Don't crash if we try to save the window layout while we are closing, and try to avoid saving at all.
Might impact #11354
## Detailed Description of the Pull Request / Additional comments
- Revoke the event handler/save throttler so we don't even try to get the window layout when we are closing
- Try to check for nullptrs, but then apply `try {} CATCH_LOG()` liberally
## Validation Steps Performed
The happy path of saving normally is still fine, but I haven't been unlucky enough to trigger the crash myself.
This commit reduces the number of generated VS profiles from 6 down to just 2
per VS instance. The reason we did this is out of concern of overwhelming or
annoying new users with too many profiles. Especially since it's far easier
at the moment to add new generators compared to removing them.
As before only the latest instance is not hidden by default.
## PR Checklist
* [x] I work here
* [x] Tests added/passed
* [x] As discussed in a Team Sync meeting
## Validation Steps Performed
* Installed Visual Studio 2019 and 2022 Preview
* A profile for both is generated, while the 2019 one is hidden by default ✔️
* $env:VSCMD_ARG_TGT_ARCH is x64 on my AMD64 machine ✔️
Considering the number of reports of "defterm isn't working (mysteriously)", I figured more logging current hurt. I also added a wprp profile for the defterm logging as well, which should capture conhost side things as well.
From an elevated conhost:
```
wpr -start path\to\Terminal.wprp!Defterm.Verbose
wpr -stop %USERPROFILE%\defterm-trace.etl
```
* [x] I work here
* [x] relevant to: #10594, #11529, #11524.
## Summary of the Pull Request
Currently when configuring the action
```json
{ "command": { "action": "closeTabsAfter" } }
```
we get a schema error in VSCode: `Matches multiple schemas when only one must validate.`.
The problem is that it matches both `closeTabsAfter` and `closeTab`, since the schema uses regex patterns to match instead of plain strings. I swapped the usage of `"pattern"` with `"const"` for all actions.
## PR Checklist
* [ ] Closes #xxx
* [x] CLA signed
* [ ] Tests added/passed
* [ ] Documentation updated
* [x] 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
## Detailed Description of the Pull Request / Additional comments
## Validation Steps Performed
I checked and this action configuration no longer errors.
Up until this commit PSReadline caused OutputDebugString to be called
with a complex log message to on every newline. At the time of writing,
Visual Studio's Output window is fairly slow and after this change newlines
feel a fair bit snappier when running under Visual Studio's debugger.
## Validation Steps Performed
* pwsh.exe continues to work correctly ✔️
ControlCore::FontFaceName() is called 10/s by TSFInputControl.
The getter was modified to cache the STL string in a hstring allowing
us to return a value without temporary allocations during runtime.
## PR Checklist
* [x] I work here
* [x] Tests added/passed
## Validation Steps Performed
* Font face and size changes properly update TSFInputControl ✔️
## Summary of the Pull Request
Currently when configuring the action
```json
{ "command": { "action": "commandPalette", "launchMode": "commandLine" }, "key": "ctrl+shift+p" }
```
or
```json
{ "command": { "action": "multipleActions", "actions": [{ "action": "paste" }] }, "key": "ctrl+shift+v" }
```
we get a schema error in VSCode. These object variants of the actions were not configured properly in the schema, so I fixed it.
## PR Checklist
* [ ] Closes #xxx
* [x] CLA signed
* [ ] Tests added/passed
* [ ] Documentation updated
* [x] 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
## Detailed Description of the Pull Request / Additional comments
In the schema there is a big `oneOf` for the `command` of an action under `actions`.
Commands that also accept extra arguments have an object type defined for it.
The `commandPalette` and `multipleActions` commands accept extra arguments, and also have matching `CommandPaletteAction` and `MultipleActionsAction` object types defined, but they are unused.
So I added them to the `oneOf` array in the correct placement.
## Validation Steps Performed
## Summary of the Pull Request
The `settings.json` schema had `"default"`s for some boolean settings set as quoted strings (`"true"` / `"false"`), so I removed the quotes.
