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
(cherry picked from commit eab1c239a9)
Service-Card-Id: 87690177
Service-Version: 1.16
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
(cherry picked from commit 3dd40791c9)
Service-Card-Id: 87728855
Service-Version: 1.16
We originally needed this library (or a separate DLL in our own project)
to handle hooking up the XAML resource loader to the providers that our
application needed. It was introduced in its nascent form in 2019, in a
PR titled "Make XAML files work."
It appears we no longer need it, and the provider hookup is being
handled by our `AppT2` base class override. I've tested this in Windows
10 Vb running unpackaged, and it seems to work totally fine. Crazy.
Removing this dependency saves us a couple hundred kilobytes on disk and
removes one consumer of the App CRT from our package.
(cherry picked from commit 68cce101bc)
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#13294Closes#12695Closes#9670
(cherry picked from commit 72be9a95af)
Service-Card-Id: 87690424
Service-Version: 1.16
Some investigation revealed that `_keyEvents` would get a `NULL_POINTER_READ` error. On the main thread, `RecordKeyEvent()` would be called, which mainly updates `_keyEvents`. On the renderer thread, `NotifyNewOutput()` would be called, which starts by iterating through `_keyEvents` and slowly clearing it out. On occasion, these two threads are modifying `_keyEvents` simultaneously, causing a crash.
The fix is to add a mutex on this variable.
Closes#14592
(cherry picked from commit 8e041692b3)
Service-Card-Id: 87649541
Service-Version: 1.16
Basically what it says on the tin. For transparent tabs, we should layer on to the tab row to evaluate what the actual color of the tab will be. We did this for deselected tabs, but not for selected ones.
Gif below.
Closes#14561
(cherry picked from commit 4c7879bfb5)
Service-Card-Id: 87567529
Service-Version: 1.16
CascadiaSettings relies on getting a JsonUtils::DeserializationException
from the various JSON Utility functions, and then formatting that into
an error message. Well, DeserializationException tries to include an
object representation in its what() message . . . and generates an
exception trying to do so. CascadiaSettings never gets the
DeserializationException, and displays a weird message.
It's safe to remove the stringification in DeserializationException
because CascadiaSettings was never using it (_and_ because
CascadiaSettings was using an even better version of the same logic.)
Fixes#14373
(cherry picked from commit 239b4d16b9)
Service-Card-Id: 87568498
Service-Version: 1.16
In order to modify the accessibility information for the PseudoConsoleWindow, it needs to have a UIA provider registered. This PR introduces `PseudoConsoleWindowAccessibilityProvider` and registers it as a UIA provider appropriately. The registration process is based on that of the `WindowUiaProvider` for ConHost.
Closes#14385
## Validation Steps Performed
Run Accessibility Insights FastPass on the window. The PseudoConsoleWindow no longer is tagged as missing a name.
(cherry picked from commit 09273be1c8)
Service-Card-Id: 87568605
Service-Version: 1.16
Grab all paths from `DROPFILES` struct provided in drag event data
`GetStorageItemsAsync()` only giving up to 16 items when items are dropped from any archives
- When this occurs, we should look into `FileDrop` key for a stream of the [`DROPFILES struct`](https://learn.microsoft.com/en-us/windows/win32/shell/clipboard#cf_hdrop)
- This struct contains a null-character delimited string of paths which we can just read out
## Validation Steps Performed
* [X] Unit tests pass locally
* [X] Drag and drop paths work for both archives and non-archives files, folders, shortcuts, etc.
Closes#14628
(cherry picked from commit b7e537e5e7)
Service-Card-Id: 87548939
Service-Version: 1.16
This commit is just a slight refactor of `ConsoleProcessList` which I've noticed
was in a poor shape. It replaces iterators with for-range loops, etc.
Additionally this fixes a bug in `SetConsoleWindowOwner`, where it used to
default to the newest client process instead of the oldest.
Finally, it changes the process container type from a doubly linked list
over to a simple array/vector, because using linked lists for heap allocated
elements felt quite a bit silly. To preserve the previous behavior of
`GetProcessList`, it simply iterates through the vector backwards.
## Validation Steps Performed
* All unit/feature tests pass ✅
* Launching a TUI application inside pwsh inside cmd
and exiting kills all 3 applications ✅
(cherry picked from commit 391abafc2e)
Service-Card-Id: 87207825
Service-Version: 1.16
This changeset consists of two parts:
* Refactor `PtySignalInputThread` to move more code from `_InputThread`
into the various `_Do*` handlers. This allows us to precisely control
console locking behavior which is the cause of this bug.
* Add the 1-line fix to `_DoSetWindowParent` to unlock the console before
calling foreign functions (`SetWindowLongPtrW` in this case).
This fix is theoretical in nature, based on a memory dump from an affected user
and most likely fixes: https://developercommunity.visualstudio.com/t/10199439
## Validation Steps Performed
* ConPTY tests complete. ✅
(cherry picked from commit 3c78e01ab5)
Service-Card-Id: 87207821
Service-Version: 1.16
This is a partial fix for the Get-Credential issue. While investigating it, I found that the pseudoconsole window is not marked as being "owned" (in NTUSER) by the PID/TID of the console application that is "hosted" "in" it. Doing this does not (and cannot) fix `Get-Credential` failing in DefTerm scenarios.
ConsoleSetWindowOwner is one of the operations that can be done by a conhost to any window, and so the RemoteConsoleControl can call through to the Win32 ConsoleControl to pull it off.
I chose to add SetWindowOwner to the IConsoleControl interface instead of moving ConsoleControl::Control into the interface to reduce the amount of churn and better separate interface responsibilities.
References #14119
(cherry picked from commit 62ffa4ba41)
Service-Card-Id: 87207849
Service-Version: 1.16
This fixes two issues with `experimental.useBackgroundImageForWindow` I discovered while looking at #14260
* It looks like the opacity of the whole-window BG image wouldn't hot reload if the path didn't.
* > set useBGForWindow:true, focus a pane with an image, then set it to useBGForWindow:false, and observe a pane with <100 opacity. You'll be able to see the BG image left behind!
These are pretty easy to miss, so I can see how it happened.
I don't think this _technically_ closes that thread, though. Ultimately, I think OP's settings were just wrong (and possible didn't hot-reload). There's another, trickier bit I'm discussing in that thread, that might deserve its own separate follow-up for discussion.
(cherry picked from commit f2eed92345)
Service-Card-Id: 87207792
Service-Version: 1.16
#14160 didn't fix#14132 entirely. There seems to be a race condition left
where (on my system) 9 out of 10 times everything works correctly,
but sometimes OpenConsole exits, while pwsh and bash keep running.
My leading theory is that the new code is exiting OpenConsole faster than the
old code. This prevents clients from calling console APIs, etc. causing them
to get stuck. The old code (and new code) calls `ExitProcess` when the ConPTY
pipes break and I think this is wrong: In conhost when you close the window we
only call `CloseConsoleProcessState` via the `WM_CLOSE` event and that's it.
Solution: Remove the call to `RundownAndExit` for ConPTY.
During testing I found that continuously printing text inside msys2 will cause
child processes to only exit slowly one by one every 5 seconds.
This happens because `CloseConsoleProcessState` calls `HandleCtrlEvent` without
holding the console lock. This creates a race condition where most of the time
the console IO thread is the one picking up the `CTRL_CLOSE_EVENT`. But that's
problematic because the `CTRL_CLOSE_EVENT` leads to a `ConsoleControl` call of
type `ConsoleEndTask` which calls back into conhost's IO thread and
so you got the IO thread waiting on itself to respond.
Solution: Don't race conditions.
## Validation Steps Performed
* `Enter-VsDevShell` and close the tab
Everything exits after 5s ✅
* Run msys2 bash from within pwsh and close the tab
Everything exits instantly ✅
* Run `cat bigfile.txt` and close the tab
Everything exits instantly ✅
* Patch `conhost.exe` with `sfpcopy`, as well as `KernelBase.dll`
with the recent changes to `winconpty`, then launch and exit
shells and applications via VS Code's terminal ✅
* On the main branch without this modification remove the call to
`TriggerTeardown` in `RundownAndExit` (this speeds up the shutdown).
Run (msys2's) `bash.exe --login` and hold enter and then press Ctrl+Shift+W
simultaneously. The tab should close and randomly OpenConsole should exit
early while pwsh/bash keep running. Then retry this with this branch and
observe how the child processes don't stick around forever anymore. ✅
(cherry picked from commit b6b1ff8b2c)
Service-Card-Id: 87207832
Service-Version: 1.16
This is just a quick drive-by improvement. Switching from double to float
roughly doubles performance on a contemporary x86 CPU with `/fp:fast`.
## Validation Steps Performed
* Patch `RenderSettings.hpp` to include `Mode::AlwaysDistinguishableColors`
* Run a color intense application in AtlasEngine and observe CPU usage
(cherry picked from commit bbc14a0baf)
Service-Card-Id: 87207830
Service-Version: 1.16
When we added support for the `DECAC1` control sequence, which
determines whether `C1` controls are accepted or not, the intention was
that conhost would be making that determination, and Windows Terminal
would always be expected to accept any passed-through `C1` controls.
However, this didn't take into account that a passed-through `RIS`
sequence could end up disabling `DECAC1`, and that would leave Windows
Terminal incapable of processing any `C1` controls. This PR attempts to
fix that oversight.
The `DECAC1` sequence was added in PR #11690, when we disabled `C1`
acceptance by default.
This is a bit of a hack, but I've added a new `AlwaysAcceptC1` mode to
the state machine, which is enabled at startup in the Terminal, and is
never disabled. The parser then just needs to check whether either
`AcceptC1` or `AlwaysAcceptC1` are set.
## Validation Steps Performed
I've manually confirmed the test case in #13968 now works as expected.
Closes#13968
(cherry picked from commit f2b361c146)
Service-Card-Id: 87207767
Service-Version: 1.16
There is a condition which causes the console host process (conhost.exe)
to crash with a `FAIL_FAST` in `WriteCharsLegacy`.
**_Repro Scenario:_** Two conditions need to be met for crash to happen:
1. The `ENABLE_PROCESSED_OUTPUT` console mode needs to be disabled. This
condition is met through the race condition, explained in the section
below.
2. We are printing a string where there is a full width character
(character that requires two spaces on the screen) being printed at
the edge of the console window. That is, we have one character space
available, and the character requires 2 spaces.
Running following script (attached to bug) causes a crash:
`for /l %%A in (0, 1, 10000) do start /B C:\test.bat`
The script `test.bat` repeatedly prints a console-width line
with a DBCS character that doesn't fit.
_**Race:**_ Normally, we get into `WriteCharsLegacy` with
`PROCESSED_OUTPUT` enabled. However, during the initialization of a new
CMD session, `cmd!ResetConsoleMode()` is called, which first sets the
output console mode to the value of `curOutputMode` (which is a static
variable initialized to 0) by calling `SetConsoleMode()`, before then
setting it to the desired output mode with processed output enabled:
```c++
void ResetConsoleMode( void )
{
static DWORD desOutMode = ENABLE_PROCESSED_OUTPUT | /* ... */;
SetConsoleMode(conOut, curOutputMode); // <------ sets console mode to 0, disabling processed output
if (GetConsoleMode(conOut, &curOutputMode)) {
if ((curOutputMode & desOutMode) != desOutMode) {
curOutputMode |= desOutMode;
if (!SetConsoleMode(conOut, curOutputMode) && /* ... */) // <----- enables processed output
```
If there is another instance of CMD that is producing output in between
these two `SetConsoleMode()` calls, then we may end up in
`WriteCharsLegacy` with processed output mode disabled.
This fix removes a `FAIL_FAST_IF` that checks for `PROCESSED_OUTPUT`
mode after the initial character processing loop.
Before RS5, this was an `ASSERT()`. This FAIL_FAST was added in RS5 in
PR !1794053 (which changed all `ASSERT`-likes to `FAIL_FAST_IF`.)
We believe this assertion guarded only the "processing" of Backspace,
Tab, CR and LF, and did not expect that we would get out of the
character processing loop with unprocessed glyph characters. The
`FAIL_FAST` is redundant in this case, as the handlers for the
Backspace, Tab, CR and LF characters we are already checking for
`PROCESSED_OUTPUT`. Therefore, it is safe to remove this `FAIL_FAST`.
It turns out that we *can* exit the loop with unprocessed glyph
characters. In these cases, we don't want to `FAIL_FAST`.
# Validation
* Basic sanity testing to confirm strings are correctly being printed.
* Repro scenario script no longer crashes conhost.exe
Retrieved from https://microsoft.visualstudio.com os.2020 OS official/rs_we_adept_e4d2 eb5d8064dc0f09d8be92efb5e6efa69983f30a3f
Related work items: MSFT-42055103
(cherry picked from commit 86928bb48d)
(cherry picked from commit 788e42f482)
This changeset includes various guards against resizing the terminal down to 0
columns/rows: The 2 `TextBuffer` locations that accept new sizes, as well as
the `HwndTerminal::Refresh` which was the entrypoint causing the issue.
Closes#14404
(cherry picked from commit c5d417fcaf)
Service-Card-Id: 87020511
Service-Version: 1.16
Watson reports show that Visual Studio terminal attempts to render a string that is null causing the renderer to crash.
More specifically, we see "NULL_POINTER_WRITE_c0000005_PublicTerminalCore.dll!TextBuffer::WriteLine" with the following stack:
PublicTerminalCore!TextBuffer::WriteLine+0x1da
PublicTerminalCore!TextBuffer::Write+0x191
PublicTerminalCore!Microsoft::Terminal::Core::Terminal::_WriteBuffer+0x1d3
PublicTerminalCore!Microsoft::Terminal::Core::Terminal::PrintString+0x9
PublicTerminalCore!TerminalDispatch::PrintString+0x22
PublicTerminalCore!Microsoft::Console::VirtualTerminal::OutputStateMachineEngine::ActionPrintString+0x42
PublicTerminalCore!Microsoft::Console::VirtualTerminal::StateMachine::ProcessString+0x123
PublicTerminalCore!TerminalSendOutput+0x68
Microsoft_DotNet_MSBuildSdkResolver!DomainBoundILStubClass.IL_STUB_PInvoke(IntPtr, System.String)+0x8f
Microsoft_Terminal_Wpf!Microsoft.Terminal.Wpf.TerminalContainer.Connection_TerminalOutput(System.Object, Microsoft.Terminal.Wpf.TerminalOutputEventArgs)+0x20
Microsoft_VisualStudio_Terminal_Implementation!Microsoft.VisualStudio.Terminal.TerminalWindowBase+<>c__DisplayClass59_0+<<BeginProcessingPtyData>b__0>d.MoveNext()+0x55f
Internal bug: [Bug 1614709](https://devdiv.visualstudio.com/DevDiv/_workitems/edit/1614709): [Watson] crash64: NULL_POINTER_WRITE_c0000005_PublicTerminalCore.dll!TextBuffer::WriteLine
Added a null check before PInvoking TerminalSendOutput.
Validated locally that the check prevents null strings from rendering.
(cherry picked from commit 79c47f64a2)
Service-Card-Id: 87204854
Service-Version: 1.16
This commit adds the setting "confirmCloseAllTabs" to SUI.
The setting was added to the Interactions pane of Global Settings.
Closes#14413Closes#14033
(cherry picked from commit 437b5ac595)
Service-Card-Id: 87207138
Service-Version: 1.16
We already were setting the automation properties on the expander, however, we were not setting it on the content when an expander was present. This change applies the automation properties to both the expander and the child content (i.e. TextBox).
Closes#13827
(cherry picked from commit 62b34cf6f7)
Service-Card-Id: 87207144
Service-Version: 1.16
## Summary of the Pull Request
- Pipe the `ShowWindow` value through to `ConptyConnection`
- When `TerminalPage` receives the new connection, it checks the `ShowWindow` value and maximizes *IF* there were no other pre-existing tabs (in glomming mode, we don't want to maximize sessions that did not ask for it)
## References
#12154
## PR Checklist
* [ ] Closes #xxx
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [ ] Tests added/passed
* [ ] Documentation updated. If checked, please file a pull request on [our docs repo](https://github.com/MicrosoftDocs/terminal) and link it here: #xxx
* [ ] Schema updated.
* [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
## Detailed Description of the Pull Request / Additional comments
This is just a temporary solution until we change our defterm handoff process. Because of the way the process currently works, we have no way of knowing that the connection has requested the window to be maximized until after we have already started a terminal session. This means that we have to manually maximize the window upon receiving the connection, instead of having the session _start_ maximized, as it probably should.
## Validation Steps Performed
`start /max python` with defterm enabled opens up python in a maximized WT window
(cherry picked from commit 85ca8f556c)
Service-Card-Id: 87207152
Service-Version: 1.16
Override the center on launch setting when a position is specified on the commandline.
## Validation Steps Performed
1. Set center on launch in the SUI.
2. Run `wtd` - the new window is centered.
3. Run `wtd --pos 100,200` - the new window is positioned at (100,200).
Closes#14176
(cherry picked from commit 42befa7b58)
Service-Card-Id: 87207151
Service-Version: 1.16
In the most recent compiler ingestion into Windows ("LKG14"), we found
that this particular construction--checking an optional for a value
during this range-for loop--resulted in bad code generation.
When optimized, it generates code that looks effectively like this:
```c++
if (!newRowWidth.has_value()) {
while (true) {
// do the row stuff...
++it;
}
}
```
The loop never exits, and `_RefreshRowIDs` walks off the end of the
buffer. Whoops.
This commit fixes that issue by tricking the optimizer to go another
way. Leonard tells me it's harmless to call `Resize` a bunch of times,
even if it's a no-op, so I trust that this change results in the right
outcome with none of the crashing.
Fixes MSFT-41456525
Retrieved from https://microsoft.visualstudio.com os.2020 OS official/rs_we_adept_e4d2 c2b3697c867bddf5660da8b222e99ff4bfd1ea5b
(cherry picked from commit 3c104440a8)
Let's find out!
* [x] This is all the current bullets in #14324. I'm going to leave that open till we signoff on the release.
(cherry picked from commit 52cc523ed2)
Service-Card-Id: 86660872
Service-Version: 1.16
The following members were not initialized during construction:
* `CursorType _defaultCursorShape`
* `bool _suppressApplicationTitle`
* `bool _bracketedPasteMode`
* `size_t _hyperlinkPatternId`
* `SelectionExpansion _multiClickSelectionMode`
* `til::CoordType _scrollbackLines`
Unlike gcc and clang, MSVC is fairly tame when it comes to removing code
tainted by undefined behavior, so the most likely affect this had is that
we were reading uninitialized memory.
Related to #14129.
(cherry picked from commit a798a603e1)
Service-Card-Id: 86603221
Service-Version: 1.16
`TextBuffer` is buggy and allows a `Trailing` `DbcsAttribute` to be written
into the first column. Since other code then blindly assumes that there's a
preceding `Leading` character, we'll get called with a X coordinate of -1.
This issue will be fixed by #13626 and this commit fixes it in the meantime.
Additionally fixes an unimportant crash when the window height is 0px,
because it was annoying during testing and doesn't hurt to be fixed.
## Validation Steps Performed
* Run a stress test that prints random Unicode at random positions
* Resize the window furiously at the same time
* Doesn't crash / fail-fast ✅
(cherry picked from commit bb4711de54)
Service-Card-Id: 86353783
Service-Version: 1.16
XAML has an issue in windows 10 where `Width="*"` does not work properly
inside the `DataTemplate` for a `ListView`. Because of this, the color
scheme list view items looked very strange in windows 10 (see #14187).
Thanks to that, we thought up a few new designs for the color schemes
page and selected a new one that blends the color chips with a region
that shows the foreground and background color with the text preview.
Closes#14187
(cherry picked from commit 23a02c5218)
Service-Card-Id: 86518549
Service-Version: 1.16
Silent MIDI notes can be used to seemingly deny a user's input for long
durations (multiple minutes). This commit improves the situation by ignoring
all DECPS sequences for a second when Ctrl+C/Ctrl+Break is pressed.
Additionally it fixes a regression introduced in 666c446:
When we close a tab we need to unblock/shutdown `MidiAudio` early,
so that `ConptyConnection::Close()` can run down as fast as possible.
* In pwsh in Windows Terminal 1.16 run ``while ($True) { echo "`e[3;8;3,~" }``
* Ctrl+C doesn't do anything ❎
* Closing the tab doesn't do anything ❎
* With these modifications in Windows Terminal:
* Ctrl+C stops the output ✅
* Closing the tab completes instantly ✅
* With these modifications in OpenConsole:
* Ctrl+C stops the output ✅
* Closing the window completes instantly ✅
(cherry picked from commit b4fce27203)
Service-Card-Id: 86518585
Service-Version: 1.16
If the opacity is set to 100%, the background becomes solid instead of 'fully opaque acrylic'. If the opacity is below 100% the acrylic material is re-enabled (depending on the user's settings).
