debug",
"command": "powershell.exe",
"args": [
"-Command",
"Import-Module ${workspaceFolder}\\tools\\OpenConsole.psm1;",
"Set-MsBuildDevEnvironment;",
"Set-Location -Path ${workspaceFolder}\\src\\cascadia\\CascadiaPackage\\AppPackages\\CascadiaPackage_0.0.1.0_x64_Debug_Test;",
"if ((Get-AppxPackage -Name 'WindowsTerminalDev*') -ne $null) { Remove-AppxPackage 'WindowsTerminalDev_0.0.1.0_x64__8wekyb3d8bbwe'};",
"New-Item ..\\loose -Type Directory -Force;",
"makeappx unpack /v /o /p .\\CascadiaPackage_0.0.1.0_x64_Debug.msix /d ..\\Loose\\;",
"Add-AppxPackage -Path ..\\loose\\AppxManifest.xml -Register -ForceUpdateFromAnyVersion -ForceApplicationShutdown"
],
"problemMatcher": ["$msCompile"],
"group": {
"kind": "build"
}
},
{
"type": "process",
"label": "Run Windows Terminal Dev",
"command": "wtd.exe",
"args": [
],
"problemMatcher": ["$msCompile"],
},
{
"type": "process",
"label": "Run Code Format",
"command": "powershell.exe",
"args": [
"-Command",
"Import-Module ${workspaceFolder}\\tools\\OpenConsole.psm1;",
"Set-MsBuildDevEnvironment;",
"Invoke-CodeFormat",
],
"problemMatcher": ["$msCompile"],
}
],
"inputs":[
{
"id": "platformChoice",
"type": "pickString",
"description": "Processor architecture choice",
"options":[
"x64",
"x86",
"arm64"
],
"default": "x64"
},
{
"id": "configChoice",
"type": "pickString",
"description": "Debug or release?",
"options":[
"Debug",
"Release"
],
"default": "Debug"
},
{
"id": "buildModeChoice",
"type": "pickString",
"description": "Build, rebuild, or clean?",
"options":[
"Build",
"Rebuild",
"Clean"
],
"default": "Build"
},
{
"id": "buildProjectChoice",
"type": "pickString",
"description": "OpenConsole or Terminal?",
"options":[
"OpenConsole",
"Terminal",
"TermControl"
],
"default": "Terminal"
}
]
}

View File

@@ -17,7 +17,7 @@
"Microsoft.Net.Component.4.5.TargetingPack",
"Microsoft.VisualStudio.Component.DiagnosticTools",
"Microsoft.VisualStudio.Component.Debugger.JustInTime",
"Microsoft.VisualStudio.Component.Windows10SDK.19041",
"Microsoft.VisualStudio.Component.Windows10SDK.18362",
"Microsoft.VisualStudio.ComponentGroup.UWP.Support",
"Microsoft.VisualStudio.Component.VC.CoreIde",
"Microsoft.VisualStudio.ComponentGroup.NativeDesktop.Core",

View File

@@ -43,7 +43,7 @@ If no existing item describes your issue/feature, great - please file a new issu
* Have a question that you don't see answered in docs, videos, etc.? File an issue
* Want to know if we're planning on building a particular feature? File an issue
* Got a great idea for a new feature? File an issue/request/idea
* Don't understand how to do something? File an issue
* Don't understand how to do something? File an issue/Community Guidance Request
* Found an existing issue that describes yours? Great - upvote and add additional commentary / info / repro-steps / etc.
When you hit "New Issue", select the type of issue closest to what you want to report/ask/request:
@@ -111,13 +111,13 @@ However, some issues/features will require careful thought & formal design befor
Specs help collaborators discuss different approaches to solve a problem, describe how the feature will behave, how the feature will impact the user, what happens if something goes wrong, etc. Driving towards agreement in a spec, before any code is written, often results in simpler code, and less wasted effort in the long run.
Specs will be managed in a very similar manner as code contributions so please follow the "[Fork, Branch and Create your PR](CONTRIBUTING.md#fork-clone-branch-and-create-your-pr)" section below.
Specs will be managed in a very similar manner as code contributions so please follow the "Fork, Branch and Create your PR" below.
### Writing / Contributing-to a Spec
To write/contribute to a spec: fork, branch and commit via PRs, as you would with any code changes.
Specs are written in markdown, stored under the [`\doc\specs`](./doc/specs) folder and named `[issue id] - [spec description].md`.
Specs are written in markdown, stored under the `\doc\spec` folder and named `[issue id] - [spec description].md`.
👉 **It is important to follow the spec templates and complete the requested information**. The available spec templates will help ensure that specs contain the minimum information & decisions necessary to permit development to begin. In particular, specs require you to confirm that you've already discussed the issue/idea with the team in an issue and that you provide the issue ID for reference.
@@ -125,7 +125,7 @@ Team members will be happy to help review specs and guide them to completion.
### Help Wanted
Once the team has approved an issue/spec, development can proceed. If no developers are immediately available, the spec can be parked ready for a developer to get started. Parked specs' issues will be labeled "Help Wanted". To find a list of development opportunities waiting for developer involvement, visit the Issues and filter on [the Help-Wanted label](https://github.com/microsoft/terminal/labels/Help%20Wanted).
Once the team have approved an issue/spec, development can proceed. If no developers are immediately available, the spec can be parked ready for a developer to get started. Parked specs' issues will be labeled "Help Wanted". To find a list of development opportunities waiting for developer involvement, visit the Issues and filter on [the Help-Wanted label](https://github.com/microsoft/terminal/labels/Help%20Wanted).
---

View File

@@ -117,6 +117,7 @@ LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
```
## dynamic_bitset
@@ -147,6 +148,7 @@ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
```
## \{fmt\}
@@ -213,6 +215,7 @@ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
```
@@ -246,78 +249,5 @@ SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE
FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE,
ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
DEALINGS IN THE SOFTWARE.
```
## PCG Random Number Generation
**Source**: [https://github.com/imneme/pcg-cpp](https://github.com/imneme/pcg-cpp)
### License
```
Copyright (c) 2014-2017 Melissa O'Neill and PCG Project contributors
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
```
# Microsoft Open Source
This product also incorporates source code from other Microsoft open source projects, all licensed under the MIT license.
## `GSL`
**Source**: [https://github.com/microsoft/GSL](https://github.com/microsoft/GSL)
## `Microsoft-UI-XAML`
**Source**: [https://github.com/microsoft/Microsoft-UI-XAML](https://github.com/microsoft/Microsoft-UI-XAML)
## `VirtualDesktopUtils`
**Source**: [https://github.com/microsoft/PowerToys](https://github.com/microsoft/PowerToys)
## `wil`
**Source**: [https://github.com/microsoft/wil](https://github.com/microsoft/wil)
### License
```
The MIT License
Copyright (c) Microsoft Corporation. All rights reserved.
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
```

View File

@@ -1,17 +1,25 @@
<?xml version="1.0" encoding="utf-8"?>
<configuration>
<packageSources>
<clear />
<!-- Dependencies that we can turn on to force override for testing purposes before uploading. -->
<add key="NuGet.org" value="https://api.nuget.org/v3/index.json" />
<!-- Add repositories here to the list of available repositories -->
<!-- Dependencies that we must carry because they're not on public nuget feeds right now. -->
<!--<add key="Static Package Dependencies" value="dep\packages" />-->
<!-- Use our own NuGet Feed -->
<add key="TerminalDependencies" value="https://pkgs.dev.azure.com/ms/terminal/_packaging/TerminalDependencies/nuget/v3/index.json" />
<!-- Temporarily? use the feeds from our friends in MUX for Helix test stuff -->
<add key="dotnetfeed" value="https://dotnetfeed.blob.core.windows.net/dotnet-core/index.json" />
<add key="dnceng" value="https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-eng/nuget/v3/index.json" />
<add key="MUX-Dependencies" value="https://pkgs.dev.azure.com/ms/microsoft-ui-xaml/_packaging/MUX-Dependencies/nuget/v3/index.json" />
<!-- Internal NuGet feeds that may not be accessible outside Microsoft corporate network -->
<!--<add key="TAEF - internal" value="https://microsoft.pkgs.visualstudio.com/DefaultCollection/_packaging/Taef/nuget/v3/index.json" />
<add key="OpenConsole - Internal" value="https://microsoft.pkgs.visualstudio.com/_packaging/OpenConsole/nuget/v3/index.json" />-->
</packageSources>
<disabledPackageSources>
<clear />
</disabledPackageSources>
<config>
<!-- TODO: TEMPORARY UNTIL PGO-Helpers is updated to search both. We shouldn't need to keep the globals path here. -->
<add key="globalPackagesFolder" value=".\packages" />
<add key="repositorypath" value=".\packages" />
</config>
</configuration>

File diff suppressed because it is too large Load Diff

View File

@@ -1,5 +1,3 @@
![terminal-logos](https://user-images.githubusercontent.com/48369326/115790869-4c852b00-a37c-11eb-97f1-f61972c7800c.png)
# Welcome to the Windows Terminal, Console and Command-Line repo
This repository contains the source code for:
@@ -35,23 +33,10 @@ This is our preferred method.
#### Via GitHub
For users who are unable to install Windows Terminal from the Microsoft Store,
released builds can be manually downloaded from this repository's [Releases
For users who are unable to install Terminal from the Microsoft Store, Terminal
builds can be manually downloaded from this repository's [Releases
page](https://github.com/microsoft/terminal/releases).
Download the `Microsoft.WindowsTerminal_<versionNumber>.msixbundle` file from
the **Assets** section. To install the app, you can simply double-click on the
`.msixbundle` file, and the app installer should automatically run. If that
fails for any reason, you can try the following command at a PowerShell prompt:
```powershell
# NOTE: If you are using PowerShell 7+, please run
# Import-Module Appx -UseWindowsPowerShell
# before using Add-AppxPackage.
Add-AppxPackage Microsoft.WindowsTerminal_<versionNumber>.msixbundle
```
> 🔴 Note: If you install Terminal manually:
>
> * Terminal will not auto-update when new builds are released so you will need
@@ -243,7 +228,7 @@ Visual Studio.
## Documentation
All project documentation is located at [aka.ms/terminal-docs](https://aka.ms/terminal-docs). If you would like
All project documentation is located at aka.ms/terminal-docs. If you would like
to contribute to the documentation, please submit a pull request on the [Windows
Terminal Documentation repo](https://github.com/MicrosoftDocs/terminal).
@@ -278,7 +263,6 @@ If you would like to ask a question that you feel doesn't warrant an issue
* Carlos Zamora, Developer: [@cazamor_msft](https://twitter.com/cazamor_msft)
* Leon Liang, Developer: [@leonmsft](https://twitter.com/leonmsft)
* Pankaj Bhojwani, Developer
* Leonard Hecker, Developer: [@LeonardHecker](https://twitter.com/LeonardHecker)
## Developer Guidance
@@ -289,7 +273,6 @@ If you would like to ask a question that you feel doesn't warrant an issue
* You must [enable Developer Mode in the Windows Settings
app](https://docs.microsoft.com/en-us/windows/uwp/get-started/enable-your-device-for-development)
to locally install and run Windows Terminal
* You must have [PowerShell 7 or later](https://github.com/PowerShell/PowerShell/releases/latest) installed
* You must have the [Windows 10 1903
SDK](https://developer.microsoft.com/en-us/windows/downloads/windows-10-sdk)
installed

View File

@@ -1,221 +0,0 @@
Microsoft Visual Studio Solution File, Format Version 12.00
# Visual Studio Version 16
VisualStudioVersion = 16.0.31205.134
MinimumVisualStudioVersion = 10.0.40219.1
Project("{C7167F0D-BC9F-4E6E-AFE1-012C56B48DB5}") = "Package", "scratch\ScratchIslandApp\Package\Package.wapproj", "{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}"
EndProject
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "SampleAppLib", "scratch\ScratchIslandApp\SampleApp\SampleAppLib.vcxproj", "{A4394404-37F7-41C1-802B-49788D3720E3}"
EndProject
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "SampleApp", "scratch\ScratchIslandApp\SampleApp\dll\SampleApp.vcxproj", "{26C51792-41A3-4FE0-AB5E-8B69D557BF91}"
ProjectSection(ProjectDependencies) = postProject
{A4394404-37F7-41C1-802B-49788D3720E3} = {A4394404-37F7-41C1-802B-49788D3720E3}
EndProjectSection
EndProject
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "WindowExe", "scratch\ScratchIslandApp\WindowExe\WindowExe.vcxproj", "{B4427499-9FDE-4208-B456-5BC580637633}"
ProjectSection(ProjectDependencies) = postProject
{26C51792-41A3-4FE0-AB5E-8B69D557BF91} = {26C51792-41A3-4FE0-AB5E-8B69D557BF91}
EndProjectSection
EndProject
Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "Common Props", "Common Props", "{53DD5520-E64C-4C06-B472-7CE62CA539C9}"
ProjectSection(SolutionItems) = preProject
src\common.build.post.props = src\common.build.post.props
src\common.build.pre.props = src\common.build.pre.props
src\common.build.tests.props = src\common.build.tests.props
common.openconsole.props = common.openconsole.props
src\cppwinrt.build.post.props = src\cppwinrt.build.post.props
src\cppwinrt.build.pre.props = src\cppwinrt.build.pre.props
src\wap-common.build.post.props = src\wap-common.build.post.props
src\wap-common.build.pre.props = src\wap-common.build.pre.props
EndProjectSection
EndProject
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "fmt", "src\dep\fmt\fmt.vcxproj", "{6BAE5851-50D5-4934-8D5E-30361A8A40F3}"
EndProject
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "Types", "src\types\lib\types.vcxproj", "{18D09A24-8240-42D6-8CB6-236EEE820263}"
EndProject
Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "dependencies", "dependencies", "{75AC9360-76FD-4ABC-AFEC-EF342BD2B3E9}"
EndProject
Global
GlobalSection(SolutionConfigurationPlatforms) = preSolution
AuditMode|ARM64 = AuditMode|ARM64
AuditMode|x64 = AuditMode|x64
AuditMode|x86 = AuditMode|x86
Debug|ARM64 = Debug|ARM64
Debug|x64 = Debug|x64
Debug|x86 = Debug|x86
Fuzzing|ARM64 = Fuzzing|ARM64
Fuzzing|x64 = Fuzzing|x64
Fuzzing|x86 = Fuzzing|x86
Release|ARM64 = Release|ARM64
Release|x64 = Release|x64
Release|x86 = Release|x86
EndGlobalSection
GlobalSection(ProjectConfigurationPlatforms) = postSolution
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.AuditMode|ARM64.ActiveCfg = Debug|ARM64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.AuditMode|ARM64.Build.0 = Debug|ARM64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.AuditMode|ARM64.Deploy.0 = Debug|ARM64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.AuditMode|x64.ActiveCfg = Debug|x64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.AuditMode|x64.Build.0 = Debug|x64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.AuditMode|x64.Deploy.0 = Debug|x64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.AuditMode|x86.ActiveCfg = Debug|x86
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.AuditMode|x86.Build.0 = Debug|x86
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.AuditMode|x86.Deploy.0 = Debug|x86
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Debug|ARM64.ActiveCfg = Debug|ARM64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Debug|ARM64.Build.0 = Debug|ARM64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Debug|ARM64.Deploy.0 = Debug|ARM64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Debug|x64.ActiveCfg = Debug|x64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Debug|x64.Build.0 = Debug|x64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Debug|x64.Deploy.0 = Debug|x64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Debug|x86.ActiveCfg = Debug|x86
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Debug|x86.Build.0 = Debug|x86
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Debug|x86.Deploy.0 = Debug|x86
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Fuzzing|ARM64.ActiveCfg = Debug|ARM64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Fuzzing|ARM64.Build.0 = Debug|ARM64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Fuzzing|ARM64.Deploy.0 = Debug|ARM64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Fuzzing|x64.ActiveCfg = Debug|x64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Fuzzing|x64.Build.0 = Debug|x64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Fuzzing|x64.Deploy.0 = Debug|x64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Fuzzing|x86.ActiveCfg = Debug|x86
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Fuzzing|x86.Build.0 = Debug|x86
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Fuzzing|x86.Deploy.0 = Debug|x86
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Release|ARM64.ActiveCfg = Release|ARM64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Release|ARM64.Build.0 = Release|ARM64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Release|ARM64.Deploy.0 = Release|ARM64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Release|x64.ActiveCfg = Release|x64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Release|x64.Build.0 = Release|x64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Release|x64.Deploy.0 = Release|x64
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Release|x86.ActiveCfg = Release|x86
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Release|x86.Build.0 = Release|x86
{CF31505E-3BAE-4C0A-81D7-F1EB279F40BB}.Release|x86.Deploy.0 = Release|x86
{A4394404-37F7-41C1-802B-49788D3720E3}.AuditMode|ARM64.ActiveCfg = AuditMode|ARM64
{A4394404-37F7-41C1-802B-49788D3720E3}.AuditMode|ARM64.Build.0 = AuditMode|ARM64
{A4394404-37F7-41C1-802B-49788D3720E3}.AuditMode|x64.ActiveCfg = AuditMode|x64
{A4394404-37F7-41C1-802B-49788D3720E3}.AuditMode|x64.Build.0 = AuditMode|x64
{A4394404-37F7-41C1-802B-49788D3720E3}.AuditMode|x86.ActiveCfg = AuditMode|Win32
{A4394404-37F7-41C1-802B-49788D3720E3}.AuditMode|x86.Build.0 = AuditMode|Win32
{A4394404-37F7-41C1-802B-49788D3720E3}.Debug|ARM64.ActiveCfg = Debug|ARM64
{A4394404-37F7-41C1-802B-49788D3720E3}.Debug|ARM64.Build.0 = Debug|ARM64
{A4394404-37F7-41C1-802B-49788D3720E3}.Debug|x64.ActiveCfg = Debug|x64
{A4394404-37F7-41C1-802B-49788D3720E3}.Debug|x64.Build.0 = Debug|x64
{A4394404-37F7-41C1-802B-49788D3720E3}.Debug|x86.ActiveCfg = Debug|Win32
{A4394404-37F7-41C1-802B-49788D3720E3}.Debug|x86.Build.0 = Debug|Win32
{A4394404-37F7-41C1-802B-49788D3720E3}.Fuzzing|ARM64.ActiveCfg = Fuzzing|ARM64
{A4394404-37F7-41C1-802B-49788D3720E3}.Fuzzing|ARM64.Build.0 = Fuzzing|ARM64
{A4394404-37F7-41C1-802B-49788D3720E3}.Fuzzing|x64.ActiveCfg = Fuzzing|x64
{A4394404-37F7-41C1-802B-49788D3720E3}.Fuzzing|x64.Build.0 = Fuzzing|x64
{A4394404-37F7-41C1-802B-49788D3720E3}.Fuzzing|x86.ActiveCfg = Fuzzing|Win32
{A4394404-37F7-41C1-802B-49788D3720E3}.Fuzzing|x86.Build.0 = Fuzzing|Win32
{A4394404-37F7-41C1-802B-49788D3720E3}.Release|ARM64.ActiveCfg = Release|ARM64
{A4394404-37F7-41C1-802B-49788D3720E3}.Release|ARM64.Build.0 = Release|ARM64
{A4394404-37F7-41C1-802B-49788D3720E3}.Release|x64.ActiveCfg = Release|x64
{A4394404-37F7-41C1-802B-49788D3720E3}.Release|x64.Build.0 = Release|x64
{A4394404-37F7-41C1-802B-49788D3720E3}.Release|x86.ActiveCfg = Release|Win32
{A4394404-37F7-41C1-802B-49788D3720E3}.Release|x86.Build.0 = Release|Win32
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.AuditMode|ARM64.ActiveCfg = AuditMode|ARM64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.AuditMode|ARM64.Build.0 = AuditMode|ARM64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.AuditMode|x64.ActiveCfg = AuditMode|x64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.AuditMode|x64.Build.0 = AuditMode|x64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.AuditMode|x86.ActiveCfg = AuditMode|Win32
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.AuditMode|x86.Build.0 = AuditMode|Win32
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Debug|ARM64.ActiveCfg = Debug|ARM64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Debug|ARM64.Build.0 = Debug|ARM64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Debug|x64.ActiveCfg = Debug|x64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Debug|x64.Build.0 = Debug|x64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Debug|x86.ActiveCfg = Debug|Win32
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Debug|x86.Build.0 = Debug|Win32
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Fuzzing|ARM64.ActiveCfg = Fuzzing|ARM64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Fuzzing|ARM64.Build.0 = Fuzzing|ARM64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Fuzzing|x64.ActiveCfg = Fuzzing|x64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Fuzzing|x64.Build.0 = Fuzzing|x64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Fuzzing|x86.ActiveCfg = Fuzzing|Win32
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Fuzzing|x86.Build.0 = Fuzzing|Win32
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Release|ARM64.ActiveCfg = Release|ARM64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Release|ARM64.Build.0 = Release|ARM64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Release|x64.ActiveCfg = Release|x64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Release|x64.Build.0 = Release|x64
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Release|x86.ActiveCfg = Release|Win32
{26C51792-41A3-4FE0-AB5E-8B69D557BF91}.Release|x86.Build.0 = Release|Win32
{B4427499-9FDE-4208-B456-5BC580637633}.AuditMode|ARM64.ActiveCfg = AuditMode|ARM64
{B4427499-9FDE-4208-B456-5BC580637633}.AuditMode|ARM64.Build.0 = AuditMode|ARM64
{B4427499-9FDE-4208-B456-5BC580637633}.AuditMode|x64.ActiveCfg = AuditMode|x64
{B4427499-9FDE-4208-B456-5BC580637633}.AuditMode|x64.Build.0 = AuditMode|x64
{B4427499-9FDE-4208-B456-5BC580637633}.AuditMode|x86.ActiveCfg = AuditMode|Win32
{B4427499-9FDE-4208-B456-5BC580637633}.AuditMode|x86.Build.0 = AuditMode|Win32
{B4427499-9FDE-4208-B456-5BC580637633}.Debug|ARM64.ActiveCfg = Debug|ARM64
{B4427499-9FDE-4208-B456-5BC580637633}.Debug|ARM64.Build.0 = Debug|ARM64
{B4427499-9FDE-4208-B456-5BC580637633}.Debug|x64.ActiveCfg = Debug|x64
{B4427499-9FDE-4208-B456-5BC580637633}.Debug|x64.Build.0 = Debug|x64
{B4427499-9FDE-4208-B456-5BC580637633}.Debug|x86.ActiveCfg = Debug|Win32
{B4427499-9FDE-4208-B456-5BC580637633}.Debug|x86.Build.0 = Debug|Win32
{B4427499-9FDE-4208-B456-5BC580637633}.Fuzzing|ARM64.ActiveCfg = Fuzzing|ARM64
{B4427499-9FDE-4208-B456-5BC580637633}.Fuzzing|ARM64.Build.0 = Fuzzing|ARM64
{B4427499-9FDE-4208-B456-5BC580637633}.Fuzzing|x64.ActiveCfg = Fuzzing|x64
{B4427499-9FDE-4208-B456-5BC580637633}.Fuzzing|x64.Build.0 = Fuzzing|x64
{B4427499-9FDE-4208-B456-5BC580637633}.Fuzzing|x86.ActiveCfg = Fuzzing|Win32
{B4427499-9FDE-4208-B456-5BC580637633}.Fuzzing|x86.Build.0 = Fuzzing|Win32
{B4427499-9FDE-4208-B456-5BC580637633}.Release|ARM64.ActiveCfg = Release|ARM64
{B4427499-9FDE-4208-B456-5BC580637633}.Release|ARM64.Build.0 = Release|ARM64
{B4427499-9FDE-4208-B456-5BC580637633}.Release|x64.ActiveCfg = Release|x64
{B4427499-9FDE-4208-B456-5BC580637633}.Release|x64.Build.0 = Release|x64
{B4427499-9FDE-4208-B456-5BC580637633}.Release|x86.ActiveCfg = Release|Win32
{B4427499-9FDE-4208-B456-5BC580637633}.Release|x86.Build.0 = Release|Win32
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.AuditMode|ARM64.ActiveCfg = AuditMode|ARM64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.AuditMode|ARM64.Build.0 = AuditMode|ARM64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.AuditMode|x64.ActiveCfg = AuditMode|x64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.AuditMode|x64.Build.0 = AuditMode|x64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.AuditMode|x86.ActiveCfg = AuditMode|Win32
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.AuditMode|x86.Build.0 = AuditMode|Win32
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Debug|ARM64.ActiveCfg = Debug|ARM64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Debug|ARM64.Build.0 = Debug|ARM64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Debug|x64.ActiveCfg = Debug|x64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Debug|x64.Build.0 = Debug|x64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Debug|x86.ActiveCfg = Debug|Win32
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Debug|x86.Build.0 = Debug|Win32
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Fuzzing|ARM64.ActiveCfg = Fuzzing|ARM64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Fuzzing|ARM64.Build.0 = Fuzzing|ARM64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Fuzzing|x64.ActiveCfg = Fuzzing|x64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Fuzzing|x64.Build.0 = Fuzzing|x64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Fuzzing|x86.ActiveCfg = Fuzzing|Win32
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Fuzzing|x86.Build.0 = Fuzzing|Win32
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Release|ARM64.ActiveCfg = Release|ARM64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Release|ARM64.Build.0 = Release|ARM64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Release|x64.ActiveCfg = Release|x64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Release|x64.Build.0 = Release|x64
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Release|x86.ActiveCfg = Release|Win32
{6BAE5851-50D5-4934-8D5E-30361A8A40F3}.Release|x86.Build.0 = Release|Win32
{18D09A24-8240-42D6-8CB6-236EEE820263}.AuditMode|ARM64.ActiveCfg = AuditMode|ARM64
{18D09A24-8240-42D6-8CB6-236EEE820263}.AuditMode|ARM64.Build.0 = AuditMode|ARM64
{18D09A24-8240-42D6-8CB6-236EEE820263}.AuditMode|x64.ActiveCfg = AuditMode|x64
{18D09A24-8240-42D6-8CB6-236EEE820263}.AuditMode|x64.Build.0 = AuditMode|x64
{18D09A24-8240-42D6-8CB6-236EEE820263}.AuditMode|x86.ActiveCfg = AuditMode|Win32
{18D09A24-8240-42D6-8CB6-236EEE820263}.AuditMode|x86.Build.0 = AuditMode|Win32
{18D09A24-8240-42D6-8CB6-236EEE820263}.Debug|ARM64.ActiveCfg = Debug|ARM64
{18D09A24-8240-42D6-8CB6-236EEE820263}.Debug|ARM64.Build.0 = Debug|ARM64
{18D09A24-8240-42D6-8CB6-236EEE820263}.Debug|x64.ActiveCfg = Debug|x64
{18D09A24-8240-42D6-8CB6-236EEE820263}.Debug|x64.Build.0 = Debug|x64
{18D09A24-8240-42D6-8CB6-236EEE820263}.Debug|x86.ActiveCfg = Debug|Win32
{18D09A24-8240-42D6-8CB6-236EEE820263}.Debug|x86.Build.0 = Debug|Win32
{18D09A24-8240-42D6-8CB6-236EEE820263}.Fuzzing|ARM64.ActiveCfg = Fuzzing|ARM64
{18D09A24-8240-42D6-8CB6-236EEE820263}.Fuzzing|ARM64.Build.0 = Fuzzing|ARM64
{18D09A24-8240-42D6-8CB6-236EEE820263}.Fuzzing|x64.ActiveCfg = Fuzzing|x64
{18D09A24-8240-42D6-8CB6-236EEE820263}.Fuzzing|x64.Build.0 = Fuzzing|x64
{18D09A24-8240-42D6-8CB6-236EEE820263}.Fuzzing|x86.ActiveCfg = Fuzzing|Win32
{18D09A24-8240-42D6-8CB6-236EEE820263}.Fuzzing|x86.Build.0 = Fuzzing|Win32
{18D09A24-8240-42D6-8CB6-236EEE820263}.Release|ARM64.ActiveCfg = Release|ARM64
{18D09A24-8240-42D6-8CB6-236EEE820263}.Release|ARM64.Build.0 = Release|ARM64
{18D09A24-8240-42D6-8CB6-236EEE820263}.Release|x64.ActiveCfg = Release|x64
{18D09A24-8240-42D6-8CB6-236EEE820263}.Release|x64.Build.0 = Release|x64
{18D09A24-8240-42D6-8CB6-236EEE820263}.Release|x86.ActiveCfg = Release|Win32
{18D09A24-8240-42D6-8CB6-236EEE820263}.Release|x86.Build.0 = Release|Win32
EndGlobalSection
GlobalSection(SolutionProperties) = preSolution
HideSolutionNode = FALSE
EndGlobalSection
GlobalSection(NestedProjects) = preSolution
{6BAE5851-50D5-4934-8D5E-30361A8A40F3} = {75AC9360-76FD-4ABC-AFEC-EF342BD2B3E9}
{18D09A24-8240-42D6-8CB6-236EEE820263} = {75AC9360-76FD-4ABC-AFEC-EF342BD2B3E9}
EndGlobalSection
GlobalSection(ExtensibilityGlobals) = postSolution
SolutionGuid = {05EAE315-9188-4D7B-B889-7D5F480A8915}
EndGlobalSection
EndGlobal

View File

@@ -1,41 +0,0 @@
{
"AttributesTolerance": 1,
"KeepFirstAttributeOnSameLine": true,
"MaxAttributeCharactersPerLine": 0,
"MaxAttributesPerLine": 1,
"NewlineExemptionElements": "RadialGradientBrush, GradientStop, LinearGradientBrush, ScaleTransform, SkewTransform, RotateTransform, TranslateTransform, Trigger, Condition, Setter",
"SeparateByGroups": false,
"AttributeIndentation": 0,
"AttributeIndentationStyle": 1,
"RemoveDesignTimeReferences": false,
"EnableAttributeReordering": true,
"AttributeOrderingRuleGroups": [
"x:Class",
"xmlns, xmlns:x",
"xmlns:*",
"x:Key, Key, x:Name, Name, x:Uid, Uid, Title",
"Grid.Row, Grid.RowSpan, Grid.Column, Grid.ColumnSpan, Canvas.Left, Canvas.Top, Canvas.Right, Canvas.Bottom",
"Width, Height, MinWidth, MinHeight, MaxWidth, MaxHeight",
"Margin, Padding, HorizontalAlignment, VerticalAlignment, HorizontalContentAlignment, VerticalContentAlignment, Panel.ZIndex",
"*:*, *",
"PageSource, PageIndex, Offset, Color, TargetName, Property, Value, StartPoint, EndPoint",
"mc:Ignorable, d:IsDataSource, d:LayoutOverrides, d:IsStaticText",
"Storyboard.*, From, To, Duration"
],
"FirstLineAttributes": "",
"OrderAttributesByName": true,
"PutEndingBracketOnNewLine": false,
"RemoveEndingTagOfEmptyElement": true,
"SpaceBeforeClosingSlash": true,
"RootElementLineBreakRule": 0,
"ReorderVSM": 2,
"ReorderGridChildren": false,
"ReorderCanvasChildren": false,
"ReorderSetters": 0,
"FormatMarkupExtension": true,
"NoNewLineMarkupExtensions": "x:Bind, Binding",
"ThicknessSeparator": 2,
"ThicknessAttributes": "Margin, Padding, BorderThickness, ThumbnailClipMargin",
"FormatOnSave": true,
"CommentPadding": 2,
}

View File

@@ -29,147 +29,4 @@ function GetQueryTestRunsUri
$baseUri = GetAzureDevOpsBaseUri -CollectionUri $CollectionUri -TeamProject $TeamProject
$queryUri = "$baseUri/_apis/test/runs?buildUri=$BuildUri$includeRunDetailsParameter&api-version=5.0"
return $queryUri
}
function Get-HelixJobTypeFromTestRun
{
Param ($testRun)
$testRunSingleResultUri = "$($testRun.url)/results?`$top=1&`$skip=0&api-version=5.1"
$singleTestResult = Invoke-RestMethod -Uri $testRunSingleResultUri -Method Get -Headers $azureDevOpsRestApiHeaders
$count = $singleTestResult.value.Length
if($count -eq 0)
{
# If the count is 0, then results have not yet been reported for this run.
# We only care about completed runs with results, so it is ok to just return 'UNKNOWN' for this run.
return "UNKNOWN"
}
else
{
$info = ConvertFrom-Json $singleTestResult.value.comment
$helixJobId = $info.HelixJobId
$job = Invoke-RestMethodWithRetries "https://helix.dot.net/api/2019-06-17/jobs/${helixJobId}?access_token=${HelixAccessToken}"
return $job.Type
}
}
function Append-HelixAccessTokenToUrl
{
Param ([string]$url, [string]$token)
if($url.Contains("?"))
{
$url = "$($url)&access_token=$($token)"
}
else
{
$url = "$($url)?access_token=$($token)"
}
return $url
}
# The Helix Rest api is sometimes unreliable. So we call these apis with retry logic.
# Note: The Azure DevOps apis are stable and do not need to be called with this retry logic.
$helixApiRetries = 0
$helixApiRetriesMax = 10
function Download-StringWithRetries
{
Param ([string]$fileName, [string]$url)
$result = ""
$done = $false
while(!($done))
{
try
{
Write-Host "Downloading $fileName"
$result = (New-Object System.Net.WebClient).DownloadString($url)
$done = $true
}
catch
{
Write-Host "Failed to download $fileName $($PSItem.Exception)"
$helixApiRetries = $helixApiRetries + 1
if($helixApiRetries -lt $helixApiRetriesMax)
{
Write-Host "Sleep and retry download of $fileName"
Start-Sleep 60
}
else
{
throw "Failed to download $fileName"
}
}
}
return $result
}
function Invoke-RestMethodWithRetries
{
Param ([string]$url,$Headers)
$result = @()
$done = $false
while(!($done))
{
try
{
$result = Invoke-RestMethod -Uri $url -Method Get -Headers $Headers
$done = $true
}
catch
{
Write-Host "Failed to invoke Rest method $($PSItem.Exception)"
$helixApiRetries = $helixApiRetries + 1
if($helixApiRetries -lt $helixApiRetriesMax)
{
Write-Host "Sleep and retry invoke"
Start-Sleep 60
}
else
{
throw "Failed to invoke Rest method"
}
}
}
return $result
}
function Download-FileWithRetries
{
Param ([string]$fileurl, [string]$destination)
$done = $false
while(!($done))
{
try
{
Write-Host "Downloading $destination"
$webClient.DownloadFile($fileurl, $destination)
$done = $true
}
catch
{
Write-Host "Failed to download $destination $($PSItem.Exception)"
$helixApiRetries = $helixApiRetries + 1
if($helixApiRetries -lt $helixApiRetriesMax)
{
Write-Host "Sleep and retry download of $destination"
Start-Sleep 60
}
else
{
throw "Failed to download $destination"
}
}
}
}

View File

@@ -21,7 +21,7 @@ Write-Host "Checking test results..."
$queryUri = GetQueryTestRunsUri -CollectionUri $CollectionUri -TeamProject $TeamProject -BuildUri $BuildUri -IncludeRunDetails
Write-Host "queryUri = $queryUri"
$testRuns = Invoke-RestMethodWithRetries $queryUri -Headers $azureDevOpsRestApiHeaders
$testRuns = Invoke-RestMethod -Uri $queryUri -Method Get -Headers $azureDevOpsRestApiHeaders
[System.Collections.Generic.List[string]]$failingTests = @()
[System.Collections.Generic.List[string]]$unreliableTests = @()
[System.Collections.Generic.List[string]]$unexpectedResultTest = @()
@@ -31,7 +31,7 @@ $totalTestsExecutedCount = 0
# We assume that we only have one testRun with a given name that we care about
# We only process the last testRun with a given name (based on completedDate)
# The name of a testRun is set to the Helix queue that it was run on (e.g. windows.10.amd64.client21h1.xaml)
# The name of a testRun is set to the Helix queue that it was run on (e.g. windows.10.amd64.client19h1.xaml)
# If we have multiple test runs on the same queue that we care about, we will need to re-visit this logic
foreach ($testRun in ($testRuns.value | Sort-Object -Property "completedDate" -Descending))
{
@@ -50,7 +50,7 @@ foreach ($testRun in ($testRuns.value | Sort-Object -Property "completedDate" -D
$totalTestsExecutedCount += $testRun.totalTests
$testRunResultsUri = "$($testRun.url)/results?api-version=5.0"
$testResults = Invoke-RestMethodWithRetries "$($testRun.url)/results?api-version=5.0" -Headers $azureDevOpsRestApiHeaders
$testResults = Invoke-RestMethod -Uri "$($testRun.url)/results?api-version=5.0" -Method Get -Headers $azureDevOpsRestApiHeaders
foreach ($testResult in $testResults.value)
{

View File

@@ -9,7 +9,7 @@ $payloadDir = "HelixPayload\$Configuration\$Platform"
$repoDirectory = Join-Path (Split-Path -Parent $script:MyInvocation.MyCommand.Path) "..\..\"
$nugetPackagesDir = Join-Path (Split-Path -Parent $script:MyInvocation.MyCommand.Path) "packages"
# Create the payload directory. Remove it if it already exists.
If(test-path $payloadDir)
{
@@ -19,13 +19,11 @@ New-Item -ItemType Directory -Force -Path $payloadDir
# Copy files from nuget packages
Copy-Item "$nugetPackagesDir\microsoft.windows.apps.test.1.0.181203002\lib\netcoreapp2.1\*.dll" $payloadDir
Copy-Item "$nugetPackagesDir\Microsoft.Taef.10.60.210621002\build\Binaries\$Platform\*" $payloadDir
Copy-Item "$nugetPackagesDir\Microsoft.Taef.10.60.210621002\build\Binaries\$Platform\NetFx4.5\*" $payloadDir
Copy-Item "$nugetPackagesDir\taef.redist.wlk.10.57.200731005-develop\build\Binaries\$Platform\*" $payloadDir
Copy-Item "$nugetPackagesDir\taef.redist.wlk.10.57.200731005-develop\build\Binaries\$Platform\CoreClr\*" $payloadDir
New-Item -ItemType Directory -Force -Path "$payloadDir\.NETCoreApp2.1\"
Copy-Item "$nugetPackagesDir\runtime.win-$Platform.microsoft.netcore.app.2.1.0\runtimes\win-$Platform\lib\netcoreapp2.1\*" "$payloadDir\.NETCoreApp2.1\"
Copy-Item "$nugetPackagesDir\runtime.win-$Platform.microsoft.netcore.app.2.1.0\runtimes\win-$Platform\native\*" "$payloadDir\.NETCoreApp2.1\"
New-Item -ItemType Directory -Force -Path "$payloadDir\content\"
Copy-Item "$nugetPackagesDir\Microsoft.Internal.Windows.Terminal.TestContent.1.0.1\content\*" "$payloadDir\content\"
function Copy-If-Exists
{
@@ -54,13 +52,3 @@ Copy-Item "build\helix\HelixTestHelpers.cs" "$payloadDir"
Copy-Item "build\helix\runtests.cmd" $payloadDir
Copy-Item "build\helix\InstallTestAppDependencies.ps1" "$payloadDir"
Copy-Item "build\Helix\EnsureMachineState.ps1" "$payloadDir"
# Copy the APPX package from the 'drop' artifact dir
Copy-Item "$repoDirectory\Artifacts\$ArtifactName\appx\CascadiaPackage_0.0.1.0_$Platform.msix" $payloadDir\CascadiaPackage.zip
# Rename it to extension of ZIP because Expand-Archive is real sassy on the build machines
# and refuses to unzip it because of its file extension while on a desktop, it just
# does the job without complaining.
# Extract the APPX package
Expand-Archive -LiteralPath $payloadDir\CascadiaPackage.zip -DestinationPath $payloadDir\appx