## PR Checklist
* [ ] Closes #xxx
* [x] CLA signed
* [ ] Tests added/passed
* [ ] Documentation updated
* [x] 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
## Detailed Description of the Pull Request / Additional comments
VSCode autocompletes the default value when you select the setting in intellisense, so it autocompleted a string which caused a schema error. Booleans should be JSON booleans, not quoted.
## Validation Steps Performed
## Summary of the Pull Request
As a part of the Interactivity split, `TermControlAutomationPeer` had to be split into `TermControlAutomationPeer` (TCAP) and `InteractivityAutomationPeer` (IAP). Just about all of the functions in `InterativityAutomationPeer` operate by calling the non-XAML UIA Provider then wrapping the resulting `UIATextRange` into a XAML format (a `XamlUiaTextRange` [XUTR]). As a part of that XUTR constructor, we need a reference to the parent provider.
We generally get that via `ProviderFromPeer()`, but IAP's `ProviderFromPeer()` returned null (presumably because IAP isn't in the UI tree, whereas TCAP is directly registered as the automation peer for the `TermControl`).
It looks like some screen readers didn't care (like NVDA, though there may be a chance we just didn't encounter an issue just yet), but Narrator definitely did.
The fix was to provide XUTR constructors the `ProviderFromPeer` from TCAP, _not_ IAP. To accomplish this, IAP now holds a weak reference to TCAP, and provides the `ProviderFromPeer` when needed. We can't cache this result because there is no guarantee that it won't change.
Some miscellaneous changes include:
- `TermControl::OnCreateAutomationPeer` now returns the existing auto peer instead of always creating a new one
- `TCAP::WrapArrayOfTextRangeProviders` was removed as it was unused (normally, this would be directly affected by the main `ProviderFromPeer` change here)
- `XUTR::GetEnclosingElement` is now hooked up to trace logging for debugging purposes
## References
Introduced in #10051Closes#11488
## Validation Steps Performed
✅ Narrator scan mode now works (verified with character, word, and line navigation)
✅ NVDA movement still works (verified with word and line navigation)
## Summary of the Pull Request
When the window moves, hide any visible content dialog (only one can be shown at a time) and ensure its associated async operation is terminated.
#10922 dismisses any open popups when the window is moved or any scroll viewer scrolls. However, if you just close a Popup from the UI tree, the async operation associated to a ContentDialog (started with `dialog.ShowAsync`) does not terminate. The dialog lock that prevents opening multiple dialogs at the same time is not released, and no further dialog can be shown.
Explicitly dismissing the only visible ContentDialog using its `Hide` method terminates the operation.
## Validation Steps Performed
Manual tests, open up dialogs and move the window (like in #11425)
References #10922Closes#11425
This commit enables /fp:fast. This doubles the performance of the Delta E
computation in #11095 for instance. Additionally it re-enables two options for
debug builds which are normally enabled by default by Visual Studio.
## PR Checklist
* [x] I work here
* [x] Tests added/passed
## Validation Steps Performed
* No change in binary size
* No obvious change in behavior
This change enables access to the Defaults page in stable builds of
terminal. It is intended that we backport this feature flag edit to
1.11, so that Defaults can roll out with 1.11 when it becomes stable.
If we find that the settings file doesn't exist, or is empty, then let's quick
delete the state file as well. If the user does have a state file, and not a
settings, then they probably tried to reset their settings. It might have data
in it that was only relevant for a previous iteration of the settings file. If
we don't, we'll load the old state and ignore all dynamic profiles (for
example)!
We'll remove all of the data in the `ApplicationState` object and reset it to
the defaults.
This will delete the state file!
That's the sure-fire way to make sure the data doesn't come back. If we leave
it untouched, then when we go to write the file back out, we'll first re-read
it's contents and try to overlay our new state. However, nullopts won't remove
keys from the JSON, so we'll end up with the original state in the file.
* [x] Closes#11119
* [x] Tested on a cold launch of the Terminal with an existing `state.json`
and an empty `settings.json`
* [x] Tested a hot-reload of deleting the `settings.json`
This implements command line matching for `CascadiaSettings::GetProfileForArgs`.
The command lines for all user profiles are resolved to absolute file paths,
argument quotes are standardized ("canonicalized") and the results are cached.