## Validation Steps Performed
I updated two unit tests to reflect the change in behavior and manually tested the transition from <100% opacity to 100% opacity (and vice versa) on win11.
Steps:
1. Start with 100% opacity and acrylic material enabled.
2. Decrease opacity and observe acrylic effect.
3. Increase opacity back to 100% and disable the acrylic effect.
4. Decrease opacity and notice that acrylic effect is no longer there.
Closes#12880
(cherry picked from commit 8ea3cb9972)
Service-Card-Id: 86444093
Service-Version: 1.16
## Summary of the Pull Request
This removes all of the redundant tooltips from the settings UI. Since all of the settings are added through the SettingsContainer, it's a pretty simple change.
Closes#14184
## Validation Steps Performed
- [X] hover over all settings in the settings UI
- [X] hover over all entries in the SUI nav view
(cherry picked from commit 3eaa781499)
Service-Card-Id: 86390069
Service-Version: 1.16
This commit makes the following improvements:
* Only adjust block characters that come from fallback fonts. This ensures
that the glyphs of the chosen font all look exactly as they were designed.
* When adjusting the size, use the fallback font's full block glyph U+2588
to determine the size that the given glyph should have.
Closes#14098
## Validation Steps Performed
* Print `UTF-8-demo.txt` in Consolas.
* All block glyphs look uniform. ✅
(cherry picked from commit 97abc3df1d)
Service-Card-Id: 86159056
Service-Version: 1.16
The `BSTR` arguments originate from COM calls who own them.
Found via AppVerifier.
(cherry picked from commit 2a5ce4900f)
Service-Card-Id: 86230696
Service-Version: 1.16
We'll just ping the window and give it a chance to respond before we
bunk with it.
Fixes#14131
(cherry picked from commit ab04067e49)
Service-Card-Id: 86230024
Service-Version: 1.16
It cannot be known why there is no site.
We should at least not crash.
Fixes MSFT-41571451
(cherry picked from commit 18e4e22394)
Service-Card-Id: 86206546
Service-Version: 1.16
Fix a missing entry in the JSON schema for `intenseTextStyle`.
The JSON schema was missing an entry in the Profile section for
`intenseTextStyle`. I have added it as it appears in AppearanceConfig.
Note that this is currently duplicated in the schema -- however, this is
the pattern used already in Profile as AppearanceConfig entries have
alternate descriptions (and I have updated the description in that
section to make it clear it applies to unfocused terminals).
Longer-term, it likely makes sense to consolidate all entires into
ApperanceConfig and rely on the description for the `unfocusedApperance`
object/the name of the object to make the limited scope of those keys
clear, so that Profile can simply extend ApperanceConfig and the
duplication in the schema can be reduced.
## Validation Steps Performed
Validation with schema verifying tools including VS Code.
Partially addresses #13387
(cherry picked from commit 33cb0eb05f)
Service-Card-Id: 86228743
Service-Version: 1.16
This pull request operates on the same theory as #14217, but at a lower
level. Carlos and I discovered that TerminalPage *already* has an
action-dispatching key preview handler, and that my implementation of
`IDirectKeyListener` handles focus-tree bubbling mostly correctly.
Because of that discovery, we learned we could move the
`IDirectKeyListener` into TerminalPage itself and not have to complicate
the SUI or the Command Palette with the DirectKey interface.
Validation:
When bound to Alt+Space, the system menu works in the command palette,
the settings UI, and in read-only panes.
Fixes#11970Closes#14217
Fixes MSFT-41390832
(cherry picked from commit d319d479c7)
Service-Card-Id: 86228470
Service-Version: 1.16
Problem:
* Calling `RundownAndExit` tries to flush out the last frame from `VtEngine`
* `VtEngine` indirectly calls `RundownAndExit` if the pipe is gone via `VtIo`
* `RundownAndExit` is called by other parts of OpenConsole
* `RundownAndExit` must be `[[noreturn]]` for most parts of OpenConsole
* `VtIo` itself has a mutex ensuring orderly shutdown
* In short, doing a thread safe orderly shutdown requires us to hold both,
a lock in `RundownAndExit` and `VtIo` at the same time, but while other parts
need a blocking `RundownAndExit`, `VtIo` needs a non-blocking one
* Refactoring the code to use optionally non-blocking `RundownAndExit`
requires refactoring and might prove to be just as buggy
Solution:
* Simply don't call `RundownAndExit` in `VtEngine` at all
* In case the write pipe breaks:
* `VtEngine` will close the handle
* The client should notice that their read pipe is broken and
close their write pipe sooner or later
* Once we notice that our read pipe is broken, we call `RundownAndExit`
* `RundownAndExit` might call back into `VtEngine` but
without a pipe it won't do anything
* In case the read pipe breaks or any other part calls `RundownAndExit`:
* We call `RundownAndExit`
* `RundownAndExit` might call back into `VtEngine` and depending on whether
the write pipe is broken or not it will simply write into it or ignore it
Closes#14132
Pretty sure this also applies to #1810
## Validation Steps Performed
* Open 5 tabs and run MSYS2's `bash --login` in each of them
* `Enter-VsDevShell` in another tab
* Close window
* 5 tab processes are killed instantly, 1 after ~3s ✅
* Replace conhost with OpenConsole via sfpcopy
* Launch Dozens of Git Bash tabs in VS Code
* Close them randomly
* Remaining ones still work, processes are gone ✅
(cherry picked from commit 1774cfd843)
Service-Card-Id: 86174637
Service-Version: 1.16
c0b2f488c1
Unlike iTerm2, we're not planning on making it configurable.
This commit also adds a max length of 1024 characters on the "display URI" that we pass up to XAML, so as to not inundate the UI.
Fixes#14200
(cherry picked from commit 07201d6cd1)
Service-Card-Id: 86182618
Service-Version: 1.16
Instead of using the currently focused tab when an unfocused tab is duplicated, the `_MakePane(...)` function now uses an optional source tab argument that points to the correct tab being duplicated.
## Validation Steps Performed
Manually tested on multiple tabs with different profiles. Performed steps:
* Construct at least two tabs with different profiles.
* Select `Duplicate Tab` option from the dropdown menu of the unfocused tab.
* Verify that the new tab has the same profile as the tab it was duplicated from.
Closes#13942
(cherry picked from commit 8f08bb04d0)
Service-Card-Id: 86159037
Service-Version: 1.16
The breadcrumbs in the SUI were not readable by screen readers because they are represented as a button with a text block inside of it. Turns out, if you make the DataTemplate's item `IStringable` (meaning it has a `ToString()`), it all magically works! Allowing the screen reader to read the button as text.
Closes#13826
(cherry picked from commit 30046dd4a7)
Service-Card-Id: 86158952
Service-Version: 1.16
It turns out that the negative margin for the progress ring is causing
the clipping in case the tab title gets too long:
43dbbd590f/src/cascadia/TerminalApp/TabHeaderControl.xaml (L18-L27)
The negative margin was introduced in #8113 because the progress ring is
supposed to replace the tab icon but the `TabView` still reserves space
even if no icon is set (see
https://github.com/microsoft/terminal/pull/8133#issuecomment-739098014).
However, it is not actually the `TabView` reserving space even when
there is no icon, but a workaround for a crash in the
`IconPathConverter` that returns a `BitmapIconSource` with a `nullptr`
source instead of a `nullptr` `IconSource`:
43dbbd590f/src/cascadia/TerminalSettingsModel/IconPathConverter.cpp (L143-L154)
The workaround in `IconPathConverter` could probably be removed as I did
not find any instance where it is still used in a way that could trigger
the mentioned crash, but I did not dare to just remove it as I do not
know enough about the code by far. Hence, I opted to just locally
instantiate the `IconSource` with a `nullptr` directly in `TerminalTab`.
Fixes#8910
(cherry picked from commit 21a9c55752)
Service-Card-Id: 86209056
Service-Version: 1.16
Adds UIA events to the WPF control for the following items:
- selection changed
- text changed (and output)
- cursor changed
### Automation Peer
Similar to the architecture of the UWP TermControl, we added a
`HwndTerminalAutomationPeer` which acts as the
`TermControlAutomationPeer` in UWP. However, we don't need a XAML
wrapper here, so really we just need it to inherit from
`TermControlUiaProvider` (the `ITextProvider` implementation shared
across conhost and WT) and `IUiaEventDispatcher` (the event dispatching
interface that is responsible for signaling the screen reader that
something has changed).
### Removing the local echo
As with WT, we need to record key events to remove the local echo. These
recorded events are matched up with the output text. Each sequential
match is removed in the output text so that it's not read by the screen
reader.
### Detecting what to send events for
As with WT, a `UiaEngine` was added to the renderer and it is set up
when a UIA client is detected. WT would normally stop sending events
when focus was lost from the control. We do the same here.
### Automation properties
`TermControlUiaProvider` was upgraded to support property values. Such
properties include class name and control type. These align with those
set in `TermControlAutomationPeer`. Realistically, those should point to
these, but that requires a lot more work and a localization burden
(because we need to move the localized word "terminal").
`HwndTerminalAutomationPeer` takes this a step further and overrides the
class name to be `WPFTermControl`. This allows screen readers to provide
special handling for the `WPFTermControl` vs the UWP term control since
they will be updating at different speeds.
### Build fixes
To build the WPF test app, I had to mess with the dependencies a little
bit. Really just add the atlas engine and uia renderer to the build
steps.
### HwndTerminal initialization
The initialization order with `WM_NCCREATE` was changed to match that of
Windows Terminal (BaseWindow/IslandWindow). This is safer now. I also
removed the `static` window because it was unnecessary.
### Handling `WM_GETOBJECT`
WPF's HwndHost likes to mark the `WM_GETOBJECT` message as handled to
force the usage of the WPF automation peer. We now explicitly mark it as
not handled and don't return an automation peer. This forces the message
to go down to the HwndTerminal where we return terminal's UiaProvider.
### Remove TermControl layer from UIA tree
TermContol (the top-most layer in the UIA tree) would pop up and not do
anything. This PR also overrides the automation peer at that layer and
marks IsContentElement/IsControlElement=false (the equivalent to
AccessibilityView=Raw). This makes the layer only appear in the UIA tree
if you are using the raw view (i.e. you know what you're doing and you
want to see each individual layer even if you can't directly interact
with it).
## Validation Steps Performed
Tested with Narrator/NVDA using WpfTerminalTestNetCore project in our
repo.
- [X] New output is read out (not just key events, but also other output
text)
- [X] Local echo does not occur (i.e. pressing 'A' should only read 'A'
once, not twice [key event and rendered letter]).
- [X] selection events are read out properly
- [X] cursor change events are read out properly (tested with text
cursor indicator preview in Settings App > Accessibility > Text
Cursor)
NOTE: test this with Release builds. Debug builds may be too slow and
not read out properly
Closes#12642
(cherry picked from commit 5608cf15a3)
Service-Card-Id: 86081129
Service-Version: 1.16
Paths that contain illegal path components will be dropped
in the dispatcher.
(cherry picked from commit fc0ef37977)
Service-Card-Id: 85900830
Service-Version: 1.16
This change fixes issues with certain fonts that draw _way_ outside the advance
width/height black box and expect to remain centered on the baseline.
## Validation Steps Performed
* Use MesloLGM Nerd Font
* Print U+E0B0
* Ensure it's centered, even if it's cut off ✅
(cherry picked from commit 2de475b36a)
Service-Card-Id: 85876935
Service-Version: 1.16
This commit changes the glyph scale algorithm to prefer aligning glyphs to
their baseline. This improves the visual appearance of simulated italic glyphs.
However wide Emojis in narrow cells now look slightly worse without centering.
Closes#13987
## Validation Steps Performed
* Use FiraCode which has no italic variant and instead uses simulated italics
* Write italic text
* Baseline is consistent ✅
(cherry picked from commit 97dc5c8d75)
Service-Card-Id: 85767343
Service-Version: 1.16
`ConptyClosePseudoConsole` blocks until OpenConsole exits.
This is problematic for the changes in 666c446, which stopped calling that
function on a background thread to solve a race condition. This commit fixes
the potential lags/deadlocks from waiting on OpenConsole's exit, by adding
`ConptyClosePseudoConsoleNoWait` which only closes the IO handles and allows
OpenConsole to exit naturally. This uncovered another potential deadlock
in `ServiceLocator::RundownAndExit` which might call itself recursively.
Closes#14032
## Validation Steps Performed
* Print tons of text and concurrently close the tab.
Tab closes, OpenConsole/pwsh exits instantly ✅
* Use `Enter-VsDevShell` and close the tab.
Tab closes instantly, OpenConsole/pwsh exits after ~5 seconds ✅
(cherry picked from commit 274bdb31da)
Service-Card-Id: 85767341
Service-Version: 1.16
When we first introduced the shell extension, it didn't work properly
for some folders (such as the Desktop, or perhaps any "background"
click) due to a bug in Windows. We worked around that bug with the help
of an awesome community member, who contributed code that would pull up
the topmost Explorer window and query its location.
That Windows bug was eventually fixed, but we still had trouble with
items appearing correctly. On Windows 11, the Open in Terminal menu item
appears and disappears at random when you right-click the desktop, but
it always appears when you right-click a folder. It sometimes appears
for Quick Access, even though it shouldn't.
We tried to fix that in #13206, but the fix caused more issues than it
solved. We reverted it for 1.15 and 1.16.
At the end of the day, it turns out that getting the path from the
toplevel explorer window is fragile. Fortunately, the shell does offer
us a way to get that information: the site chain.
This pull request replaces GetPathFromExplorer() with an implementation
of `IObjectWithSite`, which allows us to use the site chain to look up
from whence a context menu request was initiated. It also makes item
lookup generally more robust.
* ✅ Tested on Windows 11
* ✅ Desktop
* ✅ Folder Background
* ✅ Folder Selected
* ✅ Quick Access (does not appear)
* ✅ This PC (does not appear)
* ✅ Tested on Windows 10
* ✅ Desktop
* ✅ Folder Background
* ✅ Folder Selected
* ✅ Quick Access (does not appear)
* ✅ This PC (does not appear)
References #13206
References #13523Closes#12578
Co-authored-by: John Lueders <johnlue@microsoft.com>
(cherry picked from commit 5027c8031d)
Service-Card-Id: 85788409
Service-Version: 1.16
There's a different behavior regarding cursors between conhost and Windows
Terminal. In case of the latter we don't necessarily call `PaintCursor`
during cursor movement, because the cursor blinker never stops "blinking".
Closes#14028
## Validation Steps Performed
* Enter text until after the line wraps
* Hold backspace until the line unwraps
* No leftover cursor on the second line ✅
(cherry picked from commit 08096b2343)
Service-Card-Id: 85767353
Service-Version: 1.16
This fixes an issue with c51bb3a, where some fractional font
sizes are displayed as something like 13.600000000001.
Closes#14024
## Validation Steps Performed
* Enter a font size of 13.6 and save
* NumberBox displays "13.6" ✅
(cherry picked from commit f79276b21b)
Service-Card-Id: 85740785
Service-Version: 1.16
This commit fixes several issues:
* Some fonts set a line-gap even though they behave as if they
don't want any line-gaps. Since Terminals don't really have
any gaps anyways, it'll now not be taken into account anymore.
* Center alignment breaks bitmap glyphs which expect left-alignment.
* Automatic "opsz" axis makes Terminus TTF's italic glyphs look quite
weird. I disabled this feature as we might not need it anyways.
A complete fix depends on #14013Closes#14006
## Validation Steps Performed
* Use Terminus TTF at 13.5pt
* Print UTF-8-demo.txt
* No gaps between block characters ✅
(cherry picked from commit bea13bddf1)
Service-Card-Id: 85767355
Service-Version: 1.16
After this commit a user may specify fractional font sizes.
Support was only implemented for AtlasEngine however.
DxEngine continues to use rounded (integer) font sizes.
Closes#6678
## Validation Steps Performed
* Install a bitmap font that requires fractional font sizes
(e.g. Terminus TTF, https://files.ax86.net/terminus-ttf/)
* Set font size to something integer (e.g. 14pt)
Glyphs are blurry ✅
* Set font size to something fractional (e.g. 13.5pt)
Glyphs are crisp ✅
(cherry picked from commit c51bb3a7a6)
Service-Card-Id: 85708971
Service-Version: 1.16
If a rendering engine constantly throws error we'll effectively
denial-of-service our users by drowning them in warning popups.
This commit fixes the issue by limiting the retries in all cases.
Issue found in: #13985
## Validation Steps Performed
* Add a `THROW_HR(E_INVALIDARG);` in `AtlasEngine::StartPaint()`
* Launch Windows Terminal
* Only one warning popup shows up ✅
* Rendering is disabled until one clicks "resume" ✅
(cherry picked from commit 81e2bc98d1)
Service-Card-Id: 85653502
Service-Version: 1.16
The `TileHashMap` refresh via `makeNewest()` in `StartPaint()` depends
on us filling the entire `cellGlyphMapping` row with valid data.
This commit makes sure to initialize the `cellGlyphMapping` buffer.
Additionally it clears the rest of the row with whitespace
until proper `LineRendition` support is added.
Closes#13962
## Validation Steps Performed
* vttest's "Test of double-sized characters" stops crashing ✅
* No weird leftover characters ✅
(cherry picked from commit 16aa79d78d)
Service-Card-Id: 85653281
Service-Version: 1.16
Direct3D 10.0 and 10.1 only have optional support for shader model 4.
This commit fixes our assumption that it's always present by checking
`ComputeShaders_Plus_RawAndStructuredBuffers_Via_Shader_4_x` first.
Closes#13985
## Validation Steps Performed
* Set feature level to 10.1 via `dxcpl`
* `CheckFeatureSupport` is called and doesn't throw ✅
(cherry picked from commit e2b2d9b92c)
Service-Card-Id: 85653388
Service-Version: 1.16
Disables strictness and warnings as errors for custom pixel shaders in
RELEASE. Windows terminal is not telling the user why the shader won't
compile which makes it very frustrating for the shader hacker.
After trying the recent preview none of my shaders loaded anymore in
Windows Terminal Preview which made me very sad. I had no idea what was
wrong with them. After cloning the git repo, building it, fighting an
issue that prevent DEBUG SDK from being used I finally was able to
identify some issues that were blocking my shaders.
> error X3556: integer modulus may be much slower, try using uints if possible.
> error X4000: use of potentially uninitialized variable (rayCylinder)
While the first one is a good warning I don't think it is an error and
the tools I use didn't flag it so was hard to know.
The second one I was staring at the code and was unable to identify what
exactly was causing the issues, I fumbled with the code a few times and
just felt the fun drain away.
IMHO: I want it to be fun to develop shaders for windows terminal.
Fighting invisible errors are not fun. I am not after building
production shaders for Windows Terminal, I want some cool effects. So
while I am as a .NET developer always runs with Warning as errors I
don't think it's the right option here. Especially since Windows
Terminal doesn't tell what is the problem.
However, I understand if the shaders you ship with Windows Terminal
should be free of errors and silly mistakes, so I kept the stricter
setting in DEBUG mode.
## Validation Steps Performed
Loaded Windows Terminal in RELEASE and DEBUG mode and validated that
RELEASE mode had reduced strictness but DEBUG retained the previous more
restrictive mode.
(cherry picked from commit b4b6636b49)
Service-Card-Id: 85660397
Service-Version: 1.16
Our Windows branch name changed, and I took this opportunity to resolve
an issue where vpack builds would occasionally fail due to GitHub rate
limiting the Azure DevOps IP addresses.
(cherry picked from commit 3958c938af)
Service-Card-Id: 85567589
Service-Version: 1.16
`ATLAS_POD_OPS` doesn't check for `has_unique_object_representations` and so a
bug exists where `CachedCursorOptions` comparisons invoke undefined behavior.
The color of inactive tab text is incorrect since #13689 due to the introduction
of `til::color::layer_over` which incorrectly calculated the RGB values.
## Validation Steps Performed
* Added unit tests ✅
Co-authored-by: Leonard Hecker <lhecker@microsoft.com>
This commit reduces the amount of telemetry during general usage by about half.
8 events that weren't really used anymore were removed.
1 new event was added ("AppInitialized") which will help us investigate #5907.
During review 9 events were found that were incorrectly tagged as perf. data.
## Validation Steps Performed
* Launch Windows Terminal
* The "latency" field "AppInitialized" matches the approx. launch time ✅
When Leonard updated CLI11 in #12658, he imported a version that no
longer requires RTTI. Since CLI11 was the only reason we had enabled
RTTI in any of our projects, we're finally able to turn it off.
| | TerminalApp.dll | MSIX Package |
| ------ | ---------------:| ------------:|
| Before | 3180KiB | 6545KiB |
| After | 1970KiB | 6426KiB |
| Delta | **-1210KiB** | **-119KiB** |
With AtlasEngine being enabled by default in 1.16 Preview it would look weird
if the `useAtlasEngine` option would still be called "experimental".