View File

@@ -9,6 +9,11 @@ Param(
$helixLinkFile = "$OutputFolder\LinksToHelixTestFiles.html"
$accessTokenParam = ""
if($HelixAccessToken)
{
$accessTokenParam = "?access_token=$HelixAccessToken"
}
function Generate-File-Links
{
@@ -20,31 +25,13 @@ function Generate-File-Links
Out-File -FilePath $helixLinkFile -Append -InputObject "<ul>"
foreach($file in $files)
{
$url = Append-HelixAccessTokenToUrl $file.Link "{Your-Helix-Access-Token-Here}"
Out-File -FilePath $helixLinkFile -Append -InputObject "<li>$($url)</li>"
Out-File -FilePath $helixLinkFile -Append -InputObject "<li><a href=$($file.Link)>$($file.Name)</a></li>"
}
Out-File -FilePath $helixLinkFile -Append -InputObject "</ul>"
Out-File -FilePath $helixLinkFile -Append -InputObject "</div>"
}
}
function Append-HelixAccessTokenToUrl
{
Param ([string]$url, [string]$token)
if($token)
{
if($url.Contains("?"))
{
$url = "$($url)&access_token=$($token)"
}
else
{
$url = "$($url)?access_token=$($token)"
}
}
return $url
}
#Create output directory
New-Item $OutputFolder -ItemType Directory
@@ -58,71 +45,66 @@ $azureDevOpsRestApiHeaders = @{
$queryUri = GetQueryTestRunsUri -CollectionUri $CollectionUri -TeamProject $TeamProject -BuildUri $BuildUri -IncludeRunDetails
Write-Host "queryUri = $queryUri"
$testRuns = Invoke-RestMethodWithRetries $queryUri -Headers $azureDevOpsRestApiHeaders
$testRuns = Invoke-RestMethod -Uri $queryUri -Method Get -Headers $azureDevOpsRestApiHeaders
$webClient = New-Object System.Net.WebClient
[System.Collections.Generic.List[string]]$workItems = @()
foreach ($testRun in $testRuns.value)
{
Write-Host "testRunUri = $testRun.url"
$testResults = Invoke-RestMethodWithRetries "$($testRun.url)/results?api-version=5.0" -Headers $azureDevOpsRestApiHeaders
$testResults = Invoke-RestMethod -Uri "$($testRun.url)/results?api-version=5.0" -Method Get -Headers $azureDevOpsRestApiHeaders
$isTestRunNameShown = $false
foreach ($testResult in $testResults.value)
{
$info = ConvertFrom-Json $testResult.comment
$helixJobId = $info.HelixJobId
$helixWorkItemName = $info.HelixWorkItemName
$workItem = "$helixJobId-$helixWorkItemName"
Write-Host "Helix Work Item = $workItem"
if (-not $workItems.Contains($workItem))
if ("comment" -in $testResult)
{
$workItems.Add($workItem)
$filesQueryUri = "https://helix.dot.net/api/2019-06-17/jobs/$helixJobId/workitems/$helixWorkItemName/files"
$filesQueryUri = Append-HelixAccessTokenToUrl $filesQueryUri $helixAccessToken
$files = Invoke-RestMethodWithRetries $filesQueryUri
$info = ConvertFrom-Json $testResult.comment
$helixJobId = $info.HelixJobId
$helixWorkItemName = $info.HelixWorkItemName
$screenShots = $files | where { $_.Name.EndsWith(".jpg") }
$dumps = $files | where { $_.Name.EndsWith(".dmp") }
$pgcFiles = $files | where { $_.Name.EndsWith(".pgc") }
if ($screenShots.Count + $dumps.Count + $pgcFiles.Count -gt 0)
$workItem = "$helixJobId-$helixWorkItemName"
if (-not $workItems.Contains($workItem))
{
if(-Not $isTestRunNameShown)
$workItems.Add($workItem)
$filesQueryUri = "https://helix.dot.net/api/2019-06-17/jobs/$helixJobId/workitems/$helixWorkItemName/files$accessTokenParam"
$files = Invoke-RestMethod -Uri $filesQueryUri -Method Get
$screenShots = $files | where { $_.Name.EndsWith(".jpg") }
$dumps = $files | where { $_.Name.EndsWith(".dmp") }
$pgcFiles = $files | where { $_.Name.EndsWith(".pgc") }
if ($screenShots.Count + $dumps.Count + $pgcFiles.Count -gt 0)
{
Out-File -FilePath $helixLinkFile -Append -InputObject "<h2>$($testRun.name)</h2>"
$isTestRunNameShown = $true
}
Out-File -FilePath $helixLinkFile -Append -InputObject "<h3>$helixWorkItemName</h3>"
Generate-File-Links $screenShots "Screenshots"
Generate-File-Links $dumps "CrashDumps"
Generate-File-Links $pgcFiles "PGC files"
$misc = $files | where { ($screenShots -NotContains $_) -And ($dumps -NotContains $_) -And ($visualTreeVerificationFiles -NotContains $_) -And ($pgcFiles -NotContains $_) }
Generate-File-Links $misc "Misc"
foreach($pgcFile in $pgcFiles)
{
$flavorPath = $testResult.automatedTestName.Split('.')[0]
$archPath = $testResult.automatedTestName.Split('.')[1]
$fileName = $pgcFile.Name
$fullPath = "$OutputFolder\PGO\$flavorPath\$archPath"
$destination = "$fullPath\$fileName"
Write-Host "Copying $($pgcFile.Name) to $destination"
if (-Not (Test-Path $fullPath))
if(-Not $isTestRunNameShown)
{
New-Item $fullPath -ItemType Directory
Out-File -FilePath $helixLinkFile -Append -InputObject "<h2>$($testRun.name)</h2>"
$isTestRunNameShown = $true
}
Out-File -FilePath $helixLinkFile -Append -InputObject "<h3>$helixWorkItemName</h3>"
Generate-File-Links $screenShots "Screenshots"
Generate-File-Links $dumps "CrashDumps"
Generate-File-Links $pgcFiles "PGC files"
$misc = $files | where { ($screenShots -NotContains $_) -And ($dumps -NotContains $_) -And ($visualTreeVerificationFiles -NotContains $_) -And ($pgcFiles -NotContains $_) }
Generate-File-Links $misc "Misc"
$link = $pgcFile.Link
foreach($pgcFile in $pgcFiles)
{
$flavorPath = $pgcFile.Name.Split('.')[0]
$archPath = $pgcFile.Name.Split('.')[1]
$fileName = $pgcFile.Name.Remove(0, $flavorPath.length + $archPath.length + 2)
$fullPath = "$OutputFolder\PGO\$flavorPath\$archPath"
$destination = "$fullPath\$fileName"
Write-Host "Downloading $link to $destination"
Write-Host "Copying $($pgcFile.Name) to $destination"
$link = Append-HelixAccessTokenToUrl $link $HelixAccessToken
Download-FileWithRetries $link $destination
if (-Not (Test-Path $fullPath))
{
New-Item $fullPath -ItemType Directory
}
$link = "$($pgcFile.Link)$accessTokenParam"
$webClient.DownloadFile($link, $destination)
}
}
}
}

View File

@@ -13,8 +13,6 @@
</ItemGroup>
<!-- These .proj files are generated by the build machine prior to running tests via GenerateTestProjFile.ps1. -->
<Import Project="$(ProjFilesPath)\$(Configuration)\$(Platform)\RunTestsInHelix-TerminalAppLocalTests.proj" Condition=" '$(TestSuite)'=='DevTestSuite' " />
<Import Project="$(ProjFilesPath)\$(Configuration)\$(Platform)\RunTestsInHelix-SettingsModelLocalTests.proj" Condition=" '$(TestSuite)'=='DevTestSuite' " />
<Import Project="$(ProjFilesPath)\$(Configuration)\$(Platform)\RunTestsInHelix-HostTestsUIA.proj" Condition=" '$(TestSuite)'=='DevTestSuite' " />
<Import Project="$(ProjFilesPath)\$(Configuration)\$(Platform)\RunTestsInHelix-WindowsTerminalUIATests.proj" Condition=" '$(TestSuite)'=='PgoInstrumentationSuite' " />
<Import Project="$(ProjFilesPath)\RunTestsInHelix-TerminalAppLocalTests.proj" Condition=" '$(TestSuite)'=='DevTestSuite' " />
<Import Project="$(ProjFilesPath)\RunTestsInHelix-HostTestsUIA.proj" Condition=" '$(TestSuite)'=='DevTestSuite' " />
</Project>

View File

@@ -23,7 +23,7 @@ Write-Host "queryUri = $queryUri"
# To account for unreliable tests, we'll iterate through all of the tests associated with this build, check to see any tests that were unreliable
# (denoted by being marked as "skipped"), and if so, we'll instead mark those tests with a warning and enumerate all of the attempted runs
# with their pass/fail states as well as any relevant error messages for failed attempts.
$testRuns = Invoke-RestMethodWithRetries $queryUri -Headers $azureDevOpsRestApiHeaders
$testRuns = Invoke-RestMethod -Uri $queryUri -Method Get -Headers $azureDevOpsRestApiHeaders
$timesSeenByRunName = @{}
@@ -32,10 +32,10 @@ foreach ($testRun in $testRuns.value)
$testRunResultsUri = "$($testRun.url)/results?api-version=5.0"
Write-Host "Marking test run `"$($testRun.name)`" as in progress so we can change its results to account for unreliable tests."
Invoke-RestMethod "$($testRun.url)?api-version=5.0" -Method Patch -Body (ConvertTo-Json @{ "state" = "InProgress" }) -Headers $azureDevOpsRestApiHeaders -ContentType "application/json" | Out-Null
Invoke-RestMethod -Uri "$($testRun.url)?api-version=5.0" -Method Patch -Body (ConvertTo-Json @{ "state" = "InProgress" }) -Headers $azureDevOpsRestApiHeaders -ContentType "application/json" | Out-Null
Write-Host "Retrieving test results..."
$testResults = Invoke-RestMethodWithRetries $testRunResultsUri -Headers $azureDevOpsRestApiHeaders
$testResults = Invoke-RestMethod -Uri $testRunResultsUri -Method Get -Headers $azureDevOpsRestApiHeaders
foreach ($testResult in $testResults.value)
{
@@ -54,8 +54,7 @@ foreach ($testRun in $testRuns.value)
Write-Host " Test $($testResult.testCaseTitle) was detected as unreliable. Updating..."
# The errorMessage field contains a link to the JSON-encoded rerun result data.
$resultsJson = Download-StringWithRetries "Error results" $testResult.errorMessage
$rerunResults = ConvertFrom-Json $resultsJson
$rerunResults = ConvertFrom-Json (New-Object System.Net.WebClient).DownloadString($testResult.errorMessage)
[System.Collections.Generic.List[System.Collections.Hashtable]]$rerunDataList = @()
$attemptCount = 0
$passCount = 0

View File

@@ -1,8 +1,7 @@
<?xml version="1.0" encoding="utf-8"?>
<packages>
<package id="MUXCustomBuildTasks" version="1.0.48" targetFramework="native" />
<package id="Microsoft.Internal.Windows.Terminal.TestContent" version="1.0.1" />
<package id="Microsoft.Taef" version="10.60.210621002" targetFramework="native" />
<package id="TAEF.Redist.Wlk" version="10.57.200731005-develop" targetFramework="native" />
<package id="microsoft.windows.apps.test" version="1.0.181203002" targetFramework="native" />
<package id="runtime.win-x86.microsoft.netcore.app" version="2.1.0" targetFramework="native" />
<package id="runtime.win-x64.microsoft.netcore.app" version="2.1.0" targetFramework="native" />

View File

@@ -28,7 +28,7 @@ echo %TIME%
powershell -ExecutionPolicy Bypass .\InstallTestAppDependencies.ps1
echo %TIME%
set testBinaryCandidates=TerminalApp.LocalTests.dll SettingsModel.LocalTests.dll Conhost.UIA.Tests.dll WindowsTerminal.UIA.Tests.dll
set testBinaryCandidates=TerminalApp.LocalTests.dll SettingsModel.LocalTests.dll Remoting.LocalTests.dll Conhost.UIA.Tests.dll
set testBinaries=
for %%B in (%testBinaryCandidates%) do (
if exist %%B (
@@ -46,6 +46,7 @@ move te.wtl te_original.wtl
copy /y te_original.wtl %HELIX_WORKITEM_UPLOAD_ROOT%
copy /y WexLogFileOutput\*.jpg %HELIX_WORKITEM_UPLOAD_ROOT%
for /f "tokens=* delims=" %%a in ('dir /b *.pgc') do ren "%%a" "%testnameprefix%.%%~na.pgc"
copy /y *.pgc %HELIX_WORKITEM_UPLOAD_ROOT%
set FailedTestQuery=

View File

@@ -1,24 +0,0 @@
{
"Branch": [
{
"collection": "microsoft",
"project": "OS",
"repo": "os.2020",
"name": "official/rs_wdx_dxp_windev",
"workitem": "38106206",
"CheckinFiles": [
{
"source": "WindowsTerminal.app.man",
"path": "/redist/mspartners/ipa/WindowsTerminal",
"type": "File"
}
]
}
],
"Email": [
{
"sendTo": "condev",
"sendOnErrorOnly": "False"
}
]
}

View File

@@ -1,11 +0,0 @@
<PoliCheckExclusions>
<!-- All strings must be UPPER CASE -->
<!--Each of these exclusions is a folder name -if \[name]\exists in the file path, it will be skipped -->
<Exclusion Type="FolderPathFull">winrt|.git|oss|packages</Exclusion>
<!--Each of these exclusions is a folder name -if any folder or file starts with "\[name]", it will be skipped -->
<!--<Exclusion Type="FolderPathStart">ABC|XYZ</Exclusion>-->
<!--Each of these file types will be completely skipped for the entire scan -->
<Exclusion Type="FileType">.PNG|.SVG|.BMP|.ICO</Exclusion>
<!--The specified file names will be skipped during the scan regardless which folder they are in -->
<!--<Exclusion Type="FileName">ABC.TXT|XYZ.CS</Exclusion>-->
</PoliCheckExclusions>

View File

@@ -1,6 +1,5 @@
<?xml version="1.0" encoding="utf-8"?>
<packages>
<package id="MUXCustomBuildTasks" version="1.0.48" targetFramework="native" />
<package id="Microsoft.Taef" version="10.60.210621002" targetFramework="native" />
<package id="Microsoft.Internal.PGO-Helpers.Cpp" version="0.2.34" targetFramework="native" />
<package id="TAEF.Redist.Wlk" version="10.57.200731005-develop" targetFramework="native" />
</packages>

View File

@@ -1,4 +0,0 @@
<Project xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
<Import Project="Terminal.PGO.props" />
<Import Project="$(PkgMicrosoft_PGO_Helpers_Cpp)\build\Microsoft.PGO-Helpers.Cpp.targets" />
</Project>

View File

@@ -1,16 +0,0 @@
<?xml version="1.0"?>
<package xmlns="http://schemas.microsoft.com/packaging/2010/07/nuspec.xsd">
<metadata>
<id>Microsoft.Internal.Windows.Terminal.PGODatabase</id>
<version>0</version>
<title>Windows Terminal PGO Database</title>
<authors>Microsoft</authors>
<owners>Microsoft</owners>
<requireLicenseAcceptance>false</requireLicenseAcceptance>
<description>Windows Terminal PGO Database</description>
</metadata>
<files>
<!-- The target directories for pgd files need to remain as is. PGO optimization pass will rely on this exact directory layout. -->
<file src="x64\*.pgd" target="tools\x64"/>
</files>
</package>

View File

@@ -1,51 +0,0 @@
<Project>
<PropertyGroup>
<NuGetPackageDirectory>$(MSBuildThisFileDirectory)..\..\packages</NuGetPackageDirectory>
<PkgMicrosoft_PGO_Helpers_Cpp>$(NuGetPackageDirectory)\Microsoft.Internal.PGO-Helpers.Cpp.0.2.34</PkgMicrosoft_PGO_Helpers_Cpp>
</PropertyGroup>
<!-- Get version information -->
<Import Project="$(MSBuildThisFileDirectory)..\..\custom.props" />
<PropertyGroup>
<!-- Optional, defaults to main. Name of the branch which will be used for calculating branch point. -->
<PGOBranch>release-1.12</PGOBranch>
<!-- Mandatory. Name of the NuGet package which will contain PGO databases for consumption by build system. -->
<PGOPackageName>Microsoft.Internal.Windows.Terminal.PGODatabase</PGOPackageName>
<!-- Mandatory. Major version number of the PGO database which should match the version of the product. This can be hardcoded or obtained from other sources in build system. -->
<PGOPackageVersionMajor>$(VersionMajor)</PGOPackageVersionMajor>
<!-- Mandatory. Minor version number of the PGO database which should match the version of the product. This can be hardcoded or obtained from other sources in build system. -->
<PGOPackageVersionMinor>$(VersionMinor)</PGOPackageVersionMinor>
<!-- Mandatory, defaults to 0. Patch version number of the PGO database which should match the version of the product. This can be hardcoded or obtained from other sources in build system. -->
<PGOPackageVersionPatch>0</PGOPackageVersionPatch>
<!-- Optional, defaults to empty. Prerelease version number of the PGO database which should match the version of the product. This can be hardcoded or obtained from other sources in build system. -->
<PGOPackageVersionPrerelease></PGOPackageVersionPrerelease>
<!-- Mandatory. Path to nuget.config file for the project. Path is relative to where the props file will be. -->
<PGONuGetConfigPath>$(MSBuildThisFileDirectory)..\..\nuget.config</PGONuGetConfigPath>
<!-- Mandatory. Path to PGO database NuSpec, see section below. -->
<PGONuspecPath>$(MSBuildThisFileDirectory)Terminal.PGO.DB.nuspec</PGONuspecPath>
<!-- Optional, defaults to true. Should verification result in build failure or error? -->
<PGOVerifyFailureTreatedAsError>true</PGOVerifyFailureTreatedAsError>
<!-- Optional, defaults to $(IntDir)\Nuspec. Temporary path where PGD files are copied to. -->
<PGONuspecBasePath>$(PGDPathForAllArch)</PGONuspecBasePath>
<!-- Optional, defaults to true. Specifies whether to use default information for PGD files. -->
<PGOUseDefaultPGDFileInfo>true</PGOUseDefaultPGDFileInfo>
<!-- Copying the PGO runtime brings along a CRT. If we do that under normal circumstances, WAPPROJ will get its grubby hands on it and mess up all the CRT shenanigans we have to pull to make MSIX packages happy. So only pull it in for Instrument builds.-->
<PGOCopyRuntime>false</PGOCopyRuntime>
<PGOCopyRuntime Condition="'$(PGOBuildMode)' == 'Instrument'">true</PGOCopyRuntime>
</PropertyGroup>
<!-- Import PGO-Helpers -->
<Import Project="$(PkgMicrosoft_PGO_Helpers_Cpp)\build\Microsoft.PGO-Helpers.Cpp.props" />
</Project>

View File

@@ -1,27 +0,0 @@
trigger: none
pr: none
variables:
- name: runCodesignValidationInjectionBG
value: false
# 0.0.yyMM.dd##
# 0.0.1904.0900
name: 0.0.$(Date:yyMM).$(Date:dd)$(Rev:rr)
stages:
- stage: Build_x64
displayName: Build x64
dependsOn: []
condition: succeeded()
jobs:
- template: ./templates/build-console-pgo.yml
parameters:
platform: x64
- stage: Publish_PGO_Databases
displayName: Publish PGO databases
dependsOn: ['Build_x64']
jobs:
- template: ./templates/pgo-build-and-publish-nuget-job.yml
parameters:
pgoArtifact: 'PGO'

View File

@@ -1,587 +1,48 @@
# This build should never run as CI or against a pull request.
trigger: none
pr: none
pool:
name: WinDevPool-L
demands: ImageOverride -equals WinDevVS16-latest
parameters:
- name: branding
displayName: "Branding (Build Type)"
type: string
default: Release
values:
- Release
- Preview
- name: buildTerminal
displayName: "Build Windows Terminal MSIX"
type: boolean
default: true
- name: runCompliance
displayName: "Run Compliance and Security Build"
type: boolean
default: true
- name: publishSymbolsToPublic
displayName: "Publish Symbols to MSDL"
type: boolean
default: true
- name: buildTerminalVPack
displayName: "Build Windows Terminal VPack"
type: boolean
default: false
- name: buildWPF
displayName: "Build Terminal WPF Control"
type: boolean
default: false
- name: pgoBuildMode
displayName: "PGO Build Mode"
type: string
default: Optimize
values:
- Optimize
- Instrument
- None
- name: buildConfigurations
type: object
default:
- Release
- name: buildPlatforms
type: object
default:
- x64
- x86
- arm64
- name: buildWindowsVersions
type: object
default:
- Win10
- Win11
variables:
TerminalInternalPackageVersion: "0.0.7"
baseYearForVersioning: 2019 # Used by build-console-int
versionMajor: 0
versionMinor: 1
name: $(BuildDefinitionName)_$(date:yyMM).$(date:dd)$(rev:rrr)
resources:
repositories:
- repository: self
type: git
ref: main
# When we move off PackageES for Versioning, we'll need to switch
# name to this format. For now, though, we need to use DayOfYear.Rev
# to unique our builds, as mandated by PackageES's Setup task.
# name: '$(versionMajor).$(versionMinor).$(DayOfYear)$(Rev:r).0'
#
# Build name/version number above must end with .0 to make the
# store publication machinery happy.
name: 'Terminal_$(date:yyMM).$(date:dd)$(rev:rrr)'
# Build Arguments:
# WindowsTerminalOfficialBuild=[true,false]
# true - this is running on our build agent
# false - running locally
# WindowsTerminalBranding=[Dev,Preview,Release]
# <none> - Development build resources (default)
# Preview - Preview build resources
# Release - regular build resources
jobs:
- job: Build
strategy:
matrix:
${{ each config in parameters.buildConfigurations }}:
${{ each platform in parameters.buildPlatforms }}:
${{ each windowsVersion in parameters.buildWindowsVersions }}:
${{ config }}_${{ platform }}_${{ windowsVersion }}:
BuildConfiguration: ${{ config }}
BuildPlatform: ${{ platform }}
TerminalTargetWindowsVersion: ${{ windowsVersion }}
displayName: Build
timeoutInMinutes: 240
cancelTimeoutInMinutes: 1
steps:
- checkout: self
clean: true
submodules: true
persistCredentials: True
- task: PkgESSetupBuild@12
displayName: Package ES - Setup Build
inputs:
disableOutputRedirect: true
- task: PowerShell@2
displayName: Rationalize Build Platform
inputs:
targetType: inline
script: >-
$Arch = "$(BuildPlatform)"
- template: ./templates/build-console-audit-job.yml
parameters:
platform: x64
If ($Arch -Eq "x86") { $Arch = "Win32" }
- template: ./templates/build-console-int.yml
parameters:
platform: x64
additionalBuildArguments: /p:WindowsTerminalOfficialBuild=true;WindowsTerminalBranding=Preview
Write-Host "##vso[task.setvariable variable=RationalizedBuildPlatform]${Arch}"
- task: NuGetToolInstaller@1
displayName: Use NuGet 5.10
inputs:
versionSpec: 5.10
- task: NuGetAuthenticate@0
# In the Microsoft Azure DevOps tenant, NuGetCommand is ambiguous.
# This should be `task: NuGetCommand@2`
- task: 333b11bd-d341-40d9-afcf-b32d5ce6f23b@2
displayName: Restore NuGet packages for extraneous build actions
inputs:
command: restore
feedsToUse: config
configPath: NuGet.config
restoreSolution: build/packages.config
restoreDirectory: '$(Build.SourcesDirectory)\packages'
- task: NuGetCommand@2
displayName: NuGet custom
inputs:
command: custom
selectOrConfig: config
nugetConfigPath: NuGet.Config
arguments: restore OpenConsole.sln -SolutionDirectory $(Build.SourcesDirectory)
# Pull the Windows SDK for the developer tools like the debuggers so we can index sources later
- template: .\templates\install-winsdk-steps.yml
- task: UniversalPackages@0
displayName: Download terminal-internal Universal Package
inputs:
feedListDownload: 2b3f8893-a6e8-411f-b197-a9e05576da48
packageListDownload: e82d490c-af86-4733-9dc4-07b772033204
versionListDownload: $(TerminalInternalPackageVersion)
- task: TouchdownBuildTask@1
displayName: Download Localization Files
inputs:
teamId: 7105
authId: $(TouchdownAppId)
authKey: $(TouchdownAppKey)
resourceFilePath: >-
src\cascadia\TerminalApp\Resources\en-US\Resources.resw
- template: ./templates/build-console-int.yml
parameters:
platform: x86
additionalBuildArguments: /p:WindowsTerminalOfficialBuild=true;WindowsTerminalBranding=Preview
src\cascadia\TerminalApp\Resources\en-US\ContextMenu.resw
- template: ./templates/build-console-int.yml
parameters:
platform: arm64
additionalBuildArguments: /p:WindowsTerminalOfficialBuild=true;WindowsTerminalBranding=Preview
src\cascadia\TerminalControl\Resources\en-US\Resources.resw
- template: ./templates/check-formatting.yml
src\cascadia\TerminalConnection\Resources\en-US\Resources.resw
src\cascadia\TerminalSettingsModel\Resources\en-US\Resources.resw
src\cascadia\TerminalSettingsEditor\Resources\en-US\Resources.resw
src\cascadia\CascadiaPackage\Resources\en-US\Resources.resw
appendRelativeDir: true
localizationTarget: false
pseudoSetting: Included
- task: PowerShell@2
displayName: Move Loc files one level up
inputs:
targetType: inline
script: >-
$Files = Get-ChildItem . -R -Filter 'Resources.resw' | ? FullName -Like '*en-US\*\Resources.resw'
$Files | % { Move-Item -Verbose $_.Directory $_.Directory.Parent.Parent -EA:Ignore }
pwsh: true
- task: PowerShell@2
displayName: Copy the Context Menu Loc Resources to CascadiaPackage
inputs:
filePath: ./build/scripts/Copy-ContextMenuResourcesToCascadiaPackage.ps1
pwsh: true
- task: PowerShell@2
displayName: Generate NOTICE.html from NOTICE.md
inputs:
filePath: .\build\scripts\Generate-ThirdPartyNotices.ps1
arguments: -MarkdownNoticePath .\NOTICE.md -OutputPath .\src\cascadia\CascadiaPackage\NOTICE.html
pwsh: true
- ${{ if eq(parameters.buildTerminal, true) }}:
- pwsh: |-
./build/scripts/Patch-ManifestsToWindowsVersion.ps1 -NewWindowsVersion "10.0.22000.0"
displayName: Update manifest target version to Win11 (if necessary)
condition: and(succeeded(), eq(variables['TerminalTargetWindowsVersion'], 'Win11'))
- task: VSBuild@1
displayName: Build solution **\OpenConsole.sln
condition: true
inputs:
solution: '**\OpenConsole.sln'
vsVersion: 16.0
msbuildArgs: /p:WindowsTerminalOfficialBuild=true /p:WindowsTerminalBranding=${{ parameters.branding }};PGOBuildMode=${{ parameters.pgoBuildMode }} /t:Terminal\CascadiaPackage /p:WindowsTerminalReleaseBuild=true /bl:$(Build.SourcesDirectory)\msbuild.binlog
platform: $(BuildPlatform)
configuration: $(BuildConfiguration)
clean: true
maximumCpuCount: true
- task: PublishBuildArtifacts@1
displayName: 'Publish Artifact: binlog'
condition: failed()
continueOnError: True
inputs:
PathtoPublish: $(Build.SourcesDirectory)\msbuild.binlog
ArtifactName: binlog-$(BuildPlatform)-$(TerminalTargetWindowsVersion)
- task: PowerShell@2
displayName: Check MSIX for common regressions
inputs:
targetType: inline
script: >-
$Package = Get-ChildItem -Recurse -Filter "CascadiaPackage_*.msix"
.\build\scripts\Test-WindowsTerminalPackage.ps1 -Verbose -Path $Package.FullName
pwsh: true
- ${{ if eq(parameters.buildWPF, true) }}:
- task: VSBuild@1
displayName: Build solution **\OpenConsole.sln for PublicTerminalCore
inputs:
solution: '**\OpenConsole.sln'
vsVersion: 16.0
msbuildArgs: /p:WindowsTerminalOfficialBuild=true /p:WindowsTerminalBranding=${{ parameters.branding }};PGOBuildMode=${{ parameters.pgoBuildMode }} /p:WindowsTerminalReleaseBuild=true /t:Terminal\wpf\PublicTerminalCore
platform: $(BuildPlatform)
configuration: $(BuildConfiguration)
- task: PowerShell@2
displayName: Source Index PDBs
inputs:
filePath: build\scripts\Index-Pdbs.ps1
arguments: -SearchDir '$(Build.SourcesDirectory)' -SourceRoot '$(Build.SourcesDirectory)' -recursive -Verbose -CommitId $(Build.SourceVersion)
errorActionPreference: silentlyContinue
- task: PowerShell@2
displayName: Run Unit Tests
condition: and(succeeded(), or(eq(variables['BuildPlatform'], 'x64'), eq(variables['BuildPlatform'], 'x86')))
enabled: False
inputs:
filePath: build\scripts\Run-Tests.ps1
arguments: -MatchPattern '*unit.test*.dll' -Platform '$(RationalizedBuildPlatform)' -Configuration '$(BuildConfiguration)'
- task: PowerShell@2
displayName: Run Feature Tests
condition: and(succeeded(), eq(variables['BuildPlatform'], 'x64'))
enabled: False
inputs:
filePath: build\scripts\Run-Tests.ps1
arguments: -MatchPattern '*feature.test*.dll' -Platform '$(RationalizedBuildPlatform)' -Configuration '$(BuildConfiguration)'
- ${{ if eq(parameters.buildTerminal, true) }}:
- task: CopyFiles@2
displayName: Copy *.appx/*.msix to Artifacts
inputs:
Contents: >-
**/*.appx
**/*.msix
**/*.appxsym
!**/Microsoft.VCLibs*.appx
TargetFolder: $(Build.ArtifactStagingDirectory)/appx
OverWrite: true
flattenFolders: true
- task: AzureArtifacts.manifest-generator-task.manifest-generator-task.ManifestGeneratorTask@0
displayName: 'Generate SBOM manifest'
inputs:
BuildDropPath: '$(System.ArtifactsDirectory)/appx'
- task: PublishBuildArtifacts@1
displayName: Publish Artifact (appx)
inputs:
PathtoPublish: $(Build.ArtifactStagingDirectory)/appx
ArtifactName: appx-$(BuildPlatform)-$(BuildConfiguration)-$(TerminalTargetWindowsVersion)
- ${{ if eq(parameters.buildWPF, true) }}:
- task: CopyFiles@2
displayName: Copy PublicTerminalCore.dll to Artifacts
inputs:
Contents: >-
**/PublicTerminalCore.dll
**/api-ms-win-core-synch-l1-2-0.dll
TargetFolder: $(Build.ArtifactStagingDirectory)/wpf
OverWrite: true
flattenFolders: true
- task: PublishBuildArtifacts@1
displayName: Publish Artifact (PublicTerminalCore)
inputs:
PathtoPublish: $(Build.ArtifactStagingDirectory)/wpf
ArtifactName: wpf-dll-$(BuildPlatform)-$(BuildConfiguration)-$(TerminalTargetWindowsVersion)
- task: PublishSymbols@2
displayName: Publish symbols path
continueOnError: True
inputs:
SearchPattern: |
$(Build.SourcesDirectory)/bin/**/*.pdb
$(Build.SourcesDirectory)/bin/**/*.exe
$(Build.SourcesDirectory)/bin/**/*.dll
IndexSources: false
SymbolServerType: TeamServices
- ${{ if eq(parameters.runCompliance, true) }}:
- template: ./templates/build-console-compliance-job.yml
- ${{ if eq(parameters.buildTerminal, true) }}:
- job: BundleAndSign
strategy:
matrix:
${{ each windowsVersion in parameters.buildWindowsVersions }}:
${{ windowsVersion }}:
TerminalTargetWindowsVersion: ${{ windowsVersion }}
displayName: Create and sign AppX/MSIX bundles
dependsOn: Build
steps:
- checkout: self
clean: true
submodules: true
persistCredentials: True
- task: PkgESSetupBuild@12
displayName: Package ES - Setup Build
inputs:
disableOutputRedirect: true
- ${{ each platform in parameters.buildPlatforms }}:
- task: DownloadBuildArtifacts@0
displayName: Download Artifacts ${{ platform }} $(TerminalTargetWindowsVersion)
inputs:
artifactName: appx-${{ platform }}-Release-$(TerminalTargetWindowsVersion)
# Add 3000 to the major version component, but only for the bundle.
# This is to ensure that it is newer than "2022.xx.yy.zz" or whatever the original bundle versions were before
# we switched to uniform naming.
- pwsh: |-
$VersionEpoch = 3000
$Components = "$(XES_APPXMANIFESTVERSION)" -Split "\."
$Components[0] = ([int]$Components[0] + $VersionEpoch)
$BundleVersion = $Components -Join "."
.\build\scripts\Create-AppxBundle.ps1 -InputPath "$(System.ArtifactsDirectory)" -ProjectName CascadiaPackage -BundleVersion $BundleVersion -OutputPath "$(System.ArtifactsDirectory)\Microsoft.WindowsTerminal_$(TerminalTargetWindowsVersion)_$(XES_APPXMANIFESTVERSION)_8wekyb3d8bbwe.msixbundle"
displayName: Create WindowsTerminal*.msixbundle
- task: EsrpCodeSigning@1
displayName: Submit *.msixbundle to ESRP for code signing
inputs:
ConnectedServiceName: 9d6d2960-0793-4d59-943e-78dcb434840a
FolderPath: $(System.ArtifactsDirectory)
Pattern: Microsoft.WindowsTerminal*.msixbundle
UseMinimatch: true
signConfigType: inlineSignParams
inlineOperation: >-
[
{
"KeyCode": "Dynamic",
"CertTemplateName": "WINMSAPP1ST",
"CertSubjectName": "CN=Microsoft Corporation, O=Microsoft Corporation, L=Redmond, S=Washington, C=US",
"OperationCode": "SigntoolSign",
"Parameters": {
"OpusName": "Microsoft",
"OpusInfo": "http://www.microsoft.com",
"FileDigest": "/fd \"SHA256\"",
"TimeStamp": "/tr \"http://rfc3161.gtm.corp.microsoft.com/TSS/HttpTspServer\" /td sha256"
},
"ToolName": "sign",
"ToolVersion": "1.0"
},
{
"KeyCode": "Dynamic",
"CertTemplateName": "WINMSAPP1ST",
"CertSubjectName": "CN=Microsoft Corporation, O=Microsoft Corporation, L=Redmond, S=Washington, C=US",
"OperationCode": "SigntoolVerify",
"Parameters": {},
"ToolName": "sign",
"ToolVersion": "1.0"
}
]
- task: PublishBuildArtifacts@1
displayName: 'Publish Artifact: appxbundle-signed'
inputs:
PathtoPublish: $(System.ArtifactsDirectory)
ArtifactName: appxbundle-signed-$(TerminalTargetWindowsVersion)
- ${{ if eq(parameters.buildWPF, true) }}:
- job: PackageAndSignWPF
strategy:
matrix:
${{ each config in parameters.buildConfigurations }}:
${{ config }}:
BuildConfiguration: ${{ config }}
displayName: Create NuGet Package (WPF Terminal Control)
dependsOn: Build
steps:
- checkout: self
clean: true
submodules: true
persistCredentials: True
- task: PkgESSetupBuild@12
displayName: Package ES - Setup Build
inputs:
disableOutputRedirect: true
- ${{ each platform in parameters.buildPlatforms }}:
- task: DownloadBuildArtifacts@0
displayName: Download ${{ platform }} PublicTerminalCore
inputs:
artifactName: wpf-dll-${{ platform }}-$(BuildConfiguration)-Win10
itemPattern: '**/*.dll'
downloadPath: bin\${{ platform }}\$(BuildConfiguration)\
extractTars: false
- task: PowerShell@2
displayName: Move downloaded artifacts around
inputs:
targetType: inline
# Find all artifact files and move them up a directory. Ugh.
script: |-
Get-ChildItem bin -Recurse -Directory -Filter wpf-dll-* | % {
$_ | Get-ChildItem -Recurse -File | % {
Move-Item -Verbose $_.FullName $_.Directory.Parent.FullName
}
}
Move-Item bin\x86 bin\Win32
- task: NuGetToolInstaller@1
displayName: Use NuGet 5.10.0
inputs:
versionSpec: 5.10.0
- task: NuGetCommand@2
displayName: NuGet restore copy
inputs:
selectOrConfig: config
nugetConfigPath: NuGet.Config
- task: VSBuild@1
displayName: Build solution **\OpenConsole.sln for WPF Control
inputs:
solution: '**\OpenConsole.sln'
vsVersion: 16.0
msbuildArgs: /p:WindowsTerminalReleaseBuild=$(UseReleaseBranding);Version=$(XES_PACKAGEVERSIONNUMBER) /t:Pack
platform: Any CPU
configuration: $(BuildConfiguration)
maximumCpuCount: true
- task: PublishSymbols@2
displayName: Publish symbols path
continueOnError: True
inputs:
SearchPattern: |
$(Build.SourcesDirectory)/bin/**/*.pdb
$(Build.SourcesDirectory)/bin/**/*.exe
$(Build.SourcesDirectory)/bin/**/*.dll
IndexSources: false
SymbolServerType: TeamServices
SymbolsArtifactName: Symbols_WPF_$(BuildConfiguration)
- task: CopyFiles@2
displayName: Copy *.nupkg to Artifacts
inputs:
Contents: '**/*Wpf*.nupkg'
TargetFolder: $(Build.ArtifactStagingDirectory)/nupkg
OverWrite: true
flattenFolders: true
- task: EsrpCodeSigning@1
displayName: Submit *.nupkg to ESRP for code signing
inputs:
ConnectedServiceName: 9d6d2960-0793-4d59-943e-78dcb434840a
FolderPath: $(Build.ArtifactStagingDirectory)/nupkg
Pattern: '*.nupkg'
UseMinimatch: true
signConfigType: inlineSignParams
inlineOperation: >-
[
{
"KeyCode": "CP-401405",
"OperationCode": "NuGetSign",
"Parameters": {},
"ToolName": "sign",
"ToolVersion": "1.0"
},
{
"KeyCode": "CP-401405",
"OperationCode": "NuGetVerify",
"Parameters": {},
"ToolName": "sign",
"ToolVersion": "1.0"
}
]
- task: PublishBuildArtifacts@1
displayName: Publish Artifact (nupkg)
inputs:
PathtoPublish: $(Build.ArtifactStagingDirectory)\nupkg
ArtifactName: wpf-nupkg-$(BuildConfiguration)
- ${{ if eq(parameters.publishSymbolsToPublic, true) }}:
- job: PublishSymbols
displayName: Publish Symbols
dependsOn: BundleAndSign
steps:
- checkout: self
clean: true
fetchDepth: 1
submodules: true
- task: PkgESSetupBuild@12
displayName: Package ES - Setup Build
# Download the appx-PLATFORM-CONFIG-VERSION artifact for every platform/version combo
- ${{ each platform in parameters.buildPlatforms }}:
- ${{ each windowsVersion in parameters.buildWindowsVersions }}:
- task: DownloadBuildArtifacts@0
displayName: Download Symbols ${{ platform }} ${{ windowsVersion }}
inputs:
artifactName: appx-${{ platform }}-Release-${{ windowsVersion }}
# It seems easier to do this -- download every appxsym -- then enumerate all the PDBs in the build directory for the
# public symbol push. Otherwise, we would have to list all of the PDB files one by one.
- pwsh: |-
mkdir $(Build.SourcesDirectory)/appxsym-temp
Get-ChildItem "$(System.ArtifactsDirectory)" -Filter *.appxsym -Recurse | % {
$src = $_.FullName
$dest = Join-Path "$(Build.SourcesDirectory)/appxsym-temp/" $_.Name
mkdir $dest
Write-Host "Extracting $src to $dest..."
tar -x -v -f $src -C $dest
}
displayName: Extract symbols for public consumption
# Pull the Windows SDK for the developer tools like the debuggers so we can index sources later
- template: .\templates\install-winsdk-steps.yml
- task: PowerShell@2
displayName: Source Index PDBs (the public ones)
inputs:
filePath: build\scripts\Index-Pdbs.ps1
arguments: -SearchDir '$(Build.SourcesDirectory)/appxsym-temp' -SourceRoot '$(Build.SourcesDirectory)' -recursive -Verbose -CommitId $(Build.SourceVersion)
# Publish the app symbols to the public MSDL symbol server
# accessible via https://msdl.microsoft.com/download/symbols
- task: PublishSymbols@2
displayName: 'Publish app symbols to MSDL'
inputs:
symbolsFolder: '$(Build.SourcesDirectory)/appxsym-temp'
searchPattern: '**/*.pdb'
SymbolsMaximumWaitTime: 30
SymbolServerType: 'TeamServices'
SymbolsProduct: 'Windows Terminal Application Binaries'
SymbolsVersion: '$(XES_APPXMANIFESTVERSION)'
# The ADO task does not support indexing of GitHub sources.
indexSources: false
detailedLog: true
# There is a bug which causes this task to fail if LIB includes an inaccessible path (even though it does not depend on it).
# To work around this issue, we just force LIB to be any dir that we know exists.
# Copied from https://github.com/microsoft/icu/blob/f869c214adc87415dfe751d81f42f1bca55dcf5f/build/azure-nuget.yml#L564-L583
env:
LIB: $(Build.SourcesDirectory)
ArtifactServices_Symbol_AccountName: microsoftpublicsymbols
ArtifactServices_Symbol_PAT: $(ADO_microsoftpublicsymbols_PAT)
- ${{ if eq(parameters.buildTerminalVPack, true) }}:
- job: VPack
displayName: Create Windows vPack
dependsOn: BundleAndSign
steps:
- checkout: self
clean: true
submodules: true
- task: PkgESSetupBuild@12
displayName: Package ES - Setup Build
- task: DownloadBuildArtifacts@0
displayName: Download Build Artifacts
inputs:
artifactName: appxbundle-signed-Win11
extractTars: false
- task: PowerShell@2
displayName: Rename and stage packages for vpack
inputs:
targetType: inline
script: >-
# Rename to known/fixed name for Windows build system
Get-ChildItem Microsoft.WindowsTerminal_Win11_*.msixbundle | Rename-Item -NewName { 'Microsoft.WindowsTerminal_8wekyb3d8bbwe.msixbundle' }
# Create vpack directory and place item inside
mkdir WindowsTerminal.app
mv Microsoft.WindowsTerminal_8wekyb3d8bbwe.msixbundle .\WindowsTerminal.app\
workingDirectory: $(System.ArtifactsDirectory)\appxbundle-signed-Win11
- task: PkgESVPack@12
displayName: 'Package ES - VPack'
env:
SYSTEM_ACCESSTOKEN: $(System.AccessToken)
inputs:
sourceDirectory: $(System.ArtifactsDirectory)\appxbundle-signed-Win11\WindowsTerminal.app
description: VPack for the Windows Terminal Application
pushPkgName: WindowsTerminal.app
owner: conhost
- task: PublishPipelineArtifact@1
displayName: 'Copy VPack Manifest to Drop'
inputs:
targetPath: $(XES_VPACKMANIFESTDIRECTORY)
artifactName: VPackManifest
- task: PkgESFCIBGit@12
displayName: 'Submit VPack Manifest to Windows'
inputs:
configPath: '$(Build.SourcesDirectory)\build\config\GitCheckin.json'
artifactsDirectory: $(XES_VPACKMANIFESTDIRECTORY)
prTimeOut: 5
...
- template: ./templates/release-sign-and-bundle.yml

