Commit Graph

3464 Commits

Author SHA1 Message Date
Mike Griese
57d1dd4358 [ainulindale] Clean tear down the App when the process exits 2023-02-03 09:15:53 -06:00
Mike Griese
713ea71725 Revert "this is important for moving, yknow, the panes"
This reverts commit 0a6c21fed0.
2023-02-03 09:10:17 -06:00
Mike Griese
0f33913423 actually start drawing to the new control when we move it 2023-02-03 09:10:04 -06:00
Mike Griese
1f56807fa0 move to int IDs is okay too 2023-02-02 12:28:46 -06:00
Mike Griese
0a6c21fed0 this is important for moving, yknow, the panes 2023-02-02 12:09:25 -06:00
Mike Griese
ebe5fd8163 can I get a whoop whoop (Adds suuport for moving tabs to another window)
cherry-picked from 4acbffb0e
2023-02-02 11:58:05 -06:00
Mike Griese
d143f312d3 Add comments
This is basically the comment from aaefdf4408. The other comment wasn't relevant anymore
2023-02-02 11:30:01 -06:00
Mike Griese
3bd6957a6d This works to open a new pane in the new window...
but it doesn't render, and the old pane keeps on choochin.

  And it only works with a window name, not and ID

  And the `Dispatcher` definitely needs to get re-wired for the new thread, cause dragging it across a DPI boundary (aka, resize) crashes the window
2023-02-02 11:22:36 -06:00
Mike Griese
6a2239d487 Pane officially opened via the serialized actions, via across the process boundary
cherry-picked from 1f2bb760e
2023-02-02 10:52:45 -06:00
Mike Griese
4dcc67b4d7 Serialize the pane to a Content string, instead of just the guid.
There's a lot of renaming, signature changes here. But StartupActions is a good mechanism for passing panes around, more or less.

cherry-picked from cfe879da9
2023-02-02 08:40:45 -06:00
Mike Griese
37a8fde29c Holy shit all the plumbing worked on the first try
cherry-picked from 9870eb1d39
2023-02-02 08:16:37 -06:00
Mike Griese
1b4f1efb09 I literally think this is the whole thing 2023-02-01 17:08:37 -06:00
Mike Griese
e623299ac6 notes 2023-02-01 14:12:31 -06:00
Mike Griese
274d62d8a8 it's working 2023-02-01 13:56:40 -06:00
Mike Griese
f655296c1e This does successfully get a window on the screen, which is pretty impressive
It exits itself after 30s, but hey it worked
2023-02-01 13:15:44 -06:00
Mike Griese
7a3e2e098d Start rewriting WindowManager to facilitate the new process model
At this point, I determined that I would need to make some big changes to
  AppHost and decided that it was time to commit before moving on.
2023-02-01 11:48:43 -06:00
Mike Griese
2c4613a494 wow it just... worked 2023-02-01 08:07:29 -06:00
Mike Griese
f5b030c28c it runs! 2023-01-31 15:21:29 -06:00
Mike Griese
d456210f37 Use the single AppLogic for all windows. Remove the _app references. It launches and crashes immediately. We'll keep shuffling code. 2023-01-31 12:59:30 -06:00
Mike Griese
af14c2b751 [TO PARENT] Move the page ctor call, so that it can happen after the XAML island is started.
I think this can work in the parent at least
2023-01-31 12:58:39 -06:00
Mike Griese
ef7e2edfa5 Merge branch 'dev/migrie/oop/3/foreword' into dev/migrie/oop/3/ainulindale
# Conflicts:
#	src/cascadia/WindowsTerminal/AppHost.h
2023-01-31 11:11:35 -06:00
Mike Griese
5116ca1e77 I think the todo's that are left, we can move on without them for now. 2023-01-31 10:47:46 -06:00
Mike Griese
2195515937 it launches 2023-01-31 10:44:38 -06:00
Mike Griese
99bc280207 It doesn't crash on launch. That's something. There's no startupActions though, so it immediately exits 2023-01-31 09:34:02 -06:00
Mike Griese
439b21f879 this is dangerously close to compiling 2023-01-30 17:06:31 -06:00
Mike Griese
936c01f948 Start splitting AppLogic into AppLogic and Window logic
We'll need this for #5000, for ainulindale. This refactoring will be annoying
  enough as it is so we may as well do it as a first, separate PR.
2023-01-30 14:52:19 -06:00
Mike Griese
e6220b7fe7 Revert "I don't think I want any of these"
This reverts commit a5255ba8ed.
2023-01-30 12:37:35 -06:00
Mike Griese
a5255ba8ed I don't think I want any of these 2023-01-30 12:37:26 -06:00
Mike Griese
581acd40d9 I definitely want all of this. But I started down a path for refactoring AppLogic that I hate so I'm gonna start over 2023-01-30 12:37:12 -06:00
James Holderness
282c583731 Make all console output modes more strictly buffer state (#14735)
The original console output modes were considered attributes of the
buffer, while later additions were treated as global state, and yet both
were accessed via the same buffer-based API. This could result in the
reported modes being out of sync with the way the system was actually
behaving, and a call to `SetConsoleMode` without updating anything could
still trigger unpredictable changes in behavior.