Additionally we're interested in how many users opt out of useAtlasEngine,
indicating major issues that would require us to disable it by default again.
Related to #13936
## Validation Steps Performed
* Toggling `useAtlasEngine` works as expected ✅
* Observe new event with TVPP ✅
- fix: if a selection exists, mark mode should promote the existing selection instead of creating one at the cursor
- fix: mark mode --> move to middle of a word --> ExpandSelectionToWord --> Shift+Right/Left had some weird behavior. Fix that
- fix: Ctrl+enter on random selection clears selection. It should try to treat selection as a URL.
## References
#13854
NOTE: 3b53f3c can be serviced
As noted by the `winrt::event` documentation:
> [...] But for asynchronous events, even after revoking [...], an in-flight
> event might reach your object after it has started destructing.
This is because while adding/removing/calling event handlers might be
thread-safe, there's no guarantee that they run mutually exclusive.
This commit fixes the issue by reverting 6f0f245. Since we never checked
the result of closing a terminal connection anyways, this commit simply drops
the wait on the connection being teared down to ensure #1996 doesn't regress.
Closes#13880
## Validation Steps Performed
* Open tab, close tab, open tab, close tab, open tab, close tab
* ConPTY ✅
* Azure ✅
* Closing a tab with a huge amount of panes ✅
* Opening a bunch of tabs and then closing the window ✅
* Closing a tab while it's busy with VT ✅
* `wtd -w 0 nt cmd /c exit` ✅
* `wtd -w -1 cmd /c exit`
* No WerFault spawns ✅
This PR adds support for the `DECBKM` sequence, which provides a way for
applications to specify whether the `Backspace` key should generate a
`BS` character or a `DEL` character.
On the original VT100 terminals, the `Backarrow` key generated a `BS`,
and there was a separate `Delete` key that generated `DEL`. However, on
the VT220 and later, there was only the `Backarrow` key. By default it
generated `DEL`, but from the VT320 onwards there was an option to make
it generate `BS`, and the `DECBKM` sequence provided a way to control
that behavior programmatically.
On modern terminals, the `Backspace` key serves as the equivalent of the
VT `Backarrow`, and typically generates `DEL`, while `Ctrl`+`Backspace`
generates `BS`. When `DECBKM` is enabled (for those that support it),
that behavior is reversed, i.e. `Backspace` generates `BS` and
`Ctrl`+`Backspace` generates `DEL`.
This PR also gets the other `Backspace` modifiers more closely aligned
with the expected behavior for modern terminals. The `Shift` modifier
typically has no effect, and the `Alt` modifier typically prefixes the
generated sequence with an `ESC` character.
While not strictly related to `DECBKM`, I noticed while testing that the
`_SetInputMode` method was doing unnecessary work that was specific to
the `FocusEvent` mode. I've now moved that additional processing into
the `EnableFocusEventMode` method, which I think makes things somewhat
simpler.
## Validation Steps Performed
I've tested the basic `DECBKM` functionality in Vttest, and I've
manually tested all the modifier key combinations to make sure they
match what most modern terminals generate.
I've also added a unit test that confirms that the expected sequences
are generated correctly when the `DECBKM` mode is toggled.
Closes#13884
In testing the following issues were found in AtlasEngine and fixed:
1. "Toggle terminal visual effects" action not working
2. `d2dMode` failed to work with transparent backgrounds
3. `GetSwapChainHandle()` is thread-unsafe due to it being called outside
of the console lock and with single-threaded Direct2D enabled
4. 2 swap chain buffers are less performant than 3
5. Flip-Discard and `Present()` is less energy efficient than
Flip-Sequential and `Present1()`
6. `d2dMode` used to copy the front to back buffer for partial rendering,
but always redraw the entire dirty region anyways
7. Added support for DirectX 9 hardware
8. If custom shaders are used not all pixels would be presented
Closes#13906
## Validation Steps Performed
1. Toggling visual effects runs retro shader ✅
With a custom shader set, it toggles the shader ✅
Toggling `experimental.rendering.software` toggles the shader ✅
2. `"backgroundImage": "desktopWallpaper"` works with D2D ✅ and D3D ✅
3. Adding a `Sleep(3000)` in `_AttachDxgiSwapChainToXaml` doesn't break
Windows 10 ✅ nor Windows 11 ✅
4. Screen animations run at 144 FPS ✅ even while moving the window ✅
5. No weird artefacts during cursor movement or scrolling ✅
6. No weird artefacts during cursor movement or scrolling ✅
7. Forcing DirectX 9.3 in `dxcpl` runs fine ✅
This reverts commit f785168aac (PR #13244)
The error logged to NVDA was caused by the following line of code in `_getTextValue()`:
`THROW_HR_IF(E_FAIL, !bufferSize.IsInBounds(_start) || !bufferSize.IsInBounds(_end));`
NVDA would expand a text range to encompass the document in the alt buffer. This means that the "end" would be set to the dangling "endExclusive" point (x = left, y = one past the end of the buffer). This is a valid range!
However, upon extracting the text, we would hit the code above. The exclusive end doesn't actually point to anything in the buffer, so we would falsly throw `E_FAIL`.
Though this could be fixed by adding a special check for the `endExclusive` in the line above, I suspect there are other places throughout the UIA code that hit this problem too. The safest course of action is to revert this commit entirely since it was a code health commit (it doesn't actually close an issue).
Closes#13866
This is conjecture - I was totally unable to repro the original crash here.
Based on the stacks in MSFT:39994969, it looks like we try to fire off a
`RaiseAutomationEvent`, which calls through UIA core, eventually to the point of
calling `ComPtr<WUX::Automation::Peers::IAutomationPeer>::{dtor}`. I'm guessing
based on the stacks that the TermControl has already been released and cleaned
up. However, the lambda in the `RunAsync` calls here only takes a ref to the
TCAP. The TCAP has an outstanding reference (maybe on the other side of the UIA
fence), and gets successfully resolved as strong, but when calling to
`RaiseAutomationEvent`, the `owner` we passed in is gonezo.
This explicitly passes a `weak_ref` to `TermControlAutomationPeer`, rather than
a raw ptr, so we can actually check if the control is still alive before _we_
dereference it. If it is, great, we've got a strong ref to it now and it won't
get torn down.
Again, this is hearsay. Without a repro, the only way we can confirm this is
gone is by just hoping the crashes go away. 🤞
* Might close#13357 (we'll reopen if it doesn't?)
* narrator still works
## Summary of the Pull Request
Fix a bug where if you pressed the "Save" button, WT would crash. This was caused by adding the possibility that no color scheme is selected in the main page. With no "current scheme", attempting to get its "name" would cause a null ptr exception.
The fix is simple: just check if we actually have a current scheme.
Bonus points: if we don't have a current scheme, don't bother looking throught the color schemes for a match because we'll never find one.
## References
#13269 - Color Schemes Rejuv
## Summary of the Pull Request
As described in #9583, this change implements the legacy conhost "EnableColorSelection" feature.
## Detailed Description of the Pull Request / Additional comments
@zadjii-msft was super nice and provided the outline/plumbing (WinRT classes and such) as a hackathon-type project (thank you!)--a "SelectionColor" runtimeclass, a ColorSelection method on the ControlCore runtimeclass, associated plumbing through the layers; plus the action-and-args plumbing to allow hooking up a basic "ColorSelection" action, which allows you to put actions in your settings JSON like so:
```json
{
"command":
{
"action": "experimental.colorSelection",
"foreground": "#0f3"
},
"keys": "alt+4"
},
```
On top of that foundation, I added a couple of things:
* The ability to specify indexes for colors, in addition to RGB and RRGGBB colors.
- It's a bit hacky, because there are some conversions that fight against sneaking an "I'm an index" flag in the alpha channel.
* A new "matchMode" parameter on the action, allowing you to say if you want to only color the current selection ("0") or all matches ("1").
- I made it an int, because I'd like to enable at least one other "match mode" later, but it requires me/someone to fix up search.cpp to handle regex first.
- Search used an old UIA "ColorSelection" method which was previously `E_NOTIMPL`, but is now wired up. Don't know what/if anything else uses this.
* An uber-toggle setting, "EnableColorSelection", which allows you to set a single `bool` in your settings JSON, to light up all the keybindings you would expect from the legacy "EnableColorSelection" feature:
- alt+[0..9]: color foreground
- alt+shift+[0..9]: color foreground, all matches
- ctrl+[0..9]: color background
- ctrl+shift+[0..9]: color background, all matches
* A few of the actions cannot be properly invoked via their keybindings, due to #13124. `*!*` But they work if you do them from the command palette.
* If you have "`EnableColorSelection : true`" in your settings JSON, but then specify a different action in your JSON that uses the same key binding as a color selection keybinding, your custom one wins, which I think is the right thing.
* I fixed what I think was a bug in search.cpp, which also affected the legacy EnableColorSelection feature: due to a non-inclusive coordinate comparison, you were not allowed to color a single character; but I see no reason why that should be disallowed. Now you can make all your `X`s red if you like.
"Soft" spots:
* I was a bit surprised at some of the helpers I had to provide in textBuffer.cpp. Perhaps there are existing methods that I didn't find?
* Localization? Because there are so many (40!) actions, I went to some trouble to try to provide nice command/arg descriptions. But I don't know how localization works…
<!-- Please review the items on the PR checklist before submitting-->
## PR Checklist
* [x] Closes#9583
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [ ] Tests added/passed *(what would be the right place to add tests for this?)*
* [ ] Documentation updated. If checked, please file a pull request on [our docs repo](https://github.com/MicrosoftDocs/terminal) and link it here: #xxx
* [ ] Schema updated. *(is this needed?)*
* [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
Just manual testing.
This commit stores a hash of the `settings.json` file in `ApplicationState`
with which we can detect whether the settings contents actually changed.
Since I only use a small 64-bit hash as opposed to SHA2 for instance,
I'm taking the last write time of the file into account as well.
This allows us to skip calling `UpdateJumplist` at least the majority of app
launches which hopefully improves launch performance on devices with slower IO.
Part of #5907.
## Validation Steps Performed
* Delete some profiles (see above), save settings, tasks are gone ✅
FYI For some (...) inexplicable reason, shell task lists are preserved forever
even if msix applications are uninstalled, etc. So to test whether tasks are
properly written on first app launch we have to delete some profiles/tasks
first, otherwise we can't see whether they're actually written later.
* Now exit Windows Terminal, delete `settings.json` and relaunch
* All tasks are back ✅
* With a debugger, ensure that `CascadiaSettings::WriteSettingsToDisk`
generates the same hash that `LoadAll` reads. ✅
Update the color schemes page with the new Win 11 style.
With the rejuvenated experience, we now have two pages:
- `ColorSchemes.xaml`: "color scheme selection" page
- This is the starting page for the "color schemes" nav view
item. It is intended to have the user select a color scheme
they want to modify. The user can also click the "Add new"
button to add a new color scheme or the "delete" button to
delete the selected scheme.
- If a scheme cannot be deleted, the delete button is disabled
and a disclaimer is shown.
- A "set as default" button sets the selected scheme as the
color scheme for profiles.default (aka "base layer").
- The list view item for each scheme includes the name of the
scheme, the default tag (if the scheme is the one set in base
layer), a text preview of the foreground/background, and a
grid of 16 color chips showing the colors for the scheme.
- Implementation details:
- View - `ColorSchemes`:
- "Enter" --> edit the selected scheme
- "Delete" --> delete the selected scheme
- if the selected scheme cannot be deleted, we
show the disclaimer
- View model - `ColorSchemesPageViewModel`
- when possible, the XAML binds directly to the
view model functions. Thus, we include logic
to delete, edit, and set the selected scheme
as default.
- store the current page, so that we know which
page to navigate to upon saving/discarding
changes.
- `EditColorScheme.xaml`: "color scheme modification" page
- a terminal preview of the color scheme is shown at the top of
the page.
- all colors for the scheme are displayed as color chips in an
expander that starts as expanded.
- renaming a color scheme is also inside an expander, but
there's no need for "renaming mode" anymore
- Implementation details:
- View - `EditColorScheme`:
- include logic to display the disclaimer and
add the automation properties
- include logic for "Enter" and "Escape" in the
rename editor
- View Model - `ColorSchemeViewModel`:
- as before, when possible, the XAML binds
directly to the view model functions.
- To enable the "default" tag functionality, we
had to expose knowledge of being the default
via "IsDefaultScheme()" which compares the
current name to the one in the settings model.
- Miscellaneous implementation details:
- to get the expander to start as expanded, we had to modify the
setting container style.
- Since "set as default" is a button on the selector page, we
needed a way to refresh the view model's knowledge of being
the default. So we added a `RefreshIsDefault` API to the
`ColorSchemeViewModel` to notify changes.
- With the new layout, we no longer need an 'enter rename mode'
button, so all that logic has been removed
- Add logic to `MainPage` to handle navigating to the correct
page upon saving/discarding changes.
Closes#9775
Co-authored-by: Carlos Zamora <cazamor@microsoft.com>
See also:
33732458ed/dev/SplitButton/SplitButton.xaml (L290-L293)
We need to
* Also set `SplitButtonForegroundSecondary` cause SplitButton's use that resource separately from the `Foreground()` property
* Manually trigger the visual state change, to refresh the brushes. We do something similar in TabBase
This is one of the bullet points in #13725
This commit implements support for `experimental.rendering.software`.
There's not much to it. It's just another 2 if conditions.
## Validation Steps Performed
* `"experimental.rendering.software": false` renders with D3D ✅
* `"experimental.rendering.software": true` triggers the new code path ✅
This commit implements support for custom shaders in AtlasEngine
(`experimental.retroTerminalEffect` and `experimental.pixelShaderPath`).
Setting these properties invalidates the device because that made it
the easiest to implement this less often used feature.
The retro shader was slightly rewritten so that it compiles without warnings.
Additionally we noticed that AtlasEngine works well with D3D 10.0 hardware,
so support for that was added bringing feature parity with DxRenderer.
Closes#13853
## Validation Steps Performed
* Default settings (Independent Flip) ✅
* ClearType (Independent Flip) ✅
* Retro Terminal Effect (Composed Flip) ✅
* Use wallpaper as background image (Composed Flip) ✅
* Running `color 40` draws everything red ✅
* With Retro Terminal Effect ✅
When selecting candidate history, priority is given to histories with empty commands list.
I think lazily calling `s_Allocate` will require a ton of rewriting. Also I think it is not an unexpended behavior that buffer is allocated when an application connects.
Closes#13571
## Summary of the Pull Request
Introduces a new action `expandSelectionToWord()` which expands the beginning and end of the selection to encompass the word(s) it's on. This was implemented as a conditional keybinding where the key chord is passed through to the terminal if no selection is active (similar to `copy()`).
It is not bound to anything by default.
## PR Checklist
* [x] Closes#8274
* [x] Schema updated.
## Validation Steps Performed
- Scenario in #8274:
- search for some text in the find dialog
- ESC to close the dialog
- execute `expandSelectionToWord()`
- the new selection encompasses the whole word
- mark mode
- move onto a word
- execute `expandSelectionToWord()`
- mouse selection (same as above)
- select a portion of two words --> new selection fully encompasses both words
Direct2D is able to detect remote connections and will switch to sending
draw commands across RDP instead of rendering the data on the server.
This reduces the amount of data that needs to be transmitted as well
as the CPU load of the server, if it has no GPU installed.
This commit changes `AtlasEngine` to render with just Direct2D if a software or
remote device was chosen by `D3D11CreateDevice`. Selecting the DXGI adapter the
window is on explicitly in the future would allow us to be more precise here.
This new rendering mode doesn't implement some of the more fancy features just
yet, like inverted cursors or coloring a single wide glyph in multiple colors.
It reuses most existing facilities and uses the existing tile hash map to cache
DirectWrite text layouts to improve performance. Unfortunately this does incur
a fairly high memory overhead of approximately 25MB for a 120x30 viewport.
Additional drive-by changes include:
* Treat the given font size exactly as its given without rounding
Apparently we don't really need to round the font size to whole pixels
* Stop updating the const buffer on every frame
* Support window resizing if `debugGeneralPerformance` is enabled
Closes#13079
## Validation Steps Performed
* Tested fairly exhaustively over RDP ✅
More or less, as in #13554
* Dark theme by default (was `system`)
* Justification:
> I think the interesting thing that we have today is that the color
> scheme is dark by default, but our window theme is set to system.
> System theme in Windows 11 is defaulted to light unless changed by
> the user. Now, we have a conflict between the theme and color scheme
> in Terminal.
>
> I think our options become, make the color scheme match "default"
> and set it to a light color scheme if the system theme is light, or
> manually match the theme to the color scheme by setting it to dark.
>
> Given that Terminal has historically had a black background with
> its Campbell color scheme, for a consistent UI experience, I'm
> voting to change the window theme to be dark by default as well.
* `tab.background: terminalBackground`
* Change the tab row colors back to "what they were pre-Controlsv2" more
or less
- The ramp is based off of "Light tab row background from Edge's
unfocused titlebar color, inactive tab row color BG from the
controlsv1 colors"
* `tabRow.unfocusedBackground` set `Transparent`
Closes#13554
Does what it sounds like on the label.
This is important, because when unset, the tab will use the active `Background` color to create an inactive BG, which maybe isn't what we want. Consider:
- a bright cyan active BG,
- and `terminalBG` for the `tabRow` bg.
Without an unfocusedBackground setting, all the tabs will still appear cyan when unfocused, which is extra gross.
As a judgement call, I made `terminalBackground` and `accent` use 30% opacity when set, to match the existing coloration.
See also #13554. If we want to make the default theme `tab.background: terminalBackground`, we should make `tab.unfocusedBackground` transparent (`#00000000`) by default. Otherwise, a Campell (`#0c0c0c`) tab on _any_ tab row will still have a faint tab visible.
* closes#13684
* closes#13246
* tested manually
This also does a lot of code shuffling, to get SettingsUI tabs to behave sensibly. We want those tabs to have (`#0c0c0c`, `#ffffff`) colored BGs for `terminalBackground` (see mail thread).
We also don't want dark focused tabs colors, combined with light tab row colors, combined with transparent unfocused tabs, to result in unfocused tabs having white-on-white text. That's gross. So that's been added to this PR in b38b704.
This pull request adds the JSON schema for `cgmanifest.json`.
## FAQ
### Why?
A JSON schema helps you to ensure that your `cgmanifest.json` file is valid.
JSON schema validation is a build-in feature in most modern IDEs like Visual Studio and Visual Studio Code.
Most modern IDEs also provide code-completion for JSON schemas.
### How can I validate my `cgmanifest.json` file?
Most modern IDEs like Visual Studio and Visual Studio Code have a built-in feature to validate JSON files.
You can also use [this small script](https://github.com/JamieMagee/verify-cgmanifest) to validate your `cgmanifest.json` file.
### Why does it suggest camel case for the properties?
Component Detection is able to read camel case and pascal case properties.
However, the JSON schema doesn't have a case-insensitive mode.
We therefore suggest camel case as it's the most common format for JSON.
### Why is the diff so large?
To deserialize the `cgmanifest.json` file, we use [`JSON.parse()`](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/JSON/parse).
However, to serialize the JSON again we use [`prettier`](https://prettier.io/).
We found that, in general, it gave smaller diffs than the default [`JSON.stringify()`](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/JSON/stringify) function.
Since locking the console can take a non-trivial amount of time,
the main thread might have already released the ControlCore instance, while the
`UpdatePatternLocations` background thread is holding the last active reference.
If the function call goes out of scope, we destroy the instance, which might
not be safe, considering its members are usually only used by the main thread.
This commit fixes the issue by only holding a reference of the `Terminal`.
## Validation Steps Performed
* Patterns are recognized ✅
See https://github.com/microsoft/terminal/issues/12176#issuecomment-1199488906, and MSFT:39723014.
I have literally no idea how to repro this one, or debug it. The dump I looked at looked like there was a `SwapChainScaleChanged` that was being dispatched as the app was tearing down. The `ControlCore` had started closing, but the `TermControl` hadn't yet. Apparently, just none of the `_refreshSizeUnderLock` callers checked if we were already closing.
All the callers appear to be on the main thread.
Closes#12176
Since there's no real way for me to repro this manually, I'm thinking we fire this fix off to the OS terminal build, where we'll pretty quickly be able to see if this fixed it or not.
While having a debugger attached, opening the settings tab generates an
uncomfortable amount of exceptions. This change reduces this by a lot.
## Validation Steps Performed
* Icons still appear ✅
This PR introduces a new action for closing all unfocused / other panes within a tab.