View File

@@ -8,12 +8,9 @@ jobs:
variables:
BuildConfiguration: AuditMode
BuildPlatform: ${{ parameters.platform }}
pool:
${{ if eq(variables['System.CollectionUri'], 'https://dev.azure.com/ms/') }}:
name: WinDevPoolOSS-L
${{ if ne(variables['System.CollectionUri'], 'https://dev.azure.com/ms/') }}:
name: WinDevPool-L
demands: ImageOverride -equals WinDevVS16-latest
pool: "windevbuildagents"
# The public pool is also an option!
# pool: { vmImage: windows-2019 }
steps:
- checkout: self
@@ -25,17 +22,6 @@ jobs:
inputs:
versionSpec: 4.8.1
# In the Microsoft Azure DevOps tenant, NuGetCommand is ambiguous.
# This should be `task: NuGetCommand@2`
- task: 333b11bd-d341-40d9-afcf-b32d5ce6f23b@2
displayName: Restore NuGet packages for extraneous build actions
inputs:
command: restore
feedsToUse: config
configPath: NuGet.config
restoreSolution: build/packages.config
restoreDirectory: '$(Build.SourcesDirectory)\packages'
# In the Microsoft Azure DevOps tenant, NuGetCommand is ambiguous.
# This should be `task: NuGetCommand@2`
- task: 333b11bd-d341-40d9-afcf-b32d5ce6f23b@2

View File

@@ -11,32 +11,21 @@ jobs:
variables:
BuildConfiguration: ${{ parameters.configuration }}
BuildPlatform: ${{ parameters.platform }}
pool:
${{ if eq(variables['System.CollectionUri'], 'https://dev.azure.com/ms/') }}:
name: WinDevPoolOSS-L
${{ if ne(variables['System.CollectionUri'], 'https://dev.azure.com/ms/') }}:
name: WinDevPool-L
demands: ImageOverride -equals WinDevVS16-latest
pool: "windevbuildagents"
# The public pool is also an option!
# pool: { vmImage: windows-2019 }
steps:
- template: build-console-steps.yml
parameters:
additionalBuildArguments: ${{ parameters.additionalBuildArguments }}
# It appears that the Component Governance build task that gets automatically injected stopped working
# when we renamed our main branch.
- task: ms.vss-governance-buildtask.governance-build-task-component-detection.ComponentGovernanceComponentDetection@0
displayName: 'Component Detection'
condition: and(succeededOrFailed(), not(eq(variables['Build.Reason'], 'PullRequest')))
- template: helix-runtests-job.yml
parameters:
name: 'RunTestsInHelix'
dependsOn: Build${{ parameters.platform }}${{ parameters.configuration }}
condition: and(succeeded(), and(eq('${{ parameters.platform }}', 'x64'), not(eq(variables['Build.Reason'], 'PullRequest'))))
testSuite: 'DevTestSuite'
platform: ${{ parameters.platform }}
configuration: ${{ parameters.configuration }}
rerunPassesRequiredToAvoidFailure: ${{ parameters.rerunPassesRequiredToAvoidFailure }}
- template: helix-processtestresults-job.yml

View File

@@ -1,222 +0,0 @@
jobs:
- job: Compliance
# We don't *need* a matrix but there's no other way to set parameters on a "job"
# in the AzDO YAML syntax. It would have to be a "stage" or a "template".
# Doesn't matter. We're going to do compliance on Release x64 because
# that's the one all the tooling works against for sure.
strategy:
matrix:
Release_x64:
BuildConfiguration: Release
BuildPlatform: x64
displayName: Validate Security and Compliance
timeoutInMinutes: 240
steps:
- checkout: self
clean: true
submodules: true
persistCredentials: True
- task: PkgESSetupBuild@12
displayName: Package ES - Setup Build
inputs:
disableOutputRedirect: true
- task: PowerShell@2
displayName: Rationalize Build Platform
inputs:
targetType: inline
script: >-
$Arch = "$(BuildPlatform)"
If ($Arch -Eq "x86") { $Arch = "Win32" }
Write-Host "##vso[task.setvariable variable=RationalizedBuildPlatform]${Arch}"
- task: NuGetToolInstaller@1
displayName: Use NuGet 5.10
inputs:
versionSpec: 5.10
- task: NuGetAuthenticate@0
# In the Microsoft Azure DevOps tenant, NuGetCommand is ambiguous.
# This should be `task: NuGetCommand@2`
- task: 333b11bd-d341-40d9-afcf-b32d5ce6f23b@2
displayName: Restore NuGet packages for extraneous build actions
inputs:
command: restore
feedsToUse: config
configPath: NuGet.config
restoreSolution: build/packages.config
restoreDirectory: '$(Build.SourcesDirectory)\packages'
- task: NuGetCommand@2
displayName: NuGet custom
inputs:
command: custom
selectOrConfig: config
nugetConfigPath: NuGet.Config
arguments: restore OpenConsole.sln -SolutionDirectory $(Build.SourcesDirectory)
- task: UniversalPackages@0
displayName: Download terminal-internal Universal Package
inputs:
feedListDownload: 2b3f8893-a6e8-411f-b197-a9e05576da48
packageListDownload: e82d490c-af86-4733-9dc4-07b772033204
versionListDownload: $(TerminalInternalPackageVersion)
- task: TouchdownBuildTask@1
displayName: Download Localization Files
inputs:
teamId: 7105
authId: $(TouchdownAppId)
authKey: $(TouchdownAppKey)
resourceFilePath: >-
src\cascadia\TerminalApp\Resources\en-US\Resources.resw
src\cascadia\TerminalControl\Resources\en-US\Resources.resw
src\cascadia\TerminalConnection\Resources\en-US\Resources.resw
src\cascadia\TerminalSettingsModel\Resources\en-US\Resources.resw
src\cascadia\TerminalSettingsEditor\Resources\en-US\Resources.resw
src\cascadia\CascadiaPackage\Resources\en-US\Resources.resw
appendRelativeDir: true
localizationTarget: false
pseudoSetting: Included
- task: PowerShell@2
displayName: Move Loc files one level up
inputs:
targetType: inline
script: >-
$Files = Get-ChildItem . -R -Filter 'Resources.resw' | ? FullName -Like '*en-US\*\Resources.resw'
$Files | % { Move-Item -Verbose $_.Directory $_.Directory.Parent.Parent -EA:Ignore }
pwsh: true
# 1ES Component Governance onboarding (Detects open source components). See https://docs.opensource.microsoft.com/tools/cg.html
- task: ms.vss-governance-buildtask.governance-build-task-component-detection.ComponentGovernanceComponentDetection@0
displayName: Component Detection
# # PREfast and PoliCheck need Node. Install that first.
- task: NodeTool@0
# !!! NOTE !!! Run PREfast first. Some of the other tasks are going to run on a completed build.
# PREfast is going to build the code as a part of its analysis and the generated sources
# and output binaries will be sufficient for the rest of the analysis.
# If you disable this, the other tasks won't likely work. You would have to add a build
# step instead that builds the code normally before calling them.
# Also... PREfast will rebuild anyway so that's why we're not running a normal build first.
# Waste of time to build twice.
# PREfast. See https://www.1eswiki.com/wiki/SDL_Native_Rules_Build_Task
# The following 1ES tasks all operate completely differently and have a different syntax for usage.
# Most notable is every one of them has a different way of excluding things.
# Go see their 1eswiki.com pages to figure out how to exclude things.
# When writing exclusions, try to make them narrow so when new projects/binaries are added, they
# cause an error here and have to be explicitly pulled out. Don't write an exclusion so broad
# that it will catch other new stuff.
# https://www.1eswiki.com/wiki/PREfast_Build_Task
# Builds the project with C/C++ static analysis tools to find coding flaws and vulnerabilities
# !!! WARNING !!! It doesn't work with WAPPROJ packaging projects. Build the sub-projects instead.
- task: securedevelopmentteam.vss-secure-development-tools.build-task-prefast.SDLNativeRules@3
displayName: 'Run the PREfast SDL Native Rules for MSBuild'
condition: succeededOrFailed()
inputs:
msBuildCommandline: msbuild.exe /nologo /m /p:WindowsTerminalOfficialBuild=true /p:WindowsTerminalBranding=${{ parameters.branding }} /p:WindowsTerminalReleaseBuild=true /p:platform=$(BuildPlatform) /p:configuration=$(BuildConfiguration) /t:Terminal\Window\WindowsTerminal /p:VisualStudioVersion=16.0 $(Build.SourcesDirectory)\OpenConsole.sln
# Copies output from PREfast SDL Native Rules task to expected location for consumption by PkgESSecComp
- task: CopyFiles@1
displayName: 'Copy PREfast xml files to SDLNativeRulesDir'
inputs:
SourceFolder: '$(Agent.BuildDirectory)'
Contents: |
**\*.nativecodeanalysis.xml
TargetFolder: '$(Agent.BuildDirectory)\_sdt\logs\SDLNativeRules'
# https://www.1eswiki.com/index.php?title=PoliCheck_Build_Task
# Scans the text of source code, comments, and content for terminology that could be sensitive for legal, cultural, or geopolitical reasons.
# (Also finds vulgarities... takes all the fun out of everything.)
- task: securedevelopmentteam.vss-secure-development-tools.build-task-policheck.PoliCheck@2
displayName: 'Run PoliCheck'
inputs:
targetType: F
targetArgument: $(Build.SourcesDirectory)
result: PoliCheck.xml
optionsFC: 1
optionsXS: 1
optionsUEPath: $(Build.SourcesDirectory)\build\config\PolicheckExclusions.xml
optionsHMENABLE: 0
continueOnError: true
# https://www.1eswiki.com/wiki/CredScan_Azure_DevOps_Build_Task
# Searches through source code and build outputs for a credential left behind in the open
- task: securedevelopmentteam.vss-secure-development-tools.build-task-credscan.CredScan@3
displayName: 'Run CredScan'
inputs:
outputFormat: pre
# suppressionsFile: LocalSuppressions.json
batchSize: 20
debugMode: false
continueOnError: true
# https://www.1eswiki.com/wiki/BinSkim_Build_Task
# Searches managed and unmanaged binaries for known security vulnerabilities.
- task: securedevelopmentteam.vss-secure-development-tools.build-task-binskim.BinSkim@4
displayName: 'Run BinSkim'
inputs:
TargetPattern: guardianGlob
# See https://aka.ms/gdn-globs for how to do match patterns
AnalyzeTargetGlob: $(Build.SourcesDirectory)\bin\**\*.dll;$(Build.SourcesDirectory)\bin\**\*.exe;-:file|**\Microsoft.UI.Xaml.dll;-:file|**\Microsoft.Toolkit.Win32.UI.XamlHost.dll;-:file|**\vcruntime*.dll;-:file|**\vcomp*.dll;-:file|**\vccorlib*.dll;-:file|**\vcamp*.dll;-:file|**\msvcp*.dll;-:file|**\concrt*.dll;-:file|**\TerminalThemeHelpers*.dll;-:file|**\cpprest*.dll
continueOnError: true
# Set XES_SERIALPOSTBUILDREADY to run Security and Compliance task once per build
- powershell: Write-Host “##vso[task.setvariable variable=XES_SERIALPOSTBUILDREADY;]true”
displayName: 'Set XES_SERIALPOSTBUILDREADY Vars'
# https://www.osgwiki.com/wiki/Package_ES_Security_and_Compliance
# Does a few things:
# - Ensures that Windows-required compliance tasks are run either inside this task
# or were run as a previous step prior to this one
# (PREfast, PoliCheck, Credscan)
# - Runs Windows-specific compliance tasks inside the task
# + CheckCFlags - ensures that compiler and linker flags meet Windows standards
# + CFGCheck/XFGCheck - ensures that Control Flow Guard (CFG) or
# eXtended Flow Guard (XFG) are enabled on binaries
# NOTE: CFG is deprecated and XFG isn't fully ready yet.
# NOTE2: CFG fails on an XFG'd binary
# - Brokers all security/compliance task logs to "Trust Services Automation (TSA)" (https://aka.ms/tsa)
# which is a system that maps all errors into the appropriate bug database
# template for each organization since they all vary. It should also suppress
# new bugs when one already exists for the product.
# This one is set up to go to the OS repository and use the given parameters
# to file bugs to our AzDO product path.
# If we don't use PkgESSecComp to do this for us, we need to install the TSA task
# ourselves in this pipeline to finalize data upload and bug creation.
# !!! NOTE !!! This task goes *LAST* after any other compliance tasks so it catches their logs
- task: PkgESSecComp@10
displayName: 'Security and Compliance tasks'
inputs:
fileNewBugs: false
areaPath: 'OS\WDX\DXP\WinDev\Terminal'
teamProject: 'OS'
iterationPath: 'OS\Future'
bugTags: 'TerminalReleaseCompliance'
scanAll: true
errOnBugs: false
failOnStdErr: true
taskLogVerbosity: Diagnostic
secCompConfigFromTask: |
# Overrides default build sources directory
sourceTargetOverrideAll: $(Build.SourcesDirectory)
# Overrides default build binaries directory when "Scan all" option is specified
binariesTargetOverrideAll: $(Build.SourcesDirectory)\bin
# Set the tools to false if they should not run in the build
tools:
- toolName: CheckCFlags
enable: true
- toolName: CFGCheck
enable: true
- toolName: Policheck
enable: false
- toolName: CredScan
enable: false
- toolName: XFGCheck
enable: false

View File

@@ -0,0 +1,30 @@
parameters:
configuration: 'Release'
platform: ''
additionalBuildArguments: ''
jobs:
- job: Build${{ parameters.platform }}${{ parameters.configuration }}
displayName: Build ${{ parameters.platform }} ${{ parameters.configuration }}
variables:
BuildConfiguration: ${{ parameters.configuration }}
BuildPlatform: ${{ parameters.platform }}
pool:
name: Package ES Lab E
demands:
- msbuild
- visualstudio
- vstest
steps:
- task: PkgESSetupBuild@10
displayName: 'Package ES - Setup Build'
inputs:
useDfs: false
productName: WindowsTerminal
disableOutputRedirect: true
- template: build-console-steps.yml
parameters:
additionalBuildArguments: "/p:XesUseOneStoreVersioning=true;XesBaseYearForStoreVersion=$(baseYearForVersioning) ${{ parameters.additionalBuildArguments }}"

View File

@@ -1,55 +0,0 @@
parameters:
configuration: 'Release'
platform: ''
additionalBuildArguments: ''
minimumExpectedTestsExecutedCount: 1 # Sanity check for minimum expected tests to be reported
rerunPassesRequiredToAvoidFailure: 5
jobs:
- job: Build${{ parameters.platform }}${{ parameters.configuration }}
displayName: Build ${{ parameters.platform }} ${{ parameters.configuration }}
variables:
BuildConfiguration: ${{ parameters.configuration }}
BuildPlatform: ${{ parameters.platform }}
PGOBuildMode: 'Instrument'
pool:
${{ if eq(variables['System.CollectionUri'], 'https://dev.azure.com/ms/') }}:
name: WinDevPoolOSS-L
${{ if ne(variables['System.CollectionUri'], 'https://dev.azure.com/ms/') }}:
name: WinDevPool-L
demands: ImageOverride -equals WinDevVS16-latest
steps:
- template: build-console-steps.yml
parameters:
additionalBuildArguments: '${{ parameters.additionalBuildArguments }}'
- template: helix-runtests-job.yml
parameters:
name: 'RunTestsInHelix'
dependsOn: Build${{ parameters.platform }}${{ parameters.configuration }}
condition: succeeded()
testSuite: 'PgoInstrumentationSuite'
taefQuery: '@IsPgo=true'
configuration: ${{ parameters.configuration }}
platform: ${{ parameters.platform }}
rerunPassesRequiredToAvoidFailure: ${{ parameters.rerunPassesRequiredToAvoidFailure }}
- template: helix-processtestresults-job.yml
parameters:
name: 'ProcessTestResults'
pgoArtifact: 'PGO'
dependsOn:
- RunTestsInHelix
condition: succeededOrFailed()
rerunPassesRequiredToAvoidFailure: ${{ parameters.rerunPassesRequiredToAvoidFailure }}
minimumExpectedTestsExecutedCount: ${{ parameters.minimumExpectedTestsExecutedCount }}
- template: pgo-merge-pgd-job.yml
parameters:
name: 'MergePGD'
dependsOn:
- ProcessTestResults
pgoArtifact: 'PGO'
platform: ${{ parameters.platform }}
configuration: ${{ parameters.configuration }}

View File

@@ -12,10 +12,17 @@ steps:
inputs:
versionSpec: 5.2.0
- task: NuGetAuthenticate@0
# In the Microsoft Azure DevOps tenant, NuGetCommand is ambiguous.
# This should be `task: NuGetCommand@2`
- task: 333b11bd-d341-40d9-afcf-b32d5ce6f23b@2
displayName: Restore NuGet packages for solution
inputs:
command: restore
feedsToUse: config
configPath: NuGet.config
restoreSolution: OpenConsole.sln
restoreDirectory: '$(Build.SourcesDirectory)\packages'
- task: 333b11bd-d341-40d9-afcf-b32d5ce6f23b@2
displayName: Restore NuGet packages for extraneous build actions
inputs:
@@ -25,30 +32,6 @@ steps:
restoreSolution: build/packages.config
restoreDirectory: '$(Build.SourcesDirectory)\packages'
- task: 333b11bd-d341-40d9-afcf-b32d5ce6f23b@2
displayName: Restore NuGet packages for solution
inputs:
command: restore
feedsToUse: config
configPath: NuGet.config
restoreSolution: OpenConsole.sln
restoreDirectory: '$(Build.SourcesDirectory)\packages'
# The environment variable VCToolsInstallDir isn't defined on lab machines, so we need to retrieve it ourselves.
- script: |
"%ProgramFiles(x86)%\Microsoft Visual Studio\Installer\vswhere.exe" -Latest -requires Microsoft.Component.MSBuild -property InstallationPath > %TEMP%\vsinstalldir.txt
set /p _VSINSTALLDIR15=<%TEMP%\vsinstalldir.txt
del %TEMP%\vsinstalldir.txt
call "%_VSINSTALLDIR15%\Common7\Tools\VsDevCmd.bat"
echo VCToolsInstallDir = %VCToolsInstallDir%
echo ##vso[task.setvariable variable=VCToolsInstallDir]%VCToolsInstallDir%
displayName: 'Retrieve VC tools directory'
- task: CmdLine@1
displayName: 'Display build machine environment variables'
inputs:
filename: 'set'
- task: VSBuild@1
displayName: 'Build solution **\OpenConsole.sln'
inputs:
@@ -56,15 +39,12 @@ steps:
vsVersion: 16.0
platform: '$(BuildPlatform)'
configuration: '$(BuildConfiguration)'
msbuildArgs: "${{ parameters.additionalBuildArguments }} /p:PGOBuildMode=$(PGOBuildMode) /bl:$(Build.SourcesDirectory)\\msbuild.binlog"
msbuildArgs: "${{ parameters.additionalBuildArguments }}"
clean: true
maximumCpuCount: true
- task: PowerShell@2
displayName: 'Check MSIX for common regressions'
# PGO runtime needs its own CRT and it's in the package for convenience.
# That will make this script mad so skip since we're not shipping the PGO Instrumentation one anyway.
condition: ne(variables['PGOBuildMode'], 'Instrument')
inputs:
targetType: inline
script: |
@@ -73,7 +53,6 @@ steps:
- task: powershell@2
displayName: 'Source Index PDBs'
condition: ne(variables['PGOBuildMode'], 'Instrument')
inputs:
targetType: filePath
filePath: build\scripts\Index-Pdbs.ps1
@@ -95,7 +74,7 @@ steps:
targetType: filePath
filePath: build\scripts\Run-Tests.ps1
arguments: -MatchPattern '*unit.test*.dll' -Platform '$(RationalizedBuildPlatform)' -Configuration '$(BuildConfiguration)' -LogPath '${{ parameters.testLogPath }}'
condition: and(and(succeeded(), ne(variables['PGOBuildMode'], 'Instrument')), or(eq(variables['BuildPlatform'], 'x64'), eq(variables['BuildPlatform'], 'x86')))
condition: and(succeeded(), or(eq(variables['BuildPlatform'], 'x64'), eq(variables['BuildPlatform'], 'x86')))
- task: PowerShell@2
displayName: 'Run Feature Tests (x64 only)'
@@ -103,7 +82,7 @@ steps:
targetType: filePath
filePath: build\scripts\Run-Tests.ps1
arguments: -MatchPattern '*feature.test*.dll' -Platform '$(RationalizedBuildPlatform)' -Configuration '$(BuildConfiguration)' -LogPath '${{ parameters.testLogPath }}'
condition: and(and(succeeded(), ne(variables['PGOBuildMode'], 'Instrument')), eq(variables['BuildPlatform'], 'x64'))
condition: and(succeeded(), eq(variables['BuildPlatform'], 'x64'))
- task: PowerShell@2
displayName: 'Convert Test Logs from WTL to xUnit format'
@@ -111,14 +90,13 @@ steps:
targetType: filePath
filePath: build\Helix\ConvertWttLogToXUnit.ps1
arguments: -WttInputPath '${{ parameters.testLogPath }}' -WttSingleRerunInputPath 'unused.wtl' -WttMultipleRerunInputPath 'unused2.wtl' -XUnitOutputPath 'onBuildMachineResults.xml' -TestNamePrefix '$(BuildConfiguration).$(BuildPlatform)'
condition: and(ne(variables['PGOBuildMode'], 'Instrument'),or(eq(variables['BuildPlatform'], 'x64'), eq(variables['BuildPlatform'], 'x86')))
condition: or(eq(variables['BuildPlatform'], 'x64'), eq(variables['BuildPlatform'], 'x86'))
- task: PublishTestResults@2
displayName: 'Upload converted test logs'
condition: ne(variables['PGOBuildMode'], 'Instrument')
inputs:
testResultsFormat: 'xUnit' # Options: JUnit, NUnit, VSTest, xUnit, cTest
testResultsFiles: '**/onBuildMachineResults.xml'
testResultsFiles: '**/onBuildMachineResults.xml'
#searchFolder: '$(System.DefaultWorkingDirectory)' # Optional
#mergeTestResults: false # Optional
#failTaskOnFailedTests: false # Optional
@@ -149,57 +127,24 @@ steps:
TargetFolder: '$(Build.ArtifactStagingDirectory)/appx'
OverWrite: true
flattenFolders: true
condition: succeeded()
condition: and(succeeded(), ne(variables['Build.Reason'], 'PullRequest'))
- task: CopyFiles@2
displayName: 'Copy outputs needed for test runs to Artifacts'
inputs:
Contents: |
$(Build.SourcesDirectory)/bin/$(RationalizedBuildPlatform)/$(BuildConfiguration)/*.exe
$(Build.SourcesDirectory)/bin/$(RationalizedBuildPlatform)/$(BuildConfiguration)/*.dll
$(Build.SourcesDirectory)/bin/$(RationalizedBuildPlatform)/$(BuildConfiguration)/*.xml
$(Build.SourcesDirectory)/bin/$(BuildPlatform)/$(BuildConfiguration)/*.exe
$(Build.SourcesDirectory)/bin/$(BuildPlatform)/$(BuildConfiguration)/*.dll
$(Build.SourcesDirectory)/bin/$(BuildPlatform)/$(BuildConfiguration)/*.xml
**/Microsoft.VCLibs.*.appx
**/*unit.test*.dll
**/*unit.test*.manifest
**/TestHostApp/*.exe
**/TestHostApp/*.dll
**/TestHostApp/*.xml
!**/*.pdb
!**/*.ipdb
!**/*.obj
!**/*.pch
**/TestHostApp/*
TargetFolder: '$(Build.ArtifactStagingDirectory)/$(BuildConfiguration)/$(BuildPlatform)/test'
OverWrite: true
flattenFolders: true
condition: succeeded()
condition: and(and(succeeded(), eq(variables['BuildPlatform'], 'x64')), ne(variables['Build.Reason'], 'PullRequest'))
- task: PublishBuildArtifacts@1
displayName: 'Publish All Build Artifacts'
inputs:
PathtoPublish: '$(Build.ArtifactStagingDirectory)'
ArtifactName: 'drop'
- task: CopyFiles@2
displayName: 'Copy PGO databases needed for PGO instrumentation run'
inputs:
Contents: |
**/*.pgd
TargetFolder: '$(Build.ArtifactStagingDirectory)/$(BuildConfiguration)/PGO/$(BuildPlatform)'
OverWrite: true
flattenFolders: true
condition: and(succeeded(), eq(variables['PGOBuildMode'], 'Instrument'))
- task: PublishBuildArtifacts@1
displayName: 'Publish All PGO Artifacts'
inputs:
PathtoPublish: '$(Build.ArtifactStagingDirectory)/$(BuildConfiguration)/PGO'
ArtifactName: 'PGO'
condition: and(succeeded(), eq(variables['PGOBuildMode'], 'Instrument'))
- task: PublishBuildArtifacts@1
displayName: 'Publish Artifact: binlog'
condition: failed()
continueOnError: True
inputs:
PathtoPublish: $(Build.SourcesDirectory)\msbuild.binlog
ArtifactName: binlog-$(BuildPlatform)
ArtifactName: 'drop'

View File

@@ -11,7 +11,7 @@ jobs:
clean: true
- task: PowerShell@2
displayName: 'Code Formatting Check'
displayName: 'Code Formattting Check'
inputs:
targetType: filePath
filePath: '.\build\scripts\Invoke-FormattingCheck.ps1'

View File

@@ -12,4 +12,4 @@ steps:
inputs:
targetType: filePath
filePath: build\Helix\GenerateTestProjFile.ps1
arguments: -TestFile '${{ parameters.testFilePath }}' -OutputProjFile '$(Build.ArtifactStagingDirectory)\$(BuildConfiguration)\$(BuildPlatform)\${{ parameters.outputProjFileName }}' -JobTestSuiteName '${{ parameters.testSuite }}' -TaefPath '$(Build.SourcesDirectory)\build\Helix\packages\Microsoft.Taef.10.60.210621002\build\Binaries\x86' -TaefQuery '${{ parameters.taefQuery }}'
arguments: -TestFile '${{ parameters.testFilePath }}' -OutputProjFile '$(Build.ArtifactStagingDirectory)\${{ parameters.outputProjFileName }}' -JobTestSuiteName '${{ parameters.testSuite }}' -TaefPath '$(Build.SourcesDirectory)\build\Helix\packages\taef.redist.wlk.10.57.200731005-develop\build\Binaries\x86' -TaefQuery '${{ parameters.taefQuery }}'

View File

@@ -22,7 +22,6 @@ jobs:
condition: succeededOrFailed()
env:
SYSTEM_ACCESSTOKEN: $(System.AccessToken)
HelixAccessToken: $(HelixApiAccessToken)
inputs:
targetType: filePath
filePath: build\Helix\UpdateUnreliableTests.ps1
@@ -33,7 +32,6 @@ jobs:
condition: succeededOrFailed()
env:
SYSTEM_ACCESSTOKEN: $(System.AccessToken)
HelixAccessToken: $(HelixApiAccessToken)
inputs:
targetType: filePath
filePath: build\Helix\OutputTestResults.ps1

View File

@@ -10,12 +10,19 @@ parameters:
maxParallel: 4
rerunPassesRequiredToAvoidFailure: 5
taefQuery: ''
configuration: ''
platform: ''
# if 'useBuildOutputFromBuildId' is set, we will default to using a build from this pipeline:
useBuildOutputFromPipeline: $(System.DefinitionId)
openHelixTargetQueues: 'windows.10.amd64.client21h1.open.xaml'
closedHelixTargetQueues: 'windows.10.amd64.client21h1.xaml'
matrix:
# Release_x86:
# buildPlatform: 'x86'
# buildConfiguration: 'release'
# openHelixTargetQueues: 'windows.10.amd64.client19h1.open.xaml'
# closedHelixTargetQueues: 'windows.10.amd64.client19h1.xaml'
Release_x64:
buildPlatform: 'x64'
buildConfiguration: 'release'
openHelixTargetQueues: 'windows.10.amd64.client19h1.open.xaml'
closedHelixTargetQueues: 'windows.10.amd64.client19h1.xaml'
jobs:
- job: ${{ parameters.name }}
@@ -26,15 +33,13 @@ jobs:
timeoutInMinutes: 120
strategy:
maxParallel: ${{ parameters.maxParallel }}
matrix: ${{ parameters.matrix }}
variables:
buildConfiguration: ${{ parameters.configuration }}
buildPlatform: ${{ parameters.platform }}
openHelixTargetQueues: ${{ parameters.openHelixTargetQueues }}
closedHelixTargetQueues: ${{ parameters.closedHelixTargetQueues }}
artifactsDir: $(Build.SourcesDirectory)\Artifacts
taefPath: $(Build.SourcesDirectory)\build\Helix\packages\Microsoft.Taef.10.60.210621002\build\Binaries\$(buildPlatform)
taefPath: $(Build.SourcesDirectory)\build\Helix\packages\taef.redist.wlk.10.57.200731005-develop\build\Binaries\$(buildPlatform)
helixCommonArgs: '/binaryLogger:$(Build.SourcesDirectory)/${{parameters.name}}.$(buildPlatform).$(buildConfiguration).binlog /p:HelixBuild=$(Build.BuildId).$(buildPlatform).$(buildConfiguration) /p:Platform=$(buildPlatform) /p:Configuration=$(buildConfiguration) /p:HelixType=${{parameters.helixType}} /p:TestSuite=${{parameters.testSuite}} /p:ProjFilesPath=$(Build.ArtifactStagingDirectory) /p:rerunPassesRequiredToAvoidFailure=${{parameters.rerunPassesRequiredToAvoidFailure}}'
steps:
- task: CmdLine@1
displayName: 'Display build machine environment variables'
@@ -92,17 +97,9 @@ jobs:
filename: 'dir'
arguments: '/s $(Build.SourcesDirectory)\HelixPayload'
- task: PowerShell@2
displayName: 'Make artifact directories'
inputs:
targetType: inline
script: |
New-Item -ItemType Directory -Force -Path "$(Build.ArtifactStagingDirectory)\$(BuildConfiguration)\"
New-Item -ItemType Directory -Force -Path "$(Build.ArtifactStagingDirectory)\$(BuildConfiguration)\$(BuildPlatform)\"
- template: helix-createprojfile-steps.yml
parameters:
condition: and(succeeded(),eq('${{ parameters.testSuite }}','DevTestSuite'))
condition: and(succeeded(),ne('${{ parameters.testSuite }}','NugetTestSuite'))
testFilePath: '$(artifactsDir)\${{ parameters.artifactName }}\$(buildConfiguration)\$(buildPlatform)\Test\TerminalApp.LocalTests.dll'
outputProjFileName: 'RunTestsInHelix-TerminalAppLocalTests.proj'
testSuite: '${{ parameters.testSuite }}'
@@ -110,7 +107,7 @@ jobs:
- template: helix-createprojfile-steps.yml
parameters:
condition: and(succeeded(),eq('${{ parameters.testSuite }}','DevTestSuite'))
condition: and(succeeded(),ne('${{ parameters.testSuite }}','NugetTestSuite'))
testFilePath: '$(artifactsDir)\${{ parameters.artifactName }}\$(buildConfiguration)\$(buildPlatform)\Test\SettingsModel.LocalTests.dll'
outputProjFileName: 'RunTestsInHelix-SettingsModelLocalTests.proj'
testSuite: '${{ parameters.testSuite }}'
@@ -119,20 +116,12 @@ jobs:
- template: helix-createprojfile-steps.yml
parameters:
condition: and(succeeded(),eq('${{ parameters.testSuite }}','DevTestSuite'))
condition: and(succeeded(),ne('${{ parameters.testSuite }}','NugetTestSuite'))
testFilePath: '$(artifactsDir)\${{ parameters.artifactName }}\$(buildConfiguration)\$(buildPlatform)\Test\Conhost.UIA.Tests.dll'
outputProjFileName: 'RunTestsInHelix-HostTestsUIA.proj'
testSuite: '${{ parameters.testSuite }}'
taefQuery: ${{ parameters.taefQuery }}
- template: helix-createprojfile-steps.yml
parameters:
condition: and(succeeded(),or(eq('${{ parameters.testSuite }}','PgoInstrumentationSuite'),eq('${{ parameters.testSuite }}','DevTestSuite')))
testFilePath: '$(artifactsDir)\${{ parameters.artifactName }}\$(buildConfiguration)\$(buildPlatform)\Test\WindowsTerminal.UIA.Tests.dll'
outputProjFileName: 'RunTestsInHelix-WindowsTerminalUIATests.proj'
testSuite: '${{ parameters.testSuite }}'
taefQuery: ${{ parameters.taefQuery }}
- task: PublishBuildArtifacts@1
displayName: 'Publish generated .proj files'
inputs:
@@ -141,7 +130,6 @@ jobs:
- task: DotNetCoreCLI@2
displayName: 'Run tests in Helix (open queues)'
condition: and(succeeded(),eq(variables['System.CollectionUri'],'https://dev.azure.com/ms/'))
env:
SYSTEM_ACCESSTOKEN: $(System.AccessToken)
inputs:
@@ -150,14 +138,3 @@ jobs:
custom: msbuild
arguments: '$(helixCommonArgs) /p:IsExternal=true /p:Creator=Terminal /p:HelixTargetQueues=$(openHelixTargetQueues)'
- task: DotNetCoreCLI@2
displayName: 'Run tests in Helix (closed queues)'
condition: and(succeeded(),ne(variables['System.CollectionUri'],'https://dev.azure.com/ms/'))
env:
SYSTEM_ACCESSTOKEN: $(System.AccessToken)
HelixAccessToken: $(HelixApiAccessToken)
inputs:
command: custom
projects: build\Helix\RunTestsInHelix.proj
custom: msbuild
arguments: '$(helixCommonArgs) /p:HelixTargetQueues=$(closedHelixTargetQueues)'