When `GetProfileForArgs` is called with a Commandline() value, we "canonicalize"
the argument as well and find the profile that is the longest prefix.
If none could be found the default profile is returned.
## PR Checklist
* [x] Closes#9458
* [x] Closes#10952
* [x] I work here
* [ ] Tests added/passed
## Validation Steps Performed
* Open a `cmd.exe` tab in the store-version of WT
* Run `start cmd`
--> A tab with the `cmd.exe` profile opens
* Run `start pwsh.exe`
--> A tab with the PowerShell 7 profile opens
* Run PowerShell 7 from the start menu
--> A tab with the PowerShell 7 profile opens
* Create a symlink for PowerShell 7 and launch `pwsh.exe` from there
--> A tab with the PowerShell 7 profile opens
I thought that microsoft/microsoft-ui-xaml#3183 might just fix this for us, but it didn't. We've got our RadioButton's all up in SettingsContainers, so they all think they're `AutomationProperties.AccessibilityView="Raw"` for some reason. If you simply add the `Content` to these, then they all end up correct in Accessibility Insights
## PR Checklist
* [x] Will take care of #11248 but I can't be the one to close it.
* [x] I work here
* [x] Tests added/passed
* [n/a] Requires documentation to be updated
This commit adds a simple information popup about default terminals,
guiding first-time Windows 11 users into changing the default terminal.
## Validation Steps Performed
* Info bar pops up on Windows 11 ✔️
* Info bar can be dismissed persistently ✔️
This was originally in #11308. Thought we should check it in for 1.12 even
though that won't merge this release. Should slightly mitigate the number of
users that see this warning.
`CascadiaSettings::_createNewProfile` failed to call `_FinalizeInheritance`.
This commits fixes the issue and adds a stern warning for future me.
## PR Checklist
* [x] Closes#11392
* [x] I work here
* [x] Tests added/passed
## Validation Steps Performed
* Open settings UI
* Modify font size in base layer
* _Don't_ save
* Duplicate any profile with default font size
* Ensure the duplicated profile shows the modified base layer font size ✔️
In #11180 we made `opacity` independent from `useAcrylic`. We also changed the mouse wheel behavior to only change opacity, and not mess with acrylic.
However, on Windows 10, vintage opacity doesn't work at all. So there, we still need to manually enable acrylic when the user requests opacity.
* [x] Closes#11285
SUI changes in action:
It's possible that we're about to be started, _before_
our paired connection is started. Both will get Start()'ed when
their owning TermControl is finally laid out. However, if we're
started first, then we'll immediately start printing to the other
control as well, which might not have initialized yet. If we do
that, we'll explode.
Instead, wait here until the other connection is started too,
before actually starting the connection to the client app. This
will ensure both controls are initialized before the client app
is.
Fixes#11282
Tested: Opened about 100 debug taps. They all worked. :shipit:
`CascadiaSettings` is default constructed when human readable error messages are
returned. Even in such cases we need to ensure that all fields are properly
initialized, as a caller might decide to call a `GlobalSettings` getter.
Thus a crash occurred whenever a user was hot-reloading their settings file with
invalid JSON as other code then tried to compare the `GlobalSettings()`.
## PR Checklist
* [x] I work here
* [x] Tests added/passed
## Validation Steps Performed
* Start Windows Terminal and ensure the settings load fine
* Add `"commandline": 123` to any of the generated profiles in settings.json
* The application doesn't crash and shows a warning message
WinUI/XAML requires the `SelectedItem` to be member of the list of
`ItemsSource`. `CascadiaSettings::DefaultTerminals()` is such an `ItemsSource`
and is called every time the launch settings page is visited.
It calls `DefaultTerminal::Available()` which in turn calls `Refresh()`.
While the `SelectedItem` was cached in `CascadiaSettings`, the value of
`DefaultTerminals()` wasn't. Thus every time the page was visited, it refreshed
the `ItemsSource` list without invalidating the current `SelectedItem`.
This commit prevents such accidental mishaps from occurring in the future,
by moving the responsibility of caching solely to the `CascadiaSettings` class.