**Edit (17-08-2022):** Read-only panes are ignored when this action is being used (i.e., left open).
## Validation Steps Performed
The action was manually tested by applying it to various 'compositions' of panes. This includes tests on read-only panes.
Closes#12216
This PR by itself doesn't _really_ change much. Technically, now the Terminal will respect the Title of a `.lnk` when started for defterm, but we don't do anything else yet. Primarily, the goal of this PR is to just wire up startup info in OpenConsole to the connected Terminal.
* This required a bit of changes in `srvinit.cpp:ConsoleEstablishHandoff` to replicate other bits of startup, where we crack open the connect message to get the relevant bits of info.
* We pack that all into a `TERMINAL_STARTUP_INFO`, which we pass along to the registered terminal application.
* `ConptyConnection` accepts the handoff, and gathers that information out of the `TERMINAL_STARTUP_INFO`
* Some other updates to the scratch sln were made to make it build again (related, but unimportant).
* This is a precursor to:
* #13111
* #12154
* Closes#9458
* Tested manually
* I work here
## Summary of the Pull Request
The command palette (and tab search by extension) doesn't ever tell screen readers what is selected. Here, we simply hook up the selection changed event to a function that tells the screen reader what is selected. With this, the user no longer has to tab into the list view to know what is selected!
Will resolve the following bug upon validation from a11y team: #12065
## Validation Steps Performed
Performed repro steps from #12065.
NOTE: we do NOT read the selected item when the command palette is first opened. I think that's ok.
I got tired of renaming the packages that came out of our build
pipeline. I also got tired of the fact that every appxbundle artifact we
upload comes with another whole copy of Terminal for every architecture,
plus all their symbols. Those are reflected in other artifacts, so
there's no reason to duplicate them.
## Summary of the Pull Request
Polishes #13405 with some additional feedback found in testing and post-mortem reviews. Such feedback includes:
- [x] ControlInteractivity vs ControlCore split ([link](https://github.com/microsoft/terminal/pull/13405#discussion_r919365435))
- [x] clearing the selection should be under lock when copying text via "Enter"
- [x] move mark mode keybindings into a helper function
- [x] decide if "Enter" should be configurable or non-configurable ([link](https://github.com/microsoft/terminal/pull/13405#discussion_r919379305))
- [x] rename `_isTargetingUrl`
- [x] bugfix: ctrl+enter when the link is outside of the viewport
## References
Original PR: #13405
Relevant issue: #6649
Epic: #4993
Apparently, calling `GetText(INT_MAX)` causes a HUGE memory spike for a few seconds each time this is called. The [UIA docs](https://docs.microsoft.com/en-us/windows/win32/api/uiautomationcore/nf-uiautomationcore-itextrangeprovider-gettext#parameters) say to put `-1` if no limit is required, but I assume a few people have been hit by this before.
This addresses this issue (and similar ones) in two ways:
1. as we iterate over the lines of text, if we're already past the max length, just break out of the loop
2. _only_ resize if the max length is actually less than the current length. This prevents us padding the string with `L'\0'` erroneously (which is probably what causes the memory spike).
While debugging #13694, we discovered a very subtle bug we had accidentally introduced in a few places. `ViewModelHelper` defines a `PropertyChanged` event, backed by a `_propertyChangedHandlers` `event`. All opbservable properties in the viewmodels are supposed to run through that event. However, if you do `WINRT_CALLBACK(PropertyChanged, Windows::UI::Xaml::Data::PropertyChangedEventHandler)` in a derived class, it'll override that original method with the new one. XAML will subscribe to the second one, which is backed by `_PropertyChangedHandlers`, but the properties will still raise notifications on the callbacks registered to `_propertyChangedHandlers`.
This change makes it more explicit in these derived classes, that the `PropertyChanged` method exposed by these classes is indeed the one that's implemented in the base class.
This is a bit of a footgun, for sure. AuditMode would have apparently caught this, as we'd be overriding a method without using the `override` keyword.
Unblocks #13694.
In Global Appearance SUI, the toggle switch for "Always show tabs" is now grayed out when "Hide the title bar" is enabled (the default).
Also, the order of the settingcontainers was changed to help show the cause and effect more. See [this comment](https://github.com/microsoft/terminal/issues/12873#issuecomment-1094928881).
Closes#12873
After some deliberation I noticed that rounding the glyph advance up to yield
the cell width is at least just as wrong as rounding it. This is because
we draw glyphs centered, meaning that (at least in theory) anti-aliased
pixels might clip outside of the layout box on _both_ sides of the glyph
adding not 1 but 2 extra pixels to the glyph size. Instead of just `ceilf`
we would have had to use `ceilf(advanceWidth / 2) * 2` to account for that.
This commit simplifies our issue by just going with what other applications do:
Round all sizes (cell width and height) to the nearest pixel size.
Closes#13812
## Validation Steps Performed
* Set a breakpoint on `scaling Required == true` in `AtlasEngine::_drawGlyph`
* Test an assortment of Cascadia Mono, Consolas, MS Gothic, Lucida Console
at various font sizes (6, 7, 8, 10, 12, 24, ...)
* Ensure breakpoint isn't hit ✅
This tells us that no glyph resizing was necessary
This commit fixes two race conditions:
* `SetPseudoWindowCallback` set the `_pseudoWindowMessageCallback`
callback after the Win32 message thread was already spawned.
This issue was fixed by instead using the `ServiceLocator` to get
a hold of the global `VtIo` instance which is created statically.
* `XtermEngine::SetWindowVisibility` was called without holding the
console lock. This issue was fixed by simply acquiring it first.
Closes MSFT:40913882
## Validation Steps Performed
* Add `IsConsoleLocked()` assertion in `VtEngine::_Write`
* Run Windows Terminal
* No assertion failures ✅
## Summary of the Pull Request
This PR moves the key handling for mark mode into a helper method that is then called before an action/key binding is attempted.
## References
Epic: #4993Closes#13533
## Validation Steps Performed
- Add custom keybinding to "down" arrow key
- in mark mode --> selection updates appropriately
- out of mark mode --> keybinding executed
In #13024, we removed `Terminal::GetCursorPosition` from TerminalCore.
This has been widely regarded as a good move.
Now, you might rightly be wondering: why didn't compilation immediately
fail? Well. It turns out that there were _two_ copies of
`GetCursorPosition`. One for `const Terminal`, and one for `Terminal`.
This is important.
`Terminal::GetCursorPosition()` returned the cursor position relative to
the viewport. `Terminal::GetCursorPosition() const`, however, returns
the cursor position in absolute.
We removed the non-`const` one. Fortunately, thanks to the lookup rules
for `const`-qualified members, this didn't matter. Code that called
`GetCursorPosition()` still called `GetCursorPosition()`, and everything
was fine.
Except that part about the relative coordinates. That was not fine.
The TSF control is the _only_ consumer of `ControlCore.CursorPosition`,
and that was the _only_ consumer of relative-`GetCursorPosition()`.
This commit restores equilibrium by introducing a new
`GetViewportRelativeCursorPosition()` member to `Terminal` and switching
over the only consumer of relative cursor position to use it.
Closes#13769.
The absolute disgrace of a fix called 65b71ff failed to account for `std::move`
being unsafe to use for overlapping ranges. While `std::move` works for trivial
types (it happens to delegate to `memmove`), we need to dynamically switch
between that and `std::move_backward` to be correct.
Without this fix the LRU refresh is incorrect and might lead to crashes.
## Validation Steps Performed
I'm working on a new, pure D2D renderer inside AtlasEngine, which uses
the iterators contained in `_r.cellGlyphMapping` to draw text.
I noticed the bug, because scrolling up caused the text to be garbled
and with this fix applied it works as expected.
Fixes MSFT:38775539
Might also fix MSFT:38614563
Looking at this code should be pretty clear what's going on. On exit, the XAML root is already nulled out. But here, we're just yolo'ing and assuming it exists (why wouldn't it). So yea. This is like weirdly high percent of crashes internally, but that's not from real users. Real users, I suspect hit this as like .3% of our crashes. Not zero, but _low_.
* [x] tested manually
<hr>
May also be related to...
* MSFT:40602905
* MSFT:40602904
* MSFT:40412800
* MSFT:35213459 <---has links
Fixes MSFT:40853556
There's a small race here. The renderer thread in ConPTY might notice the terminal is gone, call CloseOutput, and release the vt renderer, and then the window proc fires and decides to minimize/restore the window, triggering an A/V.
I'm 100% confident that this has NEVER happened to a real user. But the test labs hit it so much that it makes up ~26% of our crashes.
I haven't tested this cause again, _it doesn't hit in the wild_
Rename `til::form_wchars` to `til::to_ulong` and
allow it to work with narrow `char`s.
This change will be used in #13429.
## Validation Steps Performed
* Loads `sc(...)` key bindings as expected ✅
* The change is thankfully fairly trivial if viewn with whitespace suppressed
We shouldn't add URLs into our binaries that we can't directly control.
This commit fixes the issue for URLs recently introduced in #13510.
Closes#13541
## Validation Steps Performed
This change is trivial enough that I simply opened the new redirects
in my browser, ensuring that they open the expected websites.
`debugGlyphGenerationPerformance` used to only test the performance of
text segmentation/parsing, so I renamed it to `debugTextParsingPerformance`.
The new `debugGlyphGenerationPerformance` actually clears the glyph atlas now.
Additionally this fixes a bug with `debugGeneralPerformance`:
If a `DXGI_SWAP_CHAIN_FLAG_FRAME_LATENCY_WAITABLE_OBJECT` is requested,
it needs to be used. Since `debugGeneralPerformance` is for testing without
V-Sync, we need to ensure that the waitable object is properly disabled.
## Summary of the Pull Request
Skips whitespace removal if pasted string is multiline
## PR Checklist
* [x] Closes#12387
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [x] Tests added/passed
* [ ] Documentation updated. If checked, please file a pull request on [our docs repo](https://github.com/MicrosoftDocs/terminal) and link it here: #xxx
* [ ] Schema updated.
* [ ] 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
Tests passed
The last command in multiline paste now executes
Does two things:
* the first two commits: shakes up the way we reference MUX in our projects so we can actually just
```xml
<PropertyGroup Label="NuGet Dependencies">
<TerminalMUX>true</TerminalMUX>
</PropertyGroup>
```
Like every other dependency we have
* the last commit: update to MUX.2.7.3
This is the 1.14 PR, which should be appropriately cherry-picked through to `release-1.15` and `main`
(cherry picked from commit a277b56f6a)
Repurpose `Feature_AtlasEngine` to enable the atlas engine by default in Dev and Preview builds.
Introduce `Feature_ConhostAtlasEngine` to solely control atlas engine inclusion in conhost.
Closes#13745
We have a number of theories why #12607 is happening, one of which is that
some GPU drivers somehow rely on Win32 messages or similar which we process
on the main thread. If we then try to acquire the console lock on the main
thread, while the GPU-driver thread itself is holding that lock, we've got
ourselves a deadlock. This PR makes this less likely by running the repeat
offender `UpdatePatternLocations` on a background thread instead.
We have a number of other locations which acquire the console lock on the
main thread and a thorough bug fix must be done in a different way.
## Validation Steps Performed
* After pasting an URL it gets underlined on hover ✅
Currently "Inverse Cursor" is actually simply bitwise inversed.
It works fine, except when it does not, namely in the middle of the spectrum:
Anything close enough to dark grey (index 8, RGB(128, 128, 128) in the
classic palette) will for obvious reasons become almost the same dark grey
again after the inversion.
The issue is addressed by additionally `XOR`ing the inverted color with
RGB(63, 63, 63). This distorts the result enough to avoid collisions
in the middle. Ultimately this restores the behavior that was in
Windows Console since the Middle Ages (and still exists in ConhostV1.dll).
## PR Checklist
* [x] Closes#3647
* [x] CLA signed
* [x] Tests added/passed
* [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: #3647
## Validation Steps Performed
1. Open OpenConsole.
2. Properties, Terminal, Cursor Colors, choose Inverse Color.
3. Optionally set the RGB value of the 8th color in the standard palette to RGB(128, 128, 128) on the Colors tab, but the default one will do too.
4. Type `color 80` to see black text on dark grey background.
5. Make sure the inverted cursor is visible.
6. Repeat in WT with both default and experimental renderers.
## Summary of the Pull Request
Changes the way `_addCommandsForArg` determines if the delimiter was at the beginning of the argument so that it accounts for the fact that match includes the last character of the string before it.
## PR Checklist
* [x] Closes#13277
## Validation Steps Performed
`wt -p "u"; nt -p "u"` does not cause an error
This commit builds directly on the changes made in #13677 and fixes:
* TSF resetting to AlphaNumeric ("ASCII") input mode when pressing enter
* Vietnamese IME not composing a new word after pressing whitespace, etc.
Closes#11479Closes#13398
## Validation Steps Performed
* Japanese IME (Full-Width Katakana)
Typing "saitama" produces "サイタマ" ✅
* Korean IME
Typing "gksrmf" produces "한글" ✅
* Vietnamese IME
Typing "xin chaof" produces "xin chào" ✅
* Emoji Picker (Win+.)
✅
Adds a variable `_isBackgroundLight` that is updated when the background
color is changed. When it is `true`, the BEL indicator flash will darken
the screen instead of brightening.
`_isColorLight(bg)` returns `true` if the average of `r`, `g`, and `b`
is >127
I was unsure of an appropriate way to change the color of the
`CompositionLight` based on the background, so I changed it to always be
gray and adjusted the intensity values of the original animation to have
roughly the same visual effect as the white.
## Validation Steps Performed
* Tested the two flashes on the default color schemes and some custom
background colors to see if they look consistent
* Used tracepoints and visual to check that the right animation is used
(including multiple tabs, split windows with different themes, and
changing settings while window is open)
References #9270Closes#13450
wyhash was chosen based on the results found in `smhasher`, were it proved
itself as an algorithm with little flaws and fairly high output quality.
While I have a personal preference for xxhash (XXH3 specifically), wyhash is a
better fit for this project as its source code is multiple magnitudes smaller,
simplifying the review and integration into the header-only `hash.h` file.
For use with hashmaps the hash quality doesn't actually matter much for
optimal performance and instead the binary size usually matters more.
But even in that scenario wyhash is fairly close to FNV1a (aka "FNV64").
The result is that this new hash algorithm will only have little impact on
hashmap performance if used over the standard FNV1a as used in the STL,
while simultaneously offering a vastly better hash quality.
This partially solves #13124.
## Validation Steps Performed
* Added test cases ✅
Other settings model classes call `JsonUtils::SetValueForKey` with the
private `_##value` member as the value. Since `_##value` is an optional,
this prevented writing out unset, optional fields. The new `Themes` class
deviated from this and this commit brings it back in line with the others.
Closes#13544
## Validation Steps Performed
* Create a `{ "name": "test" }` theme
* Save settings via the UI
* Optional/Defaulted fields aren't written ✅
## Summary of the Pull Request
VC++ v14 Descktop Framework package is required and not installed automatically when installing package manually.
## PR Checklist
* [X] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
While working on #13398 I felt that `TSFInputControl` wasn't up to sniff.
This commit is a minor cleanup of the class:
* default member initializers
* Simplified use of STL classes which already perform boundary checks
* Correctly check text buffer emptiness in `_SendAndClearText`
* Track selection range as mandated by the API
## Validation Steps Performed
* Japanese IME (Full-Width Katakana)
Typing "saitama" produces "サイタマ" ✅
* Korean IME
Typing "gksrmf" produces "한글" ✅
* Vietnamese IME
Typing "xin chaof" continues to produce broken "xin xinchaof"
(It's supposed to produce "xin chào")
* Emoji Picker (Win+.)
✅
## Summary of the Pull Request
- We only ever have 1 color picker now, instead of each tab having its own
- `TerminalPage` constructs this color picker (upon first request for it)
- `TerminalPage` attaches the color picker to the tab that requested for it
- `TerminalTab` detaches the color picker when it is done with it, so that `TerminalPage` can attach it to another tab later on
## References
#5907
## Validation Steps Performed
User-end behaviour is the same
We must use 65535 as `MAX_PARAMETER_VALUE` in order for us to properly parse
win32-input-mode sequences, which transmit UTF-16 characters as parameters.
Closes#12977
## Validation Steps Performed
* Call `SendInput` with 🙁 (`L'\xD83D'`, `L'\xDE41'`)
* 🙁 appears on the input line ✅
## Summary of the Pull Request
In #13560 we added a member to `Pane` that lets it know if it was spawned as a default terminal session, but did not propagate that value when the pane gets split or when the pane closes. This commit fixes that.
## Validation Steps Performed
A session spawned by a def term invocation remembers it even as it goes through splits
## Summary of the Pull Request
In general, when a selection marker is shown, we should scroll to it. The `selectAll` action adds a selection marker, but we don't scroll to it. This PR makes it such that we do do that.
Epic: #4993Closes#13485
Updates the schema such that the scroll mark settings are defined as profile settings instead of global settings (because they're actually profile settings).
Separately (but still relevant), I've also updated the release notes.
Closes#13583
This fixes an issue were overwriting parts of a row would only trigger
that specific portion of the row to be redrawn. This isn't just
problematic for combining characters, but also for things like
the new `TestDbcsBisectWriteCells` test introduced in #13626.
Benchmarks showed no impact on performance whatsoever.
## Validation Steps Performed
* Pick this commit into #13626
* Run the `TestDbcsBisectWriteCells` test and break before OpenConsole exits
* A correct "QいかなZYXWVUTに" output is visible ✅
This commit makes AtlasEngine recognize Powerline glyphs as box drawing ones.
The extra pixel offsets when determining the `scale` caused weird artifacts
and thus were removed. It seems like this causes no noticeable regressions.
Closes#13029
## Validation Steps Performed
* Run all values of `wchar_t` through `isInInversionList`
and ensure it produces the expected value ✅
* Powerline glyphs are correctly scaled with Cascadia Code PL ✅
07d58a8 contains a regression where the settings' `Themes()` property is
accessed without checking whether it's a `nullptr`. This can happen because
the invalid settings modal is shown with a empty settings model object.
This commit fixes the issue by deferring the update of `_settings` until
after we ensured that the `_settings` object is valid (besides warnings).
Closes#13543
## Validation Steps Performed
* Replace any string value with `123`
* Application doesn't crash ✅
## Summary of the Pull Request
- When 'discard changes' is hit, we re-initialize our list of color scheme view models but forgot to tell xaml about it, this commit fixes that.
- Make sure to exit rename mode when 'update settings' gets called
## References
color schemes mvvm added in #13179
## PR Checklist
* [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
Hitting discard changes doesn't cause an inconsistency with the currently selected scheme anymore
## Summary of the Pull Request
Adds a new mode to `CloseOnExit`: `Automatic`. In this mode, if a process handed off by defterm terminates for whatever reason, we always close (i.e. we treat the mode as `Always`), but for processes launched by Terminal we terminate as with the `Graceful` behaviour.
## PR Checklist
* [x] Closes#13325
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [ ] Tests added/passed
* [ ] Documentation updated. If checked, please file a pull request on [our docs repo](https://github.com/MicrosoftDocs/terminal) and link it here: #xxx
* [ ] Schema updated.
* [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
## Detailed Description of the Pull Request / Additional comments
- Adds a new enum value to `CloseOnExit`
- Adds a new function to `Pane`: `FinalizeConfigurationGivenDefault`: this is a function that should be called when the pane is created via default terminal handoff, and can contain any special configurations we should set given that the pane was created via handoff
## Validation Steps Performed
Curiously, at least on Windows 10 (and rarely on Windows 11), if you minimize the Terminal by clicking on the taskbar, then alt-tab to try and restore the window, the Taskbar will decide to call `SwitchToWindow` on the invisible, owned ConPTY window instead of the main window. When that happens, ConPTY'll get a `WM_SIZE(SIZE_RESTORED, lParam=0)`. The main window will NOT get a `SwitchToWindow` called. If ConPTY doesn't actually inform the hosting process about this, then the main HWND might stay hidden.
* Refer to #13158 where we disabled this.
* Closes#13589
* Closes#13248
* Tested manually on a Windows 10 VM.
* Confirmed that opening tabs while maximized/snapped doesn't restore down.
* `[Native]::ShowWindow([Native]::GetConsoleWindow(), 6)` still works
## Summary of the Pull Request
Adds support for the `tab.showCloseButton` property to themes. This accepts three values:
* `"always"` (default): The close button acts like it does today.
* `"hover"`: The close button is always visible on the active tab. On inactive tabs, the close button only appears on mouse over.
* `"never"`: The close button is never visible. You can't close the tab with middle-click, but you can still use keyboard shortcuts to close the tab.