View File

@@ -1,9 +0,0 @@
parameters:
sdkVersion: 18362
steps:
- task: powershell@2
inputs:
targetType: filePath
filePath: build\scripts\Install-WindowsSdkISO.ps1
arguments: ${{ parameters.sdkVersion }}
displayName: 'Install Windows SDK (${{ parameters.sdkVersion }})'

View File

@@ -1,66 +0,0 @@
# From our friends at MUX: https://github.com/microsoft/microsoft-ui-xaml/blob/main/build/AzurePipelinesTemplates/MUX-BuildAndPublishPGONuGet-Job.yml
parameters:
dependsOn: ''
pgoArtifact: PGO
jobs:
- job: BuildAndPublishPGONuGet
dependsOn: ${{ parameters.dependsOn }}
pool:
vmImage: 'windows-2019'
variables:
artifactsPath: $(Build.SourcesDirectory)\Artifacts
pgoToolsPath: $(Build.SourcesDirectory)\build\PGO
nuspecPath: $(pgoToolsPath)\NuSpecs
nuspecFilename: PGO.nuspec
steps:
- task: DownloadBuildArtifacts@0
inputs:
artifactName: ${{ parameters.pgoArtifact }}
downloadPath: $(artifactsPath)
- task: NuGetAuthenticate@0
inputs:
nuGetServiceConnections: 'Terminal Public Artifact Feed'
- task: NuGetToolInstaller@0
displayName: 'Use NuGet 5.8.0'
inputs:
versionSpec: 5.8.0
# In the Microsoft Azure DevOps tenant, NuGetCommand is ambiguous.
# This should be `task: NuGetCommand@2`
- task: 333b11bd-d341-40d9-afcf-b32d5ce6f23b@2
displayName: Restore NuGet packages for extraneous build actions
inputs:
command: restore
feedsToUse: config
configPath: NuGet.config
restoreSolution: build/packages.config
restoreDirectory: '$(Build.SourcesDirectory)\packages'
- task: MSBuild@1
displayName: 'Create PGO Nuget'
inputs:
solution: $(pgoToolsPath)\PGO.DB.proj
msbuildArguments: '/t:CreatePGONuGet /p:PGOBuildMode=Instrument /p:PGDPathForAllArch=$(artifactsPath)\${{ parameters.pgoArtifact }} /p:PGOOutputPath=$(Build.ArtifactStagingDirectory)'
- task: PublishBuildArtifacts@1
inputs:
pathToPublish: $(Build.ArtifactStagingDirectory)
artifactName: ${{ parameters.pgoArtifact }}
- task: 333b11bd-d341-40d9-afcf-b32d5ce6f23b@2
displayName: 'NuGet push'
inputs:
command: push
nuGetFeedType: external
packagesToPush: $(Build.ArtifactStagingDirectory)/*.nupkg
# The actual URL and PAT for this feed is configured at
# https://microsoft.visualstudio.com/Dart/_settings/adminservices
# This is the name of that connection
publishFeedCredentials: 'Terminal Public Artifact Feed'
feedsToUse: config
nugetConfigPath: '$(Build.SourcesDirectory)/NuGet.config'

View File

@@ -1,70 +0,0 @@
parameters:
dependsOn: ''
pgoArtifact: PGO
platform: ''
configuration: ''
jobs:
- job: MergePGD
dependsOn: ${{ parameters.dependsOn }}
pool:
vmImage: 'windows-2019'
variables:
artifactsPath: $(Build.SourcesDirectory)\Artifacts
pgoArtifactsPath: $(artifactsPath)\${{ parameters.pgoArtifact }}
buildPlatform: ${{ parameters.platform }}
buildConfiguration: ${{ parameters.configuration }}
steps:
# The environment variable VCToolsInstallDir isn't defined on lab machines, so we need to retrieve it ourselves.
- script: |
"%ProgramFiles(x86)%\Microsoft Visual Studio\Installer\vswhere.exe" -Latest -requires Microsoft.Component.MSBuild -property InstallationPath > %TEMP%\vsinstalldir.txt
set /p _VSINSTALLDIR15=<%TEMP%\vsinstalldir.txt
del %TEMP%\vsinstalldir.txt
call "%_VSINSTALLDIR15%\Common7\Tools\VsDevCmd.bat"
echo VCToolsInstallDir = %VCToolsInstallDir%
echo ##vso[task.setvariable variable=VCToolsInstallDir]%VCToolsInstallDir%
displayName: 'Retrieve VC tools directory'
- task: NuGetToolInstaller@0
displayName: 'Use NuGet 5.2.0'
inputs:
versionSpec: 5.2.0
- task: NuGetAuthenticate@0
# In the Microsoft Azure DevOps tenant, NuGetCommand is ambiguous.
# This should be `task: NuGetCommand@2`
- task: 333b11bd-d341-40d9-afcf-b32d5ce6f23b@2
displayName: Restore NuGet packages for extraneous build actions
inputs:
command: restore
feedsToUse: config
configPath: NuGet.config
restoreSolution: build/packages.config
restoreDirectory: '$(Build.SourcesDirectory)\packages'
- task: DownloadBuildArtifacts@0
inputs:
artifactName: ${{ parameters.pgoArtifact }}
downloadPath: $(artifactsPath)
- task: MSBuild@1
displayName: Merge counts into PGD
inputs:
solution: $(Build.SourcesDirectory)\OpenConsole.sln
platform: $(buildPlatform)
configuration: $(buildConfiguration)
msbuildArguments: '/t:MergePGOCounts /p:PGOBuildMode=Instrument /p:PGDPath=$(pgoArtifactsPath)\$(buildPlatform) /p:PGCRootPath=$(pgoArtifactsPath)\$(buildPlatform)'
- task: CopyFiles@2
displayName: 'Copy merged pgd to artifact staging'
inputs:
sourceFolder: $(pgoArtifactsPath)
contents: '**\$(buildPlatform)\*.pgd'
targetFolder: $(Build.ArtifactStagingDirectory)
- task: PublishBuildArtifacts@1
inputs:
pathToPublish: $(Build.ArtifactStagingDirectory)
artifactName: ${{ parameters.pgoArtifact }}

View File

@@ -0,0 +1,74 @@
parameters:
configuration: 'Release'
jobs:
- job: SignDeploy${{ parameters.configuration }}
displayName: Sign and Deploy for ${{ parameters.configuration }}
dependsOn:
- Buildx64AuditMode
- Buildx64Release
- Buildx86Release
- Buildarm64Release
- CodeFormatCheck
condition: |
and
(
in(dependencies.Buildx64AuditMode.result, 'Succeeded', 'SucceededWithIssues', 'Skipped'),
in(dependencies.Buildx64Release.result, 'Succeeded', 'SucceededWithIssues', 'Skipped'),
in(dependencies.Buildx86Release.result, 'Succeeded', 'SucceededWithIssues', 'Skipped'),
in(dependencies.Buildarm64Release.result, 'Succeeded', 'SucceededWithIssues', 'Skipped'),
in(dependencies.CodeFormatCheck.result, 'Succeeded', 'SucceededWithIssues', 'Skipped')
)
variables:
BuildConfiguration: ${{ parameters.configuration }}
AppxProjectName: CascadiaPackage
AppxBundleName: Microsoft.WindowsTerminal_8wekyb3d8bbwe.msixbundle
pool:
name: Package ES Lab E
steps:
- checkout: self
clean: true
- task: PkgESSetupBuild@10
displayName: 'Package ES - Setup Build'
inputs:
useDfs: false
productName: WindowsTerminal
disableOutputRedirect: true
- task: ms.vss-governance-buildtask.governance-build-task-component-detection.ComponentGovernanceComponentDetection@0
displayName: 'Component Detection'
- task: DownloadBuildArtifacts@0
displayName: Download AppX artifacts
inputs:
artifactName: 'appx-$(BuildConfiguration)'
itemPattern: |
**/*.appx
**/*.msix
downloadPath: '$(Build.ArtifactStagingDirectory)\appx'
- task: PowerShell@2
displayName: 'Create $(AppxBundleName)'
inputs:
targetType: filePath
filePath: '.\build\scripts\Create-AppxBundle.ps1'
arguments: |
-InputPath "$(Build.ArtifactStagingDirectory)\appx" -ProjectName $(AppxProjectName) -BundleVersion 0.0.0.0 -OutputPath "$(Build.ArtifactStagingDirectory)\$(AppxBundleName)"
- task: PkgESCodeSign@10
displayName: 'Package ES - SignConfig.WindowsTerminal.xml'
inputs:
signConfigXml: 'build\config\SignConfig.WindowsTerminal.xml'
inPathRoot: '$(Build.ArtifactStagingDirectory)'
outPathRoot: '$(Build.ArtifactStagingDirectory)\signed'
- task: PublishBuildArtifacts@1
displayName: 'Publish Signed AppX'
inputs:
PathtoPublish: '$(Build.ArtifactStagingDirectory)\signed'
ArtifactName: 'appxbundle-signed-$(BuildConfiguration)'

View File

@@ -1,97 +0,0 @@
<?xml version="1.0" encoding="utf-8"?>
<Project DefaultTargets="Build" ToolsVersion="16.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
<!-- THIS PROJECT CANNOT BE LOADED INTO THE SOLUTION. -->
<Import Project="$(MSBuildExtensionsPath)\$(MSBuildToolsVersion)\Microsoft.Common.props" Condition="Exists('$(MSBuildExtensionsPath)\$(MSBuildToolsVersion)\Microsoft.Common.props')" />
<ItemGroup Label="ProjectConfigurations">
<ProjectConfiguration Include="Release|Any CPU">
<Configuration>Release</Configuration>
<Platform>AnyCPU</Platform>
</ProjectConfiguration>
<ProjectConfiguration Include="Fuzzing|Any CPU">
<Configuration>Fuzzing</Configuration>
<Platform>AnyCPU</Platform>
</ProjectConfiguration>
<ProjectConfiguration Include="AuditMode|Any CPU">
<Configuration>AuditMode</Configuration>
<Platform>AnyCPU</Platform>
</ProjectConfiguration>
<ProjectConfiguration Include="Debug|Any CPU">
<Configuration>Debug</Configuration>
<Platform>AnyCPU</Platform>
</ProjectConfiguration>
</ItemGroup>
<PropertyGroup Label="Globals">
<ProjectGuid>d97c3c61-53cd-4e72-919b-9a0940e038f9</ProjectGuid>
</PropertyGroup>
<PropertyGroup>
<IntermediateOutputPath>$(SolutionDir)obj\$(Configuration)\GenerateFeatureFlags\</IntermediateOutputPath>
<OpenConsoleCommonOutDir>$(SolutionDir)bin\$(Configuration)\</OpenConsoleCommonOutDir>
<_WTBrandingName Condition="'$(WindowsTerminalBranding)'=='Preview'">Preview</_WTBrandingName>
<_WTBrandingName Condition="'$(WindowsTerminalBranding)'=='Release'">Release</_WTBrandingName>
<_WTBrandingName Condition="'$(_WTBrandingName)'==''">Dev</_WTBrandingName>
</PropertyGroup>
<Target Name="_GenerateBranchAndBrandingCache">
<Exec Command="git.exe rev-parse --abbrev-ref HEAD"
CustomWarningRegularExpression="^fatal:.*"
ConsoleToMsBuild="true"
IgnoreExitCode="true">
<Output TaskParameter="ConsoleOutput" ItemName="_GitBranchLines" />
</Exec>
<ItemGroup>
<_BrandingLines Include="$(_WTBrandingName)" />
</ItemGroup>
<WriteLinesToFile File="$(IntermediateOutputPath)branch_branding_cache.txt"
Lines="@(_GitBranchLines);@(_BrandingLines)"
Overwrite="true"
WriteOnlyWhenDifferent="true" />
<ItemGroup>
<FileWrites Include="$(IntermediateOutputPath)branch_branding_cache.txt" />
<_BranchBrandingCacheFiles Include="$(IntermediateOutputPath)branch_branding_cache.txt" />
</ItemGroup>
</Target>
<Target Name="_RunFeatureFlagScript"
Inputs="@(FeatureFlagFile);@(_BranchBrandingCacheFiles)"
Outputs="$(OpenConsoleCommonOutDir)\inc\TilFeatureStaging.h"
DependsOnTargets="_GenerateBranchAndBrandingCache">
<MakeDir Directories="$(OpenConsoleCommonOutDir)\inc" />
<Exec
Command="powershell -NoLogo -NoProfile -NonInteractive -ExecutionPolicy ByPass -Command &quot;$(SolutionDir)\tools\Generate-FeatureStagingHeader.ps1&quot; -Path &quot;%(FeatureFlagFile.FullPath)&quot; -Branding $(_WTBrandingName)"
ConsoleToMsBuild="true"
StandardOutputImportance="low">
<Output TaskParameter="ConsoleOutput" ItemName="_FeatureFlagFileLines" />
</Exec>
<!--
We gather the feature flag output in MSBuild and emit the file so that we can take advantage of
WriteOnlyWhenDifferent. Doing this ensures that we don't rebuild the world when the branch changes
(if it results in a new TilFeatureStaging.h that would have had the same content/features as the previous one)
-->
<WriteLinesToFile File="$(OpenConsoleCommonOutDir)\inc\TilFeatureStaging.h"
Lines="@(_FeatureFlagFileLines)"
Overwrite="true"
WriteOnlyWhenDifferent="true" />
<ItemGroup>
<FileWrites Include="$(OpenConsoleCommonOutDir)\inc\TilFeatureStaging.h" />
</ItemGroup>
</Target>
<Target Name="Build" DependsOnTargets="_RunFeatureFlagScript" />
<Target Name="Clean">
<Delete Files="$(OpenConsoleCommonOutDir)\inc\TilFeatureStaging.h" />
</Target>
<ItemGroup>
<FeatureFlagFile Include="$(SolutionDir)\src\features.xml" />
</ItemGroup>
</Project>

View File

@@ -1,34 +0,0 @@
# Copyright (c) Microsoft Corporation.
# Licensed under the MIT license.
$LocalizationsFromContextMenu = Get-ChildItem ./src/cascadia/TerminalApp/Resources -Recurse -Filter ContextMenu.resw
$Languages = [System.Collections.HashTable]::New()
$LocalizationsFromContextMenu | ForEach-Object {
$Languages[$_.Directory.Name] = $_
}
ForEach ($pair in $Languages.GetEnumerator()) {
$LanguageDir = "./src/cascadia/CascadiaPackage/Resources/$($pair.Key)"
$ResPath = "$LanguageDir/Resources.resw"
$PreexistingResw = Get-Item $ResPath -EA:Ignore
If ($null -eq $PreexistingResw) {
Write-Host "Copying $($pair.Value.FullName) to $ResPath"
New-Item -type Directory $LanguageDir -EA:Ignore
Copy-Item $pair.Value.FullName $ResPath
} Else {
# Merge Them!
Write-Host "Merging $($pair.Value.FullName) into $ResPath"
$existingXml = [xml](Get-Content $PreexistingResw.FullName)
$newXml = [xml](Get-Content $pair.Value.FullName)
$newDataKeys = $newXml.root.data.name
$existingXml.root.data | % {
If ($_.name -in $newDataKeys) {
$null = $existingXml.root.RemoveChild($_)
}
}
$newXml.root.data | % {
$null = $existingXml.root.AppendChild($existingXml.ImportNode($_, $true))
}
$existingXml.Save($PreexistingResw.FullName)
}
}

View File

@@ -22,7 +22,7 @@ Param(
[Parameter(HelpMessage="Path to makeappx.exe")]
[ValidateScript({Test-Path $_ -Type Leaf})]
[string]
$MakeAppxPath = "C:\Program Files (x86)\Windows Kits\10\bin\10.0.19041.0\x86\MakeAppx.exe"
$MakeAppxPath = "C:\Program Files (x86)\Windows Kits\10\bin\10.0.17763.0\x86\MakeAppx.exe"
)
If ($null -Eq (Get-Item $MakeAppxPath -EA:SilentlyContinue)) {

View File

@@ -23,7 +23,6 @@ $mappedFiles = New-Object System.Collections.ArrayList
foreach ($file in (Get-ChildItem -r:$recursive "$SearchDir\*.pdb"))
{
$mappedFiles = New-Object System.Collections.ArrayList
Write-Verbose "Found $file"
$ErrorActionPreference = "Continue" # Azure Pipelines defaults to "Stop", continue past errors in this script.
@@ -51,7 +50,7 @@ foreach ($file in (Get-ChildItem -r:$recursive "$SearchDir\*.pdb"))
if ($relative)
{
$mapping = $allFiles[$i] + "*$relative"
$ignore = $mappedFiles.Add($mapping)
$mappedFiles.Add($mapping)
Write-Verbose "Mapped path $($i): $mapping"
}
@@ -79,26 +78,7 @@ $($mappedFiles -join "`r`n")
SRCSRV: end ------------------------------------------------
"@ | Set-Content $pdbstrFile
Write-Host
Write-Host
Write-Host (Get-Content $pdbstrFile)
Write-Host
Write-Host
Write-Host "$pdbstrExe -p:""$file"" -w -s:srcsrv -i:$pdbstrFile"
& $pdbstrExe -p:"$file" -w -s:srcsrv -i:$pdbstrFile
Write-Host
Write-Host
Write-Host "$pdbstrExe -p:""$file"" -r -s:srcsrv"
& $pdbstrExe -p:"$file" -r -s:srcsrv
Write-Host
Write-Host
Write-Host "$srctoolExe $file"
& $srctoolExe "$file"
Write-Host
Write-Host
}
# Return with exit 0 to override any weird error code from other tools

View File

@@ -1,346 +0,0 @@
[CmdletBinding()]
param([Parameter(Mandatory=$true, Position=0)]
[string]$buildNumber)
# Ensure the error action preference is set to the default for PowerShell3, 'Stop'
$ErrorActionPreference = 'Stop'
# Constants
$WindowsSDKOptions = @("OptionId.UWPCpp", "OptionId.DesktopCPPx64", "OptionId.DesktopCPPx86", "OptionId.DesktopCPPARM64", "OptionId.DesktopCPPARM", "OptionId.WindowsDesktopDebuggers")
$WindowsSDKRegPath = "HKLM:\Software\WOW6432Node\Microsoft\Windows Kits\Installed Roots"
$WindowsSDKRegRootKey = "KitsRoot10"
$WindowsSDKVersion = "10.0.$buildNumber.0"
$WindowsSDKInstalledRegPath = "$WindowsSDKRegPath\$WindowsSDKVersion\Installed Options"
$StrongNameRegPath = "HKLM:\SOFTWARE\Microsoft\StrongName\Verification"
$PublicKeyTokens = @("31bf3856ad364e35")
if ($buildNumber -notmatch "^\d{5,}$")
{
Write-Host "ERROR: '$buildNumber' doesn't look like a windows build number"
Write-Host
Exit 1
}
function Download-File
{
param ([string] $outDir,
[string] $downloadUrl,
[string] $downloadName)
$downloadPath = Join-Path $outDir "$downloadName.download"
$downloadDest = Join-Path $outDir $downloadName
$downloadDestTemp = Join-Path $outDir "$downloadName.tmp"
Write-Host -NoNewline "Downloading $downloadName..."
$retries = 10
$downloaded = $false
while (-not $downloaded)
{
try
{
$webclient = new-object System.Net.WebClient
$webclient.DownloadFile($downloadUrl, $downloadPath)
$downloaded = $true
}
catch [System.Net.WebException]
{
Write-Host
Write-Warning "Failed to fetch updated file from $downloadUrl : $($error[0])"
if (!(Test-Path $downloadDest))
{
if ($retries -gt 0)
{
Write-Host "$retries retries left, trying download again"
$retries--
start-sleep -Seconds 10
}
else
{
throw "$downloadName was not found at $downloadDest"
}
}
else
{
Write-Warning "$downloadName may be out of date"
}
}
}
Unblock-File $downloadPath
$downloadDestTemp = $downloadPath;
# Delete and rename to final dest
Write-Host "testing $downloadDest"
if (Test-Path $downloadDest)
{
Write-Host "Deleting: $downloadDest"
Remove-Item $downloadDest -Force
}
Move-Item -Force $downloadDestTemp $downloadDest
Write-Host "Done"
return $downloadDest
}
function Get-ISODriveLetter
{
param ([string] $isoPath)
$diskImage = Get-DiskImage -ImagePath $isoPath
if ($diskImage)
{
$volume = Get-Volume -DiskImage $diskImage
if ($volume)
{
$driveLetter = $volume.DriveLetter
if ($driveLetter)
{
$driveLetter += ":"
return $driveLetter
}
}
}
return $null
}
function Mount-ISO
{
param ([string] $isoPath)
# Check if image is already mounted
$isoDrive = Get-ISODriveLetter $isoPath
if (!$isoDrive)
{
Mount-DiskImage -ImagePath $isoPath -StorageType ISO | Out-Null
}
$isoDrive = Get-ISODriveLetter $isoPath
Write-Verbose "$isoPath mounted to ${isoDrive}:"
}
function Dismount-ISO
{
param ([string] $isoPath)
$isoDrive = (Get-DiskImage -ImagePath $isoPath | Get-Volume).DriveLetter
if ($isoDrive)
{
Write-Verbose "$isoPath dismounted"
Dismount-DiskImage -ImagePath $isoPath | Out-Null
}
}
function Disable-StrongName
{
param ([string] $publicKeyToken = "*")
reg ADD "HKLM\SOFTWARE\Microsoft\StrongName\Verification\*,$publicKeyToken" /f | Out-Null
if ($env:PROCESSOR_ARCHITECTURE -eq "AMD64")
{
reg ADD "HKLM\SOFTWARE\Wow6432Node\Microsoft\StrongName\Verification\*,$publicKeyToken" /f | Out-Null
}
}
function Test-Admin
{
$identity = [Security.Principal.WindowsIdentity]::GetCurrent()
$principal = New-Object Security.Principal.WindowsPrincipal $identity
$principal.IsInRole([Security.Principal.WindowsBuiltInRole]::Administrator)
}
function Test-RegistryPathAndValue
{
param (
[parameter(Mandatory=$true)]
[ValidateNotNullOrEmpty()]
[string] $path,
[parameter(Mandatory=$true)]
[ValidateNotNullOrEmpty()]
[string] $value)
try
{
if (Test-Path $path)
{
Get-ItemProperty -Path $path | Select-Object -ExpandProperty $value -ErrorAction Stop | Out-Null
return $true
}
}
catch
{
}
return $false
}
function Test-InstallWindowsSDK
{
$retval = $true
if (Test-RegistryPathAndValue -Path $WindowsSDKRegPath -Value $WindowsSDKRegRootKey)
{
# A Windows SDK is installed
# Is an SDK of our version installed with the options we need?
$allRequiredSdkOptionsInstalled = $true
foreach($sdkOption in $WindowsSDKOptions)
{
if (!(Test-RegistryPathAndValue -Path $WindowsSDKInstalledRegPath -Value $sdkOption))
{
$allRequiredSdkOptionsInstalled = $false
}
}
if($allRequiredSdkOptionsInstalled)
{
# It appears we have what we need. Double check the disk
$sdkRoot = Get-ItemProperty -Path $WindowsSDKRegPath | Select-Object -ExpandProperty $WindowsSDKRegRootKey
if ($sdkRoot)
{
if (Test-Path $sdkRoot)
{
$refPath = Join-Path $sdkRoot "References\$WindowsSDKVersion"
if (Test-Path $refPath)
{
$umdPath = Join-Path $sdkRoot "UnionMetadata\$WindowsSDKVersion"
if (Test-Path $umdPath)
{
# Pretty sure we have what we need
$retval = $false
}
}
}
}
}
}
return $retval
}
function Test-InstallStrongNameHijack
{
foreach($publicKeyToken in $PublicKeyTokens)
{
$key = "$StrongNameRegPath\*,$publicKeyToken"
if (!(Test-Path $key))
{
return $true
}
}
return $false
}
Write-Host -NoNewline "Checking for installed Windows SDK $WindowsSDKVersion..."
$InstallWindowsSDK = Test-InstallWindowsSDK
if ($InstallWindowsSDK)
{
Write-Host "Installation required"
}
else
{
Write-Host "INSTALLED"
}
$StrongNameHijack = Test-InstallStrongNameHijack
Write-Host -NoNewline "Checking if StrongName bypass required..."
if ($StrongNameHijack)
{
Write-Host "REQUIRED"
}
else
{
Write-Host "Done"
}
if ($StrongNameHijack -or $InstallWindowsSDK)
{
if (!(Test-Admin))
{
Write-Host
throw "ERROR: Elevation required"
}
}
if ($InstallWindowsSDK)
{
# Static(ish) link for Windows SDK
# Note: there is a delay from Windows SDK announcements to availability via the static link
$uri = "https://software-download.microsoft.com/download/sg/Windows_InsiderPreview_SDK_en-us_$($buildNumber)_1.iso";
if ($env:TEMP -eq $null)
{
$env:TEMP = Join-Path $env:SystemDrive 'temp'
}
$winsdkTempDir = Join-Path (Join-Path $env:TEMP ([System.IO.Path]::GetRandomFileName())) "WindowsSDK"
if (![System.IO.Directory]::Exists($winsdkTempDir))
{
[void][System.IO.Directory]::CreateDirectory($winsdkTempDir)
}
$file = "winsdk_$buildNumber.iso"
Write-Verbose "Getting WinSDK from $uri"
$downloadFile = Download-File $winsdkTempDir $uri $file
Write-Verbose "File is at $downloadFile"
$downloadFileItem = Get-Item $downloadFile
# Check to make sure the file is at least 10 MB.
if ($downloadFileItem.Length -lt 10*1024*1024)
{
Write-Host
Write-Host "ERROR: Downloaded file doesn't look large enough to be an ISO. The requested version may not be on microsoft.com yet."
Write-Host
Exit 1
}
# TODO Check if zip, exe, iso, etc.
try
{
Write-Host -NoNewline "Mounting ISO $file..."
Mount-ISO $downloadFile
Write-Host "Done"
$isoDrive = Get-ISODriveLetter $downloadFile
if (Test-Path $isoDrive)
{
Write-Host -NoNewLine "Installing WinSDK..."
$setupPath = Join-Path "$isoDrive" "WinSDKSetup.exe"
Start-Process -Wait $setupPath "/features $WindowsSDKOptions /q"
Write-Host "Done"
}
else
{
throw "Could not find mounted ISO at ${isoDrive}"
}
}
finally
{
Write-Host -NoNewline "Dismounting ISO $file..."
Dismount-ISO $downloadFile
Write-Host "Done"
}
}
if ($StrongNameHijack)
{
Write-Host -NoNewline "Disabling StrongName for Windows SDK..."
foreach($key in $PublicKeyTokens)
{
Disable-StrongName $key
}
Write-Host "Done"
}

View File

@@ -3,24 +3,12 @@
# Checks for code formatting errors. Will throw exception if any are found.
function Invoke-CheckBadCodeFormatting() {
Import-Module ./tools/OpenConsole.psm1
# Don't run the XAML formatter in this step - even if it changes nothing,
# it'll still touch all the .xaml files.
Invoke-CodeFormat -IgnoreXaml
Invoke-CodeFormat
# returns a non-zero exit code if there are any diffs in the tracked files in the repo
git diff-index --quiet HEAD --
if ($lastExitCode -eq 1) {
# Write the list of files that need updating to the log
git diff-index --name-only HEAD
throw "code formatting bad, run Invoke-CodeFormat on branch"
}
# Manually check the formatting of our .xaml files, without touching them.
Test-XamlFormat
}
Invoke-CheckBadCodeFormatting

View File

@@ -1,14 +0,0 @@
# Copyright (c) Microsoft Corporation.
# Licensed under the MIT license.
Param(
[string]$NewWindowsVersion = "10.0.22000.0"
)
Get-ChildItem src/cascadia/CascadiaPackage -Recurse -Filter *.appxmanifest | ForEach-Object {
$xml = [xml](Get-Content $_.FullName)
$xml.Package.Dependencies.TargetDeviceFamily | Where-Object Name -Like "Windows*" | ForEach-Object {
$_.MinVersion = $NewWindowsVersion
}
$xml.Save($_.FullName)
}

View File

@@ -8,7 +8,7 @@ Param(
[Parameter(HelpMessage="Path to Windows Kit")]
[ValidateScript({Test-Path $_ -Type Leaf})]
[string]
$WindowsKitPath = "C:\Program Files (x86)\Windows Kits\10\bin\10.0.19041.0"
$WindowsKitPath = "C:\Program Files (x86)\Windows Kits\10\bin\10.0.18362.0"
)
$ErrorActionPreference = "Stop"
@@ -58,7 +58,7 @@ Try {
### Check the activatable class entries for a few DLLs we need.
$inProcServers = $Manifest.Package.Extensions.Extension.InProcessServer.Path
$RequiredInProcServers = ("TerminalApp.dll", "Microsoft.Terminal.Control.dll", "Microsoft.Terminal.Remoting.dll", "Microsoft.Terminal.Settings.Editor.dll", "Microsoft.Terminal.Settings.Model.dll", "TerminalConnection.dll")
$RequiredInProcServers = ("TerminalApp.dll", "TerminalControl.dll", "TerminalConnection.dll")
Write-Verbose "InProc Servers: $inProcServers"

View File

@@ -10,18 +10,4 @@
<OpenConsoleDir>$(MSBuildThisFileDirectory)</OpenConsoleDir>
</PropertyGroup>
<PropertyGroup>
<!--
For the Windows 10 build, we're targeting the prerelease version of Microsoft.UI.Xaml.
This version emits every XAML DLL directly into our package.
This is a workaround for us not having deliverable MSFT-21242953 on this version of Windows.
-->
<TerminalMUXVersion>2.7.2-prerelease.220406002</TerminalMUXVersion>
<!--
For the Windows 11-specific build, we're targeting the public version of Microsoft.UI.Xaml.
This version emits a package dependency instead of embedding the dependency in our own package.
-->
<TerminalMUXVersion Condition="'$(TerminalTargetWindowsVersion)'=='Win11'">2.7.1</TerminalMUXVersion>
</PropertyGroup>
</Project>

View File

@@ -19,13 +19,9 @@
"/.github/",
"/samples/",
"/res/terminal/",
"/res/fonts/",
"/doc/specs/",
"/doc/cascadia/",
"/doc/user-docs/",
"/src/tools/MonarchPeasantSample/",
"/scratch/",
"Scratch.sln",
"/doc/user-docs/"
],
"SuffixFilters": [
".dbb",
@@ -41,7 +37,6 @@
".wrn",
".rec",
".err",
"XamlStyler.json",
".xlsx"
]
]
}

View File

@@ -2,22 +2,10 @@
<Project ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
<!-- This file is read by XES, which we use in our Release builds. -->
<PropertyGroup Label="Version">
<!--
The Windows 11 build is going to have the same package name, so it *must* have a different version.
The easiest way for us to do this is to add 1 to the revision field.
In short, for a given Terminal build 1.11, we will emit two different versions (assume this is build
4 on day 23 of the year):
- 1.11.234.0 for Windows 10
- 1.11.235.0 for Windows 11
This presents a potential for conflicts if we want to ship two builds produced back to back on the
same day... which is terribly unlikely.
-->
<VersionBuildRevision Condition="'$(TerminalTargetWindowsVersion)'=='Win11' and '$(VersionBuildRevision)'!=''">$([MSBuild]::Add($(VersionBuildRevision), 1))</VersionBuildRevision>
<XesUseOneStoreVersioning>true</XesUseOneStoreVersioning>
<XesBaseYearForStoreVersion>2021</XesBaseYearForStoreVersion>
<XesBaseYearForStoreVersion>2020</XesBaseYearForStoreVersion>
<VersionMajor>1</VersionMajor>
<VersionMinor>12</VersionMinor>
<VersionMinor>6</VersionMinor>
<VersionInfoProductName>Windows Terminal</VersionInfoProductName>
</PropertyGroup>
</Project>

Submodule dep/wil updated: 2e225973d6...3c00e7f1d8

View File

@@ -11,7 +11,7 @@
- Yes, the large majority of the `DEFINE_PROPERTYKEY` defs are the same, it's only the last byte of the guid that changes
2. Add matching fields to Settings.hpp
- Add getters, setters, the whole drill.
- add getters, setters, the whole drill.
3. Add to the propsheet
- We need to add it to *reading and writing* the registry from the propsheet, and *reading* the link from the propsheet. Yes, that's weird, but the propsheet is smart enough to re-use ShortcutSerialization::s_SetLinkValues, but not smart enough to do the same with RegistrySerialization.

View File

@@ -189,7 +189,7 @@ I think there might be a bit of a misunderstanding here - there are two differen
* shell applications, like `cmd.exe`, `powershell`, `zsh`, etc. These are text-only applications that emit streams of characters. They don't care at all about how they're eventually rendered to the user. These are also sometimes referred to as "commandline client" applications.
* terminal applications, like the Windows Terminal, gnome-terminal, xterm, iterm2, hyper. These are graphical applications that can be used to render the output of commandline clients.
On Windows, if you just run `cmd.exe` directly, the OS will create an instance of `conhost.exe` as the _terminal_ for `cmd.exe`. The same thing happens for `powershell.exe`, the system will create a new conhost window for any client that's not already connected to a terminal of some sort. This has lead to an enormous amount of confusion for people thinking that a conhost window is actually a "`cmd` window". `cmd` can't have a window, it's just a commandline application. Its window is always some other terminal.
On Windows, if you just run `cmd.exe` directly, the OS will create an instance of `conhost.exe` as the _terminal_ for `cmd.exe`. The same thing happens for `powershell.exe`, the system will creates a new conhost window for any client that's not already connected to a terminal of some sort. This has lead to an enormous amount of confusion for people thinking that a conhost window is actually a "`cmd` window". `cmd` can't have a window, it's just a commandline application. Its window is always some other terminal.
Any terminal can run any commandline client application. So you can use the Windows Terminal to run whatever shell you want. I use mine for both `cmd` and `powershell`, and also WSL:

View File

@@ -12,11 +12,11 @@ Use the [TAEF Verify Macros for C++](https://docs.microsoft.com/en-us/windows-ha
### Running Tests
If you have Visual Studio and related C++ components installed, and you have successfully restored NuGets, you should have the TAEF test runner `te.exe` available locally as part of the `Microsoft.Taef` package.
If you have Visual Studio and related C++ components installed, and you have successfully restored NuGets, you should have the TAEF test runner `te.exe` available locally as part of the `Taef.Redist.Wlk` package.
> Note that you cannot easily run TAEF tests directly through Visual Studio. The `Microsoft.Taef` NuGet package comes with an adapter that will let you browse and execute TAEF tests inside of Visual Studio, but its performance and reliability prevent us from recommending it here.
> Note that you cannot easily run TAEF tests directly through Visual Studio. The `Taef.Redist.Wlk` NuGet package comes with an adapter that will let you browse and execute TAEF tests inside of Visual Studio, but its performance and reliability prevent us from recommending it here.
In a "normal" CMD environment, `te.exe` may not be directly available. Try the following command to set up the development environment first:
In a "normal" CMD environment, `te.exe` may not be directly available. Try the following command to set up the development enviroment first:
```shell
.\tools\razzle.cmd
@@ -48,17 +48,3 @@ Invoke-OpenConsoleTests
```
`Invoke-OpenConsoleTests` supports a number of options, which you can enumerate by running `Invoke-OpenConsoleTests -?`.
### Debugging Tests
If you want to debug a test, you can do so by using the TAEF /waitForDebugger flag, such as:
runut *Tests.dll /name:TextBufferTests::TestInsertCharacter /waitForDebugger
Replace the test name with the one you want to debug. Then, TAEF will begin executing the test and output something like this:
TAEF: Waiting for debugger - PID <some PID> @ IP <some IP address>
You can then attach to that PID in your debugger of choice. In Visual Studio, you can use Debug -> Attach To Process, or you could use WinDbg or whatever you want.
Once the debugger attaches, the test will execute and your breakpoints will be hit.