This PR attempts to address that problem by making all modes part of the
buffer state, and giving them predictable default values.

While this won't solve all the tmux layout-breaking issues in #6987, it
does at least fix one case which was the result of an unexpected change
in the `DISABLE_NEWLINE_AUTO_RETURN` mode.

All access to the output modes is now done via the `OutputMode` field in
`SCREEN_INFORMATION`. The fields that were tracking global state in the
`Settings` class (`_fAutoReturnOnNewline` and  `_fRenderGridWorldwide`)
have now been removed.

We still have a global `_dwVirtTermLevel` field, though, but that now
serves as a default value for the `ENABLE_VIRTUAL_TERMINAL_PROCESSING`
mode when creating a new buffer. It's enabled for conpty mode, and when
the VT level in the registry is not 0. That default doesn't change.

For the VT alternate buffer, things works slightly differently, since
there is an expectation that VT modes are global. So when creating an
alt buffer, we copy the current modes from the main buffer, and when
it's closed, we copy them back again.

## Validation Steps Performed

I've manually confirmed that this fixes the problem described in issue
#14690. I've also added a basic feature test that confirms the modes are
initialized as expected when creating new buffers, and changes to the
modes in one buffer do not impact any other buffers.

Closes #14690
2023-01-27 20:38:37 +00:00
Dustin L. Howett
fc960e3327 Generalize OpenConsoleProxy's HybridCRT logic (#14733)
This pull request moves the Hybrid CRT logic out of the Host.Proxy
project and makes it available for all other projects in our solution.
2023-01-27 19:17:30 +00:00
Dustin L. Howett
6a610334af When generating a stacked changelog, note which branch/es contain a commit (#14728)
New-TerminalStackedChangelog used to generate logs that looked like this:

```
* [3] A commit that was seen 3 times
* A commit that was only seen once
* [2] Some other commit
```

Now it will generate logs that look like this:

```
   / base..branch-1
   |/ base..branch-2
   ||/ base..branch-3
* [XXX] A commit that was seen 3 times
* [ X ] A commit that was only seen once
* [XX ] Some other commit
```

This format is more expressive, as it indicates _which branches_ contain which commits.

As a reminder, my release note writing style starts with a stacked changelog. It's how I tell (1) which commits are in the new preview release only, (2) which commits are in the new preview and the new stable release and (3) which commits were already released in a previous stable release.

Changes from 1 get included in the new changelog, changes from 2 get included in both and changes from 3 get deleted because they have already been released.
2023-01-25 10:52:11 -08:00
Dustin Howett
7d0baa7946 Revert "Manually set the automation name of the default color scheme for screen reader (#14704)"
This reverts commit 47f38e31a1.
2023-01-23 19:01:49 -06:00
Carlos Zamora
a0e830cc1a [UIA] Dispatch a TextChanged event on new output (#14723)
For some reason, Windows Terminal stops dispatching UIA TextChanged events sometimes. There isn't a reliable repro for this bug.

However, under NVDA's logger, it appears that when the bug does occur, we still dispatch UIA notifications (which may be ignored by NVDA in some configurations). A "quick fix" here is to dispatch a TextChanged event if we're going to dispatch a notification. Since we're just enabling a flag, we won't send two events at once.

Closes #10911
2023-01-23 21:53:09 +00:00
Mike Griese
a4cf4e2761 Resize our ContentDialog's when the window resizes (#14722)
Major thanks to @dongle-the-gadget in https://github.com/microsoft/microsoft-ui-xaml/issues/3577#issuecomment-1399250405 for coming up with this workaround.

This PR will manually forward a `WM_SIZE` message to our `CoreWindow`, to trigger the `ContentDialog` to resize itself. 