## References
* See #3327
* ⚠️ targets #13178⚠️
## PR Checklist
* [x] Closes#3335
* [x] I work here
* [ ] Tests added/passed
* [n/a] Requires documentation to be updated - YUP
## Detailed Description of the Pull Request / Additional comments
See the following two properties in WInUI that we're leveraging here.
* [`TabViewCloseButtonOverlayMode.OnPointerOver`](https://docs.microsoft.com/en-us/windows/winui/api/microsoft.ui.xaml.controls.tabviewclosebuttonoverlaymode?view=winui-2.7&viewFallbackFrom=winui-2.2)
* [`TabViewItem.IsClosable`](https://docs.microsoft.com/en-us/windows/winui/api/microsoft.ui.xaml.controls.tabviewitem.isclosable?view=winui-2.2#microsoft-ui-xaml-controls-tabviewitem-isclosable)
One is a tabview-level property, the other is a per-tab-item property, hence why this code is a little wacky.
## Validation Steps Performed
gifs below
This commit contains 3 improvements for glyph rendering:
* Scale block element and box drawing characters to fit the cell size
"perfectly" without leaving pixel gaps between cells.
* Use `IDWriteTextLayout::GetOverhangMetrics` to determine whether glyphs
are outside the given layout box and if they are, offset their position
to fit them back in. If that still fails to fit, we downscale them.
* Always scale up glyphs that are more than 2 cells wide
This ensures that long ligatures that mimic box drawing characters like
"===" under Cascadia Code are upscaled just like regular box drawings.
Unfortunately this results in ligature-heavy text (like Myanmar) to get an
"uneven" appearance because some ligatures can suddenly appear too large.
It's difficult to come up with a good heuristic here.
Closes#12512
## Validation Steps Performed
* Print UTF-8-demo.txt
* Block characters don't leave gaps ✅
* Print a lorem-ipsum in Myanmar
* Glyphs aren't cut off anymore ✅
* Print a long "===" ligature under Cascadia Code
* The ligature is as wide as the number of cells used ✅
## Summary of the Pull Request
Implements the MVVM style for the Color Schemes editor
## PR Checklist
* [x] Closes #xxx
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [x] I work here
## Detailed Description of the Pull Request / Additional comments
Introduces:
- `ColorSchemesPageViewModel`: The view model responsible for the entire color schemes page. Handles what the current scheme is, adding/deleting/renaming schemes
- `ColorSchemeViewModel`: A view model class for individual color schemes
## Validation Steps Performed
Manually tested:
- Edit a color scheme
- Add new color scheme
- Rename a color scheme
- Delete a color scheme
This is an experiment, as discussed in https://github.com/microsoft/terminal/issues/11790#issuecomment-1179143049. We don't know what for sure causes these crashes, but it seems that blindly throwing, so that it gets picked up by Watson, is probably not the move. Instead, we're just gonna do our fallback, REGARDLESS of what the exception was.
See #11790, MSFT:38542548, MSFT:38572983, MSFT:38542574 et. al.
Refer to https://docs.microsoft.com/en-us/windows/win32/rstmgr/guidelines-for-applications
The OS will send us a WM_QUERYENDSESSION when it's preparing an
update for our app. It will then send us a WM_ENDSESSION, which gives
us a small timeout (~30s) to actually shut down gracefully. After
that timeout, it will send us a WM_CLOSE. If we still don't close
after the WM_CLOSE, it'll force-kill us (causing a crash which will be
bucketed to moapphang).
We will manually start a quit, so that we can persist the state. If we refuse to
gracefully shut down, the OS will crash us to focefully terminate us. We
choose to quit here, rather than just close, to skip over any warning dialogs
(e.g. "Are you sure you want to close all tabs?") which might prevent a WM_CLOSE
from cleanly closing the window.
This will cause a appHost._RequestQuitAll, which will notify the
monarch to collect up all the window state and save it.
This "crash" caused by the OS force killing us constitutes 80% of all our crashes. 80%. See MSFT:38947155, MSFT:38877540, MSFT:21058878, MSFT:31710054, MSFT:39764652, MSFT:26883776.
Closes#13569
It also fixes the issue where if you've got Terminal Dev running (outside VS), and you try to Deploy, you have to make sure to close the "Are you sure you want to close all tabs" dialog before the deployment can proceed. A deploy in VS sends the same sequence of messages as a real update.
This commit implements the remaining 5 of 8 grid lines:
left/top/right/bottom (COMMON_LVB) borders and double underline
`AtlasEngine::_resolveFontMetrics` was partially refactored to use `float`s
instead of `double`s, because that's what the remaining code uses as well.
It also helps the new, slightly more complex double underline calculation.
## Validation Steps Performed
* Print characters with the `COMMON_LVB_GRID_HORIZONTAL`, `GRID_LVERTICAL`,
`GRID_RVERTICAL` and `UNDERSCORE` attributes via `WriteConsoleOutputW`
* All 4 grid lines are visible ✅
* Grid lines correctly scale according to the `lineWidth` ✅
* Print a double underline with `printf "\033[21mtest\033[0m"`
* A double underline is fully visible ✅
Cleans up a couple local test failures.
* [x] Closes#13474: So, I clearly hadn't ran the local tests at the end of the themes PR. We needed a sensible fallback to SOME theme, even if there wasn't one provided in the user json. This is only really hit in the tests (that don't also include `defaults.json`.
* [x] Closes#13323: Meh, the ordering of the keys in this test doesn't matter. Ordering is a map implementation detail. This is fine.
* [x] Ran tests locally
We recently figured that we can drop support for Windows 7. Coincidentally
AtlasEngine never actually supported Windows 7 properly, because it called
`ResizeBuffers` with `DXGI_SWAP_CHAIN_FLAG_FRAME_LATENCY_WAITABLE_OBJECT`
no matter whether the swap chain was created with it enabled.
The new minimally supported version is Windows 8.1.
66f4f9d had another bug: Just like how we scroll our viewport by `memmove`ing
the `_r.cells` array, we also have to `memmove` the new `_r.cellGlyphMapping`.
Without this fix drawing lots of glyphs while only scrolling slightly
(= not invalidating the entire viewport), would erroneously call
`makeNewest` on glyphs now outside of the viewport. This would cause
actually visible glyphs to be recycled and overwritten by new ones.
## Validation Steps Performed
* Switch to Cascadia Code
* Print some text that fills the glyph atlas
* Scroll down by a few rows
* Write a long "==========" ligature (this quickly fills up
any remaining space in the atlas and exacerbates the issue)
* Unrelated rows don't get corrupted ✅
While working on another PR related to this I noticed that my VS
generates `.vcxproj` files that are a bit different to the ones we have.
This commit is a quick search & replace of all our project files to make
(primarily) their `ToolsVersion` more in line with what VS does itself:
No `ToolsVersion` for `.vcxproj`, `ToolsVersion="15.0"`
for `.csproj` and `ToolsVersion="4.0"` for `.filters` files.
## Summary of the Pull Request
The xaml file no longer directly accesses the settings model object, and the settings model object is no longer exposed on the view model
## References
#13377
## PR Checklist
* [ ] Closes #xxx
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [ ] Tests added/passed
* [ ] Documentation updated. If checked, please file a pull request on [our docs repo](https://github.com/MicrosoftDocs/terminal) and link it here: #xxx
* [ ] Schema updated.
* [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
Still works
I was messing around with trying to build & deploy from the commandline. I discovered this, which is progress. However, the inner-loop commandline build for the Terminal is still egregiously long.
* just a docs update
* is EIM work
This change adds support for the `IntenseIsBold` rendering setting.
Windows Terminal for instance defaults to `false` here, causing
intense colors to only be bright but not bold.
## Validation Steps Performed
* Set "Intense text style" to "Bright colors"
* Enable AtlasEngine
* Print ``echo "`e[1mtest`e[0m"``
* "test" appears as bright without being bold ✅
## Summary of the Pull Request
This is a spec specifically dedicated to Mark Mode. It's an addition to the Keyboard Selection spec. I felt that it makes the most sense to make this a separate PR because there's a lot of ideas that are very specific to Mark Mode, and this gives us the space to modify some of that behavior and get a good look at how other terminal emulators designed this feature.
## References
#2840 - Keyboard Selection Spec (base spec/branch/PR)
## PR Checklist
* [X] Contributes to #715
#8000 will change the way we store text from a strict grid/matrix where
one UTF16 character or surrogate pair always equals 1 column with the
possibility of joining exactly 2 to a wide character pair, to a dynamic
buffer where 1 or more characters can form 1 or more columns in any
arbitrary combination. Our long term goal is to properly support both
complex grapheme clusters like Emojis and complex ligatures that a wider
than 2 columns. This change requires us to break our API as
`ReadConsoleOutputA/W` assumes the existence of exactly this grid/matrix
storage. Since we store wide characters like "い" as a single codepoint
that is simply marked as being 2 columns wide in the future, we cannot
reconstruct trailing DBCS characters that were written to the buffer
like we used to. On the other hand this new behavior allows us to
implement better Unicode support and most likely significantly improve
our performance.
### Minor breaking changes
* `ReadConsoleOutputA` will now always **zero** the high byte in
`(CHAR_INFO).Char.UnicodeChar`. Only the `.AsciiChar` can be used
then. This prevents users from storing "additional" data in the
terminal buffer.
* `ReadConsoleOutputA` will now **zero** the `.AsciiChar` if it fails to
convert the Unicode character into an appropriate DBCS.
* Example: It's possible to write "い" into a narrow column despite
being a wide character. In these cases `WriteConsoleOutputA` will
now return `0x00` instead of `0x44` (the lower half of い's code
point `0x3044`).
### Major breaking changes
* `ReadConsoleOutputW` will now repeat the leading Unicode character
twice and ignore the trailing one.
* Example 1: Writing the pair `0x3044 0xabcd` with
`WriteConsoleOutputW` used to yield the same `0x3044 0xabcd` if read
back with `ReadConsoleOutputW`. This worked because conhost
effectively ignored the trailing codepoint, allowing one to
"smuggle" data. In the future this trailing character will be
discarded and produce `0x3044 0x3044` instead.
* Example 2: Writing い with `WriteConsoleOutputA` can be done with
code page 932 (Shift-JIS) and the DBCS `0x82 0xa2`. If read back
with `ReadConsoleOutputW` this would previously yield the two
Unicode characters `0x3044 0xffff`. After this commit it'll yield
`0x3044 0x3044`.
### Alternative approaches
It's possible to "tag"/"mark" written data as originating from
`WriteConsoleOutputA/W` so that it can be reconstructed accurately later
on. However this lead to implementation complexities that we're actively
trying to avoid in the new buffer implementation. Effectively
_everything_ that touches the buffer's text would have to handle these
marks and either write or clear them. Given the most likely small amount
of users who depend on the current quirky behavior, it'd be an
unwarranted maintenance and performance burden and prevent Windows
Terminal to ever truly migrate to full Unicode support.
## Validation Steps Performed
* Adjusted feature tests complete successfully ✅
On occasion, when we submit to the store we get a package rejection
because the app name has changed for the `qps-*` locales. Instead of
constantly reserving new pseudolocalized app names every time the
pseudolocalization seed changes, we should just lock our app name so
that it does not get pseudolocalized.
Upgrade check-spelling to [v0.0.20](https://github.com/check-spelling/check-spelling/releases/tag/v0.0.20)
This upgrade includes a refresh of the workflow
key new features:
* the previous comment is collapsed
* duplicate words are flagged (see `alone` and `the`)
* forbidding patterns (see `nonexistent`, `preexisting`, and `greater than`)
Each of these features can be tuned
- comment collapsing is controlled by the `followup` bits in the workflow-- but I can't imagine why one would want to turn it off
- duplicate words can be masked in `patterns.txt` (see `Guid` and `that`)
- forbidding patterns (especially duplicates) is in `.github/actions/spelling/line_forbidden.patterns`
Fwiw, I'm slowly moving towards not using `.txt` in filenames, but it's a long term project and I have a bunch of other goals for the near term.
The refresh of advice is of course flexible -- I'm still evolving my default text. Note that the default now includes some `curl` and I'm still working on how I want to consume the output. I'm getting close to the point where I might be able to provide a tool that could reliably consume the output (including on Windows).
This code has been used internally for a while, but I tested it for this repository here:
https://github.com/check-spelling/terminal/pull/2
## Summary of the Pull Request
`AdjustIndistinguishableColors` can now be set to:
- `Never`: Never adjust the colors
- `Indexed`: Only adjust colors that are part of the color scheme
- `Always`: Always adjust the colors
## References
#13343
## PR Checklist
* [ ] Closes #xxx
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [ ] Tests added/passed
* [ ] Documentation updated. If checked, please file a pull request on [our docs repo](https://github.com/MicrosoftDocs/terminal) and link it here: #xxx
* [ ] Schema updated.
* [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
## Detailed Description of the Pull Request / Additional comments
For legacy purposes, `true` and `false` map to `Indexed` and `Never` respectively
## Validation Steps Performed
Setting still works
The behaviour of the 'go back' button in the command palette was changed to return to the previously selected element rather than the root.
Instead of returning to the root, the go back button now returns to the previously selected item in the filtered action list. The previously selected item is selected by default and the view is scoped to the item.
## Validation Steps Performed
Manually tested by going back and forth between nested actions in the command palette.
Closes#13457
<!-- Enter a brief description/summary of your PR here. What does it fix/what does it change/how was it tested (even manually, if necessary)? -->
## Summary of the Pull Request
Added two new buttons to the About dialog. Source Code and Send Feedback buttons link directly to the Terminal project on GitHub and to the Issues page respectively.
<!-- Other than the issue solved, is this relevant to any other issues/existing PRs? -->
## References
#13371
<!-- Please review the items on the PR checklist before submitting-->
## PR Checklist
* [ ] Closes#13371
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [ ] Tests added/passed
* [ ] Documentation updated. If checked, please file a pull request on [our docs repo](https://github.com/MicrosoftDocs/terminal) and link it here: #xxx
* [ ] Schema updated.
* [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: #13371
<!-- Describe how you validated the behavior. Add automated tests wherever possible, but list manual validation steps taken as well -->
## Validation Steps Performed
Performed manual testing and confirmed that the implementation works.
The current TextBuffer implementation will happily overwrite the
leading/trailing half of a wide glyph with a narrow one without
padding the other half with whitespace. This could crash AtlasEngine
which aggressively guarded against such inconsistencies.
Closes#13522
## Validation Steps Performed
* Run .bat file linked in #13522
(Override wide glyph with a single space.)
* `AtlasEngine` doesn't crash ✅
### Disappearing glyphs
We only process glyphs within the dirtyRect, but glyphs outside of the dirtyRect
are still in use and shouldn't be discarded. This is critical if someone uses
a tool like tmux to split the terminal horizontally. If they then print a lot
of Unicode text on just one side, we have to ensure that the (for example)
plain ASCII glyphs on the other half of the viewport are still retained.
### Black viewport after font changes
The cursor was drawn without a clip rect, causing the entire atlas
texture to be filled with black. This just so happened to work fine
in Windows Terminal but relied on a race condition.
Closes#13490
## Validation Steps Performed
* Disappearing glyphs
* Start `tmux` in `wsl`
* Split horizontally with `Ctrl+B`, `"`
* `cat` a huge Unicode text file on the bottom
* Ensure ASCII glyphs in the top half don't disappear ✅
* Black viewport after font changes
* Start `OpenConsole` with `AtlasEngine`
* Open Properties dialog and click "Ok"
* Viewport content doesn't disappear ✅
2b202ce6 changed this code to fix 2-phase name lookup, but accidentally
changed `&m_tInit` to `&CFuzzType::m_tInit` (pointer-to-member¹)
instead of `&this->m_tInit` (a regular pointer to some value).
This should not be confused with `&(CFuzzType::m_tInit)`
of course which is _not_ a pointer-to-member.
¹ To simplify things, a pointer-to-member is basically the byte offset of a
member within a struct. For instance given `struct T { char a, b, c, d; }` then
`&T::c` would commonly "store" the value 2, equivalent to `offsetof(T, c)`.
Closes#13501
## Summary of the Pull Request
Re-enable the setting to adjust indistinguishable text, with some changes:
- We now pre-calculate the nudged values for 'default bright foreground' as well
- When a color table entry is manually set (i.e. via VT sequences) we re-calculate the nudged values
## References
#11095#12160
## PR Checklist
* [x] Closes#11917
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [ ] Tests added/passed
* [ ] Documentation updated. If checked, please file a pull request on [our docs repo](https://github.com/MicrosoftDocs/terminal) and link it here: #xxx
* [ ] Schema updated.
## Validation Steps Performed
Indistinguishable text gets adjusted again
The min/max/close buttons now use the same font glyphs used in Windows instead of paths.
They will also look different depending on whether you use Windows 10 or 11.
With certain scaling levels, the new fluent icons are always blurry on titlebars (win32, UWP, terminal).
I didn't check, but I assume that the glyphs will not be blurry on Windows 10 because they use the old font.
However, the glyps seem to have some alignment issues, making them even more blurry, and I'm not sure if I can fix that.
I looked into the minimize button problem:
* The UWP titlebar has 4px/5px, just like Terminal, but win32 titlebars have 5px/4px
* The new fluent icons have rounded caps, but the win32 minimize one doesn't (it's very subtle)
* The win32 minimize button is not blurry when the other buttons are
From this I presume that the minimize button is still using the old icon. So perhaps the Terminal/UWP vertical alignment is correct.
I was able to improve the rendering by wrapping the icon inside a Viewbox.
However, it won't perfectly match UWP, because scaling is calculated differently.
The icon width is 10px, which on 1.25x scaling would become 12.5px.
UWP titlebars truncate that to 12px, while Xaml renders that as 13px.
This is probably the best that can be done for now.
I found these icons [here](https://docs.microsoft.com/en-us/windows/apps/design/style/segoe-fluent-icons-font).
## Summary of the Pull Request
This PR adds support for downloadable soft fonts in the DirectX
renderer, potentially enabling them to be used in Windows Terminal.
## References
Soft fonts were first implemented in conhost (with the GDI renderer) in
PR #10011.
## PR Checklist
* [ ] Closes #xxx
* [x] CLA signed.
* [ ] Tests added/passed
* [ ] Documentation updated.
* [ ] Schema updated.
* [ ] 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
## Detailed Description of the Pull Request / Additional comments
The way the DirectX implementation works is by building up a bitmap
containing all of the glyphs, and then drawing an appropriate subsection
of that bitmap for each character that needs to be rendered. The current
text color is applied with a color matrix effect, and the glyphs are
automatically scaled up to the current font size with a scaling effect.
By default the scaling uses a high quality cubic interpolation, which
gives it a smoother antialiased effect. But if the *Text antialiasing*
option is configured as *Aliased*, we use a simpler nearest-neighbor
interpolation, which more closely matches the rendering of the original
GDI implementation.
## Validation Steps Performed
I've manually tested the renderer in conhost with the `UseDx` registry
entry. I've also tested in Windows Terminal using the experimental
passthrough mode.
## Summary of the Pull Request
The original `DECPS` implementation made use of the Windows MIDI APIs to
generate the sound, but that required a 3MB package dependency for the
GS wavetable DLS. This PR reimplements the `MidiAudio` class using
`DirectSound`, so we can avoid that dependency.
## References
The original `DECPS` implementation was added in PR #13208, but was
hidden behind a velocity flag in #13258.
## PR Checklist
* [x] Closes#13252
* [x] CLA signed.
* [ ] Tests added/passed
* [ ] Documentation updated.
* [ ] Schema updated.
* [x] I've discussed this with core contributors already. Issue number
where discussion took place: #13252
## Detailed Description of the Pull Request / Additional comments
The way it works is by creating a sound buffer with a single triangle
wave that is played in a loop. We generate different notes simply by
adjusting the frequency at which that buffer is played.
When we need a note to end, we just set the volume to its minimum value
rather than stopping the buffer. If we don't do that, the repeated
starting and stopping tends to produce a lot of static in the output. We
also use two buffers, which we alternate between notes, as another way
to reduce that static.
One other thing worth mentioning is the handling of the buffer position.
At the end of each note we save the current position, and then use an
offset from that position when starting the following note. This helps
produce a clearer separation between tones when repeating sequences of
the same note.
In an ideal world, we should really have something like an attack-decay-
sustain-release envelope for each note, but the above hack seems to work
reasonably well, and keeps the implementation simple.
## Validation Steps Performed
I've manually tested both conhost and Terminal with the sample tunes
listed in issue #8687, as well as a couple of games that I have which
make use of `DECPS` sound effects.
<!-- Enter a brief description/summary of your PR here. What does it fix/what does it change/how was it tested (even manually, if necessary)? -->
## Summary of the Pull Request
Added setting for for hide window when it loses focus.
Works on normal window and quake window.