## PR Checklist
* [x] Closes#11424
* [x] I work here
* [x] Tests added/passed
## Validation Steps Performed
* Navigating between SUI pages maintains the current dropdown selection ✔️
* Saving the settings saves the correct terminal at `HKCU:\Console\%%Startup` ✔️
## Summary of the Pull Request
Fixes two issues related to SUI's Actions page:
1. Crash when adding an action and setting key chord to one that is already taken
- **Cause**: the new key binding that was introduced with the "Add new" button appears in `_KeyBindingList` that we're iterating over. This has no `CurrentKeys()`, resulting in a null pointer exception.
- **Fix**: null-check it
2. There's an action that appears as being nameless in the dropdown
- **Cause**: The culprit seems to be `MultipleActions`. We would register it, but it wouldn't have a name, so it would appear as a nameless option.
- **Fix**: if it has no name, don't register it. This is also future-proof in that any new nameless actions won't be automatically added.
Closes#10981
Part of #11353
This commit fixes various failing TestHostApp unit tests.
Most of these broke as part of 168d28b (#11184).
## PR Checklist
* [x] Closes#11339
* [x] I work here
* [x] Tests added/passed
I've had a hard time with the tab colors this week.
Turns out that setting the background to nullptr will make the tabviewitem invisible to hit tests. `Transparent`, on the other hand, is totally valid, and the expected default.
Tabs as of this commit:
## PR Checklist
* [x] Closes#11382
* [x] I work here
This low-key reverts a bit of #11369, which fixed#11294, which regressed in #11240
This fixes an issue that Touch Keyboard is not invoked when user taps on the PowerShell.
Before this change, it was returning small rectangle on the right of the cursor. Touch Keyboard should be invoked by tapping anywhere inside the console.
## 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
ITfContextOwner::GetScreenExt is used to define rectangle that can invoke Touch Keyboard.
https://docs.microsoft.com/en-us/windows/win32/api/msctf/nf-msctf-itfcontextowner-getscreenext
## Validation Steps Performed
* [x] Touch keyboard was invoked by tapping inside the Console while Hardware Keyboard was not attached.
* [x] Selecting text worked as expected without invoking touch keyboard.
* [x] Long tapping the console invoked Touch Keyboard. I would like to confirm if this is the expected behavior.
## Summary of the Pull Request
In `settings.json` there's an `actions` array to configure keybindings.
The action `globalSummon` has an argument called `dropdownDuration`.
The settings editor deleted this argument from the settings because of a typo in `ActionArgs.h`.
## PR Checklist
* [x] Closes#11400
* [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.
## Detailed Description of the Pull Request / Additional comments
There was a `JsonUtils::GetValueForKey` instead of a `JsonUtils::SetValueForKey`.
This is what happens when such code is not autogenerated.
## Validation Steps Performed
None
## Summary of the Pull Request
Fixes the "Reset to inherited value" button for the opacity slider and removes the unwanted padding between the header and the control.
## PR Checklist
* [x] Closes#11352
* [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
## Validation Steps Performed
Manually tested
## Summary of the Pull Request
The code saved `args.DropdownDuration()` to a local and then called the function again, instead of using the local.
Changed to use the local.
## 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.
## Detailed Description of the Pull Request / Additional comments
I think this getter simply accesses a member on `args`, it doesn't parse the settings or anything, so compiler optimizes it, but seemed to make more sense to use the local.
<!-- 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
Try to save the working directory if we know what it is (just copied what was done in duplicating a pane). I overlooked this in my original implementation that always used the settings StartingDirectory.
<!-- Other than the issue solved, is this relevant to any other issues/existing PRs? -->
## References
#9800
<!-- Please review the items on the PR checklist before submitting-->
## 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.
* [ ] 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
Tried setting the working directory using the OSC 9;9 escape and confirmed that the directory saves correctly.
## Summary of the Pull Request
The type of the `"id"` argument of the `focusPane` action under `"actions"` in the `settings.json` schema was incorrectly set to a string.
It's actually expecting a non-negative number, and defaults to 0.
So I fixed the schema.
## PR Checklist
* [x] Closes#11393
* [x] CLA signed
* [ ] Tests added/passed
* [ ] Documentation updated
* [x] 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.