We always closed these issues as dupes of the upstream one, so this doesn't actually close anything.
HOWEVER, these are the following issues that reported this bug:
- #2380
- #4463
- #5252
- #5810
- #6181
- #7113
- #7225
- #8245
- #8496
- #8643
- #9336
- #9563
- #5808
- #10351
- #10634
- #10995
- #11770
- #13796
2023-01-23 12:42:27 -06:00
Carlos Zamora
3dd40791c9 Ensure TermControl is not closing when firing UIA events (#14714)
The `SignalTextChanged` crash seems to be occurring due to the `TermControlAutomationPeer` being destructed by the time the UIA event is actually dispatched. Even though we're already checking if TCAP and TermControl still exist, it could be that the TermControl is being closed as text is being output.

The proposed fix here is to record when the closing process starts and exposing that information directly to the TCAP. If TCAP sees that we're in the process of closing, don't bother sending a UIA event.


Closes #13978
2023-01-21 15:48:33 +00:00
Dustin Howett
8f1960d0b4 version: bump to 1.18 on main 2023-01-20 15:24:08 -06:00
Dustin L. Howett
72be9a95af Code sign the contents of the Terminal package (#14710)
Up until now, we have been relying on the catalog signature produced for our MSIX package.
There are some things (Packaged COM, Process Explorer as of 2022) that cannot handle catalog-signed
files. It's easier and safer for us to simply sign all the executables we produce before packaging them.

Unfortunately, we can't do it before we package them. We have to unpack and re-pack our package.

In the future, this will allow us to provide a codesigned distribution that is not in an MSIX package.

TEST=Ran a build and checked out the contents of the package. They were all signed!

Closes #13294
Closes #12695
Closes #9670
2023-01-20 11:47:18 -06:00
Mike Griese
596d0c5155 Update the titlebar visibility when we gain/lose focus too (#14708)
We forgot to updateTheme again when we change the titlebar color. That would result in us setting the titlebar visibility only when the settings were reloaded, but if the unfocused BG was opaque, and the focused was transparent, we'd use the current _unfocused_ color to set the titlebar visibility. 

Also, this fixes a bug with `tabRow.BG: terminalBackground`

from teams for brevity 

> tabRow.BG = terminalBackground might not work, and here's why
>
> the termcontrol's BG brush might be (r,g,b, 255), with an Opacity of 0
> 
> so my "use mica when the brush's A is <1.0" doesn't work
> 

closes #14563
2023-01-20 01:38:34 +00:00
Mike Griese
79eb9b3d3c Use a dark titlebar when we've requested dark theme (#14536)
This applies to `"showTabsInTitlebar": false,`. We literally never set that for the `IslandWindow` before, only ever the NCIW. 

Closes #11589


![update-titlebar-for-theme](https://user-images.githubusercontent.com/18356694/207109370-a63a8b19-4c42-4b1f-8d39-8c3abdf1b403.gif)


For a dramatic example - here's the IW with mica enabled, in dark mode:

![image](https://user-images.githubusercontent.com/18356694/207109465-a6165637-31a5-45a4-bff0-51ac79404cd6.png)

Theme json:

```json
        {
            "name": "chonk",
            "tab":
            {
                "background": "#00000000",
                "unfocusedBackground": "#00000000",
                "showCloseButton": "never"
            },
            "tabRow":
            {
                "background": "#00000000",
                "unfocusedBackground": "#00000000",
            },
            "window":
            {
                "applicationTheme": "dark",
                "useMica": true
            }
        },
```
2023-01-19 23:51:21 +00:00
Joshua Boelter
eab1c239a9 Launch elevated instances via shell:AppFolder (#14637)
This uses `shell:AppsFolder` to launch elevated instances of the app via
`ShellExecuteEx` and `runas` in elevate-shim.exe. The app to launch is
discovered via the `GetCurrentApplicationUserModelId` API.

e.g. `shell:AppsFolder\WindowsTerminalDev_8wekyb3d8bbwe!App`

This will fallback to launching `WindowsTerminal.exe` if it fails to
discover the app user model id to launch.

This also fixes a bug in elevate-shim where the first argument of
WinMain was lost (e.g. `new-tab`). 

Curiously, `AppLogic::RunAsUwp()` is never called and
`AppLogic::IsUwp()` is always false when running debug builds locally
(e.g. WindowsTerminalDev). It's not clear if this is an artifact of
development packages or something else.

## Validation Steps Performed

Various manual debug/execution scenarios.

Verified the fallback path by running the unbundled app by extracting
the `CascadiaPackage_0.0.1.0_x64.msix` from the 'drop' build artifact.