<!-- Other than the issue solved, is this relevant to any other issues/existing PRs? -->
## References
<!-- Please review the items on the PR checklist before submitting-->
## PR Checklist
* [x] Closes#10660
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [ ] Tests added/passed
* [ ] Documentation updated. If checked, please file a pull request on [our docs repo](https://github.com/MicrosoftDocs/terminal) and link it here: #xxx
* [ ] Schema updated.
* [ ] 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
<!-- Provide a more detailed description of the PR, other things fixed or any additional comments/features here -->
## Detailed Description of the Pull Request / Additional comments
<!-- Describe how you validated the behavior. Add automated tests wherever possible, but list manual validation steps taken as well -->
## Validation Steps Performed
- Enable **Settings > Appearance > Automatically hide window**
- Terminal window should be minimized on taskbar when it loses focus
- Quake window should be minimized on system tray when it loses focus
- Enable also **Settings > Appearance > Hide Terminal in the notification area when it is minimized**
- Terminal window should be minimized on system tray when it loses focus
- Quake window should be minimized on system tray when it loses focus
⚠️ This spec is going into the `specs/drafts/` folder, because it's CLEARLY not done yet.
I discussed this a bit with Dustin. We felt it would be valuable to have these thoughts committed as a durable artifact. Better to have our Mica thoughts written down somewhere, with the context they belong in. That of course includes the bigger Theming spec, which never got finished.
I don't think we need to go through the fill spec review for this. Theming is clearly still a WIP. But committing this draft should give a better picture of what the vision is.
See also:
* #3327
* #10509
### TODOs
* [ ] The many that are straight up in the doc
* [ ] The fact that there's multiple Mica's now
* [ ] GO look at MSFT:39027976
Moving to VS 2022 and C++20 in #13368 introduced a new dependency on an
as-yet-nonexistent VC Runtime library: msvcp140_atomic_wait.dll. That
library is not installed on the 21H1 machines that run our PGO tests.
Fortunately, the app platform will--largely--handle this dependency for
us on machines in the wild.
Fixes#13500
The debugging tools folks helpfully published PdbStr and SrcTool as
NuGet packages for us to take a dependency on! Now we can end this
wasteful practice once and for all...
NOTE: This change introduces a harmless error message at the end of
symbol indexing. Instead of printing out the list of mapped sources,
`srctool` will complain that `srcsrv.dll` is missing. This is only cosmetic.
Yep this is my bad. I was trying to fix a bug where only having one of `background` or `unfocusedBackground` would cause the other to never get set to the default. In attempting to force all the logic into an if statement, I messed up the order of operations. My bad 😢
Regressed in #13456
## Summary of the Pull Request
Adds support to navigate to clickable hyperlinks using only the keyboard. When in mark mode, the user can press [shift+]tab to go the previous/next hyperlink in the text buffer. Once a hyperlink is selected, the user can press <kbd>Ctrl+Enter</kbd> to open the hyperlink.
## References
#4993
## PR Checklist
* [x] Closes#6649
* [x] Documentation updated at https://github.com/MicrosoftDocs/terminal/pull/558
## Detailed Description of the Pull Request / Additional comments
- Main change
- The `OpenHyperlink` event needs to be piped down to `ControlCore` now so that we can open a hyperlink at that layer.
- `SelectHyperlink(dir)` searches the buffer in a given direction and finds the first hyperlink, then selects it.
- "finding the hyperlink" is the tough part because the pattern tree takes in buffer coordinates, searches through the buffer in that given space, then stores everything relative to the defined space. Normally, the given buffer coordinates would align with the viewport's start and end. However, we're now trying to search outside of the viewport (sometimes), so we need to manage two coordinate systems at the same time.
- `convertToSearchArea()` lambda was used to convert a given coordinate into the search area coordinate system. So if the search area is the visible viewport, we spit out a viewport position. If the search area is the _next_ viewport, we spit out a position relative to that.
- `extractResultFromList()` lambda takes the list of patterns from the pattern tree and spits out the hyperlink we want. If we're searching forwards, we get the next one. Otherwise, we get the previous one. We explicitly ignore the one we're already on. If we don't find any, we return `nullopt`.
- Now that we have all these cool tools, we use them to progressively search through the buffer to find the next/previous hyperlink. Start by searching the visible viewport _after_ (or _before_) the current selection. If we can't find anything, go to the next "page" (viewport scrolled up/down). Repeat this process until something comes up.
- If we can't find anything, nothing happens. We don't wrap around.
- Other relevant changes
- the `copy` action is no longer bound to `Enter`. Instead, we manually handle it in `ControlCore.cpp`. This also lets us handle <kbd>Shift+Enter</kbd> appropriately without having to take another key binding.
- `_ScrollToPoint` was added. It's a simple function that just scrolls the viewport such that the provided buffer position is in view. This was used to de-duplicate code.
- `_ScrollToPoint` was added into the `ToggleMarkMode()` function. Turns out, we don't scroll to the new selection when we enter mark mode (whoops!). We _should_ do that and we should backport this part of the change too. I'll handle that.
- add some clarity when some functions are using the viewport position vs the buffer position. This is important because the pattern interval tree works in the viewport space.
## Validation Steps Performed
- case: all hyperlinks are in the view
- ✅ get next link
- ✅ get prev link
- ✅ case: need to scroll down for next link
- ✅ case: need to scroll up for next link
Does what it says on the tin. When we get focused, temporarily turn off readonly mode, as to not pop the dialog when the focus sequence is eventually sent to the connection.
* closes#13461
The main fix here is for the caption button colors. If you had a dark OS/app theme, and a light titlebar, we'd end up with light glyphs, so the caption buttons would be impossible to find.
There's also a pile of nits from #12992 in here. Probably enough to close#13456 out, but I'll let Dustin be the judge.
Filing today, to get in for 1.16 selfhost (@DHowett)
`TestDbcsWriteRead` failed to properly test all text input scenarios.
In order to write wide characters with `CHAR_INFO` structs using the `W` APIs
one must still repeat the `CHAR_INFO` twice, once with `COMMON_LVB_LEADING_BYTE`
and once with `COMMON_LVB_TRAILING_BYTE` set in the attributes. This is because
`CHAR_INFO` APIs are column oriented. `TestDbcsWriteRead` on the other hand
only tested the scenario of sending wide characters without those flags.
This commit fixes the problem by introducing a `UnicodeDoubled` mode and
increases the number of test patterns from 14 to 17. Due to the existing
code having been written with the false assumption in mind, a simpler
modification resulted in +1100 LOC. As such I opted to rewrite the code instead
and replaced the potentially buggy output generators with static arrays.
Code page 437 tests were removed, as it contains no DBCS characters anyways.
This is preliminary work for #8000.
## Summary of the Pull Request
When the debug tap converts control characters into visible glyphs, it
ends up losing the structure of the output, and that can sometimes make
things difficult to read. This PR attempts to alleviate that problem by
reinjecting an actual line break in the debug stream whenever an `LF`
control is received.
## PR Checklist
* [x] Closes#12312
* [x] CLA signed.
* [ ] Tests added/passed
* [ ] Documentation updated.
* [ ] Schema updated.
* [x] I've discussed this with core contributors already. Issue number
where discussion took place: #12312
## Validation Steps Performed
I've tested the updated debug tab with a number of different shells, and
also a couple of different apps. When there aren't many linefeeds in the
output, it's obviously not going to make much of a difference, but when
there are, I think it definitely improves the readability.
If launch mode is set to full screen quake window is opened in full
screen.
## Validation Steps Performed
- Set startup > launch mode > full screen
- Launch quake window
- Quake window shouldn't be opened in full screen
Closes#12894
#13458 added the ability to reuse tiles from our glyph atlas texture
so that we stop running out of GPU memory for complex Unicode.
This however can result in our glyph generation being a performance issue in
edge cases, to the point that the application may feel outright unuseable.
CJK glyphs for instance can easily exceed the maximum atlas texture size
(twice the window size), but take a significant amount of CPU and GPU time to
rasterize and draw, which results in "jelly scrolling" down to ~1 FPS.
This PR improves the situation of the latter half by directly drawing
glyphs into the texture atlas without an intermediate scratchpad texture.
This reduces GPU usage by 96% on my system (33% -> 2%) which improves general
render performance by ~100% (15 -> 30 FPS). CPU usage remains the same however,
but that's not really something we can do anything about at this time.
The atlas texture is already our primary means to reduce the CPU cost after all.
## Validation Steps Performed
* Disable V-Sync for OpenConsole in NVIDIA Control Panel
* Enable `debugGlyphGenerationPerformance`
* Print the entire CJK block U+4E00..U+9FFF
* Measure the above GPU usage and FPS improvements ✅
(Alternatively: Just scroll around and judge the "jellyness".)
## Summary of the Pull Request
When you create a console alias that overrides an existing command, it
should still be possible to execute the original command by prefixing it
with a space. However, at some point in the past, there was an attempt
to improve the usability by trimming leading spaces, and that ended up
breaking this functionality. This PR reverts that change, so leading
spaces can once again be used to bypass an alias.
## PR Checklist
* [x] Closes#4189
* [x] CLA signed.
* [x] Tests added/passed
* [ ] Documentation updated.
* [ ] Schema updated.
* [x] I've discussed this with core contributors already. Issue number
where discussion took place: #4189
## Validation Steps Performed
I've updated the existing alias unit test for leading spaces to match
the new behavior, i.e. it now confirms that a command with leading
spaces will not match the alias.
I've also manually confirmed that the `doskey` test case reported in
issue #4189 is now working as expected.
So far AtlasEngine would only grow the backing texture atlas once it gets full,
without the ability to reuse tiles once it gets full. This commit adds LRU
capabilities to the glyph-to-tile hashmap, allowing us to reuse the least
recently used tiles for new ones once the atlas texture is full.
This commit uses a quadratic growth factor with power-of-2 textures,
resulting in a backing atlas of 1x to 2x the size of the window.
While AtlasEngine is still incapable of shrinking the texture, it'll now at
least not grow to 128MB or result in weird glitches under most circumstances.
## Validation Steps Performed
* Print `utf8_sequence_0-0x2ffff_assigned_printable_unseparated.txt`
from https://github.com/bits/UTF-8-Unicode-Test-Documents
* Scroll back up to the top
* PowerShell input line is still there rendering as ASCII. ✅
## Summary of the Pull Request
Implements a `RenderingViewModel` for the Rendering page in the SUI
## PR Checklist
* [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
Rendering page still works
## Summary of the Pull Request
Implements an `InteractionViewModel` for the Interaction page in the SUI
## PR Checklist
* [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
Interaction page still works
<!-- Enter a brief description/summary of your PR here. What does it fix/what does it change/how was it tested (even manually, if necessary)? -->
## Summary of the Pull Request
Adds ability to select where new tab will appear: at the end or after currently selected tab.
<!-- Please review the items on the PR checklist before submitting-->
## PR Checklist
* [x] Closes#12955
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [ ] Tests added/passed
* [ ] Documentation updated. If checked, please file a pull request on [our docs repo](https://github.com/MicrosoftDocs/terminal) and link it here: #xxx
* [ ] Schema updated.
* [ ] 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
<!-- Provide a more detailed description of the PR, other things fixed or any additional comments/features here -->
## Detailed Description of the Pull Request / Additional comments
PR adds setting "NewTabPosition" in global appearances with dropdown list and uses it then creating new tabs. This setting does not affect settings tab position (should it?).
There should also be a documentation update, I'm just waiting to see if this change even acceptable.
<!-- Describe how you validated the behavior. Add automated tests wherever possible, but list manual validation steps taken as well -->
## Validation Steps Performed
## Summary of the Pull Request
Implements a `GlobalAppearanceViewModel` for the GlobalAppearance page in the SUI
## PR Checklist
* [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
Global appearance page still works
## Summary of the Pull Request
Implements a `LaunchViewModel` for the Launch page in the SUI
## PR Checklist
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
## Validation Steps Performed
Launch page still works
Adds `tabRow.unfocusedBackground` to the theme properties.
When provided, the window will use this ThemeColor as the color of the tab row when the window is inactive.
When omitted, the window will fall back to the default tab row color, `{"key": "TabViewBackground"}` from our App.xaml.
* [ ] tests added.
* [x] Closes#4862
* [ ] Needs a whole pile of docs updates, which we'll do at the end here.
This actually helped validate #12992 quite a bit. I found a bunch of bugs concerning null colors, null objects. Json parsing is hard 😛
This one's pretty obvious. I added another scrollbar sized grid to the terminal,
but I forgot to collapse it when the user requests `"scrollbarState": "hidden"`.
* [x] Closes#13446
* [x] I work here
## Summary of the Pull Request
Adds support for the `tab.background` property to themes. This is also a ThemeColor, so it accepts, colors, `accent`, and `terminalBackground`, just like everything else.
## References
* See #3327
* ⚠️ targets #12992⚠️
## PR Checklist
* [x] Closes#702
* [x] I work here
* [ ] Tests added/passed
* [n/a] Requires documentation to be updated - YUP
## Detailed Description of the Pull Request / Additional comments
I apparently left behing an optional color in TerminalTab for theme colors some time ago, just never used it. Crazy, huh?
## Validation Steps Performed
gifs below
##### ⚠️ targeting 1.15
## Summary of the Pull Request
Adds support for Themes, a new type of customization for the Terminal. Themes allow the user to customize elements of the Terminal window itself. In this first iteration, this PR adds support for two main properties:
* enabling Mica as the window backdrop
* changing the tab row color (read: changing the titelbar color)
These represent the most important asks of theming in the Terminal. The properties added in this PR are:
* Theme color variants:
- `"#rrggbb"` or `"#aarrggbb"`
- `"accent"`
- `"terminalBackground"`
* Properties (_listed here in dot notation, but implemented as sub-objects_)
- `tabRow.background`: accepts a ThemeColor (above)
- `window.applicationTheme`: accepts one of `{"system", "light", "dark"}`
- `window.useMica`: accepts a boolean, defaults to false.
## References
* As first described in #3327
* spec'd in #12530
## PR Checklist
* [x] Sorta enables #10509, but doesn't close it. That'll need more comprehensive changes to the titlebar code.
* **update**: I totally disabled mica, but left the serialization code. It just seems silly without #10509.
* [x] Closes#1963
* [x] Closes#3774
* [x] Closes#12939
* [x] Does the bulk of the #3327 work, but I'm going to leave that open since that's become my megathread for everything related to theming.
* [x] I work here
* [x] Tests added/passed
* [ ] Requires documentation to be updated - **SURE DOES**
## Detailed Description of the Pull Request / Additional comments
### --> GO READ #12530 <--
Seriously.
These themes aren't customizable in the SUI currently. You can change the active theme, and the UI will show all of the defined themes, but they're not editable there.
They don't layer. You'll need to define your own themes.
Thay can't come from fragments. This is a really cool future idea, but not implemented in this v0.
The sub objects have some gnarly macros to generate a lot of the serialization code for you.
### TODOs
* [x] I still have yet to establish what the accent color algorithm is. This might be proprietary and require a ThemeHelpers workaround.
* [x] Make sure `terminalBackground` & the SUI result in something sensible
* [x] Make sure runtime BG changes work with `terminalBackground`. One time, they didn't. `printf "\x1b]11;rgb:ff/00/ff\x07"`
* [x] Acrylic Terminal BG's look weird, like, the opacity is always 50% or something. And the tab row looks all wrong then.
## Validation Steps Performed
This is the blob I've been testing with:
<details>
```jsonc
// "useAcrylicInTabRow": true,
"theme": "my dark",
// "theme": "Edge",
"theme": "orangey",
"theme": "WHITE",
// "theme": "terminal",
"themes": [
{
"name": "my dark",
"window": {
"applicationTheme": "dark",
"useMica": true,
},
"tabRow": {
"background": "#00000000",
},
},
{
"name": "Edge",
"tabRow": { "background": "accent" },
"window": { "applicationTheme": "system" }
},
{
"name": "orangey",
"window": {
"applicationTheme": "light",
"useMica": true,
},
"tabRow": {
"background": "#ff8800",
},
},
{
"name": "WHITE",
"window": {
"applicationTheme": "dark",
"useMica": true,
},
"tabRow": {
"background": "#FFFFFF",
},
},
{
"name": "terminal",
"window": {
"applicationTheme": "dark",
"useMica": false,
},
"tabRow": {
"background": "terminalBackground",
},
},
]
```
</details>
```
% .\tools\ReleaseEngineering\ServicingPipeline.ps1
Inferred servicing version 1.14
PICK f025c53dba: Remove the fallback to wsl.exe when HKCU\...\Lxss doesn't exist (#13436)
OK
```
The hover tab color used to be generated from the selected tab color, which would end up lighter or darker, and white-gray colors would end up pink.
It is now simply the selected tab color with 60% opacity. This is also how brushes are created for accent buttons and color buttons (although with different opacity levels).
The main result of this fallback is that we attempt to launch wsl.exe
when the user hasn't installed or interacted with WSL. On our test
machines, that results in the creation of a wsl.exe process that tells
us precisely nothing; on WDAC managed machines it results in an Event
Log entry about spawning another (possibly blocked) process.
The registry is more reliable, and if the "API" it provides changes we
can just rev terminal.
Closes#11716
## Summary of the Pull Request
In #13370, we should be notifying the renderer that the selection changed. Minor oversight and simple fix.
## References
#4993#13370Closes#13413
This commit fixes a minor race condition covered as part of #13368.
The member `_pfnHandoff` was read without the mutex `_mtx` being locked first.
The issue was solved by acquiring the lock early and running the entire
`s_StopListening` function with that lock held.
## Summary of the Pull Request
Introduces the `switchSelectionEndpoint` action which switches whichever selection endpoint is targeted when a selection is present. For example, if you are targeting "start", `switchSelectionEndpoint` makes it so that now you are targeting "end". This also updates the selection markers appropriately.
## References
Spec - #5804#13358Closes#3663
## Detailed Description of the Pull Request / Additional comments
Most of the code is just standard code of adding a new action. Other than that, we have...
- if there is no selection, the action fails and the keybinding is passed through (similar to `copy()`)
- when we update the selection endpoint, we need to also update the "pivot". This ensures that future calls of `UpdateSelection()` respect this swap.
- [Corner Case] if the cursor is being moved, we make it so that you basically "anchored" an endpoint and you don't have to hold shift anymore.
## Why is this change being made?
Our conhost OneCore backend isn't as thoroughly tested as our Win32 one and fell victim to the general "bugginess" of our shutdown handling centered around `ServiceLocator::RundownAndExit`. In the past we simply leaked all resources, but changed it so that a cleanup on exit occurs, so that we can track resource leaks for instance. This broke OneCore which has a more delicate shutdown than Win32.
## What changed?
This commit reverts changes being made in 9d7a46f64c and after.
## How was the change tested?
This change was tested in a OneCore VM by repeatedly spawning subprocesses and ensuring they exit in a timely manner and without unexpected crashes.
Related work items: MSFT-40226902, MSFT-22128499
Retrieved from https://microsoft.visualstudio.com os.2020 OS official/rs_wdx_dxp_windev 0683758a0846aefbbe50730c3cd336623328ddeb
## Summary of the Pull Request
1. [copy on select] when manually copying text (i.e. kbd or right-click) while in mark/quick-edit mode, we now dismiss the selection.
2. `Enter` is now bound to copy by default.
- This works very well with mark mode and provides a more consistent behavior with conhost's selection experience overall.
- "why not hardcode `Enter` as a way to copy when in mark mode?"
- In an effort to make this as configurable as possible, I decided to make it a configurable keybinding, but am open to suggestions.
3. selection markers
a. we now hide the selection markers when multi-clicking the terminal.
b. selection markers are now properly shown when a single cell selection exists
- Prior to this PR, any single cell selection would display both markers. Now, we actually track which endpoint we're moving and display the appropriate one.
4. ensures that when you use keyboard selection to move past the top/bottom of the scroll area, we clamp it to the origin/bottom-right respectively. The fix is also better here in that it provides consistent behavior across all of the `_MoveByX` functions.
5. adds `toggleBlockSelection` to the schema
## References
#13053
## Validation Steps Performed
Did a whole flowchart of expected behavior with copy on select:
- enable `copyOnSelect`
- make a selection with the mouse
- ✅ right-click should copy the text --> clear the selection --> paste
- use keyboard selection to quick-edit the existing selection
- ✅ `copy` action should clear the selection
- ✅ right-click should copy the text --> clear the selection --> paste
Played with selection markers a bit in mark mode and quick edit mode. Markers are updating appropriately.
FabricBot is now a [config-as-code-only] platform. As a result, while
you can still use the [FabricBot Configuration Portal] to modify your
FabricBot configuration, you can no longer save the changes. The only
way to save changes to your configuration at the moment is to _export
configuration_ from the portal and upload the exported configuration to
`.github/fabricbot.json` in your repository. In this pull request, we
are adding your FabricBot configuration to your repository at
`.github/fabricbot.json` so that you can make changes to it going
forward.