View File

@@ -127,4 +127,4 @@ When a release is created, if the PR ID number is linked inside the release desc
- Issue message: 🎉This issue was addressed in #{pull request ID}, which has now been successfully released as {release name} {release version}.🎉"
## Admin Panel
[Here](https://portal.fabricbot.ms/bot/?repo=microsoft/terminal)
[Here](https://fabric-cp.azurewebsites.net/bot/)

View File

@@ -9,13 +9,7 @@ git submodule update --init --recursive
OpenConsole.sln may be built from within Visual Studio or from the command-line using a set of convenience scripts & tools in the **/tools** directory:
When using Visual Studio, be sure to set up the path for code formatting. To download the required clang-format.exe file, follow one of the building instructions below and run:
```powershell
Import-Module .\tools\OpenConsole.psm1
Set-MsBuildDevEnvironment
Get-Format
```
After, go to Tools > Options > Text Editor > C++ > Formatting and check "Use custom clang-format.exe file" in Visual Studio and choose the clang-format.exe in the repository at /packages/clang-format.win-x86.10.0.0/tools/clang-format.exe by clicking "browse" right under the check box.
When using Visual Studio, be sure to set up the path for code formatting. This can be done in Visual Studio by going to Tools > Options > Text Editor > C++ > Formatting and checking "Use custom clang-format.exe file" and choosing the clang-format.exe in the repository at /dep/llvm/clang-format.exe by clicking "browse" right under the check box.
### Building in PowerShell

View File

@@ -1,397 +0,0 @@
# Adding Settings to Windows Terminal
Adding a setting to Windows Terminal is fairly straightforward. This guide serves as a reference on how to add a setting.
## 1. Terminal Settings Model
The Terminal Settings Model (`Microsoft.Terminal.Settings.Model`) is responsible for (de)serializing and exposing settings.
### `INHERITABLE_SETTING` macro
The `INHERITABLE_SETTING` macro can be used to implement inheritance for your new setting and store the setting in the settings model. It takes three parameters:
- `type`: the type that the setting will be stored as
- `name`: the name of the variable for storage
- `defaultValue`: the value to use if the user does not define the setting anywhere
### Adding a Profile setting
This tutorial will add `CloseOnExitMode CloseOnExit` as a profile setting.
1. In `Profile.h`, declare/define the setting:
```c++
INHERITABLE_SETTING(CloseOnExitMode, CloseOnExit, CloseOnExitMode::Graceful)
```
2. In `Profile.idl`, expose the setting via WinRT:
```c++
Boolean HasCloseOnExit();
void ClearCloseOnExit();
CloseOnExitMode CloseOnExit;
```
3. In `Profile.cpp`, add (de)serialization and copy logic:
```c++
// Top of file:
// - Add the serialization key
static constexpr std::string_view CloseOnExitKey{ "closeOnExit" };
// CopySettings() or Copy():
// - The setting is exposed in the Settings UI
profile->_CloseOnExit = source->_CloseOnExit;
// LayerJson():
// - get the value from the JSON
JsonUtils::GetValueForKey(json, CloseOnExitKey, _CloseOnExit);
// ToJson():
// - write the value to the JSON
JsonUtils::SetValueForKey(json, CloseOnExitKey, _CloseOnExit);
```
- If the setting is not a primitive type, in `TerminalSettingsSerializationHelpers.h` add (de)serialization logic for the accepted values:
```c++
// For enum values...
JSON_ENUM_MAPPER(::winrt::Microsoft::Terminal::Settings::Model::CloseOnExitMode)
{
JSON_MAPPINGS(3) = {
pair_type{ "always", ValueType::Always },
pair_type{ "graceful", ValueType::Graceful },
pair_type{ "never", ValueType::Never },
};
};
// For enum flag values...
JSON_FLAG_MAPPER(::winrt::Microsoft::Terminal::TerminalControl::CopyFormat)
{
JSON_MAPPINGS(5) = {
pair_type{ "none", AllClear },
pair_type{ "html", ValueType::HTML },
pair_type{ "rtf", ValueType::RTF },
pair_type{ "all", AllSet },
};
};
// NOTE: This is also where you can add functionality for...
// - overloaded type support (i.e. accept a bool and an enum)
// - custom (de)serialization logic (i.e. coordinates)
```
### Adding a Global setting
Follow the "adding a Profile setting" instructions above, but do it on the `GlobalAppSettings` files.
### Adding an Action
This tutorial will add the `openSettings` action.
1. In `KeyMapping.idl`, declare the action:
```c++
// Add the action to ShortcutAction
enum ShortcutAction
{
OpenSettings
}
```
2. In `ActionAndArgs.cpp`, add serialization logic:
```c++
// Top of file:
// - Add the serialization key
static constexpr std::string_view OpenSettingsKey{ "openSettings" };
// ActionKeyNamesMap:
// - map the new enum to the json key
{ OpenSettingsKey, ShortcutAction::OpenSettings },
```
3. If the action should automatically generate a name when it appears in the Command Palette...
```c++
// In ActionAndArgs.cpp GenerateName() --> GeneratedActionNames
{ ShortcutAction::OpenSettings, RS_(L"OpenSettingsCommandKey") },
// In Resources.resw for Microsoft.Terminal.Settings.Model.Lib,
// add the generated name
// NOTE: Visual Studio presents the resw file as a table.
// If you choose to edit the file with a text editor,
// the code should look something like this...
<data name="OpenSettingsCommandKey" xml:space="preserve">
<value>Open settings file</value>
</data>
```
4. If the action supports arguments...
- In `ActionArgs.idl`, declare the arguments
```c++
[default_interface] runtimeclass OpenSettingsArgs : IActionArgs
{
// this declares the "target" arg
SettingsTarget Target { get; };
};
```
- In `ActionArgs.h`, define the new runtime class
```c++
struct OpenSettingsArgs : public OpenSettingsArgsT<OpenSettingsArgs>
{
OpenSettingsArgs() = default;
// adds a getter/setter for your argument, and defines the json key
WINRT_PROPERTY(SettingsTarget, Target, SettingsTarget::SettingsFile);
static constexpr std::string_view TargetKey{ "target" };
public:
hstring GenerateName() const;
bool Equals(const IActionArgs& other)
{
auto otherAsUs = other.try_as<OpenSettingsArgs>();
if (otherAsUs)
{
return otherAsUs->_Target == _Target;
}
return false;
};
static FromJsonResult FromJson(const Json::Value& json)
{
// LOAD BEARING: Not using make_self here _will_ break you in the future!
auto args = winrt::make_self<OpenSettingsArgs>();
JsonUtils::GetValueForKey(json, TargetKey, args->_Target);
return { *args, {} };
}
IActionArgs Copy() const
{
auto copy{ winrt::make_self<OpenSettingsArgs>() };
copy->_Target = _Target;
return *copy;
}
};
```
- In `ActionArgs.cpp`, define `GenerateName()`. This is used to automatically generate a name when it appears in the Command Palette.
- In `ActionAndArgs.cpp`, add serialization logic:
```c++
// ActionKeyNamesMap --> argParsers
{ ShortcutAction::OpenSettings, OpenSettingsArgs::FromJson },
```
### Adding an Action Argument
Follow step 3 from the "adding an Action" instructions above, but modify the relevant `ActionArgs` files.
## 2. Setting Functionality
Now that the Terminal Settings Model is updated, Windows Terminal can read and write to the settings file. This section covers how to add functionality to your newly created setting.
### App-level settings
App-level settings are settings that affect the frame of Windows Terminal. Generally, these tend to be global settings. The `TerminalApp` project is responsible for presenting the frame of Windows Terminal. A few files of interest include:
- `TerminalPage`: XAML control responsible for the look and feel of Windows Terminal
- `AppLogic`: WinRT class responsible for window-related issues (i.e. the titlebar, focus mode, etc...)
Both have access to a `CascadiaSettings` object, for you to read the loaded setting and update Windows Terminal appropriately.
### Terminal-level settings
Terminal-level settings are settings that affect a shell session. Generally, these tend to be profile settings. The `TerminalApp` project is responsible for packaging this settings from the Terminal Settings Model to the terminal instance. There are two kinds of settings here:
- `IControlSettings`:
- These are settings that affect the `TerminalControl` (a XAML control that hosts a shell session).
- Examples include background image customization, interactivity behavior (i.e. selection), acrylic and font customization.
- The `TerminalControl` project has access to these settings via a saved `IControlSettings` member.
- `ICoreSettings`:
- These are settings that affect the `TerminalCore` (a lower level object that interacts with the text buffer).
- Examples include initial size, history size, and cursor customization.
- The `TerminalCore` project has access to these settings via a saved `ICoreSettings` member.
`TerminalApp` packages these settings into a `TerminalSettings : IControlSettings, ICoreSettings` object upon creating a new terminal instance. To do so, you must submit the following changes:
- Declare the setting in `IControlSettings.idl` or `ICoreSettings.idl` (whichever is relevant to your setting). If your setting is an enum setting, declare the enum here instead of in the `TerminalSettingsModel` project.
- In `TerminalSettings.h`, declare/define the setting...
```c++
// The WINRT_PROPERTY macro declares/defines a getter setter for the setting.
// Like INHERITABLE_SETTING, it takes in a type, name, and defaultValue.
WINRT_PROPERTY(bool, UseAcrylic, false);
```
- In `TerminalSettings.cpp`...
- update `_ApplyProfileSettings` for profile settings
- update `_ApplyGlobalSettings` for global settings
- If additional processing is necessary, that would happen here. For example, `backgroundImageAlignment` is stored as a `ConvergedAlignment` in the Terminal Settings Model, but converted into XAML's separate horizontal and vertical alignment enums for packaging.
### Actions
Actions are packaged as an `ActionAndArgs` object, then handled in `TerminalApp`. To add functionality for actions...
- In the `ShortcutActionDispatch` files, dispatch an event when the action occurs...
```c++
// ShortcutActionDispatch.idl
event Windows.Foundation.TypedEventHandler<ShortcutActionDispatch, Microsoft.Terminal.Settings.Model.ActionEventArgs> OpenSettings;
// ShortcutActionDispatch.h
TYPED_EVENT(OpenSettings, TerminalApp::ShortcutActionDispatch, Microsoft::Terminal::Settings::Model::ActionEventArgs);
// ShortcutActionDispatch.cpp --> DoAction()
// - dispatch the appropriate event
case ShortcutAction::OpenSettings:
{
_OpenSettingsHandlers(*this, eventArgs);
break;
}
```
- In `TerminalPage` files, handle the event...
```c++
// TerminalPage.h
// - declare the handler
void _HandleOpenSettings(const IInspectable& sender, const Microsoft::Terminal::Settings::Model::ActionEventArgs& args);
// TerminalPage.cpp --> _RegisterActionCallbacks()
// - register the handler
_actionDispatch->OpenSettings({ this, &TerminalPage::_HandleOpenSettings });
// AppActionHandlers.cpp
// - direct the function to the right place and call a helper function
void TerminalPage::_HandleOpenSettings(const IInspectable& /*sender*/,
const ActionEventArgs& args)
{
// NOTE: this if-statement can be omitted if the action does not support arguments
if (const auto& realArgs = args.ActionArgs().try_as<OpenSettingsArgs>())
{
_LaunchSettings(realArgs.Target());
args.Handled(true);
}
}
```
`AppActionHandlers` vary based on the action you want to perform. A few useful helper functions include:
- `_GetFocusedTab()`: retrieves the focused tab
- `_GetActiveControl()`: retrieves the active terminal control
- `_GetTerminalTabImpl()`: tries to cast the given tab as a `TerminalTab` (a tab that hosts a terminal instance)
## 3. Settings UI
### Exposing Enum Settings
If the new setting supports enums, you need to expose a map of the enum and the respective value in the Terminal Settings Model's `EnumMappings`:
```c++
// EnumMappings.idl
static Windows.Foundation.Collections.IMap<String, Microsoft.Terminal.Settings.Model.CloseOnExitMode> CloseOnExitMode { get; };
// EnumMappings.h
static winrt::Windows::Foundation::Collections::IMap<winrt::hstring, CloseOnExitMode> CloseOnExitMode();
// EnumMappings.cpp
// - this macro leverages the json enum mapper in TerminalSettingsSerializationHelper to expose
// the mapped values across project boundaries
DEFINE_ENUM_MAP(Model::CloseOnExitMode, CloseOnExitMode);
```
### Binding and Localizing the Enum Setting
Find the page in the Settings UI that the new setting fits best in. In this example, we are adding `LaunchMode`.
1. In `Launch.idl`, expose the bindable setting...
```c++
// Expose the current value for the setting
IInspectable CurrentLaunchMode;
// Expose the list of possible values
Windows.Foundation.Collections.IObservableVector<Microsoft.Terminal.Settings.Editor.EnumEntry> LaunchModeList { get; };
```
2. In `Launch.h`, declare the bindable enum setting...
```c++
// the GETSET_BINDABLE_ENUM_SETTING macro accepts...
// - name: the name of the setting
// - enumType: the type of the setting
// - settingsModelName: how to retrieve the setting (use State() to get access to the settings model)
// - settingNameInModel: the name of the setting in the terminal settings model
GETSET_BINDABLE_ENUM_SETTING(LaunchMode, Model::LaunchMode, State().Settings().GlobalSettings, LaunchMode);
```
3. In `Launch.cpp`, populate these functions...
```c++
// Constructor (after InitializeComponent())
// the INITIALIZE_BINDABLE_ENUM_SETTING macro accepts...
// - name: the name of the setting
// - enumMappingsName: the name from the TerminalSettingsModel's EnumMappings
// - enumType: the type for the enum
// - resourceSectionAndType: prefix for the localization
// - resourceProperty: postfix for the localization
INITIALIZE_BINDABLE_ENUM_SETTING(LaunchMode, LaunchMode, LaunchMode, L"Globals_LaunchMode", L"Content");
```
4. In `Resources.resw` for Microsoft.Terminal.Settings.Editor, add the localized text to expose each enum value. Use the following format: `<SettingGroup>_<SettingName><EnumValue>.Content`
- `SettingGroup`:
- `Globals` for global settings
- `Profile` for profile settings
- `SettingName`:
- the Pascal-case format for the setting type (i.e. `LaunchMode` for `"launchMode"`)
- `EnumValue`:
- the json key for the setting value, but with the first letter capitalized (i.e. `Focus` for `"focus"`)
- The resulting resw key should look something like this `Globals_LaunchModeFocus.Content`
- This is the text that will be used in your control
### Updating the UI
When adding a setting to the UI, make sure you follow the [UWP design guidance](https://docs.microsoft.com/windows/uwp/design/).
#### Enum Settings
Now, create a XAML control in the relevant XAML file. Use the following tips and tricks to style everything appropriately:
- Wrap the control in a `ContentPresenter` adhering to the `SettingContainerStyle` style
- Bind `SelectedItem` to the relevant `Current<Setting>` (i.e. `CurrentLaunchMode`). Ensure it's a TwoWay binding
- Bind `ItemsSource` to `<Setting>List` (i.e. `LaunchModeList`)
- Set the ItemTemplate to the `Enum<ControlType>Template` (i.e. `EnumRadioButtonTemplate` for radio buttons)
- Set the style to the appropriate one in `CommonResources.xaml`
```xml
<!--Launch Mode-->
<ContentPresenter Style="{StaticResource SettingContainerStyle}">
<muxc:RadioButtons x:Uid="Globals_LaunchMode"
SelectedItem="{x:Bind CurrentLaunchMode, Mode="TwoWay"}"
ItemsSource="{x:Bind LaunchModeList}"
ItemTemplate="{StaticResource EnumRadioButtonTemplate}"
Style="{StaticResource RadioButtonsSettingStyle}"/>
</ContentPresenter>
```
To add any localized text, add a `x:Uid`, and access the relevant property via the Resources.resw file. For example, `Globals_LaunchMode.Header` sets the header for this control. You can also set the tooltip text like this:
`Globals_DefaultProfile.[using:Windows.UI.Xaml.Controls]ToolTipService.ToolTip`.
#### Non-Enum Settings
Continue to reference `CommonResources.xaml` for appropriate styling and wrap the control with a similar `ContentPresenter`. However, instead of binding to the `Current<Setting>` and `<Setting>List`, bind directly to the setting via the state. Binding a setting like `altGrAliasing` should look something like this:
```xml
<!--AltGr Aliasing-->
<ContentPresenter Style="{StaticResource SettingContainerStyle}">
<CheckBox x:Uid="Profile_AltGrAliasing"
IsChecked="{x:Bind State.Profile.AltGrAliasing, Mode=TwoWay}"
Style="{StaticResource CheckBoxSettingStyle}"/>
</ContentPresenter>
```
#### Profile Settings
If you are specifically adding a Profile setting, in addition to the steps above, you need to make the setting observable by modifying the `Profiles` files...
```c++
// Profiles.idl --> ProfileViewModel
// - this declares the setting as observable using the type and the name of the setting
OBSERVABLE_PROJECTED_SETTING(Microsoft.Terminal.Settings.Model.CloseOnExitMode, CloseOnExit);
// Profiles.h --> ProfileViewModel
// - this defines the setting as observable off of the _profile object
OBSERVABLE_PROJECTED_SETTING(_profile, CloseOnExit);
// Profiles.h --> ProfileViewModel
// - if the setting cannot be inherited by another profile (aka missing the Clear() function), use the following macro instead:
PERMANENT_OBSERVABLE_PROJECTED_SETTING(_profile, Guid);
```
The `ProfilePageNavigationState` holds a `ProfileViewModel`, which wraps the `Profile` object from the Terminal Settings Model. The `ProfileViewModel` makes all of the profile settings observable.
### Actions
Actions are not yet supported in the Settings UI.