Fixes #14501
2023-01-19 23:40:53 +00:00
PankajBhojwani
96a9dd5209 Automatically focus the ColorSchemeListView when we navigate to the ColorSchemes page (#14631)
This regressed when we implemented ColorSchemeViewModel in #13179.
This commit fixes that by automatically focusing the ColorSchemeListView
when we navigate to the ColorSchemes page, which makes sense from a
user experience perspective anyway.

Closes #11971
2023-01-19 23:20:56 +00:00
PankajBhojwani
6278a2d9bf Fix the selected color scheme list item container background being blue in Windows 10 (#14706)
Specify the resource to use for the list view item background when selected.

References #14693 

## Validation Steps Performed
Background is no longer blue, it is light gray (like in Windows 11). Screenshots below.
2023-01-19 22:04:28 +00:00
PankajBhojwani
16fe2e5905 Add color chips to the color scheme dropdown in Appearance (#14587)
Does what it says on the tin.
2023-01-19 21:46:59 +00:00
Mike Griese
f2b82cd054 Only use Mica in the SUI if Mica is available (#14675)
This should make sure we only use Mica for the BG of the SUI when we're on a Windows build that supports it. Otherwise, we're just gonna get the emergency backstop / full transparency under it. 

Confirmed this fixes it on my win10 VM.

Closes #14667
2023-01-19 21:06:49 +00:00
James Holderness
7813953b23 Add support for IRM (Insert Replace Mode) (#14700)
This PR add support for the ANSI Insert/Replace mode (`IRM`), which
determines whether output characters are inserted at the active cursor
position, moving existing content to the right, or whether they should
overwrite the content that is already there.

The implementation is a bit of a hack. When that mode is enabled, it
first measures how many cells the string is expected to occupy, then
scrolls the target line right by that amount before writing out the new
text.

In the longer term it might be better if this was implemented entirely
in the `TextBuffer` itself, so the scrolling could take place at the
same time as the content was being written.

## Validation Steps Performed

I've added a very basic unit test that verifies the mode is working as
expected. But I've also done a lot more manual testing, confirming edge
cases like wide characters, double-width lines, and both with and
without wrapping mode enabled.

Closes #1947
2023-01-19 19:59:05 +00:00
PankajBhojwani
47f38e31a1 Manually set the automation name of the default color scheme for screen reader (#14704)
## Summary of the Pull Request
When we navigate to the color schemes page, find the default color scheme (if present) and manually set the container's automation name to include the word 'default'. 

Note that we don't want to change the actual `ColorSchemeViewModel`'s name since that name is used internally to identify schemes. We only want to change the `ListViewItem`'s name, i.e. the container in the SUI.

## PR Checklist
* [x] Closes #14401 
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [x] I've discussed this with core contributors already. If not checked, I'm ready to accept this work might be rejected in favor of a different grand plan. Issue number where discussion took place: #xxx

## Validation Steps Performed
Screen reader reads out the 'default' text now. It also correctly reads out the new default scheme if the default scheme is changed via SUI or json.
2023-01-19 19:39:08 +00:00
Mike Griese
0fe0868f98 Remove the win10/win11 digit from our version number. (#14660)
As of about 2022, the one's digit of the Build of our version is a placeholder value to differentiate the Windows 10 build from the Windows 11 build. Let's trim that out, it's only a source of confusion.

For additional clarity, let's omit the Revision, which _must_ be `.0`, and doesn't provide any value to report.

We will need to make sure we report releases as `1.17.ABC` now, instead of `1.17.ABC1.0`/`1.17.ABC2.0`

Let's not backport this. 1.17 will be the start of the new numbering scheme. Otherwise, `1.16.EFG` < `1.16.ABC1.0`, for an old version `ABC1` and a new version `EFG`.

* closes #14106
* As summarized here: https://github.com/microsoft/terminal/issues/14106#issuecomment-1289462310
2023-01-19 19:32:23 +00:00
Dustin L. Howett
179bb9bded Add TerminalStress, Mike Treit's application for breaking WT (#14701)
From Treit/TerminalStress@39c03e2d00

Co-authored-by: Dustin L. Howett <duhowett@microsoft.com>
Co-authored-by: Mike Treit <mtreit@ntdev.microsoft.com>
2023-01-19 13:20:43 -06:00
Mike Griese
90485e4c79 Fix a crash on startup with a folder entry without entries (#14629)
`_Entries` was getting default constructed to `nullptr`. We should be careful about that. 

Adds a test too, and fixes a regression in the local tests introduced in #13763.

Closes #14557
2023-01-19 19:12:40 +00:00