While the [FabricBot Configuration Portal] is the *only way* to modify
your FabricBot configuration at the moment, we have a feature on our
backlog to publish the JSON schema defining the structure of the
FabricBot configuration file. With the JSON schema, you can (1) use a
plaintext editor of your choice to modify the FabricBot configuration
file and use the schema to validate the file after editing or (2)
[configure VS Code] to use the schema when editing FabricBot
configuration file to take advantage of convenience features such as
automatic code completion and field description on mouseover.
[config-as-code-only]: https://eng.ms/docs/products/1es-data-insights/merlinbot/extensions/bot-config-as-code
[FabricBot Configuration Portal]: https://portal.fabricbot.ms/bot/?repo=microsoft/terminal
[configure VS Code]: https://code.visualstudio.com/Docs/languages/json#_json-schemas-and-settings
Co-authored-by: msftbot[bot] <48340428+msftbot[bot]@users.noreply.github.com>
Co-authored-by: Dustin Howett <duhowett@microsoft.com>
## Summary of the Pull Request
Replaces most uses of `Viewport::CompareInBounds()` with `til::point`'s `<` and `>` operators. `CompareInBounds` has been the cause of a bunch of UIA crashes over the years. Replacing them entirely ensures that the `FAILFAST_IF` isn't ever touched.
Unfortunately, we still need `IncrementInBounds` and `DecrementInBounds` to have support for that exclusive end.
## References
#13183
This commit fixes a bug causing the OpenConsole to get increasingly larger
every time the font is changed when the AtlasEngine is active.
The only impact this bug had on Windows Terminal is that the
`font-size` in HTML and RTF selection copies are too large.
## Validation Steps Performed
* OpenConsole window size doesn't change when
switching between main and alt buffer ✅
[Git2Git] Merged PR 7517171: Fix a race condition in ServiceLocator::RundownAndExit
The whole premise of RundownAndExit is that one thread enters it, runs
down the console and terminates it. One thread enters, 0 threads leave.
After some recent console locking changes, we found that on OneCore
devices it was possible for two threads (the I/O thread and the coniosrv
Input thread) to try to rundown and exit at the same time.
This SRWLOCK prevents that from happening.
Fixes#40146639
Related work items: #40146639 Retrieved from https://microsoft.visualstudio.com os.2020 OS official/rs_wdx_dxp_windev ff235c6e49e065bccc937dee91f43ff310b5743c
Related work items: #40146639
[Git2Git] Merged PR 7517072: conhost: Force the double-click tests to have a high click timeout
The conhost tests might run somewhere that does not have a double click time.
We should hardcode a value for the purposes of testing.
Retrieved from https://microsoft.visualstudio.com os.2020 OS official/rs_wdx_dxp_windev 3c2620dab577e16e8672b369e4dfcde7c02881a1
Related work items: MSFT-40150725
This introduces the build rules for midi.lib to the OS!
Retrieved from https://microsoft.visualstudio.com os.2020 OS official/rs_wdx_dxp_windev fb066544f112ae491bb1d1f73578bc47fa167b41
In #6810, we introduced a "quirk" for all known versions of PowerShell
that suppressed their requests for black background/gray foreground.
This was done to avoid an [issue in PSReadline] where it would paint
black bars all over the screen if the default background color wasn't
the same as the ANSI black color.
Years have passed since that quirk was introduced. The underlying bug
was fixed, and the fix was released broadly long ago. It's time for us
to remove the quirk... almost.
Terminal still runs on versions of Windows that ship a broken version of
PSReadline. We must maintain the quirk there -- the user can't do
anything about it, and we would make their experience worse if we
removed the quirk entirely.
PowerShell 7.0 also ships a broken version of PSReadline. It is still in
support for another 6 months, but updates have been available for some
time. We can encourage users to update.
Therefore, we only need the quirk for Windows PowerShell, and then only
for specific versions of Windows.
_Inside Windows_, we don't even need that: we're guaranteed to be built
alongside a fixed version of PowerShell!
Closes#6807
[issue in PSReadline]: https://github.com/PowerShell/PSReadLine/issues/830#issuecomment-650508857
## Summary of the Pull Request
Introduced in #10824, this fixes a bug where you could use keyboard selection to move below the scroll area. Instead, we now clamp movement to the mutable viewport (aka the scrollable area). Specifically, we clamp to the corners (i.e. 0,0 or bottom right cell of scroll area).
## Validation Steps Performed
✅ (no output) try to move past bottom of viewport
✅ (with output, at bottom of scroll area) try to move past viewport
✅ (with output, NOT at bottom of scroll area) try to move past viewport
✅ try to move past top of viewport
It's not useful to notify users that WT can be made the default if it's already
clearly being used for handoff. This commit will suppresses the banner then.
## PR Checklist
* [x] Closes#13314
* [x] I work here
## Validation Steps Performed
* Modify `TerminalPage::ShowSetAsDefaultInfoBar` to not check for
`CascadiaSettings::IsDefaultTerminalSet()`
* Set Terminal Dev as the default
* Set incoming connections to open in the latest Terminal window
* Delete `state.json` after every test below
* Launching Terminal Dev shows the banner ✅
Launching `cmd.exe` dismisses the banner in the current Terminal ✅
* Launching `cmd.exe` launches Terminal Dev without banner ✅
These files are vestigial, because we are also shipping (either as loose
files or embedded in resources.pri) precompiled xbf/xaml binary format
files.
This saves us almost 500kb on disk.
Fixes#11687
Validation
----------
I ran a local build and saw that it produced a working Terminal, packaged
and unpackaged.
## Summary of the Pull Request
This introduces a selection marker overlay that tells the user which endpoint is currently being moved by the keyboard. The selection markers are respect font size changes and `cursor` color.
## References
#715 - Keyboard Selection
#2840 - Keyboard Selection Spec
#5804 - Mark Mode Spec
## Detailed Description of the Pull Request / Additional comments
- `TermControl` layer:
- Use a canvas (similar to the one used for hyperlinks) to be able to draw the selection markers.
- If we are notified that the selection changed, update the selection markers appropriately.
- `UpdateSelectionMarkersEventArgs` lets us distinguish between mouse and keyboard selections. `ClearMarkers` is set to true in the following cases...
1. Mouse selection, via SetEndSelectionPoint
2. `LeftClickOnTerminal`, pretty self-explanatory
3. a selection created from searching for text
- `ControlCore` layer:
- Responsible for notifying `TermControl` to update the selection markers when a selection has changed.
- Transfers info (the selection endpoint positions and which endpoint we're moving) from the terminal core to the term control.
- `TerminalCore` layer:
- Provides the viewport position of the selection endpoints.
## Validation Steps Performed
- mouse selection (w/ and w/out shift) --> no markers
- keyboard selection --> markers
- markers update appropriately when we pivot the selection
- markers scroll when you hit a boundary
- markers take the color of the cursor color setting
- markers are resized when the font size changes
## Summary of the Pull Request
When you execute a `cls` in the cmd shell, or `Clear-Host` in
PowerShell, we have a pair of shims that attempt to detect those
operations and forward an `ED3` sequence to conpty to clear the
scrollback.
If there was a linefeed at the bottom of the viewport immediately
prior to those functions being called, that event might still be
pending, and only forwarded to conpty after the `ED3`. The result
then is a line pushed into the scrollback that shouldn't be there.
This PR tries to avoid that situation by forcing the renderer to
flush before the `ED3` sequence is sent.
## References
The `cls` and `Clear-Host` shims were originally added in PR #5627.
## PR Checklist
* [x] Closes#5770
* [x] Closes#13320
* [x] CLA signed.
* [ ] Tests added/passed
* [ ] Documentation updated.
* [ ] Schema updated.
* [ ] 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
I've manually tested in PowerShell with `echo Hello; Clear-Host` (this
is the only way I could reliably reproduce the original problem), and in
the cmd shell with `cls`. Both cases are now working as expected.
When we send this ShowWindow message, if we send it to it's
going to need to get processed by the window message thread before
returning. We're handling this message under lock. However, the first
thing the conhost message thread does is lock the console. That'll
deadlock us. So unlock here, first, to let the message thread deal with
this message, then re-lock so later on this thread can unlock again
safely.
* [x] Closes#13301
* [x] Tested conhost
* [x] Tested terminal
When I moved this into ControlCore, I forgot that UserScrollViewport is usually triggered by the scrollbar updating, so it doesn't ask the UI to update. Since this logic is in ControlCore, it's sorta in a weird place where it needs to communicate both up and down:
* update the `Terminal`'s viewport position
* update the `TermControl`'s scrollbar position
Checklist:
* [x] Closes a bug bash bug
* [x] Missed in #12948
* See also #11000
If we want to make Windows Terminal the default terminal under Windows,
we'll have to make conhost "handoff" incoming connections by default.
But this poses a problem: How can the seldomly updated conhost know
whether the routinely updated Windows Terminal version is actually willing
to accept such handoffs by default (it might be unwilling due to bugs, etc.)?
This commit solves the issue by introducing:
* A marker interface (`IDefaultTerminalMarker`): If it exists,
Windows Terminal indicates its willingness to accept the handoff.
* Turning the all-0 GUID from being synonymous for conhost,
to being synonymous for "Let Windows decide". Without this we wouldn't
be able to differentiate between users who consciously chose conhost
as their default terminal, vs. users who want the standard behavior.
Testing fallback behavior:
* Install "Terminal" 1.13
* Delete the 2 keys below `HKCU\Console\%%Startup`
* Enable `Feature_AttemptHandoff` in `features.xml`
Return `true` from `DefaultApp::CheckShouldTerminalBeDefault`
* Replace `conhost.exe` and `console.dll` with `sfpcopy` after building
* Launching `cmd.exe` launches as a conhost window ✅
(because "Terminal" 1.13 lacks the marker interface)
* Open properties page in `conhost.exe`
"Let Windows decide" is select by default ✅
* Changing the selection writes the new value ✅
Testing the new behavior:
* Delete the 2 keys below `HKCU\Console\%%Startup`
* Enable `Feature_AttemptHandoff` in `features.xml`
Return `true` from `DefaultApp::CheckShouldTerminalBeDefault`
* Use `CLSID_WindowsTerminalConsoleDev` and `CLSID_WindowsTerminalTerminalDev`
for the initialization of `TerminalDelegationPair`
* Replace `conhost.exe` and `console.dll` with `sfpcopy` after building
* Deploy the "Terminal Dev" package
* Launching `cmd.exe` launches "Terminal Dev" ✅
(because "Terminal Dev" has the marker interface)
* Open the settings tab
"Let Windows decide" is select by default ✅
* Changing the selection and saving writes the new value ✅
(cherry picked from commit 1b81c6540f)
Service-Card-Id: 82925080
Service-Version: Inbox
This will result in the deletion of the following directories from the OS tree, under `onecore/windows/core/console/open`:
* doc/
* src/tools/MonarchPeasantPackage/
* src/api-ms-win-core-synch-l1-2-0/
* src/tools/ansi-color/
* src/tools/ColorTool/
We have gotten some PoliCheck flags on `doc/` (thanks to Niksa.md), but also the OS build just doesn't need these folders 😄
(cherry picked from commit 27b63ad02a)
Service-Card-Id: 80783789
Service-Version: Inbox
855e136 contains a regression which breaks buffer reflow if wide surrogate
characters are present. This happens because we made use of the
`TextBufferCellIterator` whose increment operator skips 2 cells for wide
characters. This created a "misalignment" in the reflow logic which was written
for cell-wise iteration. This commit fixes the issue, by reverting back to the
previous algorithm without iterators.
Closes#12837
Closes MSFT-38904421
## Validation Steps Performed
* Run ``pwsh -noprofile -command echo "`u{D83D}`u{DE43}"``
* Resizing conhost preserves all contents ✅
* Resizing Windows Terminal doesn't crash it ✅
* Added a test covering this issue ✅
(cherry picked from commit 10b9044120)
Service-Card-Id: 80340091
Service-Version: Inbox
The `cascadia/` directory straight up isn't checked into the OS. So adding a test dependency on code in there was a BAD IDEA.
(cherry picked from commit 4e61be9cd7)
Service-Card-Id: 79863821
Service-Version: Inbox
## THE WHITE WHALE
This is a fairly naive fix for this bug. It's not terribly performant,
but neither is resize in the first place.
When the buffer gets resized, typically we only copy the text up to the
`MeasureRight` point, the last printable char in the row. Then we'd just
use the last char's attributes to fill the remainder of the row.
Instead, this PR changes how reflow behaves when it gets to the end of
the row. After we finish copying text, then manually walk through the
attributes at the end of the row, and copy them over. This ensures that
cells that just have a colored space in them get copied into the new
buffer as well, and we don't just blat the last character's attributes
into the rest of the row. We'll do a similar thing once we get to the
last printable char in the buffer, copying the remaining attributes.
This could DEFINITELY be more performant. I think this current
implementation walks the attrs _on every cell_, then appends the new
attrs to the new ATTR_ROW. That could be optimized by just using the
actual iterator. The copy after the last printable char bit is also
especially bad in this regard. That could likely be a blind copy - I
just wanted to get this into the world.
Finally, we now copy the final attributes to the correct buffer: the new
one. We used to copy them to the _old_ buffer, which we were about to
destroy.
## Validation
I'll add more gifs in the morning, not enough time to finish spinning a
release Terminal build with this tonight.
Closes#32🎉🎉🎉🎉🎉🎉🎉🎉🎉Closes#12567
(cherry picked from commit 855e1360c0)
2022-03-28 13:33:54 -05:00
565 changed files with 21401 additions and 9062 deletions
@@ -81,7 +81,7 @@ When you hit "New Issue", select the type of issue closest to what you want to r
Microsoft Windows [Version 10.0.18900.1001]
```
* What tools and apps you're using (e.g. VS 2019, VSCode, etc.)
* What tools and apps you're using (e.g. VS 2022, VSCode, etc.)
* Don't assume we're experts in setting up YOUR environment and don't assume we are experts in `<your distro/tool of choice>`. Teach us to help you!
* **We LOVE detailed repro steps!** What steps do we need to take to reproduce the issue? Assume we love to read repro steps. As much detail as you can stand is probably _barely_ enough detail for us!
* If you're reporting a particular character/glyph not rendering correctly, the specific Unicode codepoint would be MOST welcome (e.g. U+1F4AF, U+4382)
> * You may need to install the [VC++ v14 Desktop Framework Package](https://docs.microsoft.com/troubleshoot/cpp/c-runtime-packages-desktop-bridge#how-to-install-and-update-desktop-framework-packages).
> This should only be necessary on older builds of Windows 10 and only if you get an error about missing framework packages.
> * Terminal will not auto-update when new builds are released so you will need
> to regularly install the latest Terminal release to receive all the latest
> fixes and improvements!
@@ -175,7 +177,7 @@ While overhauling Windows Console, we modernized its codebase considerably,
cleanly separating logical entities into modules and classes, introduced some
key extensibility points, replaced several old, home-grown collections and
* You must install the [.NET Framework Targeting Pack](https://docs.microsoft.com/dotnet/framework/install/guide-for-developers#to-install-the-net-framework-developer-pack-or-targeting-pack) to build test projects
(yes, the cmd version is just calling powershell to do the powershell version. Too lazy to convert the rest by hand, I'm already copying from `.vscode\tasks.json`)
Building the package from VS generates the loose layout to begin with, and then registers the loose manifest, skipping the msix stop. It's a lot faster than the commandline inner loop here, unfortunately.
### 2022 Update
The following command can be used to build the terminal package, and then deploy it.
The `bx` will build just the Terminal package, critically, populating the `CascadiaPackage.build.appxrecipe` file. Once that's been built, then the `DeployAppRecipe.exe` command can be used to deploy a loose layout in the same way that Visual Studio does.
Notably, this method of building the Terminal package can't leverage the FastUpToDate check in Visual Studio, so the builds end up being considerably slower for the whole package, as cppwinrt does a lot of work before confirming that it's up to date and doing nothing.
"description":"Sets the file location of the sound played when the application emits a BEL character. If the path is invalid no sound will be played. This property also accepts an array of sounds and the terminal will pick one at random.",
"oneOf":[
{
"type":[
"string",
"null"
]
},
{
"type":"array",
"items":{
"type":"string"
{
"type":[
"string",
"null"
]
},
{
"type":"array",
"items":{
"type":"string"
}
}
}
]
},
},
"AppearanceConfig":{
"properties":{
"colorScheme":{
@@ -197,7 +203,7 @@
},
"intenseTextStyle":{
"default":"bright",
"description":"Controls how 'intense' text is rendered. Values are \"bold\", \"bright\", \"all\" and \"none\"",
"description":"Controls how 'intense' text is rendered when unfocused. Values are \"bold\", \"bright\", \"all\" and \"none\"",
"enum":[
"none",
"bold",
@@ -207,8 +213,14 @@
"type":"string"
},
"adjustIndistinguishableColors":{
"description":"When set to true, we will (when necessary) adjust the foreground color to make it more visible, based on the background color.",
"type":"boolean"
"default":"never",
"description":"Setting to adjust the foreground color to make it more visible, based on the background color. When set to \"indexed\", we will only adjust the colors if they came from the color scheme. Other possible values are \"never\" and \"always\".",
"enum":[
"never",
"indexed",
"always"
],
"type":"string"
},
"experimental.retroTerminalEffect":{
"description":"When set to true, enable retro terminal effects when unfocused. This is an experimental feature, and its continued existence is not guaranteed.",
@@ -232,7 +244,7 @@
"default":12,
"description":"Size of the font in points.",
"minimum":1,
"type":"integer"
"type":"number"
},
"weight":{
"default":"normal",
@@ -300,6 +312,7 @@
"enum":[
"adjustFontSize",
"clearBuffer",
"closeOtherPanes",
"closeOtherTabs",
"closePane",
"closeTab",
@@ -308,6 +321,7 @@
"commandPalette",
"copy",
"duplicateTab",
"expandSelectionToWord",
"exportBuffer",
"find",
"findMatch",
@@ -350,9 +364,11 @@
"switchToTab",
"tabSearch",
"toggleAlwaysOnTop",
"toggleBlockSelection",
"toggleFocusMode",
"selectAll",
"setFocusMode",
"switchSelectionEndpoint",
"toggleFullscreen",
"setFullScreen",
"setMaximized",
@@ -364,6 +380,11 @@
"quit",
"adjustOpacity",
"restoreLastClosed",
"addMark",
"scrollToMark",
"clearMark",
"clearAllMarks",
"experimental.colorSelection",
"unbound"
],
"type":"string"
@@ -383,6 +404,15 @@
],
"type":"string"
},
"ScrollToMarkDirection":{
"enum":[
"previous",
"next",
"first",
"last"
],
"type":"string"
},
"ResizeDirection":{
"enum":[
"left",
@@ -732,6 +762,30 @@
"direction"
]
},
"ScrollToMarkAction":{
"description":"Arguments corresponding to a Scroll to Mark Action",
"allOf":[
{
"$ref":"#/$defs/ShortcutAction"
},
{
"properties":{
"action":{
"type":"string",
"const":"scrollToMark"
},
"direction":{
"$ref":"#/$defs/ScrollToMarkDirection",
"default":"previous",
"description":"The direction to scroll to a mark."
}
}
}
],
"required":[
"direction"
]
},
"SendInputAction":{
"description":"Arguments corresponding to a Send Input Action",
"allOf":[
@@ -791,6 +845,43 @@
}
]
},
"ColorSelectionAction":{
"description":"Arguments corresponding to a Color Selection Action",
"allOf":[
{
"$ref":"#/$defs/ShortcutAction"
},
{
"properties":{
"action":{
"type":"string",
"const":"experimental.colorSelection"
},
"matchMode":{
"type":"string",
"default":"none",
"description":"Specifies if only the selected text should be colored (0), or all instances of selected text (case-insensitive) (1).",
"enum":[
"none",
"all"
]
},
"foreground":{
"$ref":"#/$defs/ColorOrIndex",
"type":"string",
"default":"",
"description":"The foreground color to use, as an RGB value (\"#rrggbb\"), or color index (\"iNN\"). If left unspecified it falls back to the default text foreground color."
},
"background":{
"$ref":"#/$defs/ColorOrIndex",
"type":"string",
"default":"",
"description":"The background color to use, as an RGB value (\"#rrggbb\"), or color index (\"iNN\"). If left unspecified it falls back to the default text background color."