View File

@@ -166,7 +166,7 @@ should be working just the same as before.
Now that you have a static library project, you can start building your unittest
dll. Start by creating a new directory for your unittest code, and creating a
`.vcxproj` for a TAEF unittest dll. For the Terminal solution, we use the TAEF
nuget package `Microsoft.Taef`.
nuget package `Taef.Redist.Wlk`.
### Referencing your C++/WinRT static lib
@@ -380,7 +380,7 @@ Here's the AppxManifest we're using:
</Properties>
<Dependencies>
<TargetDeviceFamily Name="Windows.Universal" MinVersion="10.0.18362.0" MaxVersionTested="10.0.19041.0" />
<TargetDeviceFamily Name="Windows.Universal" MinVersion="10.0.18362.0" MaxVersionTested="10.0.18362.0" />
<PackageDependency Name="Microsoft.VCLibs.140.00.Debug" MinVersion="14.0.27023.1" Publisher="CN=Microsoft Corporation, O=Microsoft Corporation, L=Redmond, S=Washington, C=US" />
<PackageDependency Name="Microsoft.VCLibs.140.00.Debug.UWPDesktop" MinVersion="14.0.27027.1" Publisher="CN=Microsoft Corporation, O=Microsoft Corporation, L=Redmond, S=Washington, C=US" />
</Dependencies>
@@ -517,7 +517,7 @@ This is because of a few key lines we already put in the appxmanifest:
```xml
<Dependencies>
<TargetDeviceFamily Name="Windows.Universal" MinVersion="10.0.18362.0" MaxVersionTested="10.0.19041.0" />
<TargetDeviceFamily Name="Windows.Universal" MinVersion="10.0.18362.0" MaxVersionTested="10.0.18362.0" />
<PackageDependency Name="Microsoft.VCLibs.140.00.Debug" MinVersion="14.0.27023.1" Publisher="CN=Microsoft Corporation, O=Microsoft Corporation, L=Redmond, S=Washington, C=US" />
<PackageDependency Name="Microsoft.VCLibs.140.00.Debug.UWPDesktop" MinVersion="14.0.27027.1" Publisher="CN=Microsoft Corporation, O=Microsoft Corporation, L=Redmond, S=Washington, C=US" />
</Dependencies>

File diff suppressed because it is too large Load Diff

View File

@@ -1,65 +0,0 @@
# til::feature
Feature flags are controlled by an XML document stored at `src/features.xml`.
## Example Document
```xml
<?xml version="1.0"?>
<featureStaging xmlns="http://microsoft.com/TilFeatureStaging-Schema.xsd">
<feature>
<!-- This will produce Feature_XYZ::IsEnabled() and TIL_FEATURE_XYZ_ENABLED (preprocessor) -->
<name>Feature_XYZ</name>
<description>Does a cool thing</description>
<!-- GitHub deliverable number; optional -->
<id>1234</id>
<!-- Whether the feature defaults to enabled or disabled -->
<stage>AlwaysEnabled|AlwaysDisabled</stage>
<!-- Branch wildcards where the feature should be *DISABLED* -->
<alwaysDisabledBranchTokens>
<branchToken>branch/with/wildcard/*</branchToken>
<!-- ... more branchTokens ... -->
</alwaysDisabledBranchTokens>
<!-- Just like alwaysDisabledBranchTokens, but for *ENABLING* the feature. -->
<alwaysEnabledBranchTokens>
<branchToken>...</branchToken>
</alwaysEnabledBranchTokens>
<!-- Brandings where the feature should be *DISABLED* -->
<alwaysDisabledBrandingTokens>
<!-- Valid brandings include Dev, Preview, Release, WindowsInbox -->
<brandingToken>Release</brandingToken>
<!-- ... more brandingTokens ... -->
</alwaysDisabledBrandingTokens>
<!-- Just like alwaysDisabledBrandingTokens, but for *ENABLING* the feature -->
<alwaysEnabledBrandingTokens>
<branchToken>...</branchToken>
</alwaysEnabledBrandingTokens>
<!-- Unequivocally disable this feature in Release -->
<alwaysDisabledReleaseTokens />
</feature>
</featureStaging>
```
## Notes
Features that are disabled for Release using `alwaysDisabledReleaseTokens` are
*always* disabled in Release, even if they come from a branch that would have
been enabled by the wildcard.
### Precedence
1. `alwaysDisabledReleaseTokens`
2. Enabled branches
3. Disabled branches
* The longest branch token that matches your branch will win.
3. Enabled brandings
4. Disabled brandings
5. The feature's default state

View File

@@ -9,7 +9,7 @@ issue id: #1043
## Abstract
This spec is for task #1043 “Be able to set an initial position for the terminal”. It goes over the details of a new feature that allows users to set the initial position and size of the terminal. Expected behavior and design of this feature is included. Besides, future possible follow-up works are also addressed.
This spec is for task #1043 “Be able to set an initial position for the terminal”. It goes over the details of a new feature that allows users to set the initial position and size of the terminal. Expected behavior and design of this feature is included. Besides, future possible follow-up works are also addressed.
## Inspiration
@@ -17,7 +17,7 @@ The idea is to allow users to set the initial position of the Terminal when they
## Solution Design
For now, the Terminal window is put on a default initial position. The program uses CW_USEDEFAULT in the screen coordinates for top-left corner. We have two different types of window client window and non-client window. However, code path for window creation (WM_CREATE message is shared by the two types of windows) are almost the same for the two types of windows, except that there are some differences in calculation of the width and height of the window.
For now, the Terminal window is put on a default initial position. The program uses CW_USEDEFAULT in the screen coordinates for top-left corner. We have two different types of window client window and non-client window. However, code path for window creation (WM_CREATE message is shared by the two types of windows) are almost the same for the two types of windows, except that there are some differences in calculation of the width and height of the window.
Two new properties should be added in the json settings file:
@@ -92,9 +92,9 @@ For now, this feature only allows the user to set initial position and choose wh
3. We may also consider more launch modes. Like full screen mode and minimized mode.
GitHub issue for future follow-ups: https://github.com/microsoft/terminal/issues/766
Github issue for future follow-ups: https://github.com/microsoft/terminal/issues/766
## Resources
GitHub issue:
Github issue:
https://github.com/microsoft/terminal/issues/1043

View File

@@ -1,105 +0,0 @@
---
author: Pankaj Bhojwani, pabhojwa@microsoft.com
created on: 2021-6-17
last updated: 2021-6-23
issue id: #1790
---
# Font features and axes of variation
## Abstract
This spec outlines how we can allow users to specify font features and axes of variation for fonts in Windows Terminal. Font features include things like being able to specify whether ligatures should be used as well as the specific stylistic set used for a font. Axes of variation commonly include things like weight and slant but can also include fancier things like shadow distance, depending on the font.
## Inspiration
Reference: [#1790](https://github.com/microsoft/terminal/issues/1790)
Currently, if a font has ligatures, we offer no way for a user to disable them. Many users would like the option to do so, and would also like the ability to choose stylistic sets for fonts - for example, at the time of this writing, Cascadia Code offers 4 stylistic sets but we offer no way for users to specify any of them.
In a similar vein, many fonts allow for setting variations on the font along certain attributes, commonly referred to as 'axes of variation'. We can offer users more font customization options by allowing them to configure these font variations.
## Solution Design
### Font features
It is already possible to pass in a list of [font feature structs](https://docs.microsoft.com/en-us/windows/win32/api/dwrite/ns-dwrite-dwrite_font_feature) to DWrite for it to handle. A font feature struct contains only 2 things:
1. A font feature tag
2. A parameter value
A font feature tag is constructed using a 4-character feature tag and the parameter value defines how the feature is applied. For most features, the parameter value is simply treated as a binary value - a value of 0 means the feature is not applied and a non-zero value means the feature is applied. For example, a font feature struct like {'ss03', 1} enables stylistic set 3 for the font and a font feature struct like {'liga', 0} disables ligatures. (Technically, the feature tag is _constructed_ with the 4-character tag and is not the 4-character tag itself, but they are treated the same in the example here for brevity's sake).
Currently, we pass in to DWrite a null value for the list of features to apply to the font. This causes DWrite to automatically apply a ['standard' list](https://github.com/fdwr/TextLayoutSampler/blob/master/DrawableObject.ixx#L802) of font features to the font. Naturally, passing in our own list of font features to DWrite means DWrite will _only_ apply the features we defined, and no longer apply the standard list. Since the standard list contains 11 features, we need to consider how we can allow users to specify 1 additional feature or delete 1 of the standard features without needing to redefine all the others.
We will do this by allowing users to define a dictionary in their settings.json file, where the keys are the 4-character feature tags and the values are the parameter values. This dictionary will then get applied to our internal dictionary (which will contain the standard list of 11 features with their parameter values), meaning that any new key-value pairs will get added to our dictionary and any existing key-value pairs will get updated. Finally, this 'merged' dictionary will be what we use to construct the list of features to pass into DWrite.
### Axes of variation
Specifying axes of variation is done in an extremely similar manner to the way font features are specified - a 4-character tag is used to specify which font axis is being modified and a numerical value is provided to specify the value the axis should be set to. For example, {'slnt', 20} specifies that the 'slant' axis should be set to 20.
There is also a standard list of axes of variation, and each axis has its own default. We will approach this the same way we approached font features, by allowing users to specify additional features or omit features without needing to redefine the defaults.
## UI/UX Design
Users will be able to add a new setting to their font objects (added in [#10433](https://github.com/microsoft/terminal/pull/10433)). The resultant font object may look something like this
```json
"font": {
"face": "Cascadia Code",
"size": 12,
"features": {
"ss03": 1,
"liga": 0
},
"axes": {
"slnt": 20.5
}
}
```
There is one point to note here about clashing. For example, if a user has the old "weight" setting defined _as well as_ a "wght" axis defined, we will only use the "wght" axis value. We prioritize that value for a few reasons:
1. It is the more recent addition to our settings model. Thus, it is likely that a user that has defined both values probably just forgot to remove the old value.
2. It is the more precise value, it is a specific float value whereas the the old "weight" setting is an enum (that eventually gets mapped to a float value).
## Capabilities
### Accessibility
Should not affect accessibility.
### Security
Should not affect security.
### Reliability
Aside from additional parsing required for the settings file (which inherently offers more locations for parsing to fail), we need to be careful about badly formed/non-existant feature tags or axes specified in the user-defined dictionaries. We must make sure to ignore such declarations (perhaps alongside emitting a warning to the user) and only apply those that are correctly formed and exist.
### Compatibility
Older versions of Windows may not have the DWrite updates that allow for defining font features and axes of variation. We must make sure to fallback to the current implementation in these cases.
### Performance, Power, and Efficiency
Currently when rendering a run of text, if we detect that the given run is simple we will use a shortcut to obtain the glyphs needed, skipping over an expensive `GetGlyphs` call to DWrite. However, when the default feature list is changed in any way (either by adding a new feature or removing one of the defaults), there is no way for us to detect beforehand how the font glyphs would change.
This means that as long as the user requests a change to the default font feature list, we will _always_ skip the shortcut and call the expensive `GetGlyphs` function for every run of text.
This will naturally cause a performance cost that we will have to bear for this feature. However, it is worth noting that there are a fair number of glyphs that will cause a run of text to be deemed "not simple" (and thus cause us to call `GetGlyphs` anyway), for example when using Cascadia Code, any run of text that has the letters 'i', 'j', 'l', 'n', 'w' or 'x' is not considered simple (because those glyphs have localized variants).
## Potential Issues
See performance issues above.
## Future considerations
DWrite additionally offers the ability to vary the font features across runs of text. However, for our initial implementation of this feature, we will only apply font features to the entire buffer. If/when we decide to allow specifying font features for particular runs of text, we can lean into our existing mechanisms of splitting up runs of text to implement that.
We will also need to consider how we want to represent this in the settings UI. This is slightly more complex than other settings since users should be allowed to manually input 4-character tags.
## Resources
[DWRITE_FONT_FEATURE structure](https://docs.microsoft.com/en-us/windows/win32/api/dwrite/ns-dwrite-dwrite_font_feature)
[DWRITE_FONT_AXIS_VALUE structure](https://docs.microsoft.com/en-us/windows/win32/api/dwrite_3/ns-dwrite_3-dwrite_font_axis_value)

View File

@@ -1,302 +0,0 @@
---
author: Mike Griese @zadjii-msft
created on: 2020-11-23
last updated: 2020-12-15
issue id: #2871
---
# Focus Pane Actions
## Abstract
Currently, the Terminal only allows users to navigate through panes
_directionally_. However, we might also want to allow a user to navigate through
panes in most recently used order ("MRU" order), or to navigate directly to a
specific pane. This spec proposes some additional actions in order to enable
these sorts of scenarios.
## Background
### Inspiration
`tmux` allows the user to navigate through panes using its `select-pane`
command. The `select-pane` command works in the following way:
```
select-pane [-DLlMmRU] [-T title] [-t target-pane]
Make pane target-pane the active pane in window target-window, or set its
style (with -P). If one of -D, -L, -R, or -U is used, respectively the
pane below, to the left, to the right, or above the target pane is used.
-l is the same as using the last-pane command.
-m and -M are used to set and clear the marked pane. There is one marked
pane at a time, setting a new marked pane clears the last. The marked pane
is the default target for -s to join-pane, swap-pane and swap-window.
```
_from `man tmux`_.
The Terminal currently allows the user to navigate through panes with the
`moveFocus` action, which only accepts a `direction` to move in.
Additionally, the Terminal allows movement between tabs with the `nextTab` and
`prevTab` actions, who move between tabs either in-order or in MRU order.
Furthermore, these actions may or may not display the "tab switcher" user
interface, based on the value of `tabSwitcherMode`.
### User Stories
* **Scenario 1**: A user who wants to be able to split the window into 4 equal
corners from the commandline. Currently this isn't possible, because the user
cannot move focus during the startup actions - `split-pane` actions always end
up splitting the current leaf in the tree of panes. (see [#5464])
* **Scenario 2**: A user who wants to quickly navigate to the previous pane they
had opened. (see [#2871])
* **Scenario 3**: A user who wants to bind a keybinding like <kbd>alt+1</kbd>,
<kbd>alt+2</kbd>, etc to immediately focus the first, second, etc. pane in a
tab. (see [#5803])
### Future Considerations
There's been talk of updating the advanced tab switcher to also display panes,
in addition to just tabs. This would allow users to navigate through the ATS
directly to a pane, and see all the panes in a tab. Currently, `tabSwitcherMode`
changes the behavior of `nextTab`, `prevTab` - should we just build the
`paneSwitcherMode` directly into the action we end up designing?
## Solution Design
Does using the pane switcher with a theoretical `focusPane(target=id)` action
even make sense? Certainly not! That's like `switchToTab(index=id)`, the user
already knows which tab they want to go to, there's no reason to pop an
ephemeral UI in front of them.
Similarly, it almost certainly doesn't make sense to display the pane switcher
while moving focus directionally. Consider moving focus with a key bound to the
arrow keys. Displaying another UI in front of them while moving focus with the
arrow keys would be confusing.
Addressing Scenario 1 is relatively easy. So long as we add any of the proposed
actions, including the existing `moveFocus` action as a subcommand that can be
passed to `wt.exe`, then the user should be able to navigate through the panes
they've created with the startup commandline, and build the tree of panes
however they see fit.
Scenario 2 is more complicated, because MRU switching is always more
complicated. Without a UI of some sort, there's no way to switch to another pane
in the MRU order without also updating the MRU order as you go. So this would
almost certainly necessitate a "pane switcher", like the tab switcher.
### Proposal A: Add next, prev to moveFocus
* `moveFocus(direction="up|down|left|right|next|prev")`
* **Pros**:
- Definitely gets the "MRU Pane Switching" scenario working
* **Cons**:
- Doesn't really address any of the other scenarios
- How will it play with pane switching in the UI?
- MRU switching without a dialog to track & display the MRU stack doesn't
really work - this only allows to the user to navigate to the most recently
used pane, or through all the panes in least-recently-used order. This is
because switching to the MRU pane _will update the MRU pane_.
❌ This proposal is no longer being considered.
### Proposal B: focusNextPane, focusPrevPane with order, useSwitcher args
```json
// Focus pane 1
// - This is sensible, no arguments here
{ "command": { "action": "focusPane", "id": 1 } },
// Focus the next MRU pane
// - Without the switcher, this can only go one pane deep in the MRU stack
// - presumably once there's a pane switcher, it would default to enabled?
{ "command": { "action": "focusNextPane", "order": "mru" } },
// Focus the prev inOrder pane
// - this seems straightforward
{ "command": { "action": "focusPrevPane", "order": "inOrder" } },
// Focus the next pane, in mru order, explicitly disable the switcher
// - The user opted in to only being able to MRU switch one deep. That's fine, that's what they want.
{ "command": { "action": "focusNextPane", "order": "mru", "useSwitcher": false} },
// Focus the prev inOrder pane, explicitly with the switcher
// - Maybe they disabled the switcher globally, but what it on for this action?
{ "command": { "action": "focusPrevPane", "order": "inOrder", "useSwitcher": true } },
```
_From [discussion in the implementation
PR](https://github.com/microsoft/terminal/pull/8183#issuecomment-729672645)_
Boiled down, that's three actions:
* `focusPane(target=id)`
* `focusNextPane(order="inOrder|mru", useSwitcher=true|false)`
* `focusPrevPane(order="inOrder|mru", useSwitcher=true|false)`
* **Pros**:
- Everything is explicit, including the option to use the pane switcher (when
available)
- Adds support for in-order pane switching
- No "conditional parameters" - where providing one argument makes other
arguments invalid or ambiguous.
* **Cons**:
- Doesn't really address any of the other scenarios
- What does the "next most-recently-used tab" even mean? How is it different
than "previous most-recently-used tab"? Semantically, these are the same
thing!
- No one's even asked for in-order pane switching. Is that a UX that even
really makes sense?
❌ This proposal is no longer being considered.
> 👉 **NOTE**: At this point, we stopped considering navigating in both MRU
> "directions", since both the next and prev MRU pane are the same thing. We're
> now using "last" to mean "the previous MRU pane".
### Proposal C: One actions, combine the args
* `moveFocus(target=id|"up|down|left|right|last")`
* **Pros**:
- Absolutely the least complicated action to author. There's only one
parameter, `target`.
- No "conditional parameters".
* **Cons**:
- How do we express this in the Settings UI? Mixed-type enums work fine for
the font weight, where each enum value has a distinct integer value it maps
to, but in this case, using `id` is entirely different from the other
directional values
❌ This proposal is no longer being considered.
### Proposal D: Two actions
* `focusPane(target=id)`
* `moveFocus(direction="up|down|left|right|last")`
* **Pros**:
- Each action does explicitly one thing.
* **Cons**:
- two actions for _similar_ behavior
- This now forks the "Direction" enum into "MoveFocusDirection" and
"ResizeDirection" (because `resizePane(last)` doesn't make any sense).
This proposal doesn't really have any special consideration for the pane
switcher UX. Neither of these actions would summon the pane switcher UX.
### Proposal E: Three actions
* `focusPane(target=id)`
* `moveFocus(direction="up|down|left|right")`
* `focusLastPane(usePaneSwitcher=false|true)`
In this design, neither `focusPane` nor `moveFocus` will summon the pane
switcher UI (even once it's added). However, the `focusLastPane` one _could_,
and subsequent keypresses could pop you through the MRU stack, while it's
visible? The pane switcher could then display the panes for the tab in MRU
order, and the user could just use the arrow keys to navigate the list if they
so choose.
* **Pros**:
- Each action does explicitly one thing.
- Design accounts for future pane switcher UX
* **Cons**:
- Three separate actions for similar behavior
❌ This proposal is no longer being considered.
### Proposal F: It's literally just tmux
_Also known as the "one action to rule them all" proposal_
`focusPane(target=id, direction="up|down|left|right|last")`
Previously, this design was avoided, because what does `focusPane(target=4,
direction=down)` do? Does it focus pane 4, or does it move focus down?
`tmux` solves this in one action by just doing both!
```
Make pane target-pane the active pane ... If one of -D, -L, -R, or -U is used,
respectively the pane below, to the left, to the right, or above the target pane
is used.
```
_from `man tmux`_.
So `focusPane(target=1, direction=up)` will attempt to focus the pane above pane
1. This action would not summon the pane switcher UX, even for
`focusPane(direction=last)`
* **Pros**:
- Fewest redundant actions
* **Cons**:
- Is this intuitive? That combining the params would do both, with `target`
happening "first"?
- Assumes that there will be a separate action added in the future for "Open
the pane switcher (with some given ordering)"
> 👉 **NOTE**: At this point, the author considered "Do we even want a separate
> action to engage the tab switcher with panes expanded?" Perhaps panes being
> visible in the tab switcher is just part fo the tab switcher's behavior. Maybe
> there shouldn't be a separate "open the tab switcher with the panes expanded
> to the pane I'm currently on, and the panes listed in MRU order" action.
❌ This proposal is no longer being considered.
## Conclusion
After much discussion as a team, we decided that **Proposal D** would be the
best option. We felt that there wasn't a need to add any extra configuration to
invoke the "pane switcher" as anything different than the "tab switcher". The
"pane switcher" should really just exist as a part of the functionality of the
advanced tab switcher, not as it's own thing.
Additionally, we concurred that the new "direction" value should be `prev`, not
`last`, for consistency's sake.
## UI/UX Design
The only real UX being added with the agreed upon design is allowing the user to
execute an action to move to the previously active pane within a single tab. No
additional UX (including the pane switcher) is being prescribed in this spec at
this time.
## Potential Issues
<table>
<tr>
<td><strong>Compatibility</strong></td>
<td>
We've only adding a single enum value to an existing enum. Since we're not
changing the meaning of any of the existing values, we do not expect any
compatibility issues there. Additionally, we're not changing the default value
of the `direction` param of the `moveFocus` action, so there are no further
compatibility concerns there. Furthermore, no additional parameters are being
added to the `moveFocus` action that would potentially give it a different
meaning.
</td>
</tr>
</table>
In the current design, there's no way to move through all the panes with a
single keybinding. For example, if a user wanted to bind <kbd>Alt+]</kbd> to
move to the "next" pane, and <kbd>Alt+[</kbd> to move to the "previous" one.
These movements would necessarily need to be in-order traversals, since there's
no way of doing multiple MRU steps.
Fortunately, no one's really asked for traversing the panes in-order, so we're
not really worried about this. Otherwise, it would maybe make sense for `last`
to be the "previous MRU pane", and reserve `next`/`prev` for in-order traversal.
[#2871]: https://github.com/microsoft/terminal/issues/2871
[#5464]: https://github.com/microsoft/terminal/issues/5464
[#5803]: https://github.com/microsoft/terminal/issues/5803

View File

@@ -1,128 +0,0 @@
---
author: Pankaj Bhojwani, pabhojwa@microsoft.com
created on: 2020-11-20
last updated: 2021-2-5
issue id: #8345
---
# Appearance configuration objects for profiles
## Abstract
This spec outlines how we can support 'configuration objects' in our profiles, which
will allow us to render differently depending on the state of the control. For example, a
control can be rendered differently if it's focused as compared to when it's unfocused.
## Inspiration
Reference: [#3062](https://github.com/microsoft/terminal/issues/3062)
Users want there to be a more visible indicator than the one we have currently for which
pane is focused and which panes are unfocused. This change would grant us that feature.
## Solution Design
The implementation design for appearance config objects centers around the recent change where inheritance was added to the
`TerminalSettings` class in the Terminal Settings Model - i.e. different `TerminalSettings` objects can inherit from each other.
The reason for this change was that we did not want a settings reload to erase any overrides `TermControl` may have made
to the settings during runtime. By instead passing a child of the `TerminalSettings` object to the control, we can change
the parent of the child during a settings reload without the overrides being erased (since those overrides live in the child).
The idea behind unfocused appearance configurations is similar. We will pass in another `TerminalSettings` object to the control,
which is simply a child that already has some overrides in it. When the control gains or loses focus, it simply switches between
the two settings objects appropriately.
### Allowed parameters
For now, these states are meant to be entirely appearance-based. So, not all parameters which can be
defined in a `Profile` can be defined in this new object (for example, we do not want parameters which
would cause a resize in this object.) Here is the list of parameters we will allow:
- Anything regarding colors: `colorScheme`, `foreground`, `background`, `cursorColor` etc
- Anything regarding background image: `path`, `opacity`, `alignment`, `stretchMode`
- `cursorShape`
We may wish to allow further parameters in these objects in the future (like `bellStyle`?). The addition
of further parameters can be discussed in the future and is out of scope for this spec.
### Inheritance
The inheritance model can be thought of as an 'all-or-nothing' approach in the sense that the `unfocusedAppearance` object
is considered as a *single* setting instead of an object with many settings. We have chosen this model because it is cleaner
and easier to understand than the alternative, where each setting within an `unfocusedAppearance` object has a parent from which
it obtains its value.
Note that when `TerminalApp` initializes a control, it creates a `TerminalSettings` object for that profile and passes the
control a child of that object (so that the control can store overrides in the child, as described earlier). If an unfocused
config is defined in the profile (or in globals/profile defaults), then `TerminalApp` will create a *child of that child*,
put the relevant overrides in it, and pass that into the control as well. Thus, the inheritance of any undefined parameters
in the unfocused config will be as follows:
1. The unfocused config specified in the profile (or in globals/profile defaults)
2. Overrides made by the terminal control
3. The parent profile
## UI/UX Design
Users will be able to add a new setting to their profiles that will look like this:
```
"unfocusedAppearance":
{
"colorScheme": "Campbell",
"cursorColor": "#888",
"cursorShape": "emptyBox",
"foreground": "#C0C0C0",
"background": "#000000"
}
```
When certain appearance settings are changed via OSC sequences (such as the background color), only the focused/regular
appearance will change and the unfocused one will remain unchanged. However, since the unfocused settings object inherits
from the regular one, it will still apply the change (provided it does not define its own value for that setting).
## Capabilities
### Accessibility
Does not affect accessibility.
### Security
Does not affect security.
### Reliability
This is another location in the settings where parsing/loading the settings may fail. However, this is the case
for any new setting we add so I would say that this is a reasonable cost for this feature.
### Compatibility
Should not affect compatibility.
### Performance, Power, and Efficiency
Rapidly switching between many panes, causing several successive appearance changes in a short period of time, could
potentially impact performance. However, regular/reasonable pane switching should not have a noticeable effect.
## Potential Issues
Inactive tabs will be 'rendered' in the background with the `UnfocusedRenderingParams` object, we need to make
sure that switching to an inactive tab (and so causing the renderer to update with the 'normal' parameters)
does not cause the window to flash/show a jarring indicator that the rendering values changed.
## Future considerations
We will need to decide how this will look in the settings UI.
We may wish to add more states in the future (like 'elevated'). When that happens, we will need to deal with how
these appearance objects can scale/layer over each other. We had a lot of discussion about this and could not find
a suitable solution to the problem of multiple states being valid at the same time (like unfocused and elevated).
This, along with the fact that it is uncertain if there even will be more states we would want to add led us to
the conclusion that we should only support the unfocused state for now, and come back to this issue later. If there
are no more states other than unfocused and elevated, we could allow combining them (like having an 'unfocused elevated' state).
If there are more states, we could do the implementation as an extension rather than inherently supporting it.
## Resources

View File

@@ -1,619 +0,0 @@
---
author: Mike Griese @zadjii-msft
created on: 2020-11-20
last updated: 2021-08-17
issue id: #1032
---
# Elevation Quality of Life Improvements
## Abstract
For a long time, we've been researching adding support to the Windows Terminal
for running both unelevated and elevated (admin) tabs side-by-side, in the same
window. However, after much research, we've determined that there isn't a safe
way to do this without opening the Terminal up as a potential
escalation-of-privilege vector.
Instead, we'll be adding a number of features to the Terminal to improve the
user experience of working in elevated scenarios. These improvements include:
* A visible indicator that the Terminal window is elevated ([#1939])
* Configuring the Terminal to always run elevated ([#632])
* Configuring a specific profile to always open elevated ([#632])
* Allowing new tabs, panes to be opened elevated directly from an unelevated
window
* Dynamic profile appearance that changes depending on if the Terminal is
elevated or not. ([#1939], [#8311])
## Background
_This section was originally authored in the [Process Model 2.0 Spec]. Please
refer to it there for its original context._
Let's presume that you're a user who wants to be able to open an elevated tab
within an otherwise unelevated Terminal window. We call this scenario "mixed
elevation" - the tabs within the Terminal can be running either unelevated _or_
elevated client applications.
It wouldn't be terribly difficult for the unelevated Terminal to request the
permission of the user to spawn an elevated client application. The user would
see a UAC prompt, they'd accept, and then they'd be able to have an elevated
shell alongside their unelevated tabs.
However, this creates an escalation of privilege vector. Now, there's an
unelevated window which is connected directly to an elevated process. At this
point, **any other unelevated application could send input to the Terminal's
`HWND`**. This would make it possible for another unelevated process to "drive"
the Terminal window, and send commands to the elevated client application.
It was initially theorized that the window/content model architecture would also
help enable "mixed elevation". With mixed elevation, tabs could run at different
integrity levels within the same terminal window. However, after investigation
and research, it has become apparent that this scenario is not possible to do
safely after all. There are numerous technical difficulties involved, and each
with their own security risks. At the end of the day, the team wouldn't be
comfortable shipping a mixed-elevation solution, because there's simply no way
for us to be confident that we haven't introduced an escalation-of-privilege
vector utilizing the Terminal. No matter how small the attack surface might be,
we wouldn't be confident that there are _no_ vectors for an attack.
Some things we considered during this investigation:
* If a user requests a new elevated tab from an otherwise unelevated window, we
could use UAC to create a new, elevated window process, and "move" all the
current tabs to that window process, as well as the new elevated client. Now,
the window process would be elevated, preventing it from input injection, and
it would still contains all the previously existing tabs. The original window
process could now be discarded, as the new elevated window process will
pretend to be the original window.
- However, it is unfortunately not possible with COM to have an elevated
client attach to an unelevated server that's registered at runtime. Even in
a packaged environment, the OS will reject the attempt to `CoCreateInstance`
the content process object. this will prevent elevated windows from
re-connecting to unelevated client processes.
- We could theoretically build an RPC tunnel between content and window
processes, and use the RPC connection to marshal the content process to the
elevated window. However, then _we_ would need to be responsible for
securing access the the RPC endpoint, and we feel even less confident doing
that.
- Attempts were also made to use a window-broker-content architecture, with
the broker process having a static CLSID in the registry, and having the
window and content processes at mixed elevation levels `CoCreateInstance`
that broker. This however _also_ did not work across elevation levels. This
may be due to a lack of Packaged COM support for mixed elevation levels.
It's also possible that the author forgot that packaged WinRT doesn't play
nicely with creating objects in an elevated context. The Terminal has
previously needed to manually manifest all its classes in a SxS manifest for
Unpackaged WinRT to allow the classes to be activated, rather than relying
on the packaged catalog. It's theoretically possible that doing that would
have allowed the broker to be activated across integrity levels.
Even if this approach did end up working, we would still need to be
responsible for securing the elevated windows so that an unelevated attacker
couldn't hijack a content process and trigger unexpected code in the window
process. We didn't feel confident that we could properly secure this channel
either.
We also considered allowing mixed content in windows that were _originally_
elevated. If the window is already elevated, then it can launch new unelevated
processes. We could allow elevated windows to still create unelevated
connections. However, we'd want to indicate per-pane what the elevation state
of each connection is. The user would then need to keep track themselves of
which terminal instances are elevated, and which are not.
This also marks a departure from the current behavior, where everything in an
elevated window would be elevated by default. The user would need to specify for
each thing in the elevated window that they'd want to create it elevated. Or the
Terminal would need to provide some setting like
`"autoElevateEverythingInAnElevatedWindow"`.
We cannot support mixed elevation when starting in a unelevated window.
Therefore, it doesn't make a lot of UX sense to support it in the other
direction. It's a cleaner UX story to just have everything in a single window at
the same elevation level.
## Solution Design
Instead of supporting mixed elevation in the same window, we'll introduce the
following features to the Terminal. These are meant as a way of improving the
quality of life for users who work in mixed-elevation (or even just elevated)
environments.
### Visible indicator for elevated windows
As requested in [#1939], it would be nice if it was easy to visibly identify if
a Terminal window was elevated or not.
One easy way of doing this is by adding a simple UAC shield to the left of the
tabs for elevated windows. This shield could be configured by the theme (see
[#3327]). We could provide the following states:
* Colored (the default)
* Monochrome
* Hidden, to hide the shield even on elevated windows. This is the current
behavior.
![UAC-shield-in-titlebar](UAC-shield-in-titlebar.png)
_figure 1: a monochrome UAC shield in the titlebar of the window, courtesy of @mdtauk_
We could also simplify this to only allow a boolean true/false for displaying
the shield. As we do often with other enums, we could define `true` to be the
same as the default appearance, and `false` to be the hidden option. As always,
the development of the Terminal is an iterative process, where we can
incrementally improve from no setting, to a boolean setting, to a enum-backed
one.
### Configuring a profile to always run elevated
Oftentimes, users might have a particular tool chain that only works when
running elevated. In these scenarios, it would be convenient for the user to be
able to identify that the profile should _always_ run elevated. That way, they
could open the profile from the dropdown menu of an otherwise unelevated window
and have the elevated window open with the profile automatically.
We'll be adding the `"elevate": true|false` setting as a per-profile setting,
with a default value of `false`. When set to `true`, we'll try to auto-elevate
the profile whenever it's launched. We'll check to see if this window is
elevated before creating the connection for this profile. If the window is not
elevated, then we'll create a new window with the requested elevation level to
handle the new connection.
`"elevate": false` will do nothing. If the window is already elevated, then the
profile won't open an un-elevated window.
If the user tries to open an `"elevate": true` profile in a window that's
already elevated, then a new tab/split will open in the existing window, rather
than spawning an additional elevated window.
There are three situations where we're creating new terminal instances: new
tabs, new splits, and new windows. Currently, these are all actions that are
also exposed in the `wt` commandline as subcommands. We can convert from the
commandline arguments into these actions already. Therefore, it shouldn't be too
challenging to convert these actions back into the equal commandline arguments.
For the following examples, let's assume the user is currently in an unelevated
Terminal window.
When the user tries to create a new elevated **tab**, we'll need to create a new
process, elevated, with the following commandline:
```
wt new-tab [args...]
```
When we create this new `wt` instance, it will obey the glomming rules as
specified in [Session Management Spec]. It might end up glomming to another
existing window at that elevation level, or possibly create its own window.
Similarly, for a new elevated **window**, we can make sure to pass the `-w new`
arguments to `wt`. These parameters indicate that we definitely want this
command to run in a new window, regardless of the current glomming settings.
```
wt -w new new-tab [args...]
```
However, creating a new **pane** is a little trickier. Invoking the `wt
split-pane [args...]` is straightforward enough.
<!-- Discussion notes follow:
If the current window doesn't have the same elevation level as the
requested profile, do we always want to just create a new split? If the command
ends up glomming to an existing window, does that even make sense? That invoking
an elevated split in an unelevated window would end up splitting the elevated
window? It's very possible that the user wanted a split in the tab they're
currently in, in the unelevated window, but they don't want a split in the
elevated window.
What if there's not space in the elevated window to create the split (but there
would be in the current window)? That would sure make it seem like nothing
happened, silently.
We could alternatively have cross-elevation splits default to always opening a
new tab. That might mitigate some of the odd behaviors. Until we actually have
support for running commands in existing windows, we'll always need to make a
new window when running elevated. We'll need to make the new window for new tabs
and splits, because there's no way to invoke another existing window.
A third proposal is to pop a warning dialog at the user when they try to open an
elevated split from and unelevated window. This dialog could be something like
> What you requested couldn't be completed. Do you want to:
> A. Make me a new tab instead.
> B. Forget it and cancel. I'll go fix my config.
I'm certainly leaning towards proposal 2 - always create a new tab. This is how
it's implemented in [#8514]. In that PR, this seems to work sensibly.
-->
After discussing with the team, we have decided that the most sensible approach
for handling a cross-elevation `split-pane` is to just create a new tab in the
elevated window. The user can always re-attach the pane as a split with the
`move-pane` command once the new pane in the elevated window.
#### Configure the Terminal to _always_ run elevated
`elevate` is a per-profile property, not a global property. If a user
wants to always have all instances of the Terminal run elevated, they
could set `"elevate": true` in their profile defaults. That would cause _all_
profiles they launch to always spawn as elevated windows.
#### `elevate` in Actions
Additionally, we'll add the `elevate` property to the `NewTerminalArgs` used in
the `newTab`, `splitPane`, and `newWindow` actions. This is similar to how other
properties of profiles can be overridden at launch time. This will allow
windows, tabs and panes to all be created specifically as elevated windows.
In the `NewTerminalArgs`, `elevate` will be an optional boolean, with the
following behavior:
* `null` (_default_): Don't modify the `elevate` property for this profile
* `true`: This launch should act like the profile had `"elevate": true` in its
properties.
* `false`: This launch should act like the profile had `"elevate": false` in its
properties.
We'll also add an iterable command for opening a profile in an
elevated tab, with the following json:
```jsonc
{
// New elevated tab...
"name": { "key": "NewElevatedTabParentCommandName", "icon": "UAC-Shield.png" },
"commands": [
{
"iterateOn": "profiles",
"icon": "${profile.icon}",
"name": "${profile.name}",
"command": { "action": "newTab", "profile": "${profile.name}", "elevated": true }
}
]
},
```
#### Elevation from the dropdown
Currently, the new tab dropdown supports opening a new pane by
<kbd>Alt+click</kbd>ing on a profile. We could similarly add support to open a
tab elevated with <kbd>Ctrl+click</kbd>. This is similar to the behavior of the
Windows taskbar. It supports creating an elevated instance of a program by
<kbd>Ctrl+click</kbd>ing on entries as well.
## Implementation Details
### Starting an elevated process from an unelevated process
It seems that we're able to create an elevated process by passing the `"runas"`
verb to
[`ShellExecute`](https://docs.microsoft.com/en-us/windows/win32/api/shellapi/nf-shellapi-shellexecutea).
So we could use something like
```c++
ShellExecute(nullptr,
L"runas",
L"wt.exe",
L"-w new new-tab [args...]",
nullptr,
SW_SHOWNORMAL);
```
This will ask the shell to perform a UAC prompt before spawning `wt.exe` as an
elevated process.
> 👉 NOTE: This mechanism won't always work on non-Desktop SKUs of Windows. For
> more discussion, see [Elevation on OneCore SKUs](#Elevation-on-OneCore-SKUs).
## Potential Issues
<table>
<tr>
<td><strong>Accessibility</strong></td>
<td>
The set of changes proposed here are not expected to introduce any new
accessibility issues. Users can already create elevated Terminal windows. Making
it easier to create these windows doesn't really change our accessibility story.
</td>
</tr>
<tr>
<td><strong>Security</strong></td>
<td>
We won't be doing anything especially unique, so there aren't expected to be any
substantial security risks associated with these changes. Users can already
create elevated Terminal windows, so we're not really introducing any new
functionality, from a security perspective.
We're relying on the inherent security of the `runas` verb of `ShellExecute` to
prevent any sort of unexpected escalation-of-privilege.
<hr>
One security concern is the fact that the `settings.json` file is currently a
totally unsecured file. It's completely writable by any medium-IL process. That
means it's totally possible for a malicious program to change the file. The
malicious program could find a user's "Elevated PowerShell" profile, and change
the commandline to `malicious.exe`. The user might then think that their
"Elevated PowerShell" will run `powershell.exe` elevated, but will actually
auto-elevate this attacker.
If all we expose to the user is the name of the profile in the UAC dialog, then
there's no way for the user to be sure that the program that's about to be
launched is actually what they expect.
To help mitigate this, we should _always_ pass the evaluated `commandline` as a
part of the call to `ShellExecute`. the arguments that are passed to
`ShellExecute` are visible to the user, though they need to click the "More
Details" dropdown to reveal them.
We will need to mitigate this vulnerability regardless of adding support for the
auto-elevation of individual terminal tabs/panes. If a user is launching the
Terminal elevated (i.e. from the Win+X menu in Windows 11), then it's possible
for a malicious program to overwrite the `commandline` of their default profile.
The user may now unknowingly invoke this malicious program while thinking they
are simply launching the Terminal.
To deal with this more broadly, we will display a dialog within the Terminal
window before creating **any** elevated terminal instance. In that dialog, we'll
display the commandline that will be executed, so the user can very easily
confirm the commandline.
This will need to happen for all elevated terminal instances. For an elevated
Windows Terminal window, this means _all_ connections made by the Terminal.
Every time the user opens a new profile or a new commandline in a pane, we'll
need to prompt them first to confirm the commandline. This dialog within the
elevated window will also prevent an attacker from editing the `settings.json`
file while the user already has an elevated Terminal window open and hijacking a
profile.
The dialog options will certainly be annoying to users who don't want to be
taken out of their flow to confirm the commandline that they wish to launch.
There's precedent for a similar warning being implemented by VSCode, with their
[Workspace Trust] feature. They too faced a similar backlash when the feature
first shipped. However, in light of recent global cybersecurity attacks, this is
seen as an acceptable UX degradation in the name of application trust. We don't
want to provide an avenue that's too easy to abuse.
When the user confirms the commandline of this profile as something safe to run,
we'll add it to an elevated-only version of `state.json`. (see [#7972] for more
details). This elevated version of the file will only be accessible by the
elevated Terminal, so an attacker cannot hijack the contents of the file. This
will help mitigate the UX discomfort caused by prompting on every commandline
launched. This should mean that the discomfort is only limited to the first
elevated launch of a particular profile. Subsequent launches (without modifying
the `commandline`) will work as they always have.
The dialog for confirming these commandlines should have a link to the docs for
"Learn more...". Transparency in the face of this dialog should
mitigate some dissatisfaction.
The dialog will _not_ appear if the user does not have a split token - if the
user's PC does not have UAC enabled, then they're _already_ running as an
Administrator. Everything they do is elevated, so they shouldn't be prompted in
this way.
The Settings UI should also expose a way of viewing and removing these cached
entries. This page should only be populated in the elevated version of the
Terminal.
</td>
</tr>
<tr>
<td><strong>Reliability</strong></td>
<td>
No changes to our reliability are expected as a part of this change.
</td>
</tr>
<tr>
<td><strong>Compatibility</strong></td>
<td>
There are no serious compatibility concerns expected with this changelist. The
new `elevate` property will be unset by default, so users will heed to opt-in
to the new auto-elevating behavior.
There is one minor concern regarding introducing the UAC shield on the window.
We're planning on using themes to configure the appearance of the shield. That
means we'll need to ship themes before the user will be able to hide the shield
again.
</td>
</tr>
<tr>
<td><strong>Performance, Power, and Efficiency</strong></td>
<td>
No changes to our performance are expected as a part of this change.
</td>
</tr>
</table>
### Centennial Applications
In the past, we've had a notoriously rough time with the Centennial app
infrastructure and running the Terminal elevated. Notably, we've had to list all
our WinRT classes in our SxS manifest so they could be activated using
unpackaged WinRT while running elevated. Additionally, there are plenty of
issues running the Terminal in an "over the shoulder" elevation (OTS) scenario.
Specifically, we're concerned with the following scenario:
* the current user account has the Terminal installed,
* but they aren't an Administrator,
* the Administrator account doesn't have the Terminal installed.
In that scenario, the user can run into issues launching the Terminal in an
elevated context (even after entering the Admin's credentials in the UAC
prompt).
This spec proposes no new mitigations for dealing with these issues. It may in
fact make them more prevalent, by making elevated contexts more easily
accessible.
Unfortunately, these issues are OS bugs that are largely out of our own control.
We will continue to apply pressure to the centennial app team internally as we
encounter these issues. They are are team best equipped to resolve these issues.
### Default Terminal & auto-elevation
In the future, when we support setting the Terminal as the "default terminal
emulator" on Windows. When that lands, we will use the `profiles.defaults`
settings to create the tab where we'll be hosting the commandline client. If the user has
`"elevate": true` in their `profiles.defaults`, we'd usually try to
auto-elevate the profile. In this scenario, however, we can't do that. The
Terminal is being invoked on behalf of the client app launching, instead of the
Terminal invoking the client application.
**2021-08-17 edit**: Now that "defterm" has shipped, we're a little more aware
of some of the limitations with packaged COM and elevation boundaries. Defterm
cannot be used with elevated processes _at all_ currently (see [#10276]). When
an elevated commandline application is launched, it will always just appear in
`conhost.exe`. Furthermore, An unelevated peasant can't communicate with an
elevated monarch so we can't toss the connection to the elevated monarch and
have them handle it.
The simplest solution here is to just _always_ ignore the `elevate` property for
incoming defterm connections. This is not an ideal solution, and one that we're
willing to revisit if/when [#10276] is ever fixed.
### Elevation on OneCore SKUs
This spec proposes using `ShellExecute` to elevate the Terminal window. However,
not all Windows SKUs have support for `ShellExecute`. Notably, the non-Desktop
SKUs, which are often referred to as "OneCore" SKUs. On these platforms, we
won't be able to use `ShellExecute` to elevate the Terminal. There might not
even be the concept of multiple elevation levels, or different users, depending
on the SKU.
Fortunately, this is a mostly hypothetical concern for the moment. Desktop is
the only publicly supported SKU for the Terminal currently. If the Terminal ever
does become available on those SKUs, we can use these proposals as mitigations.
* If elevation is supported, there must be some other way of elevating a
process. We could always use that mechanism instead.
* If elevation isn't supported (I'm thinking 10X is one of these), then we could
instead display a warning dialog whenever a user tries to open an elevated
profile.
- We could take the warning a step further. We could add another settings
validation step. This would warn the user if they try to mark any profiles
or actions as `"elevate":true`
## Future considerations
* If we wanted to go even further down the visual differentiation route, we
could consider allowing the user to set an entirely different theme ([#3327])
based on the elevation state. Something like `elevatedTheme`, to pick another
theme from the set of themes. This would allow them to force elevated windows
to have a red titlebar, for example.
* Over the course of discussion concerning appearance objects ([#8345]), it
became clear that having separate "elevated" appearances defined for
`profile`s was overly complicated. This is left as a consideration for a
possible future extension that could handle this scenario in a cleaner way.
* Similarly, we're going to leave [#3637] "different profiles when elevated vs
unelevated" for the future. This also plays into the design of "configure the