## Detailed Description of the Pull Request / Additional comments
## Validation Steps Performed
I've validated that a string makes Windows Terminal complain it's a string and not a number, and that a number works as expected, and that the default is indeed zero.
This commit introduces a number of poor abstractions to split
`SettingsLoader::_parse` into `_parse` for content in the format of the user's
settings.json and `_parseFragment` which is specialized for fragment files.
The latter suppresses exceptions and supports the "updates" key for profiles.
## PR Checklist
* [x] Closes#11330
* [x] I work here
* [x] Tests added/passed
## Validation Steps Performed
* Wrote the following to
`%LOCALAPPDATA%\Microsoft\Windows Terminal\Fragments\test\test.json`:
```json
{
"profiles": [
{
"name": "bad",
"unfocusedAppearance": ""
},
{
"name": "good"
},
{
"updates": "{574e775e-4f2a-5b96-ac1e-a2962a402336}",
"background": "#333"
}
]
}
```
* Ensured that "bad" is ignored ✔️
* Ensured that "good" shows up and works ✔️
* Ensured that the pwsh profile has a gray background ✔️
Just like in #9760, we can't actually use the UWP file picker API, because it will absolutely not work at all when the Terminal is running elevated. That would prevent the picker from appearing at all. So instead, we'll just use the shell32 one manually.
This also gets rid of the confirmation dialog, since the team felt we didn't really need that. We could maybe replace it with a Toast (#8592), but _meh_
* [x] closes#11356
* [x] closes#11358
* This is a lot like #9760
* introduced in #11062
* megathread: #9700
## Summary of the Pull Request
The deadlock was caused by `ScreenInfoUiaProviderBase::GetSelection()` calling `TermControlUiaProvider::GetSelectionRange` (both of which attempted to lock the console). This PR removes the lock and initialization check from `TermControlUiaProvider`. It is no longer necessary because the only one that calls it is `SIUPB::GetSelection()`.
Additionally, this adds some code that was useful in debugging this race condition. That should help us figure out any locking issues that may come up in the future.
## References
#11312Closes#11385
## Validation Steps Performed
✅ Repro steps don't cause hang
## Summary of the Pull Request
Similar to `vswhere -latest`, show only the latest Visual Studio command prompts / developer PowerShell. This was tested by deleting the local package state and testing against fresh state with both VS2019 and VS2022 Preview installed, and indeed VS2022 Preview (both cmd and powershell) show. The other profiles were generated but hidden by default.
## References
Modification of PR #7774
## PR Checklist
* [x] Closes#11307
* [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.
* [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
The sort algorithm is the same basic algorithm I used in https://github.com/microsoft/vswhere. It sorts first by installation version with a secondary sort based on the install date in case the installation versions are the same.
## Validation Steps Performed
With both VS2019 and VS2022 Preview installed, I made sure the initial state was expected, and tried different combinations of hiding and unhiding generated entries, and restarted Terminal to make sure my settings "stuck".
All these controls didn't have `Name`s assigned, and Accessibility Insights doesn't like that. Their parents did, but the actual focusable elements themselves didn't. So I've just taken the nearby headers for these things and slapped them in as the Automation names for these controls.
I verified that each of these automated tests in Accessibility Insights pass again.
* Will do the thing to #11155 but we need confirmation before that can be closed.
DESPITE the fact that there's a `Background()` API that we
could just call like:
```c++
TabViewItem().Background(deselectedTabBrush);
```
We actually can't, because it will make the part of the tab that
doesn't contain the text totally transparent to hit tests. So we
actually _do_ still need to set `TabViewItemHeaderBackground` manually.
* Regressed in #11240
* Root cause up in https://github.com/microsoft/microsoft-ui-xaml/pull/3769
* [x] closes#11294
PankajBhojwani
Schuyler Rosefield
Leonard Hecker
Mike Griese
NotWearingPants
Dmitriy Fishman
Carlos Zamora
Dustin L. Howett
MPela
Leon Liang
Yoshiko
NotWearingPants
Ian O'Neill
Heath Stewart
Michael Niksa