}
}
}
]
},
"OpenSettingsAction":{
"description":"Arguments corresponding to a Open Settings Action",
"allOf":[
@@ -839,6 +930,27 @@
}
]
},
"AddMarkAction":{
"description":"Arguments corresponding to an Add Mark Action",
"allOf":[
{
"$ref":"#/$defs/ShortcutAction"
},
{
"properties":{
"action":{
"type":"string",
"const":"addMark"
},
"color":{
"$ref":"#/$defs/Color",
"default":null,
"description":"If provided, will set the mark's color to the given value."
}
}
}
]
},
"SetColorSchemeAction":{
"description":"Arguments corresponding to a Set Color Scheme Action",
"allOf":[
@@ -1434,6 +1546,101 @@
}
]
},
"ShowCloseButton":{
"enum":[
"always",
"hover",
"never"
],
"type":"string"
},
"ThemeColor":{
"description":"A special kind of color for use in themes. Can be an #rrggbb color, #rrggbbaa color, or a special value. 'accent' is evaluated as the user's selected Accent color in the OS, and 'terminalBackground' will be evaluated as the background color of the active terminal pane.",
"description":"A set of properties for customizing the appearance of the tabs",
"properties":{
"background":{
"description":"The color of a tab when it is the active tab",
"$ref":"#/$defs/ThemeColor"
},
"unfocusedBackground":{
"description":"The color of a tab when it is not the active tab",
"$ref":"#/$defs/ThemeColor"
},
"showCloseButton":{
"description":"Controls the visibility of the close button on the tab",
"$ref":"#/$defs/ShowCloseButton"
}
}
},
"TabRowTheme":{
"additionalProperties":false,
"description":"A set of properties for customizing the appearance of the tab row",
"properties":{
"background":{
"description":"The color of the tab row when the window is the foreground window.",
"$ref":"#/$defs/ThemeColor"
},
"unfocusedBackground":{
"description":"The color of the tab row when the window is inactive",
"$ref":"#/$defs/ThemeColor"
}
}
},
"WindowTheme":{
"additionalProperties":false,
"description":"A set of properties for customizing the appearance of the window itself",
"properties":{
"applicationTheme":{
"description":"Which UI theme the Terminal should use for controls",
"enum":["light","dark","system"],
"type":"string"
}
}
},
"Theme":{
"additionalProperties":false,
"description":"A set of properties for customizing the appearance of the window. This controls things like the titlebar, the tabs, the application theme.",
"properties":{
"name":{
"type":"string",
"description":"The name of the theme. This will be displayed in the settings UI.",
"not":{
"enum":["light","dark","system"]
}
},
"tab":{
"$ref":"#/$defs/TabTheme"
},
"tabRow":{
"$ref":"#/$defs/TabRowTheme"
},
"window":{
"$ref":"#/$defs/WindowTheme"
}
}
},
"Keybinding":{
"additionalProperties":false,
"properties":{
@@ -1545,6 +1752,9 @@
{
"$ref":"#/$defs/AdjustOpacityAction"
},
{
"$ref":"#/$defs/ColorSelectionAction"
},
{
"type":"null"
}
@@ -1667,6 +1877,11 @@
"description":"When set to true, URLs will be detected by the Terminal. This will cause URLs to underline on hover and be clickable by pressing Ctrl.",
"type":"boolean"
},
"experimental.enableColorSelection":{
"default":false,
"description":"When set to true, adds preset \"Color Selection\" actions (keybindings) to allow colorizing selected text via keystroke, similar to the legacy conhost EnableColorSelection feature (such as alt+6 to color the selection red).",
"type":"boolean"
},
"disableAnimations":{
"default":false,
"description":"When set to `true`, visual animations will be disabled across the application.",
@@ -1810,15 +2025,17 @@
"type":"string"
},
"theme":{
"default":"system",
"description":"Sets the theme of the application. The special value \"system\" refers to the active Windows system theme.",
"enum":[
"light",
"dark",
"system"
],
"default":"dark",
"description":"Sets the theme of the application. This value should be the name of one of the themes defined in `themes`. The Terminal also includes the themes `dark`, `light`, and `system`.",
"type":"string"
},
"themes":{
"description":"The list of available themes",
"items":{
"$ref":"#/$defs/Theme"
},
"type":"array"
},
"showTabsInTitlebar":{
"default":true,
"description":"When set to true, the tabs are moved into the titlebar and the titlebar disappears. When set to false, the titlebar sits above the tabs.",
@@ -1898,6 +2115,20 @@
"useAnyExisting"
],
"type":"string"
},
"newTabPosition":{
"default":"afterLastTab",
"description":"Position of newly created tabs. Possible values are \"afterLastTab\" and \"afterCurrentTab\".",
"enum":[
"afterLastTab",
"afterCurrentTab"
],
"type":"string"
},
"autoHideWindow":{
"default":false,
"description":"If enabled, Terminal window will be hidden as soon as it loses focus.",
"type":"boolean"
}
},
"required":[
@@ -2011,15 +2242,20 @@
"description":"Controls what happens when the application emits a BEL character. When set to \"all\", the Terminal will play a sound, flash the taskbar icon (if the terminal window is not in focus) and flash the window. An array of specific behaviors can also be used. Supported array values include `audible`, `window` and `taskbar`. When set to \"none\", nothing will happen.",
"$ref":"#/$defs/BellStyle"
},
"bellSound":{
"description":"Sets the sound played when the application emits a BEL. When set to an array, the terminal will pick one of those sounds at random.",
"$ref":"#/$defs/BellSound"
},
"closeOnExit":{
"default":"graceful",
"description":"Sets how the profile reacts to termination or failure to launch. Possible values:\n -\"graceful\" (close when exit is typed or the process exits normally)\n -\"always\" (always close)\n -\"never\" (never close).\ntrue and false are accepted as synonyms for \"graceful\" and \"never\" respectively.",
"default":"automatic",
"description":"Sets how the profile reacts to termination or failure to launch. Possible values:\n -\"graceful\" (close when exit is typed or the process exits normally)\n -\"always\" (always close)\n -\"automatic\" (behave as \"graceful\" only for processes launched by terminal, behave as \"always\" otherwise)\n -\"never\" (never close).\ntrue and false are accepted as synonyms for \"graceful\" and \"never\" respectively.",
"oneOf":[
{
"enum":[
"never",
"graceful",
"always"
"always",
"automatic"
],
"type":"string"
},
@@ -2076,6 +2312,11 @@
"default":false,
"description":"When true, this profile should always open in an elevated context. If the window isn't running as an Administrator, then a new elevated window will be created."
},
"experimental.autoMarkPrompts":{
"default":false,
"description":"When set to true, prompts will automatically be marked.",
"type":"boolean"
},
"experimental.connection.passthroughMode":{
"description":"When set to true, directs the PTY for this connection to use pass-through mode instead of the original Conhost PTY simulation engine. This is an experimental feature, and its continued existence is not guaranteed.",
"type":"boolean"
@@ -2084,14 +2325,19 @@
"description":"When set to true, enable retro terminal effects. This is an experimental feature, and its continued existence is not guaranteed.",
"type":"boolean"
},
"experimental.showMarksOnScrollbar":{
"default":false,
"description":"When set to true, marks added to the buffer via the addMark action will appear on the scrollbar.",
"type":"boolean"
},
"experimental.pixelShaderPath":{
"description":"Use to set a path to a pixel shader to use with the Terminal. Overrides `experimental.retroTerminalEffect`. This is an experimental feature, and its continued existence is not guaranteed.",
"type":"string"
},
"experimental.useAtlasEngine":{
"description":"Enable using the experimental new rendering engine for this profile. This is an experimental feature, and its continued existence is not guaranteed.",
"useAtlasEngine":{
"description":"Windows Terminal 1.16 and later ship with a new, performant text renderer. Set this to false to revert back to the old text renderer.",
"type":"boolean",
"default":false
"default":true
},
"fontFace":{
"default":"Cascadia Mono",
@@ -2103,7 +2349,7 @@
"default":12,
"description":"[deprecated] Define 'size' within the 'font' object instead.",
"minimum":1,
"type":"integer",
"type":"number",
"deprecated":true
},
"fontWeight":{
@@ -2134,6 +2380,17 @@
],
"deprecated":true
},
"intenseTextStyle":{
"default":"bright",
"description":"Controls how 'intense' text is rendered. Values are \"bold\", \"bright\", \"all\" and \"none\"",
"enum":[
"none",
"bold",
"bright",
"all"
],
"type":"string"
},
"foreground":{
"$ref":"#/$defs/Color",
"default":"#cccccc",
@@ -2187,8 +2444,14 @@
]
},
"adjustIndistinguishableColors":{
"description":"When set to true, we will (when necessary) adjust the foreground color to make it more visible, based on the background color.",
"type":"boolean"
"default":"never",
"description":"Setting to adjust the foreground color to make it more visible, based on the background color. When set to \"indexed\", we will only adjust the colors if they came from the color scheme. Other possible values are \"never\" and \"always\".",
"enum":[
"never",
"indexed",
"always"
],
"type":"string"
},
"scrollbarState":{
"default":"visible",
@@ -2229,7 +2492,10 @@
},
"startingDirectory":{
"description":"The directory the shell starts in when it is loaded.",
"type":"string"
"type":[
"string",
"null"
]
},
"suppressApplicationTitle":{
"description":"When set to true, tabTitle overrides the default title of the tab and any title change messages from the application will be suppressed. When set to false, tabTitle behaves as normal.",
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-existent 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.
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/nonexistent 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.
@@ -138,7 +138,7 @@ The delegation repeats the same dance as above as well:
The terminal will be its own complete presentation and input solution on top of a ConPTY connection, separating the concerns between API servicing and the user experience.
Today the Terminal knows how to start and then launches a ConPTY under it. The Terminal will need to be updated to accept a pre-existing ConPTY connection on launch (or when the multi-process model arrives, as an inbound connection), and connect that to a new tab/pane instead of using the `winconpty.lib` libraries to make its own.
Today the Terminal knows how to start and then launches a ConPTY under it. The Terminal will need to be updated to accept a preexisting ConPTY connection on launch (or when the multi-process model arrives, as an inbound connection), and connect that to a new tab/pane instead of using the `winconpty.lib` libraries to make its own.
For now, I'm considering only the fresh-start scenario.
- The Terminal will have to detect the inbound connection through ~~its argument parsing (or through~~ a new entrypoint in the COM alternative ~~)~~ and store the PTY in/out/signal handles for that connection in the startup arguments information
This spec describes a new set of key bindings that allows the user to create and 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.
### Creating a selection
Mark Mode is a ConHost feature that allows the user to create a selection using only the keyboard. In CMD, pressing <kbd>ctrl+m</kbd> enters mark mode. The current cursor position becomes a selection endpoint. The user can use the arrow keys to move that endpoint. While the user then holds <kbd>shift</kbd>, the selection endpoint ('start') is anchored to it's current position, and the arrow keys move the other selection endpoint ('end').
Additionally, pressing <kbd>shift+arrow</kbd> also initiates a selection, but it anchors the first selection endpoint to the cursor position.
Other terminal emulators have different approaches to this feature. iTerm2, for example, has Copy Mode (documentation [linked here](https://iterm2.com/documentation-copymode.html)). Here, <kbd>cmd+shift+c</kbd> makes the current cursor position become a selection endpoint. The arrow keys can be used to move that endpoint. However, unlike Mark Mode, a key binding <kbd>c+space</kbd> is used to change the start/stop selecting. The first time it's pressed, the 'start' endpoint is anchored. The second time it's pressed, the 'end' endpoint is set. After this, you can still move a cursor, but the selection persists until a new selection is created (either by pressing the key binding again, or using the mouse).
Though tmux is not a terminal emulator, it does also have Copy Mode that behaves fairly similarly to that of iTerm2's.
## 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 direction of the Terminal Control. Terminal Core maintains the state of the selection.
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.
#### Abandoned 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:
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.
#### Abandoned 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.
**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.
### Mark Mode
Mark Mode is a mode where the user can create and modify a selection using only the keyboard.
When no selection is present, the user may use the `markMode` action to enter mark mode. Upon doing so, a selection will be created at the current cursor position.
When in mark mode, the user may...
- press <kbd>ESC</kbd> to clear the selection and exit mark mode
- invoke the `markMode` action to exit mark mode
- invoke the `copy` action (this includes right-clicking the terminal) to copy the selected text, clear the selection, and exit mark mode
- move the cursor in the following ways:
- arrow keys --> move by character
- ctrl + left/right --> move by word
- ctrl + home/end --> move to the beginning/end of the buffer
- home/end --> move to the beginning/end of the line respectively
- pgup/pgdn --> move up/down by viewport respectively
- expand the selection in the following ways:
- shift + arrow keys --> move the "end" endpoint by character
- ctrl + shift + left/right --> move the "end" endpoint by word
- ctrl + shift + home/end --> move the "end" endpoint to the beginning/end of the buffer
- shift + home/end --> move the "end" endpoint to the beginning/end of the line respectively
- shift + pgup/pgdn --> move the "end" endpoint up/down by viewport respectively
As with mouse selections, keybindings are still respected and pressing a key that is not bound to a keybinding (or mentioned above) will clear the selection and exit mark mode.
#### Corner cases
- In mark mode, if a selection was created via the keyboard, moving the cursor moves at the "end" endpoint. This is consistent with conhost.
- If a user creates a selection using the mouse, then enters mark mode, mark mode inherits the existing selection as if it was made using the keyboard.
- If `copyOnSelect` is enabled, the selection is copied when the selection operation is "complete". Thus, the selection is copied when the `copy` keybinding is used or the selection is copied using the mouse.
- If `copyOnSelect` is enabled, `ESC` is interpreted as "cancelling" the selection, so nothing is copied. Keys that generate input are also interpreted as "cancelling" the selection. Only the `copy` keybinding or copying using the mouse is considered "completing" the selection operation, and copying the content to the clipboard.
**NOTE** - Related to #3884:
If the user has chosen to have selections persist after a copy operation, the selection created by Copy Mode is treated no differently than one created with the mouse. The selection will persist after a copy operation. However, if the user exits Copy Mode in any of the other situations, the selection is cleared.
#### Block Selection
A user can normally create a block selection by holding <kbd>alt</kbd> then creating a selection.
If the user is in Mark Mode, and desires to make a block selection, they can use the `toggleBlockSelection()` action. `toggleBlockSelection()` takes an existing selection, and transforms it into a block selection (or vice-versa).
All selections created in Mark Mode will have block selection disabled by default.
#### Rendering during Copy Mode
Since we are just moving the selection endpoints, rendering the selection rects should operate normally. We need to ensure that we still scroll when we move a selection point past the top/bottom of the viewport.
In ConHost, output would be paused when a selection was present. Windows Terminal does not pause the output when a selection is present, however, it does not scroll to the new output.
#### Interaction with Mouse Selection
If a selection exists, the user is basically already in Copy Mode. The user should be modifying the "end" endpoint of the selection when using the `updateSelection()` bindings. The existing selection should not be cleared (contrary to prior behavior). However, the half-y-beam will not be drawn. Once the user presses the `copyMode` or `moveSelectionPoint` keybinding, the half-y-beam is drawn on the targeted endpoint (which will then be "start").
During Copy Mode, if the user attempts to create a selection using the mouse, any existing selections are cleared and the mouse creates a selection normally. However, contrary to prior behavior, the user will still be in Copy Mode. The target endpoint being modified in Copy Mode, however, will be the "end" endpoint of the selection, instead of the cursor (as explained earlier in the flowchart).
#### Abandoned Idea: Copy Mode
Copy Mode is a more complex version of Mark Mode that is intended to provide a built-in way to switch the active selection endpoint. This idea was abandoned because we would then run into a user education issue. Rather than reinventing the wheel, selection should feel natural like that of a text editor, and any diversion from that model should be introduced separately (i.e. keybindings). Doing so ensures that users can "hit the ground running" when trying to make a selection, but won't be hindered by new functionality that is available.
Copy Mode is a mode where the user can create and modify a selection using only the keyboard. The following flowchart covers how the new `copymode()` action works:
**NOTE**: `copyMode()` refers to the action, whereas `updateSelection()` refers to the underlying function that is being called in the code.
If a selection is not active, a "start" and "end" selection point is created at the cursor position. `updateSelection()` calls then move "start" and "end" together as one position.
Invoking `copyMode()` again, will then anchor "start" (meaning that it will be kept in place). Subsequent `updateSelection()` calls move the "end" selection point.
Invoking `copyMode()` essentially cycles between which selection point is targeted.
## UI/UX Design
### Key Bindings
| Action | Keybinding Args | Description |
|--|--|--|
| `selectAll` | none | Select the entire text buffer. |
| `markMode` | none | Toggle mark mode. If no selection exists, create a selection at the cursor position. Otherwise, use the existing selection as one in mark mode. |
| `toggleBlockSelection` | none | Transform the existing selection between a block selection and a line selection. |
| `switchSelectionEndpoint` | none | If a selection is present, switch which selection endpoint is targeted when in mark mode or another keyboard selection mode (i.e. modified a mouse selection using the keyboard). |
By default, the following key binding will be set:
```JS
{"command":"selectAll","keys":"ctrl+shift+a"},
// Copy Mode
{"command":"copyMode","keys":"ctrl+shift+m"},
{"command":"toggleBlockSelection"},
{"command":"switchSelectionEndpoint"},
```
### Selection Markers
A y-beam will be used to identify which selection endpoint is currently being moved when in mark mode. The y-beam will match the cursor color and the font size so that it essentially fills a cell in the buffer.
When we're moving the cursor (this happens when mark mode is entered from no existing selection), a full y-beam will be displayed at the cursor position.
When <kbd>shift</kbd> is held, we're expanding the selection. In this case, the y-beam will be split, and the relevant half will be rendered on the active endpoint functioning as a selection marker.
If the selection is up against the end (or beginning) of the line, the selection marker can't be rendered normally because there is no space to render it off the side of the terminal. Instead, the selection marker will be horizontally flipped. Alternatively, it can be rendered on the next available cell (i.e. end of line translates to the beginning of the next line), but that would cause issues when the selection is positioned at the beginning or end of the buffer, thus the idea was abandoned.
If the active endpoint crosses the inactive endpoint, the selection marker will be flipped. In a sense, the flag will always point away from the selection.
The y-beam doesn't make as much sense for block selections, so instead an L-shaped marker will be placed hugging the corner of the active selection endpoint.
**NOTE:** Both half y-beams could have been presented as shown in the image below. This idea was omitted because then there is no indication for which half y-beam is currently focused.
When mark mode is enabled, the cursor will stop blinking.
## 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.
We will be expected to send "selection changed" events via UIA every time the cursor moves or selection updates (both are considered the same event).
### 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
N/A
## 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.
### Circling the buffer
As usual, if the buffer is circling, the selection should be updated to follow the content (and "scroll up" appropriately).
In the event that one endpoint "scrolls" off the buffer, we must clamp "start" to the buffer origin. Conversely, in the event that both endpoints "scroll" off the buffer, the selection must be considered cleared.
## 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.
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:
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.
@@ -114,7 +114,7 @@ The jumplist will have to be saved each time a profile change occurs but the fre
#### Should it open a new instance of the terminal or open in a new tab?
#### What should happen if a nonexistent profile is launched
#### What should happen if a nonexistent profile is launched
The jumplist is only updated when the application is running so the profiles could be modified or deleted outside and the jumplist will not be updated. Handling will be done by whatever handles the command line parsing.
<!-- This -must- go after cppwinrt.build.post.props because that includes many VS-provided props including appcontainer.common.props, which stomps on what cppwinrt.targets did. -->
<ErrorText>This project references NuGet package(s) that are missing on this computer. Use NuGet Package Restore to download them. For more information, see http://go.microsoft.com/fwlink/?LinkID=322105. The missing file is {0}.</ErrorText>
<ErrorText>This project references NuGet package(s) that are missing on this computer. Use NuGet Package Restore to download them. For more information, see http://go.microsoft.com/fwlink/?LinkID=322105. The missing file is {0}.</ErrorText>
<!-- This -must- go after cppwinrt.build.post.props because that includes many VS-provided props including appcontainer.common.props, which stomps on what cppwinrt.targets did. -->
<ErrorText>This project references NuGet package(s) that are missing on this computer. Use NuGet Package Restore to download them. For more information, see http://go.microsoft.com/fwlink/?LinkID=322105. The missing file is {0}.</ErrorText>
<!-- This -must- go after cppwinrt.build.post.props because that includes many VS-provided props including appcontainer.common.props, which stomps on what cppwinrt.targets did. -->
<!-- We can't do dynamic cast because RTTI is off. -->
<!-- RTTI is off because Windows OS policies believe RTTI has too much binary size impact for the value and is less portable than RTTI-off modules. -->
<Rule Id="C26466" Action="None" />
<!-- This one has caught us off guard as it suddenly showed up. Re-enablement is going to be in #2941 -->
<Rule Id="C26814" Action="None" />
<!-- There are *so many* enums that should be enum classes. -->
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