new tab dropdown" ([#1571]), and reconciling those two designs is out-of-scope
for this particular release.
* Tangentially, we may want to have a separate Terminal icon we ship with the
UAC shield present on it. This would be especially useful for the tray icon.
Since there will be different tray icon instances for elevated and unelevated
windows, having unique icons may help users identify which is which.
### De-elevating a Terminal
the original version of this spec proposed that `"elevated":false` from an
elevated Terminal window should create a new unelevated Terminal instance. The
mechanism for doing this is described in [The Old New Thing: How can I launch an
unelevated process from my elevated process, redux].
This works well when the Terminal is running unpackaged. However, de-elevating a
process does not play well with packaged centennial applications. When asking
the OS to run the packaged application from an elevated context, the system will
still create the child process _elevated_. This means the packaged version of
the Terminal won't be able to create a new unelevated Terminal instance.
From an internal mail thread:
> App model intercepts the `CreateProcess` call and redirects it to a COM
> service. The parent of a packaged app is not the launching app, its some COM
> service. So none of the parent process nonsense will work because the
> parameters you passed to `CreateProcess` arent being used to create the
> process.
If this is fixed in the future, we could theoretically re-introduce de-elevating
a profile. The original spec proposed a `"elevated": bool?` setting, with the
following behaviors:
* `null` (_default_): Don't modify the elevation level when running this profile
* `true`: If the current window is unelevated, try to create a new elevated
window to host this connection.
* `false`: If the current window is elevated, try to create a new unelevated
window to host this connection.
We could always re-introduce this setting, to supercede `elevate`.
### Change profile appearance for elevated windows
In [#3062] and [#8345], we're planning on allowing users to set different
appearances for a profile whether it's focused or not. We could do similar thing
to enable a profile to have a different appearance when elevated. In the
simplest case, this could allow the user to set `"background": "#ff0000"`. This
would make a profile always appear to have a red background when in an elevated
window.
The more specific details of this implementation are left to the spec
[Configuration object for profiles].
In discussion of that spec, we decided that it would be far too complicated to
try and overload the `unfocusedAppearance` machinery for differentiating between
elevated and unelevated versions of the same profile. Already, that would lead
to 4 states: [`appearance`, `unfocusedAppearance`, `elevatedAppearance`,
`elevatedUnfocusedAppearance`]. This would lead to a combinatorial explosion if
we decided in the future that there should also be other states for a profile.
This particular QoL improvement is currently being left as a future
consideration, should someone come up with a clever way of defining
elevated-specific settings.
<!--
Brainstorming notes for future readers:
You could have a profile that layers on an existing profile, with elevated-specific settings:
{
"name": "foo",
"background": "#0000ff",
"commandline": "cmd.exe /k echo I am unelevated"
},
{
"inheritsFrom": "foo",
"background": "#ff0000",
"elevate": true,
"commandline": "cmd.exe /k echo I am ELEVATED"
}
-->
<!-- Footnotes -->
[#632]: https://github.com/microsoft/terminal/issues/632
[#1032]: https://github.com/microsoft/terminal/issues/1032
[#1571]: https://github.com/microsoft/terminal/issues/1571
[#1939]: https://github.com/microsoft/terminal/issues/1939
[#3062]: https://github.com/microsoft/terminal/issues/3062
[#3327]: https://github.com/microsoft/terminal/issues/3327
[#3637]: https://github.com/microsoft/terminal/issues/3637
[#4472]: https://github.com/microsoft/terminal/issues/4472
[#5000]: https://github.com/microsoft/terminal/issues/5000
[#7972]: https://github.com/microsoft/terminal/pull/7972
[#8311]: https://github.com/microsoft/terminal/issues/8311
[#8345]: https://github.com/microsoft/terminal/issues/8345
[#8514]: https://github.com/microsoft/terminal/issues/8514
[#10276]: https://github.com/microsoft/terminal/issues/10276
[Process Model 2.0 Spec]: https://github.com/microsoft/terminal/blob/main/doc/specs/%235000%20-%20Process%20Model%202.0.md
[Configuration object for profiles]: https://github.com/microsoft/terminal/blob/main/doc/specs/Configuration%20object%20for%20profiles.md
[Session Management Spec]: https://github.com/microsoft/terminal/blob/main/doc/specs/%234472%20-%20Windows%20Terminal%20Session%20Management.md
[The Old New Thing: How can I launch an unelevated process from my elevated process, redux]: https://devblogs.microsoft.com/oldnewthing/20190425-00/?p=102443
[Workspace Trust]: https://code.visualstudio.com/docs/editor/workspace-trust

View File

@@ -1,562 +0,0 @@
---
author: Mike Griese @zadjii-msft
created on: 2020-10-30
last updated: 2020-02-05
issue id: #4472
---
# Windows Terminal Session Management
## Abstract
This document is intended to serve as an addition to the [Process Model 2.0
Spec]. That document provides a big-picture overview of changes to the entirety
of the Windows Terminal process architecture, including both the split of
window/content processes, as well as the introduction of monarch/peasant
processes. The focus of that document was to identify solutions to a set of
scenarios that were closely intertwined, and establish these solutions would
work together, without preventing any one scenario from working. What that
document did not do was prescribe specific solutions to the given scenarios.
This document offers a deeper dive on a subset of the issues in [#5000], to
describe specifics for managing multiple windows with the Windows Terminal. This
includes features such as:
* Run `wt` in the current window ([#4472])
* Single Instance Mode ([#2227])
## Solution Design
### Monarch and Peasant Processes
This document assumes the reader is already familiar with the "Monarch and
Peasant" architecture as detailed in the [Windows Terminal Process Model 2.0
Spec]. As a quick summary:
* Every Windows Terminal window is a "Peasant" process.
* One of the Windows Terminal window processes is also the "Monarch" process.
The Monarch is picked randomly from the Terminal windows, and there is only
ever one Monarch process at a time.
* Peasants can communicate with the monarch when certain state changes (such as
their window being activated), and the monarch can send commands to any of the
peasants.
This architecture will be used to enable each of the following scenarios.
### Scenario: Open new tabs in most recently used window
A common feature of many browsers is that when a web URL is clicked somewhere,
the web page is opened as a new tab in the most recently used window of the
browser. This functionality is often referred to as "glomming", as the new tab
"gloms" onto the existing window.
Currently, the terminal does not support such a feature - every `wt` invocation
creates a new window. With the monarch/peasant architecture, it'll now be
possible to enable such a scenario.
As each window is activated, it will call a method on the `Monarch` object
(hosted by the monarch process) which will indicate that "I am peasant N, and
I've been focused". The monarch will use those method calls to update its own
internal stack of the most recently used windows.
Whenever a new `wt.exe` process is launched, that process will _first_ ask the
monarch if it should run the commandline in an existing window, or create its
own window.
![auto-glom-wt-exe](auto-glom-wt-exe.png)
If glomming is enabled, the monarch will dispatch the commandline to the
appropriate window for them to handle instead. To the user, it'll seem as if the
tab just opened in the most recent window.
Users should certainly be able to specify if they want new instances to glom
onto the MRU window or not. You could imagine that currently, we default to the
hypothetical value `"windowingBehavior": "useNew"`, meaning that each new wt gets
its own new window.
If glomming is disabled, then the Monarch will call back to the peasant and tell
it to run the provided commandline. The monarch will use the return value of
`ExecuteCommandline` to indicate that the calling process should create a window
and become a peasant process, and run the commandline itself.
#### Glomming within the same virtual desktop
When links are opened in the new Edge browser, they will only glom onto an
existing window if that window is open in the current virtual desktop. This
seems like a good idea of a feature for the Terminal to follow as well.
There must be some way for an application to determine which virtual desktop it
is open on. We could use that information to have the monarch track the last
active window _per-desktop_, and only glom when there's one on the current
desktop.
We could make the `windowingBehavior` property accept a variety of
configurations:
- `"useExisting"`: always glom to the most recent window, regardless of desktop.
- `"useExistingOnSameDesktop"`: Only glom if there's an existing window on this
virtual desktop, otherwise create a new window. This will be the new default
value.
- `"useNew"`: Never glom, always create a new window. This is technically the
current behavior of the Terminal.
### Handling the current working directory
Consider the following scenario: the user runs `wt -d .` in the address bar of
explorer, and the monarch determines that this new tab should be created in an
existing window. For clarity during this example, we will label the existing
window WT[1], and the second `wt.exe` process WT[2].
An example of this scenario is given in the following diagram:
![single-instance-mode-cwd](single-instance-mode-cwd.png)
In this scenario, we want the new tab to be spawned in the current working
directory of WT[2], not WT[1]. So when WT[1] is about to run the commands that
were passed to WT[2], WT[1] will need to:
* First, stash its own CWD
* Change to the CWD of WT[2]
* Run the commands from WT[2]
* Then return to its original CWD.
So, as a part of the interface that a peasant uses to communicate the startup
commandline to the monarch, we should also include the current working
directory.
### Scenario: Run `wt` in the current window
One often requested scenario is the ability to run a `wt.exe` commandline in the
current window, as opposed to always creating a new window. Presume we have the
ability to communicate between different window processes. The logical extension
of this scenario would be "run a `wt` commandline in _any_ given WT window".
Each window process will have its own unique ID assigned to it by the monarch.
This ID will be a positive number. Windows can also have names assigned to them.
These names are strings that the user specifies. A window will always have an
ID, but not necessarily a name. Running a command in a given window with ID N
should be as easy as something like:
```sh
wt.exe --window N new-tab ; split-pane
```
(or for shorthand, `wt -w N new-tab ; split-pane`).
More formally, we will add the following parameter to the top-level `wt`
command:
#### `--window,-w <window-id>`
Run these commands in the given Windows Terminal session. This enables opening
new tabs, splits, etc. in already running Windows Terminal windows.
* If `window-id` is `0`, run the given commands in _the current window_.
* If `window-id` is a negative number, or the reserved name `new`, run the
commands in a _new_ Terminal window.
* If `window-id` is the ID or name of an existing window, then run the
commandline in that window.
* If `window-id` is _not_ the ID or name of an existing window, create a new
window. That window will be assigned the ID or name provided in the
commandline. The provided subcommands will be run in that new window.
* If `window-id` is omitted, then obey the value of `windowingBehavior` when
determining which window to run the command in.
_Whenever_ `wt.exe` is started, it must _always_ pass the provided commandline
first to the monarch process for handling. This is important for glomming
scenarios (as noted above). The monarch will parse the commandline, determine
which window the commandline is destined for, then call `ExecuteCommandline` on
that peasant, who will then run the command.
#### Running commands in the current window:`wt --window 0`
If `wt -w 0 <commands>` is run _outside_ a WT instance, it could attempt to glom
onto _the most recent WT window_ instead. This seems more logical than something
like `wt --window last` or some other special value indicating "run this in the
MRU window".<sup>[[2]](#footnote-2)</sup>
That might be a simple, but **wrong**, implementation for "the current window".
If the peasants always raise an event when their window is focused, and the
monarch keeps track of the MRU order for peasants, then one could naively assume
that the execution of `wt -w 0 <commands>` would always return the window the
user was typing in, the current one. However, if someone were to do something
like `sleep 10 ; wt -w 0 <commands>`, then the user could easily focus another
WT window during the sleep, which would cause the MRU window to not be the same
as the window executing the command.
To solve this issue, we'll other than
attempting to use the `WT_SESSION` environment variable. If a `wt.exe` process
is spawned and that's in its environment variables, it could try and ask the
monarch for the peasant who's hosting the session corresponding to that GUID.
This is more of a theoretical solution than anything else.
In the past we've been reluctant to rely too heavily on `WT_SESSION`. However,
an environment variable does seem to be the only reliable way to be confident
where the window was created from. We could introduce another environment
variable instead - `WT_WINDOW_ID`. That would allow us to shortcut the session
ID lookup. However, I worry about exposing the window ID as an environment
variable. If we do that, users will inevitably use that instead of the `wt -0`
alias, which should take care of the work for them. Additionally, `WT_WINDOW_ID`
wouldn't update in the child processes as tabs are torn out of windows to create
new windows.
Both solutions are prone to the user changing the value of the variable to some
garbage value. If they do that, this lookup will most certainly not work as
expected. Using the session ID (a GUID) instead of the window ID (an int) makes
it less likely that they guess the ID of an existing instance.
#### Running commands in a new window:`wt --window -1` / `wt --window new`
If the user passes a negative number, or the reserved name `new` to the
`--window` parameter, then we will always create a new window for that
commandline, regardless of the value of `windowingBehavior`. This will allow
users to do something like `wt -w -1 new-tab` to _always_ create a new window.
#### `--window` in subcommands
The `--window` parameter is a setting to `wt.exe` itself, not to one of its
subcommands (like `new-tab` or `split-pane`). This means that all of the
subcommands in a particular `wt` commandline will all be handled by the same
session. For example, let us consider a user who wants to open a new tab in
window 2, and split a new pane in window 3, all at once. The user _cannot_ do
something like:
```cmd
wt -w 2 new-tab ; -w 3 split-pane
```
Instead, the user will need to separate the commands (by whatever their shell's
own command delimiter is) and run two different `wt.exe` instances:
```cmd
wt -w 2 new-tab & wt -w 3 split-pane
```
This is done to make the parsing of the subcommands easier, and for the internal
passing of arguments simpler. If the `--window` parameter were a part of each
subcommand, then each individual subcommand's parser would need to be
enlightened about that parameter, and then it would need to be possible for any
single part of the commandline to call out to another process. It would be
especially tricky then to coordinate the work being done across process here.
The source process would need some sort of way to wait for the other process to
notify the source that a particular subcommand completed, before allowing the
source to dispatch the next part of the commandline.
Overall, this is seen as unnecessarily complex, and dispatching whole sets of
commands as a simpler solution.
### Naming Windows
It's not user-friendly to rely on automatically generated, invisible numbers to
identify windows. There's not a great way of identifying which window is which.
The user would need to track the IDs in their head manually. Instead, we'll
allow the user to provide a string name for the window. This name can be used to
address a window in addition to the ID.
Names can be provided on the commandline, in the original commandline. For
example, `wt -w foo nt` would name the new window "foo". Names can also be set
with a new action, `NameWindow`<sup>[[3]](#footnote-3)</sup>. `name-window`
could also be used as a subcommand. For example, `wt -w 4 name-window bar` would
name window 4 "bar".
To keep identities mentally distinct, we will disallow names that are integers
(positive or negative). This will prevent users from renaming a window to `2`,
then having `wt -w 2` be ambiguous as to which window it refers to.
Names must also be unique. If a user attempts to set the name of the window to
an already-used name, we'll need to ignore the name change. We could also
display a "toast" or some other type of low-impact message to the user. That
message would have some text like: "Unable to rename window. Another window with
that name already exists".
The Terminal will reserve the name `new`. It will also reserve any names
starting with the character `_`. The user will not be allowed to set the window
name to any of these reserved names. Reserving `_*` allows us to add other
keywords in the future, without introducing a breaking change.
## UI/UX Design
### `windowingBehavior` details
The following list gives greater breakdown of the values of `windowingBehavior`,
and how they operate:
* `"windowingBehavior": "useExisting", "useExistingOnSameDesktop"`:
**Browser-like glomming**
- New instances open in the current window by default.
- `newWindow` opens a new window.
- Tabs can be torn out to create new windows.
- `wt -w -1` opens a new window.
* `"windowingBehavior": "useNew"`: No auto-glomming. This is **the current
behavior** of the Terminal.
- New instances open in new windows by default
- `newWindow` opens a new window
- Tabs can be torn out to create new windows.
- `wt -w -1` opens a new window.
We'll be changing the default behavior from `useNew` to
`useExistingOnSameDesktop`. This will be more consistent with other tabbed
applications.
## Concerns
<table>
<tr>
<td><strong>Accessibility</strong></td>
<td>
There is no expected accessibility impact from this feature. Each window will
handle UIA access as it normally does.
In the future, we could consider exposing the window IDs and/or names via UIA.
</td>
</tr>
<tr>
<td><strong>Security</strong></td>
<td>
Many security concerns have already be covered in greater detail in the parent
spec, [Process Model 2.0 Spec].
When attempting to instantiate the Monarch, COM will only return the object from
a server running at the same elevation level. We don't need to worry about
unelevated peasants connecting to the elevated Monarch, or vice-versa.
</td>
</tr>
<tr>
<td><strong>Reliability</strong></td>
<td>
We will need to be careful when working with objects hosted by another process.
Any work we do with it MUST be in a try/catch, because at _any_ time, the other
process could be killed. At any point, a window process could be killed. Both
the monarch and peasant code will need to be redundant to such a scenario, and
if the other process is killed, make sure to display an appropriate error and
either recover or exit gracefully.
In any and all these situations, we will want to try and be as verbose as
possible in the logging. This will make tracking which process had the error
occur easier.
</td>
</tr>
<tr>
<td><strong>Compatibility</strong></td>
<td>
We will be changing the default behavior of the Terminal to auto-glom to the
most-recently used window on the same desktop in the course of this work, which
will be a breaking UX change. This is behavior that can be reverted with the
`"windowingBehavior": "useNew"` setting.
We acknowledge that this is a pretty massive change to the default experience of
the Terminal. We're planning on doing some polling of users to determine which
behavior they want by default. Additionally, we'll be staging the rollout of
this feature, using the Preview builds of the Terminal. The release notes that
first include it will call extra attention to this feature. We'll ask that users
provide their feedback in a dedicated thread, so we have time to collect
opinions from users before rolling the change out to all users.
We may choose to only change the default to `useExistingOnSameDesktop` once tab
tear out is available, so users who are particularly unhappy about this change
can still tear out the tab (if they can't be bothered to change the setting).
</td>
</tr>
<tr>
<td><strong>Performance, Power, and Efficiency</strong></td>
<td>
There's no dramatic change expected here. There may be a minor delay in the
spawning of new terminal instances, due to requiring cross-process hops for the
communication between monarch and peasant processes.
</td>
</tr>
</table>
## Potential Issues
### Mixed Elevation Levels
As of December 2020, we're no longer pursuing a "mixed-elevation" scenario for
the Terminal. This makes many of the cross-elevation scenarios simpler. Elevated
and unelevated `wt` instances will always remain separate. The different
elevation levels will maintain separate lists of window IDs. If the user is
running both an elevated and unelevated window, then there will be two monarchs.
One elevated, and the other unelevated.
There will also be some edge cases when handling the commandline that will need
special care. Say the user wanted to open a new tab in the elevated window, from
and unelevated `explorer.exe`. That would be a commandline like:
```sh
wt -w 0 new-tab -d . --elevated
```
Typically we first determine which window the commandline is intended for, then
dispatch it to that window. In this case, the `-w 0` will cause us to pass the
commandline to the current unelevated window. Then, that window will try to open
an elevated tab, fail, and create a new `wt.exe` process. This second `wt.exe`
process will lose the `-w 0` context. It won't inform the elevated monarch that
this commandline should be run in the active session.
We will need to make sure that special care is taken when creating elevated
instances that we maintain the `--window` parameter passed to the Terminal.
### `wt` Startup Commandline Options
There are a few commandline options which can be provided to `wt.exe` which
don't make sense to pass to another session. These options include (but are not
limited to):
* `--initialSize r,c`
* `--initialPosition x,y`
* `--fullscreen`, `--maximized`, etc.
When we're passing a commandline to another instance to handle, these arguments
will be ignored. they only apply to the initial creation of a window.
`--initialSize 32, 120` doesn't make sense if the window already has a size.
On startup of a new window, we currently assume that the first command is always
`new-tab`. When passing commandlines to existing windows, we won't need to make
that assumption anymore. There will already be existing tabs.
### Monarch MRU Window Tracking
As stated above, the monarch is responsible for tracking the MRU window stack.
However, when the monarch is closed, this state will be lost. The new monarch
will be elected, but it will be unable to ask the old monarch for the MRU
order of the windows.
We had previously considered an _acceptable_ UX when this would occur. We would
randomize the order (with the new monarch becoming the MRU window). If someone
noticed this bug and complained, then we had a theoretical solution prepared.
The peasants could inform not only the monarch, but _all other peasants_ when
they become activated. This would mean all peasants are simultaneously tracking
the MRU stack. This would mean that any given peasant would be prepared always
to become the monarch.
A simpler solution though would be to not track the MRU stack in the Monarch at
all. Instead, each peasant could just track internally when they were last
activated. The Monarch wouldn't track any state itself. It would be distributed
across all the peasants. The Monarch could then iterate over the list of
peasants and find the one with the newest `LastActivated` timestamp.
Now, when a Monarch dies, the new Peasant doesn't have to come up with the stack
itself. All the other Peasants keep their state. The new Monarch can query them
and get the same answer the old Monarch would have.
We could further optimize this by having the Monarch also track the stack. Then,
the monarch could query the MRU window quickly. The `LastActivated` timestamps
would only be used by a new Monarch when it is elected, to reconstruct the MRU
stack.
## Implementation Plan
This is a list of actionable tasks generated as described by this spec:
* [ ] Add support for `wt.exe` processes to be Monarchs and Peasants, and
communicate that state between themselves. This task does not otherwise add
any user-facing features, merely an architectural update.
* [ ] Add support for the `windowingBehavior` setting as a boolean. Opening new
WT windows will conditionally glom to existing windows.
* [ ] Add support for per-desktop `windowingBehavior`, by adding the support for
the enum values `"useExisting"`, `"useExistingOnSameDesktop"` and `"useNew"`.
* [ ] Add support for `wt.exe` to pass commandlines intended for another window
to the monarch, then to the intended window, with the `--window,-w
window-id` commandline parameter.
* [ ] Add support for targeting and naming windows via the `-w` parameter on the
commandline
* [ ] Add a `NameWindow` action, subcommand that allows the user to set the name
for the window.
* [ ] Add an action that will cause all windows to briefly display a overlay
with the current window ID and name. This would be something like the
"identify" feature of the Windows "Display" settings.
## Future considerations
* What if the user wanted to pipe a command to a pane in an existing window?
```sh
man ping > wt -w 0 split-pane cat
```
Is there some way for WT to pass its stdin/out handles to the child process
it's creating? This is _not_ related to the current spec at hand, just
something the author considered while writing the spec. This likely belongs
over in [#492], or in its own spec.
- Or I suppose, with less confusion, someone could run `wt -w 0 split-pane --
man ping > cat`. That's certainly more sensible, and wouldn't require any
extra work.
* "Single Instance Mode" is a scenario in which there is only ever one single WT
window. A user might want this functionality to only ever allow a single
terminal window to be open on their desktop. This is especially frequently
requested in combination with "quake mode", as discussed in [#653]. When Single
Instance Mode is active, and the user runs a new `wt.exe` commandline, it will
always end up running in the existing window, if there is one.
An earlier version of this spec proposed a new value of `glomToLastWindow`.
(`glomToLastWindow` was later renamed `windowingBehavior`). The `always` value
would disable tab tear out<sup>[[1]](#footnote-1)</sup>. It would additionally
disable the `newWindow` action, and prevent `wt -w new` from opening a new
window.
In discussion, it was concluded that this setting didn't make sense. Why did the
`glomToLastWindow` setting change the behavior of tear out? Single Instance Mode
is most frequently requested in regards to quake mode. We're leaving the
implementation of true single instance mode to that spec.
* It was suggested in review that we could auto-generate names for windows, from
some list of words. Prior art could be the URLS for gfycat.com or
what3words.com, which use three random words. I believe `docker` also assigns
names from a random selection of `adjective`+`name`. This is an interesting
idea, and something that could be pursued in the future.
- This would be a massive pain to localize though, hence why this is left as
a future consideration.
* We will _need_ to provide a commandline tool to list windows and their IDs &
names. We're thinking a list of windows, their IDs, names, PIDs, and the title
of the window.
Currently we're stuck with `wt.exe` which is a GUI application, and cannot
print to the console. Our need is now fairly high for the ability to print
info to the console. To remedy this, we'll need to ship another helper exe as
a commandline tool for working with the terminal. The design for this is left
for the future.
## Footnotes
<a name="footnote-1"><a>[1]: While tear-out is a separate track of work from
session management in general, this setting could be implemented along with this
set of features, and later used to control tear out as well.
<a name="footnote-2"><a>[2]: Since we're reserving the keyword `new` to mean "a
new window", then we could also reserve `last` or `current` as an alias for "the
current window".
<a name="footnote-3"><a>[3]: We currently have two actions for renaming _tabs_
in the Terminal: `renameTab(name)`, and `openTabRenamer()`. We will likely
similarly need `nameWindow(name)` and `openWindowNamer()`. `openWindowNamer`
could display a dialog to allow the user to rename the current window at
runtime.
## Resources
* [Tab Tear-out in the community toolkit] - this document proved invaluable to
the background of tearing a tab out of an application to create a new window.
<!-- Footnotes -->
[#5000]: https://github.com/microsoft/terminal/issues/5000
[#1256]: https://github.com/microsoft/terminal/issues/1256
[#4472]: https://github.com/microsoft/terminal/issues/4472
[#2227]: https://github.com/microsoft/terminal/issues/2227
[#653]: https://github.com/microsoft/terminal/issues/653
[#1032]: https://github.com/microsoft/terminal/issues/1032
[#632]: https://github.com/microsoft/terminal/issues/632
[#492]: https://github.com/microsoft/terminal/issues/492
[#4000]: https://github.com/microsoft/terminal/issues/4000
[#7972]: https://github.com/microsoft/terminal/pull/7972
[#961]: https://github.com/microsoft/terminal/issues/961
[`30b8335`]: https://github.com/microsoft/terminal/commit/30b833547928d6dcbf88d49df0dbd5b3f6a7c879
[Tab Tear-out in the community toolkit]: https://github.com/windows-toolkit/Sample-TabView-TearOff
[Quake mode scenarios]: https://github.com/microsoft/terminal/issues/653#issuecomment-661370107
[`ISwapChainPanelNative2::SetSwapChainHandle`]: https://docs.microsoft.com/en-us/windows/win32/api/windows.ui.xaml.media.dxinterop/nf-windows-ui-xaml-media-dxinterop-iswapchainpanelnative2-setswapchainhandle
[Process Model 2.0 Spec]: https://github.com/microsoft/terminal/blob/main/doc/specs/%235000%20-%20Process%20Model%202.0/%235000%20-%20Process%20Model%202.0.md

Binary file not shown.

Before

Width:  |  Height:  |  Size: 565 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 83 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 54 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 25 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 37 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 52 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 44 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 53 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 46 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 93 KiB

View File

@@ -1,95 +0,0 @@
---
author: Kayla Cinnamon @cinnamon-msft
created on: 2020-01-03
last updated: 2020-01-03
issue id: 597
---
# Tab Sizing
## Abstract
This spec outlines the tab sizing feature. This is an application-level feature that is not profile-specific (at least for now).
Global properties that encompass tab sizing:
* `tabWidthMode` (accepts pre-defined values for tab sizing behavior)
* `tabWidthMin` (can never be smaller than the icon width)
* `tabWidthMax` (can never be wider than the tab bar)
Acceptable values for `tabWidthMode`:
* [default] `equal` (all tabs are sized the same, regardless of tab title length)
* `titleLength` (width of tab contains entire tab title)
## Inspiration
Other browsers and terminals have varying tab width behavior, so we should give people options.
## Solution Design
`tabWidthMode` will be a global setting that will accept the following strings:
* `equal`
* All tabs are equal in width
* If the tab bar has filled, tabs will shrink as additional tabs are added
* Utilizes the `equal` setting from WinUI's TabView
* `titleLength`
* Tab width varies depending on title length
* Width of tab will fit the whole tab title
* Utilizes the `sizeToContent` setting from WinUI's TabView
In addition to `tabWidthMode`, the following global properties will also be available:
* `tabWidthMin`
* Accepts an integer
* Value correlates to the minimum amount of pixels the tab width can be
* If value is less than the width of the icon, the minimum width will be the width of the icon
* If value is greater than the width of the tab bar, the maximum width will be the width of the tab bar
* If not set, the tab will have the system-defined minimum width
* `tabWidthMax`
* Accepts an integer
* Value correlates to the maximum amount of pixels the tab width can be
* If value is less than the width of the icon, the minimum width will be the width of the icon
* If value is greater than the width of the tab bar, the maximum width will be the width of the tab bar
* If not set, the tab will have the system-defined maximum width
If `tabWidthMode` is set to `titleLength`, the tab widths will fall between the `tabWidthMin` and `tabWidthMax` values if they are set, depending on the length of the tab title.
If `tabWidthMode` isn't set, the default experience will be `equal`. Justification for the default experience is the results from this [twitter poll](https://twitter.com/cinnamon_msft/status/1203093459055210496).
## UI/UX Design
[This tweet](https://twitter.com/cinnamon_msft/status/1203094776117022720) displays how the `equal` and `titleLength` values behave for the `tabWidthMode` property.
## Capabilities
### Accessibility
This feature could impact accessibility if the tab title isn't stored within the metadata of the tab. If the tab width is the width of the icon, then the title isn't visible. The tab title will have to be accessible by a screen reader.
### Security
This feature will not impact security.
### Reliability
This feature will not impact reliability. It provides users with additional customization options.
### Compatibility
This feature will not break existing compatibility.
### Performance, Power, and Efficiency
## Potential Issues
This feature will not impact performance, power, nor efficiency.
## Future considerations
* Provide tab sizing options per-profile
* A `tabWidthMode` value that will evenly divide the entirety of the tab bar by the number of open tabs
* i.e. One tab will take the full width of the tab bar, two tabs will each take up half the width of the tab bar, etc.

View File

@@ -123,4 +123,4 @@ This open issue tracks the phase features of Search: https://github.com/microsof
## Resources
GitHub Issue: https://github.com/microsoft/terminal/issues/605
Github Issue: https://github.com/microsoft/terminal/issues/605

View File

@@ -1,736 +0,0 @@
---
author: Mike Griese @zadjii-msft
created on: 2021-02-23
last updated: 2021-05-13
issue id: #653
---
# Quake Mode
## Abstract
Many existing terminals support a feature whereby a user can press a keybinding
anywhere in the OS, and summon their terminal application. Oftentimes the act of
summoning this window is accompanied by a "dropdown" animation, where the window
slides in to view from the top of the screen. This global summon action is often
referred to as "quake mode", a reference to the video game Quake, who's console
slid in from the top.
This spec addresses both of the following two issues:
* "Quake Mode" ([#653])
* "Minimize to tray" ([#5727])
Readers should make sure to have read the [Process Model 2.0 Spec], for
background on Monarch and Peasant processes.
## Background
### Inspiration
For an example of the original Quake console in action, take a look at the
following video (noisy video warning): [Quake 3 sample]. Additionally, plenty of
existing terminal emulators support similar functionality:
* **Tilda** allows the user to specify different keys to summon the window on
different monitors.
* **Guake** alternatively allows the user to either summon the terminal window to
a specific monitor, or whichever monitor the mouse is on. Guake only allows
one single instance, so pressing the global hotkey always summons the same
instance.
### User Stories
The original quake mode thread (#653) is absolutely _filled_ with variations on
how users want to be able to summon their terminal windows. These include, but
are not limited to:
* **Story A** Press a hotkey anywhere to activate the single Terminal window
wherever it was
* **Story B** Press a hotkey anywhere to activate the single Terminal window _on
the current monitor_. If it wasn't previously on that monitor, move it there.
* **Story C** When the Terminal is summoned using the hotkey, have it "slide in"
from the top. Similarly, slide out on deactivate.
* **Story D** <kbd>Ctrl+1</kbd> to activate the terminal on monitor 1,
<kbd>Ctrl+2</kbd> to activate the terminal on monitor 2.
* **Story E** Variable dropdown speed
* **Story F** Minimize to tray, press a hotkey to activate the terminal window
(#5727)
* **Story G** Terminal doesn't appear in alt+tab view, press a hotkey to
activate the single terminal window / the nearest terminal window (I'm not
sure this is distinct from the above)
## Solution Design
To implement this feature, we'll add the following settings:
* a new action, named `globalSummon`.
* a new global, app setting named `minimizeToTray`
* a new global, app setting named `alwaysShowTrayIcon`
* a new action, named `quakeMode`, and a specially named `_quake` window.
### `globalSummon` Action
The `globalSummon` action will be a keybinding the user can use to summon a
Terminal window from anywhere in the OS. Various arguments to the action will
specify which window is summoned, to where, and how the window should behave on
summon.
From a technical perspective, the action will work by using the
[`RegisterHotKey`] function. This API allows us to bind a particular hotkey with
the OS. Whenever that hotkey is pressed, our window message loop will receive a
`WM_HOTKEY`. We'll use the payload of that window message to lookup the action
arguments for that hotkey. Then we'll use those arguments to control which
window is invoked, where, and how the window behaves.
Since `RegisterHotKey` can only be used to register a hotkey _once_ with the OS,
we'll need to make sure it's only ever set up by the Monarch process. We know
that there will only ever be one Monarch for the Terminal at a time, so it's the
perfect process to have the responsibility of managing the global hotkey.
The Monarch will be responsible for calling `RegisterHotKey`, and processing the
`WM_HOTKEY` messages. It will then dispatch method calls to the appropriate
window to summon it. When a Monarch dies and a new process becomes the Monarch,
the new Monarch will re-register for the hotkeys.
#### Where in the settings?
Since users may want to bind multiple keys to summon different windows, we'll
need to allow the user to specify multiple keybindings simultaneously, each with
their own set of args.
We stick all the `globalSummon`s in the actions array, like they're any other
keybinding.
However, these are not keys that are handled by the TerminalApp layer itself.
These are keys that need to be registered with the OS. So while they will be in
the normal `KeyMap`, they will need to be retrieved from that object and
manually passed to the window layer.
> A previous iteration of this spec considered placing the `globalSummon`
> actions in their own top-level array of the settings file, separate from the
> keybindings. This is no longer being considered, because it would not work for
> the case where the user has something like:
> ```json
> { "keys": "ctrl+c", "command": { "action": "globalSummon", "monitor": 1 } },
> { "keys": "ctrl+v", "command": { "action": "copy" } },
> ```
#### Which window, and where?
When looking at the list of requested scenarios, there are lots of different
ways people would like to use the global summon action. Some want the most
recent window activated, always. Others want to have one window _per monitor_.
Some would like to move the window to where the user is currently interacting
with the PC, and others want to activate the window where it already exists.
Trying to properly express all these possible configurations is complex. The
settings should be unambiguous as to what will happen when you press the
keybinding.
I believe that in order to accurately support all the variations that people
might want, we'll need two properties in the `globalSummon` action. These
properties will specify _which_ window we're summoning, and _where_ to summon
the window. To try and satisfy all these scenarios, I'm proposing the following
two arguments to the `globalSummon` action:
```json
"monitor": "any"|"toCurrent"|"toMouse"|"onCurrent"|int,
"desktop": "any"|"toCurrent"|"onCurrent"
```
The way these settings can be combined is in a table below. As an overview:
* `monitor`: This controls the monitor that the window will be summoned from/to
- `"any"`: Summon the MRU window, regardless of which monitor it's currently
on.
- `"toCurrent"`/omitted: (_default_): Summon the MRU window **TO** the monitor
with the current **foreground** window.
- `"toMouse"`: Summon the MRU window **TO** the monitor where the **mouse**
cursor is.
- `"onCurrent"`: Summon the MRU window **ALREADY ON** the current monitor.
- `int`: Summon the MRU window for the given monitor (as identified by the
"Identify" displays feature in the OS settings)
* `desktop`: This controls how the terminal should interact with virtual desktops.
- `"any"`: Leave the window on whatever desktop it's already on - we'll switch
to that desktop as we activate the window.
- > NOTE: A previous version of this spec had this enum value as `null`.
This was changed to `"any"` for parity with the `monitor` property.
- `"toCurrent"`/omitted: (_default_): Move the window **to** the current
virtual desktop
- `"onCurrent"`: Only summon the window if it's **already on** the current
virtual desktop
Neither `desktop` nor `monitor` is a required parameter - if either is omitted,
the omitted property will default to `toCurrent`.
Together, these settings interact in the following ways:
<!-- This table is formatted for viewing as rendered HTML. It's too complicated
for pure markdown, sorry. -->
<table>
<tr>
<td></td>
<th colspan=3><code>"desktop"</code></th>
</tr>
<!-- ----------------------------------------------------------------------- -->
<tr>
<th><code>"monitor"</code></th>
<td><code>any</code><br><strong>Leave where it is</strong></td>
<td><code>"toCurrent"</code><br><strong>Move to current desktop</strong></td>
<td><code>"onCurrent"</code><br><strong>On current desktop only</strong></td>
</tr>
<!-- ----------------------------------------------------------------------- -->
<tr>
<td><code>"any"</code><br> Summon the MRU window</td>
<td>Go to the desktop the window is on (leave position alone)</td>
<td>Move the window to this desktop (leave position alone)</td>
<td>
If there isn't one on this desktop:
* create a new one (default position)
Else:
* activate the one on this desktop (don't move it)
</td>
</tr>
<!-- ----------------------------------------------------------------------- -->
<tr>
<td><code>"toCurrent"</code><br> Summon the MRU window TO the monitor with the foreground window</td>
<td>Go to the desktop the window is on, move to the monitor with the foreground window</td>
<td>Move the window to this desktop, move to the monitor with the foreground window</td>
<td>
If there isn't one on this desktop:
* create a new one (on the monitor with the foreground window)
Else:
* activate the one on this desktop, move to the monitor with the foreground window
</td>
</tr>
<!-- ----------------------------------------------------------------------- -->
<tr>
<td>
<code>"toMouse"</code>
<sup><a href="#footnote-2">[2]</a></sup> <br>
Summon the MRU window TO the monitor with the mouse</td>
<td>Go to the desktop the window is on, move to the monitor with the mouse</td>
<td>Move the window to this desktop, move to the monitor with the mouse</td>
<td>
If there isn't one on this desktop:
* create a new one (on the monitor with the mouse)
Else:
* activate the one on this desktop, move to the monitor with the mouse
</td>
</tr>
<!-- ----------------------------------------------------------------------- -->
<tr>
<td><code>"onCurrent"</code><br> Summon the MRU window for the current monitor</td>
<td>
If there is a window on this monitor on any desktop,
* Go to the desktop the window is on (leave position alone)
else
* Create a new window on this monitor & desktop
</td>
<td>
If there is a window on this monitor on any desktop,
* Move the window to this desktop (leave position alone)
else
* Create a new window on this monitor & desktop
</td>
<td>
If there isn't one on this desktop, (even if there is one on this monitor on
another desktop),
* create a new one on this monitor
Else if ( there is one on this desktop, not this monitor)
* create a new one on this monitor
Else (one on this desktop & monitor)
* Activate the one on this desktop (don't move)
</td>
</tr>
<!-- ----------------------------------------------------------------------- -->
<tr>
<td><code>int</code><br> Summon the MRU window for monitor N</td>
<td>
If there is a window on monitor N on any desktop,
* Go to the desktop the window is on (leave position alone)
else
* Create a new window on this monitor & desktop
</td>
<td>
If there is a window on monitor N on any desktop,
* Move the window to this desktop (leave position alone)
else
* Create a new window on this monitor & desktop
</td>
<td>
If there isn't one on this desktop, (even if there is one on monitor N on
another desktop),
* create a new one on monitor N
Else if ( there is one on this desktop, not monitor N)
* create a new one on monitor N
Else (one on this desktop & monitor N)
* Activate the one on this desktop (don't move)
</td>
</tr>
</table>
##### Stories, revisited
With the above settings, let's re-examine the original user stories, and see how
they fit into the above settings. (_Stories that are omitted aren't relevant to
the discussion of these settings_)
> When the `desktop` param is omitted below, that can be interpreted as "any
> `desktop` value will make sense here"
* **Story A** Press a hotkey anywhere to activate the single Terminal window
wherever it was
- This is `{ "monitor": "any", "desktop": "any" }`
* **Story B** Press a hotkey anywhere to activate the single Terminal window _on
the current monitor_. If it wasn't previously on that monitor, move it there.
- This is `{ "monitor": "toCurrent" }`
* **Story D** <kbd>Ctrl+1</kbd> to activate the terminal on monitor 1,
<kbd>Ctrl+2</kbd> to activate the terminal on monitor 2.
- This is `[ { "keys": "ctrl+1", monitor": 1 }, { "keys": "ctrl+2", monitor": 2 } ]`
As some additional examples:
```json
// Go to the MRU window, wherever it is
{ "keys": "win+1", "command":{ "action":"globalSummon", "monitor":"any", "desktop": "any" } },
// activate the MRU window, and move it to this desktop & this monitor
{ "keys": "win+2", "command":{ "action":"globalSummon", "monitor":"toCurrent", "desktop": "toCurrent" } },
// Since "toCurrent" & "toCurrent" are the default values, just placing a single
// entry here will bind the same behavior:
{ "keys": "win+2", "command": "globalSummon" },
// activate the MRU window on this desktop
{ "keys": "win+3", "command":{ "action":"globalSummon", "monitor":"any", "desktop": "onCurrent" } },
// Activate the MRU window on monitor 2 (from any desktop), and place it on the
// current desktop. If there isn't one on monitor 2, make a new one.
{ "keys": "win+4", "command":{ "action":"globalSummon", "monitor": 2, "desktop": "toCurrent" } },
// Activate the MRU window on monitor 3 (ONLY THIS desktop), or make a new one.
{ "keys": "win+5", "command":{ "action":"globalSummon", "monitor": 3, "desktop": "onCurrent" } },
// Activate the MRU window on this monitor (from any desktop), and place it on
// the current desktop. If there isn't one on this monitor, make a new one.
{ "keys": "win+6", "command":{ "action":"globalSummon", "monitor": "onCurrent", "desktop": "toCurrent" } },
```
#### Summoning a specific window
What if you want to press a keybinding to always summon a specific, named
window? This window might not be the most recent terminal window, nor one that
would be selected by the `monitor` and `desktop` selectors. You could name a
window "Epona", and press `win+e` to always summon the "Epona" window.
We'll add the following property to address this scenario
* `"window": string|int`
- When omitted (_default_): Use monitor and desktop to find the appropriate
MRU window to summon.
- When provided: Always summon the window who's name or ID matches the given
`window` value. If no such window exists, then create a new window with that
name/id.
When provided _with_ `monitor` and `desktop`, `window` behaves in the following
ways:
* `desktop`
- `"any"`: Go to the desktop the given window is already on.
- `"toCurrent"`: If the window is on another virtual desktop, then move it to
the currently active one.
- `"onCurrent"`: If the window is on another virtual desktop, then move it to
the currently active one.
* `monitor`
- `"any"`: Leave the window on the monitor it is already on.
- `"toCurrent"`: If the window is on another monitor, then move it to the
currently active one.
- `"onCurrent"`: If the window is on another monitor, then move it to the
currently active one.
- `<int>`: If the window is on another monitor, then move it to the specified
monitor.
> NOTE: You read that right, `onCurrent` and `toCurrent` both do the same thing
> when `window` is provided. They both already know which window to select, the
> context of moving to the "current" monitor is all that those parameters add.
#### Other properties
Some users would like the terminal to just appear when the global hotkey is
pressed. Others would like the true quake-like experience, where the terminal
window "slides-in" from the top of the monitor. Furthermore, some users would
like to configure the speed at which that dropdown happens. To support this
functionality, the `globalSummon` action will support the following property:
* `"dropdownDuration": float`
- When omitted, `0`, or a negative number: No animation is used
when summoning the window. The summoned window is focused immediately where
it is.
- When a positive number is provided, the terminal will use that value as a
duration (in seconds) to slide the terminal into position when activated.
- The default would be some sensible value. The pane animation is .2s, so
`0.2` might be a reasonable default here.
We could have alternatively provided a `"dropdownSpeed"` setting, that provided
a number of pixels per second. In my opinion, that would be harder for users to
use correctly. I believe that it's easier for users to mentally picture "I'd
like the dropdown to last 100ms" vs "My monitor is 1504px tall, so I need to set
this to 15040 to make the window traverse the entire display in .1s"
> NOTE: `dropdownDuration` will be ignored when the user has animations disabled
> in the OS. In that case, the terminal will just appear, as if it was set to 0.
Some users might want to be able to use the global hotkey to hide the window
when the window is already visible. This would let the hotkey act as a sort of
global toggle for the Terminal window. Others might not like that behavior, and
just want the action to always bring the Terminal into focus, and do nothing if
the terminal is already focused. To facilitate both these use cases, we'll add
the following property:
* `"toggleVisibility": bool`
- When `true`: (_default_) When this hotkey is pressed, and the terminal
window is currently active, minimize the window.
- When `dropdownDuration` is not `0`, then the window will slide back off
the top at the same speed as it would come down.
- When `false`: When this hotkey is pressed, and the terminal window is
currently active, do nothing.
### Quake Mode
In addition to just summoning the window from anywhere, some terminals also
support a special "quake mode" buffer or window. This window is one that closely
emulates the console from quake:
* It's docked to the top of the screen
* It takes the full width of the monitor, and only the bottom can be resized
* It often doesn't have any other UI elements, like tabs
For fun, we'll also be adding a special `"_quake"` window with the same
behavior. If the user names a window `_quake`, then it will behave in the
following special ways:
* On launch, it will ignore the `initialPosition` and
`initialRows`/`initialCols` setting, and instead resize to the top half of the
monitor.
* On launch, it will ignore the `launchMode` setting, and always launch in focus
mode.
- Users can disable focus mode on the `_quake` window if they do want tabs.
* It will not be resizable from any side except the bottom of the window, nor
will it be drag-able.
* It will not be a valid target for the "most recent window" for window
glomming. If it's the only open window, with `"windowingBehavior":
"useExisting*"`, then a new window will be created instead.
- It _is_ still a valid target for something like `wt -w _quake new-tab`
A window at runtime can be renamed to become the `_quake` window (if no other
`_quake` window exists). When it does, it will resize to the position of the
quake window, and enter focus mode.
We'll also be adding a special action `quakeMode`. This action is a special case
of the `globalSummon` action, to specifically invoke the quake window in the
current place. It is basically the same thing as the more elaborate:
```json
{
"monitor": "toCurrent",
"desktop": "toCurrent",
"window": "_quake",
"toggleVisibility": true,
"dropdownDuration": 0.5
},
```
### Minimize to Tray
Many users have requested that the terminal additionally supports minimizing the
window "to the tray icon". This is a bit like when you close the Teams window,
but Teams is actually still running in the system tray, or the "notification
area".
![The Teams tray icon](tray-icon-000.png)
_fig 1: an example of the Teams tray icon in the notification area_.
When users want to be able to "minimize to the tray", they want:
* The window to no longer appear on the taskbar
* The window to no longer appear in the alt-tab order
When minimized to the tray, it's almost as if there's no window for the Terminal
at all. This can be combined with the global hotkey (or the tray icon's context
menu) to quickly restore the window.
The tray icon could be used for a variety of purposes. As a simple start, we
could include the following three options:
```
Focus Terminal
---
Windows > Window 1 - <un-named window>
Window 2 - "This-window-does-have-a-name"
---
Quit
```
Just clicking on the icon would summon the recent terminal window. Right
clicking would show the menu with "Focus Terminal", "Windows" and "Quit" in it, and
"Windows" would have nested entries for each Terminal window.
* "Focus Terminal" would do just that - summon the most recent terminal window,
wherever it is.
* "Windows" would have nested popups for each open Terminal window. Each of
these nested entries would display the name and ID of the window. Clicking
them would summon that window (wherever it may be)
* "Quit" would be akin to quit in browsers - close all open windows
<sup>[[1]](#footnote-1)</sup>.
The tray notification would be visible always when the user has
`"minimizeToTray": true` set in their settings. If the user has that set to
false, but would still like the tray, they can specify `"alwaysShowTrayIcon":
true`. That will cause the tray icon to always be added to the system tray.
There's not a combination of settings where the Terminal is "minimized to the
tray", and there's _no tray icon visible_. We don't want to let users get into a
state where the Terminal is running, but is totally hidden from their control.
From a technical standpoint, the tray icon is managed similar to the global
hotkey. The Monarch process is responsible for setting it up, and processing the
messages. When a Monarch dies and a new process becomes the Monarch, then it
will re-create the tray icon.
## UI/UX Design
To summarize, we're proposing the following set of settings:
```jsonc
{
"minimizeToTray": bool,
"alwaysShowTrayIcon": bool,
"actions": [
{
"keys": KeyChord,
"command": {
"action": "globalSummon",
"dropdownDuration": float,
"toggleVisibility": bool,
"monitor": "any"|"toCurrent"|"onCurrent"|int,
"desktop": "any"|"toCurrent"|"onCurrent"
}
},
{
"keys": KeyChord,
"command": {
"action": "quakeMode"
}
}
]
}
```
## Potential Issues
<table>
<tr>
<td><strong>Compatibility</strong></td>
<td>
As part of this set of changes, we'll also be allowing the <kbd>Win</kbd> key in
keybindings. Generally, the OS reserves the Windows key for its own shortcuts.
For example, <kbd>Win+R</kbd> for the run dialog, <kbd>Win+A</kbd> for the
Action Center, <kbd>Win+V</kbd> for the cloud clipboard, etc. Users will now be
able to use the win key themselves, but they should be aware that the OS has
"first dibs" on any hotkeys involving the Windows key.
</td>
</tr>
<tr>
<td><strong>Mixed elevation</strong></td>
<td>
Only one app at a time gets to register for global hotkeys. However, from the
Terminal's perspective, unelevated and elevated windows will act like different
apps. Each privilege level has its own Monarch. The two are unable to
communicate across the elevation boundary.
This means that if the user often runs terminals in both contexts, then only one
will have the global hotkeys bound. The naive implementation would have the
first elevation level "win" the keybindings.
A different option would be to have elevated windows not register global hotkeys
_at all_. I don't believe that there's any sort of security implication for
having a global hotkey for an elevated window.
A third option would be to have some sort of `"whenElevated": bool?` property
for global hotkeys. This would explicitly enable a given hotkey for unelevated
vs elevated windows.
* `"whenElevated": null`: behave as normal - the first context level to run wins
* `"whenElevated": true`: only register the hotkey when running elevated
* `"whenElevated": false`: only register the hotkey when running unelevated
</td>
</tr>
<tr>
<td><strong>OneCore / Windows 10X</strong></td>
<td>
I'm fairly certain that none of these APIs would work on Windows 10X at all.
These features would have to initially be disabled in a pure UWP version of the
Terminal, until we could find workarounds. Since the window layer is the one
responsible for the management of the hotkeys and the tray icon, we're not too
worried about this.
</td>
</tr>
</table>
* If there are any other applications running that have already registered
hotkeys with `RegisterHotKey`, then it's possible that the Terminal's attempt
to register that hotkey will fail. If that should happen, then we should
display a warning dialog to the user indicating which hotkey will not work
(because it's already used for something else).
* Which is the "current" monitor? The one with the mouse or the one with the
active window? This isn't something that has an obvious answer. Guake
implements this feature where the "current monitor" is the one with the mouse
on it. At least for the first iterations of this action, that's what we'll
use.
`monitor: onCurrent|onCurrentWindow|toCurrent|<int>`
* Currently, running both the Release and Preview versions of the Terminal at
the same time side-by-side is not generally supported. (For example, `wt.exe`
can only ever point at one of two.) If a user binds the same key to a
`globalSummon` or `quakeMode` action, then only one of the apps will actually
be able to successfully claim the global hotkey.
## Implementation plan
Currently, in [`dev/migrie/f/653-QUAKE-MODE`], I have some sample rudimentary
code to implement quake mode support. It allows for only a single global hotkey
that summons the MRU window, without dropdown. That would be a good place for
anyone starting to work on this feature. From there, I imagine the following
work would be needed:
* [ ] Add a `globalSummon` action. `AppHost` would need to be able to get _all_
of these actions, and register all of them. Each one would need to be assigned
a unique ID, so `WM_HOTKEY` can identify which hotkey was pressed.
- This could be committed without any other args to the `globalHotkeys`. In
this initial version, the behavior would be summoning the MRU window,
where it is, no dropdown, to start with. From there, we'd add the
remaining properties:
* [ ] Add support for the `toggleVisibility` property
* [ ] Add support for the `desktop` property to control how window summoning
interacts with virtual desktops
* [ ] Add support for the `monitor` which monitor the window appears on.
* [ ] Add support for the `dropdownDuration` property
* [ ] Add the `minimizeToTray` setting, and implement it without any sort of flyout
* [ ] Add a list of windows to the right-click flyout on the tray icon
* [ ] Add support for the `alwaysShowTrayIcon` setting
* [ ] When the user creates a window named `_quake`, ignore the initial size,
position, and launch mode, and create the window in quake mode instead.
* [ ] Exempt the `_quake` window from window glomming
* [ ] Add the `quakeMode` action, which `globalSummon`'s the `_quake` window.
* [ ] Prevent the `_quake` window from being dragged or resized on the
top/left/right.
### Future Considerations
I don't believe there are any other tracked requests that are planned that
aren't already included in this spec.
* Should the tray icon's list of windows include window titles? Both the name
and title? Maybe something like `({name}|{id}): {title}`? I'd bet that most
people don't end up naming their windows.
* Dropdown duration could be a `float|bool`, with `true`->(whatever the default
is), `false`->0.
- We could have the setting appear as a pair of radio buttons, with the first
disabling dropdown, and the second enabling a text box for inputting an
animation duration.
* It might be an interesting idea to have the ability to dock the quake window
to a specific side of the monitor, not just the top. We could probably do that
with a global setting `"quakeModeDockSide": "top"|"left"|"bottom"|"right"` or
something like that.
* We might want to pre-load the quake window into the tray icon as an entry for
"Quake Mode", and otherwise exclude it from the list of windows in that menu.
* We might think of other things for the Quake Mode window in the future - this
spec is by no means comprehensive. For example, it might make sense for the
quake mode window to automatically open in "always on top" mode.
* It was suggested that the quake mode window could auto-hide when it loses
focus. That's a really neat idea, but we're a little worried about the
implementation. What happens when the IME window gets focus? Or the Start
Menu? Would those end up causing the quake window to prematurely minimize
itself? For that reason, we're leaving this as a future consideration.
* Perhaps there could be a top-level object in the settings like
```json
{
"quakeMode": {
"hideOnFocusLost": true,
"useFocusMode": false,
"profile": "my quake mode profile" ,
"quakeModeDockSide": "bottom"
}
}
```
That would allow the user some further customizations on the quake mode
behaviors.
- This was later converted to the idea in [#9992] - Add per-window-name global
settings
* Another proposed idea was a simplification of some of the summoning modes. `{
"monitor": "any", "desktop": "any" }` is a little long, and maybe not the most
apparent naming. Perhaps we could add another property like `summonMode` that
would act like an alias for a `monitor`, `desktop` combo.
- `"summonMode": "activateInMyFace"`: `{ "monitor": "toCurrent", "desktop": "toCurrent" }`
- `"summonMode": "activateWherever"`: `{ "monitor": "any", "desktop": "any" }`
## Resources
Docs on adding a system tray item:
* https://docs.microsoft.com/en-us/windows/win32/shell/notification-area
* https://www.codeproject.com/Articles/18783/Example-of-a-SysTray-App-in-Win32
Docs regarding hiding a window from the taskbar:
* https://docs.microsoft.com/en-us/previous-versions//bb776822(v=vs.85)#managing-taskbar-buttons
### Footnotes
<a name="footnote-1"><a>[1]: Quitting the terminal is different than closing the
windows one-by-one. Quiting implies an atomic action, for closing all the
windows. Once [#766] lands, this will give us a chance to persist the state of
_all_ open windows. This will allow us to re-open with all the user's windows,
not just the one that happened to be closed last.
<a name="footnote-2"><a>[2]: **Addenda, May 2021**: In the course of
implementation, it became apparent that there's an important UX difference
between summoning _to the monitor with the cursor_ vs _to the monitor with the
foreground window_. `"monitor": "toMouse"` was added as an option, to allow the
user to differentiate between the two behaviors.
[#653]: https://github.com/microsoft/terminal/issues/653
[#766]: https://github.com/microsoft/terminal/issues/766
[#5727]: https://github.com/microsoft/terminal/issues/5727
[#9992]: https://github.com/microsoft/terminal/issues/9992
[Process Model 2.0 Spec]: https://github.com/microsoft/terminal/blob/main/doc/specs/%235000%20-%20Process%20Model%202.0/%235000%20-%20Process%20Model%202.0.md
[Quake 3 sample]: https://youtu.be/ZmR6HQbuHPA?t=27
[`RegisterHotKey`]: https://docs.microsoft.com/en-us/windows/win32/api/winuser/nf-winuser-registerhotkey
[`dev/migrie/f/653-QUAKE-MODE`]: https://github.com/microsoft/terminal/tree/dev/migrie/f/653-QUAKE-MODE

Binary file not shown.

Before

Width:  |  Height:  |  Size: 20 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 142 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 140 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 142 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 142 KiB

View File

@@ -1,130 +0,0 @@
---
author: Kayla Cinnamon - cinnamon-msft
created on: 2021-03-04
last updated: 2021-03-09
issue id: 6900
---
# Actions Page
## Abstract
We need to represent actions inside the settings UI. This spec goes through the possible use cases and reasoning for including specific features for actions inside the settings UI.
## Background
### Inspiration
It would be ideal if we could get the settings UI to have parity with the JSON file. This will take some design work if we want every feature possible in relation to actions. There is also the option of not having parity with the JSON file in order to present a simpler UX.
### User Stories
All of these features are possible with the JSON file. This spec will go into discussion of which (possibly all) of these user stories need to be handled by the settings UI.
- Add key bindings to an action that does not already have keys assigned
- Edit key bindings for an action
- Remove key bindings from an action
- Add multiple key bindings for the same action
- Create an iterable action
- Create a nested action
- Choose which actions appear inside the command palette
- See all possible actions, regardless of keys
Commands with properties:
- sendInput has "input"
- closeOtherTabs has "index"
- closeTabsAfter has "index"
- renameTab has "title"*
- setTabColor has "color"*
- newWindow has "commandline", "startingDirectory", "tabTitle", "index", "profile"
- splitPane has "split", "commandline", "startingDirectory", "tabTitle", "index", "profile", "splitMode", "size"
- copy has "singleLine", "copyFormatting"
- scrollUp has "rowsToScroll"
- scrollDown has "rowsToScroll"
- setColorScheme has "colorScheme"
Majority of these commands listed above are intended for the command palette, so they wouldn't make much sense with keys assigned to them anyway.
### Future Considerations
One day we'll have actions that can be invoked by items in the dropdown menu. This setting will have to live somewhere. Also, once we get a status bar, people may want to invoke actions from there.
## Solution Design
### Proposal 1: Keyboard and Command Palette pages
Implement a Keyboard page in place of the Actions page. Also plan for a Command Palette page in the future if it's something that's heavily requested. The Command Palette page would cover the missing use cases listed below.
When users want to add a new key binding, the dropdown will list every action, regardless if it already has keys assigned. This page should show every key binding assigned to an action, even if there are multiple bindings to the same action.
Users will be able to view every possible action from the command palette if they'd like.
Use cases covered:
- Add key bindings to an action that does not already have keys assigned
- Edit key bindings for an action
- Remove key bindings from an action
- Add multiple key bindings for the same action
- See all actions that have keys assigned
Use cases missing:
- Create an iterable action
- Create a nested action
- Choose which actions appear inside the command palette
- See all possible actions, regardless of keys
* **Pros**:
- This allows people to edit their actions in most of their scenarios.
- This gives us some wiggle room to cover majority of the use cases we need and seeing if people want the other use cases that are missing.
* **Cons**:
- Unfortunately we couldn't cover every single use case with this design.
- You can't edit the properties that are on some commands, however the default commands from the command palette include options with properties anyway. For example "decrease font size" has the `delta` property already included.
### Proposal 2: Have everything on one Actions page
Implement an Actions page that allows you to create actions designed for the command palette as well as actions with keys.
Use cases covered:
- Add key bindings to an action that does not already have keys assigned
- Edit key bindings for an action
- Remove key bindings from an action
- Add multiple key bindings for the same action
- See all actions that have keys assigned
- Create an iterable action
- Create a nested action
- Choose which actions appear inside the command palette
- See all possible actions, regardless of keys
I could not come up with a UX design that wasn't too complicated or confusing for this scenario.
**Pros**:
- There is full parity with the JSON file.
**Cons**:
- Could not come up with a simplistic design to represent all of the use cases (which makes the settings UI not as enticing since it promotes ease of use).
## Conclusion
We considered Proposal 2, however the design became cluttered very quickly and we agreed to create two pages and start off with Proposal 1.
## UI/UX Design
![Click edit on key binding](./edit-click.png)
The Add new button is using the secondary color, to align with the button on the Color schemes page.
![Edit key binding](./edit-keys.png)
![Click add new](./add-click.png)
![Add key binding](./add-keys.png)
## Potential Issues
This design is not 1:1 with the JSON file, so actions that don't have keys will not appear on this page. Additionally, you can't add a new action without keys with this current design.
You also cannot specify properties on commands (like the `newTab` command) and these will have to be added through the JSON file. Considering there are only a few of these and we're planning to iterate on this and add a Command Palette page, we were okay with this decision.
## Resources
### Footnotes

View File

@@ -1,344 +0,0 @@
---
author: Carlos Zamora @carlos-zamora
created on: 2021-03-12
last updated: 2021-03-17
issue id: [#885]
---
# Actions in the Settings Model
## Abstract
This spec proposes a refactor of how Windows Terminal actions are stored in the settings model.
The new representation would mainly allow the serialization and deserialization of commands and keybindings.
## Inspiration
A major component that is missing from the Settings UI is the representation of keybindings and commands.
The JSON represents both of these as a combined entry as follows:
```js
{ "icon": "path/to/icon.png", "name": "Copy the selected text", "command": "copy", "keys": "ctrl+c" },
```
In the example above, the copy action is...
- bound to <kbd>ctrl+c</kbd>
- presented as "Copy the selected text" with the "path/to/icon.png" icon
However, at the time of writing this spec, the settings model represents it as...
- (key binding) a `KeyChord` to `ActionAndArgs` entry in a `KeyMapping`
- (command) a `Command` with an associated icon, name, and action (`ActionAndArgs`)
This introduces the following issues:
1. Serialization
- We have no way of knowing when a command and a key binding point to the same action. Thus, we don't
know when to write a "name" to the json.
- We also don't know if the name was auto-generated or set by the user. This can make the JSON much more bloated by
actions with names that would normally be autogenerated.
2. Handling Duplicates
- The same action can be bound to multiple key chords. The command palette combines all of these actions into one entry
because they have the same name. In reality, this same action is just being referenced in different ways.
## Solution Design
I propose that the issues stated above be handled via the following approach.
### Step 1: Consolidating actions
`Command` will be updated to look like the following:
```c++
runtimeclass Command
{
// The path to the icon (or icon itself, if it's an emoji)
String IconPath;
// The associated name. If none is defined, one is auto-generated.
String Name;
// The key binding that can be used to invoke this action.
// NOTE: We're actually holding the KeyChord instead of just the text.
// KeyChordText just serializes the relevant keychord
Microsoft.Terminal.Control.KeyChord Keys;
String KeyChordText;
// The action itself.
ActionAndArgs ActionAndArgs;
// NOTE: nested and iterable command logic will still be here,
// But they are omitted to make this section seem cleaner.
// Future Considerations:
// - [#6899]: Action IDs --> add an identifier here
}
```
The goal here is to consolidate key binding actions and command palette actions into a single class.
This will also require the following supplemental changes:
- `Command::LayerJson`
- This must combine the logic of `KeyMapping::LayerJson` and `Command::LayerJson`.
- Key Chord data
- Internally, store a `vector<KeyChord> _keyMappings` to keep track of all the key chords associated with this action.
- `RegisterKey` and `EraseKey` update `_keyMappings`, and ensure that the latest key registered is at the end of the list.
- `Keys()` simply returns the last entry of `_keyMappings`, which is the latest key chord this action is bound to.
- `KeyChordText())` is exposed to pass the text directly to the command palette.
This depends on `Keys` and, thus, propagate changes automatically.
- Observable properties
- `Command` has observable properties today, but does not need them because `Command` will never change while the app is running.
- Nested and iterable commands
- `HasNestedCommands`, `NestedCommands{ get; }`, `IterateOn` will continue to be exposed.
- A setter for these customizations will not be exposed until we find it necessary (i.e. adding support for customizing it in the Settings UI)
- Command expansion can continue to be exposed here to reduce implementation cost.
- An additional `IsNestedCommand` is necessary to record a case where a nested command is being unbound `{ "commands": null, "name": "foo" }`.
Overall, the `Command` class is simply being promoted to include the `KeyChord` it has.
This allows the implementation cost of this step to be relatively small.
Completion of this step should only cause relatively minor changes to anything that depends on `Command`, because
it is largely the same class. However, key bindings will largely be impacted because we represent key bindings as
a map of `KeyChord`s to `ActionAndArgs`. This leads us to step 2 of this process.
### Step 2: Querying actions
Key bindings and commands are deserialized by basically storing individual actions to a map.
- `KeyMapping` is basically an `IMap<KeyChord, ActionAndArgs>` with a few extra functions. In fact, it actually
stores key binding data to a `std::map<KeyChord, ActionAndArgs>` and directly interacts with it.
- `Command::LayerJson` populates an `IMap<String, Command>` during deserialization as it iterates over every action.
Note that `Command` can be interpreted as a wrapper for `ActionAndArgs` with more stuff here.
It makes sense to store these actions as maps. So, following step 1 above, we can also store and expose actions
something like the following:
```c++
runtimeclass ActionMap
{
ActionAndArgs GetActionByKeyChord(KeyChord keys);
KeyChord GetKeyBindingForAction(ShortcutAction action);
KeyChord GetKeyBindingForAction(ShortcutAction action, IActionArgs actionArgs);
IMapView<String, Command> NameMap { get; };
// Future Considerations:
// - [#6899]: Action IDs --> GetActionByID()
}
```
The getters will return null if a matching action or key chord is not found. Since iterable commands need to be expanded at in TerminalApp, we'll just expose `NameMap`, then let TerminalApp perform the expansion as they do now. Internally, we can store the actions as follows:
```c++
std::map<KeyChord, InternalActionID> _KeyMap;
std::map<InternalActionID, Command> _ActionMap;
```
`InternalActionID` will be a hash of `ActionAndArgs` such that two `ActionAndArgs` with the same `ShortcutAction` and `IActionArgs` output the same hash value.
`GetActionByKeyChord` will use `_KeyMap` to find the `InternalActionID`, then use the `_ActionMap` to find the bound `Command`.
`GetKeyBindingForAction` will hash the provided `ActionAndArgs` (constructed by the given parameters) and check `_ActionMap` for the given `InternalActionID`.
`NameMap` will need to ensure every action in `_ActionMap` is added to the output name map if it has an associated name. This is done by simply iterating over `_ActionMap`. Nested commands must be added separately because they cannot be hashed.
`ActionMap` will have an `AddAction(Command cmd)` that will update the internal state whenever a command is registered. If the given command is valid, we will check for collisions and resolve them. Otherwise, we will consider this an "unbound" action and update the internal state normally. It is important that we don't handle "unbound" actions differently because this ensures that we are explicitly unbinding a key chord.
### Step 3: Settings UI needs
After the former two steps are completed, the new representation of actions in the settings model is now on-par with
what we have today. In order to bind these new actions to the Settings UI, we need the following:
1. Exposing the maps
- `ActionMap::KeyBindings` and `ActionMap::Commands` may need to be added to pass the full list of actions to the Settings UI.
- In doing this, we can already update the Settings UI to include a better view of our actions.
2. Creating a copy of the settings model
- The Settings UI operates by binding the XAML controls to a copy of the settings model.
- Copying the `ActionMap` is fairly simple. Just copy the internal state and ensure that `Command::Copy` is called such that no reference to the original WinRT objects persist. Since we are using `InternalActionID`, we will not have to worry about multiple `Command` references existing within the same `ActionMap`.
3. Modifying the `Command`s
- `ActionMap` must be responsible for changing `Command`s so that we can ensure `ActionMap` always has a correct internal state:
- It is important that `Command` only exposes getters (not setters) to ensure `ActionMap` is up to date.
- If a key chord is being changed, update the `_KeyMap` and the `Command` itself.
- If a key binding is being deleted, add an unbound action to the given key chord.
- This is similar to how color schemes are maintained today.
- In the event that name/key-chord is set to something that's already taken, we need to propagate those changes to
the rest of `ActionMap`. As we do with the JSON, we respect the last name/key-chord set by the user. See [Modifying Actions](#modifying-actions)
in potential issues.
- For the purposes of the key bindings page, we will introduce a `KeyBindingViewModel` to serve as an intermediator between the settings UI and the settings model. The view model will be responsible for things like...
- exposing relevant information to the UI controls
- converting UI control interactions into proper API calls to the settings model
4. Serialization
- `Command::ToJson()` and `ActionMap::ToJson()` should perform most of the work for us. Simply iterate over the `_ActionMap` and call `Command::ToJson` on each action.
- See [Unbinding actions](#unbinding-actions) in potential issues.
## UI/UX Design
N/A
## Capabilities
N/A
### Accessibility
N/A
### Security
N/A
### Reliability
N/A
### Compatibility
N/A
### Performance, Power, and Efficiency
## Potential Issues
### Layering Actions
We need a way to determine where an action came from to minimize how many actions we serialize when we
write to disk. This is a two part approach that happens as we're loading the settings
1. Load defaults.json
- For each of the actions in the JSON...
- Construct the `Command` (basically the `Command::LayerJson` we have today)
- Add it to the `ActionMap`
- this should update the internal state of `ActionMap` appropriately
- if the newly added key chord conflicts with a pre-existing one,
redirect `_KeyMap` to the newly added `Command` instead,
and update the conflicting one.
2. Load settings.json
- Create a child for the `ActionMap`
- The purpose of a parent is to continue a search when the current `ActionMap` can't find a `Command` for a query. The parent is intended to be immutable.
- Load the actions array like normal into the child (see step 1)
Introducing a parent mechanism to `ActionMap` allows it to understand where a `Command` came from. This allows us to minimize the number of actions we serialize when we write to disk, as opposed to serializing the entire list of actions.
`ActionMap` queries will need to check their parent when they cannot find a matching `InternalActionID` in their `_ActionMap`.
Since `NameMap` is generated upon request, we will need to use a `std::set<InternalActionID>` as we generate the `NameMap`. This will ensure that each `Command` is only added to the `NameMap` once. The name map will be generated as follows:
1. Get an accumulated list of `Command`s from our parents
2. Iterate over the list...
- Update `NameMap` with any new `Command`s (tracked by the `std::set<InternalActionID>`)
Nested commands will be saved to their own map since they do not have an `InternalActionID`.
- During `ActionMap`'s population, we must ensure to resolve any conflicts immediately. This means that any new `Command`s that generate a name conflicting with a nested command will take priority (and we'll remove the nested command from its own map). Conversely, if a new nested command conflicts with an existing standard `Command`, we can ignore it because our generation of `NameMap` will handle it.
- When populating `NameMap`, we must first add all of the standard `Command`s. To ensure layering is accomplished correctly, we will need to start from the top-most parent and update `NameMap`. As we go down the inheritance tree, any conflicts are resolved by prioritizing the current layer (the child). This ensures that the current layer takes priority.
- After adding all of the standard `Command`s to the `NameMap`, we can then register all of the nested commands. Since nested commands have no identifier other than a name, we cannot use the `InternalActionID` heuristic. However, as mentioned earlier, we are updating our internal nested command map as we add new actions. So when we are generating the name map, we can assume that all of these nested commands now have priority. Thus, we simply add all of these nested commands to the name map. Any conflicts are resolved in favor of th nested command.
### Modifying Actions
There are several ways a command can be modified:
- change/remove the key chord
- change the name
- change the icon
- change the action
It is important that these modifications are done through `ActionMap` instead of `Command`.
This is to ensure that the `ActionMap` is always aligned with `Command`'s values. `Command` should only expose getters in the projected type to enforce this. Thus, we can add the following functions to `ActionMap`:
```c++
runtimeclass ActionMap
{
void SetKeyChord(Command cmd, KeyChord keys);
void SetName(Command cmd, String name);
void SetIcon(Command cmd, String iconPath);
void SetAction(Command cmd, ShortcutAction action, IActionArgs actionArgs);
}
```
`SetKeyChord` will need to make sure to modify the `_KeyMap` and the provided `Command`.
If the new key chord was already taken, we also need to update the conflicting `Command`
and remove its key chord.
`SetName` will need to make sure to modify the `Command` in `_ActionMap` and regenerate `NameMap`.
`SetIcon` will only need to modify the provided `Command`. We can choose to not expose this in the `ActionMap`, but doing so makes the API consistent.
`SetAction` will need to begin by updating the provided `Command`'s `ActionAndArgs`.
If the generated name is being used, the name will need to be updated. `_ActionMap` will need to be updated with a new `InternalActionID` for the new action. This is a major operation and so all views exposed will need to be regenerated.
Regarding [Layering Actions](#layering-actions), if the `Command` does not exist in the current layer,
but exists in a parent layer, we need to...
0. check if it exists
- use the hash `InternalActionID` to see if it exists in the current layer
- if it doesn't (which is the case we're trying to solve here), call `_GetActionByID(InternalActionID)` to retrieve the `Command` wherever it may be. This helper function simply checks the current layer, if none is found, it recursively checks its parents until a match is found.
1. duplicate it with `Command::Copy`
2. store the duplicate in the current layer
- `ActionMap::AddAction(duplicate)`
3. make the modification to the duplicate
This ensures that the change persists post-serialization.
TerminalApp has no reason to ever call these setters. To ensure that relationship, we will introduce an `IActionMapView` interface that will only expose `ActionMap` query functions. Conversely, `ActionMap` will be exposed to the TerminalSettingsEditor to allow for modifications.
### Unbinding actions
Removing a name is currently omitted from this spec because there
is no Settings UI use case for it at the moment. This scenario is
designed for Command Palette customization.
The only kind of unbinding currently in scope is freeing a key chord such that
no action is executed on that key stroke. To do this, simply `ActionMap::AddAction` a `Command` with...
- `ActionAndArgs`: `ShortcutAction = Invalid` and `IActionArgs = nullptr`
- `Keys` being the provided key chord
In explicitly storing an "unbound" action, we are explicitly saying that this key chord
must be passed through and this string must be removed from the command palette. `AddAction` automatically handles updating the internal state of `ActionMap` and any conflicting `Commands`.
This allows us to output something like this in the JSON:
```js
{ "command": "unbound", "keys": "ctrl+c" }
```
### Consolidated Actions
`AddAction` must be a bit smarter when it comes to the following scenario:
- Given a command that unbinds a key chord: `{ "command": "unbound", "keys": "ctrl+c" }`
- And... that key chord was set in a parent layer `{ "command": "copy", "keys": "ctrl+c" }`
- But... the action has another key chord from a parent layer `{ "command": "copy", "keys": "ctrl+shift+c" }`
`_ActionMap` does not contain any information about a parent layer; it only contains actions introduced in the current layer. Thus, in the scenario above, unbinding `ctrl+c` is what is relevant to `_ActionMap`. However, this may cause some complications for `GetKeyChordForAction`. We cannot simply check our internal `_ActionMap`, because the primary key chord in the entry may be incorrect. Again, this is because `_ActionMap` is only aware of what was bound in the current layer.
To get around this issue, we've introduced `_ConsolidatedActions`. In a way, `_ConsolidatedActions` is similar to `_ActionMap`, except that it consolidates the `Command` data into one entry constructed across the current layer and the parent layers. Specifically, in the scenario above, `_ActionMap` will say that `copy` has no key chords. In fact, `_ActionMap` has no reason to have `copy` at all, because it was not introduced in this layer. Conversely, `_ConsolidatedActions` holds `copy` with a `ctrl+shift+c` binding, which is then returned to `GetKeyChordForAction`.
To maintain `_ConsolidatedActions`, any new action added to the Action Map must also update `_ConsolidatedActions`. It is especially important to handle and propagate collisions to `_ConsolidatedActions`.
When querying Action Map for an ID, we should always check in the following order:
- `_ConsolidatedActions`
- `_ActionMap`
- repeat this process for each parent
This is to ensure that we are returning the correct and wholistic view of a `Command` on a query. Rather than acquiring a `Command` constructed in this layer, we receive one that contains all of the data acquired across the entire Action Map and its parents.
## Future considerations
There are a number of ideas regarding actions that would be fairly trivial to implement given this refactor:
- [#6899]: Action IDs
- As actions grow to become more widespread within Windows Terminal (i.e. dropdown and jumplist integration),
a formal ID system would help users reference the same action throughout the app. With the internal
ID system introduced earlier, we would simply introduce a new
`std:map<string, InternalActionID> _ExternalIDMap` that is updated like the others, and add a `String ID`
property to `Action`.
- [#8100] Source Tracking
- Identifying where a setting came from can be very beneficial in the settings model and UI. For example,
profile settings now have an `OverrideSource` getter that describes what `Profile` object the setting
came from (i.e. base layer, profile generator, etc...). A similar system can be used for `Action` in
that we record if the action was last modified in defaults.json or settings.json.
- There seems to be no desire for action inheritance (i.e. inheriting the name/key-chord from the parent).
So this should be sufficient.
## Resources
[#885]: https://github.com/microsoft/terminal/issues/885
[#6899]: https://github.com/microsoft/terminal/issues/6899
[#8100]: https://github.com/microsoft/terminal/issues/8100
[#8767]: https://github.com/microsoft/terminal/issues/8767
Other references:
[Settings UI: Actions Page]: https://github.com/microsoft/terminal/issues/6900
[Settings UI: Actions Page Design]: https://github.com/microsoft/terminal/pulls/9427
[Action ID Spec]: https://github.com/microsoft/terminal/issues/7175

View File

@@ -1,171 +0,0 @@
---
author: Carlos Zamora @carlos-zamora
created on: 2019-08-30
last updated: 2021-09-17
issue id: 715
---
# Keyboard Selection
## Abstract
This spec describes a new set of non-configurable keybindings that allows the user to update a selection without the use of a mouse or stylus.
## Inspiration
ConHost allows the user to modify a selection using the keyboard. Holding `Shift` allows the user to move the second selection endpoint in accordance with the arrow keys. The selection endpoint updates by one cell per key event, allowing the user to refine the selected region.
Mark mode allows the user to create a selection using only the keyboard, then edit it as mentioned above.
## Solution Design
The fundamental solution design for keyboard selection is that the responsibilities between the Terminal Control and Terminal Core must be very distinct. The Terminal Control is responsible for handling user interaction and directing the Terminal Core to update the selection. The Terminal Core will need to update the selection according to the preferences of the Terminal Control.
Relatively recently, TerminalControl was split into `TerminalControl`, `ControlInteractivity`, and `ControlCore`. Changes made to `ControlInteractivity`, `ControlCore`, and below propagate functionality to all consumers, meaning that the WPF terminal would benefit from these changes with no additional work required.
### Fundamental Terminal Control Changes
`ControlCore::TrySendKeyEvent()` is responsible for handling the key events after key bindings are dealt with in `TermControl`. At the time of writing this spec, there are 2 cases handled in this order:
- Clear the selection (except in a few key scenarios)
- Send Key Event
The first branch will be updated to _modify_ the selection instead of usually _clearing_ it. This will happen by converting the key event into parameters to forward to `TerminalCore`, which then updates the selection appropriately.
#### Idea: Make keyboard selection a collection of standard keybindings
One idea is to introduce an `updateSelection` action that conditionally works if a selection is active (similar to the `copy` action). For these key bindings, if there is no selection, the key events are forwarded to the application.
Thanks to Keybinding Args, there would only be 1 new command:
| Action | Keybinding Args | Description |
|--|--|--|
| `updateSelection` | | If a selection exists, moves the last selection endpoint. |
| | `Enum direction { up, down, left, right }` | The direction the selection will be moved in. |
| | `Enum mode { char, word, view, buffer }` | The context for which to move the selection endpoint to. (defaults to `char`) |
By default, the following keybindings will be set:
```JS
// Character Selection
{ "command": {"action": "updateSelection", "direction": "left", "mode": "char" }, "keys": "shift+left" },
{ "command": {"action": "updateSelection", "direction": "right", "mode": "char" }, "keys": "shift+right" },
{ "command": {"action": "updateSelection", "direction": "up", "mode": "char" }, "keys": "shift+up" },
{ "command": {"action": "updateSelection", "direction": "down", "mode": "char" }, "keys": "shift+down" },
// Word Selection
{ "command": {"action": "updateSelection", "direction": "left", "mode": "word" }, "keys": "ctrl+shift+left" },
{ "command": {"action": "updateSelection", "direction": "right", "mode": "word" }, "keys": "ctrl+shift+right" },
// Viewport Selection
{ "command": {"action": "updateSelection", "direction": "left", "mode": "view" }, "keys": "shift+home" },
{ "command": {"action": "updateSelection", "direction": "right", "mode": "view" }, "keys": "shift+end" },
{ "command": {"action": "updateSelection", "direction": "up", "mode": "view" }, "keys": "shift+pgup" },
{ "command": {"action": "updateSelection", "direction": "down", "mode": "view" }, "keys": "shift+pgdn" },
// Buffer Corner Selection
{ "command": {"action": "updateSelection", "direction": "up", "mode": "buffer" }, "keys": "ctrl+shift+home" },
{ "command": {"action": "updateSelection", "direction": "down", "mode": "buffer" }, "keys": "ctrl+shift+end" },
```
These are in accordance with ConHost's keyboard selection model.
This idea was abandoned due to several reasons:
1. Keyboard selection should be a standard way to interact with a terminal across all consumers (i.e. WPF control, etc.)
2. There isn't really another set of key bindings that makes sense for this. We already hardcoded <kbd>ESC</kbd> as a way to clear the selection. This is just an extension of that.
3. Adding 12 conditionally effective key bindings takes the spot of 12 potential non-conditional key bindings. It would be nice if a different key binding could be set when the selection is not active, but that makes the settings design much more complicated.
4. 12 new items in the command palette is also pretty excessive.
5. If proven wrong when this is in WT Preview, we can revisit this and make them customizable then. It's better to add the ability to customize it later than take it away.
#### Idea: Make keyboard selection a simulation of mouse selection
It may seem that some effort can be saved by making the keyboard selection act as a simulation of mouse selection. There is a union of mouse and keyboard activity that can be represented in a single set of selection motion interfaces that are commanded by the TermControl's Mouse/Keyboard handler and adapted into appropriate motions in the Terminal Core.
However, the mouse handler operates by translating a pixel coordinate on the screen to a text buffer coordinate. This would have to be rewritten and the approach was deemed unworthy.
### Fundamental Terminal Core Changes
The Terminal Core will need to expose a `UpdateSelection()` function that is called by the keybinding handler. The following parameters will need to be passed in:
- `enum SelectionDirection`: the direction that the selection endpoint will attempt to move to. Possible values include `Up`, `Down`, `Left`, and `Right`.
- `enum SelectionExpansion`: the selection expansion mode that the selection endpoint will adhere to. Possible values include `Char`, `Word`, `View`, `Buffer`.
#### Moving by Cell
For `SelectionExpansion = Char`, the selection endpoint will be updated according to the buffer's output pattern. For **horizontal movements**, the selection endpoint will attempt to move left or right. If a viewport boundary is hit, the endpoint will wrap appropriately (i.e.: hitting the left boundary moves it to the last cell of the line above it).
For **vertical movements**, the selection endpoint will attempt to move up or down. If a **viewport boundary** is hit and there is a scroll buffer, the endpoint will move and scroll accordingly by a line.
If a **buffer boundary** is hit, the endpoint will not move. In this case, however, the event will still be considered handled.
**NOTE**: An important thing to handle properly in all cases is wide glyphs. The user should not be allowed to select a portion of a wide glyph; it should be all or none of it. When calling `_ExpandWideGlyphSelection` functions, the result must be saved to the endpoint.
#### Moving by Word
For `SelectionExpansion = Word`, the selection endpoint will also be updated according to the buffer's output pattern, as above. However, the selection will be updated in accordance with "chunk selection" (performing a double-click and dragging the mouse to expand the selection). For **horizontal movements**, the selection endpoint will be updated according to the `_ExpandDoubleClickSelection` functions. The result must be saved to the endpoint. As before, if a boundary is hit, the endpoint will wrap appropriately. See [Future Considerations](#FutureConsiderations) for how this will interact with line wrapping.
For **vertical movements**, the movement is a little more complicated than before. The selection will still respond to buffer and viewport boundaries as before. If the user is trying to move up, the selection endpoint will attempt to move up by one line, then selection will be expanded leftwards. Alternatively, if the user is trying to move down, the selection endpoint will attempt to move down by one line, then the selection will be expanded rightwards.
#### Moving by Viewport
For `SelectionExpansion = View`, the selection endpoint will be updated according to the viewport's height. Horizontal movements will be updated according to the viewport's width, thus resulting in the endpoint being moved to the left/right boundary of the viewport.
#### Moving by Buffer
For `SelectionExpansion = Buffer`, the selection endpoint will be moved to the beginning or end of all the text within the buffer. If moving up or left, set the position to 0,0 (the origin of the buffer). If moving down or right, set the position to the last character in the buffer.
**NOTE**: In all cases, horizontal movements attempting to move past the left/right viewport boundaries result in a wrap. Vertical movements attempting to move past the top/bottom viewport boundaries will scroll such that the selection is at the edge of the screen. Vertical movements attempting to move past the top/bottom buffer boundaries will be clamped to be within buffer boundaries.
Every combination of the `SelectionDirection` and `SelectionExpansion` will map to a keybinding. These pairings are shown below in the UI/UX Design --> Keybindings section.
**NOTE**: If `copyOnSelect` is enabled, we need to make sure we **DO NOT** update the clipboard on every change in selection. The user must explicitly choose to copy the selected text from the buffer.
## UI/UX Design
### Key Bindings
There will only be 1 new command that needs to be added:
| Action | Keybinding Args | Description |
|--|--|--|
| `selectAll` | | Select the entire text buffer.
By default, the following key binding will be set:
```JS
{ "command": "selectAll", "keys": "ctrl+shift+a" },
```
## Capabilities
### Accessibility
Using the keyboard is generally a more accessible experience than using the mouse. Being able to modify a selection by using the keyboard is a good first step towards making selecting text more accessible.
### Security
N/A
### Reliability
With regards to the Terminal Core, the newly introduced code should rely on already existing and tested code. Thus no crash-related bugs are expected.
With regards to Terminal Control and the settings model, crash-related bugs are not expected. However, ensuring that the selection is updated and cleared in general use-case scenarios must be ensured.
### Compatibility
N/A
### Performance, Power, and Efficiency
## Potential Issues
### Grapheme Clusters
When grapheme cluster support is inevitably added to the Text Buffer, moving by "cell" is expected to move by "character" or "cluster". This is similar to how wide glyphs are handled today. Either all of it is selected, or none of it.
## Future considerations
### Word Selection Wrap
At the time of writing this spec, expanding or moving by word is interrupted by the beginning or end of the line, regardless of the wrap flag being set. In the future, selection and the accessibility models will respect the wrap flag on the text buffer.
## Mark Mode
This functionality will be expanded to create a feature similar to Mark Mode. This will allow a user to create a selection using only the keyboard.
## Resources
- https://blogs.windows.com/windowsdeveloper/2014/10/07/console-improvements-in-the-windows-10-technical-preview/

View File

@@ -2,16 +2,17 @@
## Overview
This document outlines the roadmap towards delivering Windows Terminal 2.0 by Winter 2021.
This document outlines the roadmap towards delivering Windows Terminal 2.0 by Spring 2021.
## Milestones
Windows Terminal is engineered and delivered as a set of 6-week milestones. New features will go into [Windows Terminal Preview](https://aka.ms/terminal-preview) first, then a month after they've been in Preview, those features will move into [Windows Terminal](https://aka.ms/terminal).
The Windows Terminal project is engineered and delivered as a set of 4-week milestones. New features will go into [Windows Terminal Preview](https://aka.ms/terminal-preview) first, then a month after they've been in Preview, those features will move into [Windows Terminal](https://aka.ms/terminal).
| Duration | Activity | Releases |
| --- | --- | --- |
| 4 weeks | Dev Work<br/> <ul><li>Fixes / Features for future Windows Releases</li><li>Fixes / Features for Windows Terminal</li></ul> | Release to Internal Selfhosters at end of week 4 |
| 1 week | Quality & Stability<br/> <ul><li>Bug Fixes</li><li>Perf & Stability</li><li>UI Polish</li><li>Tests</li><li>etc.</li></ul>| Push to Microsoft Store at end of week 5 |
| 2 weeks | Dev Work<br/> <ul><li>Fixes / Features for future Windows Releases</li><li>Fixes / Features for Windows Terminal</li></ul> | Release to Internal Selfhosters at end of week 2 |
| 1 week | Quality & Stability<br/> <ul><li>Bug Fixes</li><li>Perf & Stability</li><li>UI Polish</li><li>Tests</li><li>etc.</li></ul>| Push to Microsoft Store at end of week 3 |
| 1 week | Release <br/> <ul><li>Available from [Microsoft Store](https://aka.ms/terminal) & [GitHub Releases](https://github.com/microsoft/terminal/releases)</li><li>Release Notes & Announcement Blog published</li><li>Engineering System Maintenance</li><li>Community Engagement</li><li>Docs</li><li>Future Milestone Planning</li></ul> | Release available from Microsoft Store & GitHub Releases |
## Terminal Roadmap / Timeline
@@ -25,15 +26,12 @@ Below is the schedule for when milestones will be included in release builds of
| 2020-08-31 | [1.3] in Windows Terminal Preview<br>[1.2] in Windows Terminal | [Windows Terminal Preview 1.3 Release](https://devblogs.microsoft.com/commandline/windows-terminal-preview-1-3-release/) |
| 2020-09-30 | [1.4] in Windows Terminal Preview<br>[1.3] in Windows Terminal | [Windows Terminal Preview 1.4 Release](https://devblogs.microsoft.com/commandline/windows-terminal-preview-1-4-release/) |
| 2020-11-30 | [1.5] in Windows Terminal Preview<br>[1.4] in Windows Terminal | [Windows Terminal Preview 1.5 Release](https://devblogs.microsoft.com/commandline/windows-terminal-preview-1-5-release/) |
| 2021-01-31 | [1.6] in Windows Terminal Preview<br>[1.5] in Windows Terminal | [Windows Terminal Preview 1.6 Release](https://devblogs.microsoft.com/commandline/windows-terminal-preview-1-6-release/) |
| 2021-03-01 | [1.7] in Windows Terminal Preview<br>[1.6] in Windows Terminal | [Windows Terminal Preview 1.7 Release](https://devblogs.microsoft.com/commandline/windows-terminal-preview-1-7-release/) |
| 2021-04-14 | [1.8] in Windows Terminal Preview<br>[1.7] in Windows Terminal | [Windows Terminal Preview 1.8 Release](https://devblogs.microsoft.com/commandline/windows-terminal-preview-1-8-release/) |
| 2021-05-31 | [1.9] in Windows Terminal Preview<br>[1.8] in Windows Terminal | [Windows Terminal Preview 1.9 Release](https://devblogs.microsoft.com/commandline/windows-terminal-preview-1-9-release/) |
| 2021-07-14 | [1.10] in Windows Terminal Preview<br>[1.9] in Windows Terminal | [Windows Terminal Preview 1.10 Release](https://devblogs.microsoft.com/commandline/windows-terminal-preview-1-10-release/) |
| 2021-08-31 | [1.11] in Windows Terminal Preview<br>[1.10] in Windows Terminal | [Windows Terminal Preview 1.11 Release](https://devblogs.microsoft.com/commandline/windows-terminal-preview-1-11-release/) |
| 2021-10-31 | 1.12 in Windows Terminal Preview<br>1.11 in Windows Terminal | |
| 2021-11-30 | 2.0 RC in Windows Terminal Preview<br>2.0 RC in Windows Terminal | |
| 2021-12-31 | [2.0] in Windows Terminal Preview<br>[2.0] in Windows Terminal | |
| 2020-12-31 | [1.6] in Windows Terminal Preview<br>[1.5] in Windows Terminal | |
| 2021-01-31 | 1.7 in Windows Terminal Preview<br>[1.6] in Windows Terminal | |
| 2021-02-28 | 1.8 in Windows Terminal Preview<br>1.8 in Windows Terminal | |
| 2021-03-31 | 1.9 in Windows Terminal Preview<br>1.9 in Windows Terminal | |
| 2021-04-30 | 2.0 RC in Windows Terminal Preview<br>2.0 RC in Windows Terminal | |
| 2021-05-31 | [2.0] in Windows Terminal Preview<br>[2.0] in Windows Terminal | |
## Issue Triage & Prioritization
@@ -86,11 +84,6 @@ Feature Notes:
[1.4]: https://github.com/microsoft/terminal/milestone/28
[1.5]: https://github.com/microsoft/terminal/milestone/30
[1.6]: https://github.com/microsoft/terminal/milestone/31
[1.7]: https://github.com/microsoft/terminal/milestone/32
[1.8]: https://github.com/microsoft/terminal/milestone/33
[1.9]: https://github.com/microsoft/terminal/milestone/34
[1.10]: https://github.com/microsoft/terminal/milestone/35
[1.11]: https://github.com/microsoft/terminal/milestone/36
[2.0]: https://github.com/microsoft/terminal/milestone/22
[#1564]: https://github.com/microsoft/terminal/issues/1564
[#6720]: https://github.com/microsoft/terminal/pull/6720

View File

@@ -4,7 +4,7 @@ This was originally imported by @Austin-Lamb in December 2020.
The provenance information (where it came from and which commit) is stored in the file `cgmanifest.json` in the same directory as this readme.
Please update the provenance information in that file when ingesting an updated version of the dependent library.
That provenance file is automatically read and inventoried by Microsoft systems to ensure compliance with appropriate governance standards.
That provenance file is automatically read and inventoried by Microsoft systems to ensure compliance with appropiate governance standards.
## What should be done to update this in the future?
@@ -12,4 +12,4 @@ That provenance file is automatically read and inventoried by Microsoft systems
2. Take the parts you want, but leave most of it behind since it's HUGE and will bloat the repo to take it all. At the time of this writing, we only use small_vector.hpp and its dependencies as a header-only library.
3. Validate that the license in the root of the repository didn't change and update it if so. It is sitting in a version-specific subdirectory below this readme.
If it changed dramatically, ensure that it is still compatible with our license scheme. Also update the NOTICE file in the root of our repository to declare the third-party usage.
4. Submit the pull.
4. Submit the pull.

View File

@@ -4,7 +4,7 @@ This was originally imported by @miniksa in January 2020.
The provenance information (where it came from and which commit) is stored in the file `cgmanifest.json` in the same directory as this readme.
Please update the provenance information in that file when ingesting an updated version of the dependent library.
That provenance file is automatically read and inventoried by Microsoft systems to ensure compliance with appropriate governance standards.
That provenance file is automatically read and inventoried by Microsoft systems to ensure compliance with appropiate governance standards.
## What should be done to update this in the future?

Some files were not shown because too many files have changed in this diff Show More