## 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
Missed this in #11180. I forgot to init the BG opacity with the renderer on startup, because that matters when you have `"antialiasingMode": "cleartype",`.
Repro json
```json
{
"commandline": "cmd.exe",
"guid": "{0caa0dad-35be-5f56-a8ff-afceeeaa6101}",
"hidden": false,
"opacity": 35,
"antialiasingMode": "cleartype",
"padding": "0",
"name": "Command Prompt"
},
```
* [x] Fixes#11315
## Summary of the Pull Request
This replaces the `GridLines` enum in the renderers with a `til::enumset` type, avoiding the need for the various `WI_IsFlagSet` macros and flag operators.
## References
This is followup to PR #10492 which introduced the `enumset` class.
## 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.
## Validation Steps Performed
I've manually confirmed that all the different gridlines are still rendering correctly in both the GDI and DX renderers.
This logic was seemingly redundant. There's two cases I'm looking at here:
#### Case 1
```jsonc
"defaults":
{
"opacity": 35
},
"list":
[
{
"commandline": "cmd.exe",
"name": "Command Prompt"
},
```
In this case, we wouldn't set the `TerminalSettings` Opacity to .35, we'd set it to 1.0, because the profile didn't have an `opactity`.
#### Case 2
```jsonc
"defaults":
{
"useAcrylic": true
},
"list":
[
{
"commandline": "cmd.exe",
"name": "Command Prompt"
},
```
In this case we still want to have an acrylic effect. Previously, we'd default this effect to 50% opaque. I'm not sure that we can actually get that anymore. BUT it turns out, we _can_ have 100% opacity and HostBackdropAcrylic. It is very subtle, but is maybe something we should be allowing anyways. It kinda looks like:
* [x] Fixes#11355
* [x] Regressed in #11180
* [x] I work here
* [x] Fixes a bunch of the checkboxes in #11352
* [x] Fixes one of the boxes in #11353
* [x] The opacity warning -> error gibberish was fixed with the change to `DeserializationError` - `asCString` only works if the `JsonValue` is a string already.
This fixes two issues with profiles.schema.json:
* The `$schema` should not end in a `#`
* `$defs` is the official reserved keyword for schema re-use
See: http://json-schema.org/draft/2020-12/json-schema-core.html
## PR Checklist
* [x] I work here
* [x] Tests added/passed
* [x] Schema updated
## Validation Steps Performed
The previous schema didn't pass https://jschon.dev/, the new schema does.
This commit adds the ability to interact with subtrees of panes.
Have you ever thought that you don't have enough regression testing to
do? Boy do I have the PR for you! This breaks all kinds of assumptions
about what is or is not focused, largely complicated by the fact that a
pane is not a proper control. I did my best to cover as many cases as I
could, but I wouldn't be surprised if there are some things broken that
I am unaware of.
Done:
- Add `parent` and `child` movement directions to move up and down the
tree respectively
- When a parent pane is selected it will have borders all around it in
addition to any borders the children have.
- Fix focus, swap, split, zoom, toggle orientation, resize, and move to
all handle interacting with more than one pane.
- Similarly the actions for font size changing, closing, read-only, clearing
buffer, and changing color scheme will distribute to all children.
- This technically leaves control focus on the original control in the
focused subtree because panes aren't proper controls themselves. This
is also used to make sure we go back down the same path with the
`child` movement.
- You can zoom a parent pane, and click between different zoomed
sub-panes and it won't unzoom you until you use moveFocus or another
action. This wasn't explicitly programmed behavior so it is probably
buggy (I've quashed a couple at least). It is a natural consequence of
showing multiple terminals and allowing you to focus a terminal and a
parent separately, since changing the active pane directly does not
unzoom. This also means there can be a disconnect between what pane is
zoomed and what pane is active.
## Validation Steps Performed
Tested focus movement, swapping, moving panes, and zooming.
Closes#10733
<!-- 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
Continuation of https://github.com/microsoft/terminal/pull/10972 to handle multiple windows, requires that to be merged first.
<!-- 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] Also closes#766
* [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
* [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. 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
Rough changelog:
Normally saving is triggered to occur every 30s, or sooner if a window is created/closed. The existing behavior of saving on last close is maintained to bypass that throttling. The automatic saving allows for crash recovery. Additionally all window layouts will be saved upon taking the `quit` action.
For loading we will check if we are the first window, that there are any saved layouts, and if the setting is enabled, and then depending on if we were given command line args or startup actions.
- create a new window for each saved layout, or
- take the first layout for our self and then a new window for each other layout.
This also saves the layout when the quit action is taken.
Misc changes
- A -s,--saved argument was added to the command line to facilitate opening all of the windows with the right settings. This also means that while a terminal session is running you can do wt -s idx to open a copy of window idx. There isn't a stable ordering of which idx each window gets saved as (it is whatever the iteration order of _peasants is), so it is just a cute hack for now.
- All position calculation has been moved up to AppHost this does mean we need to awkwardly pass around positions in a couple of unexpected places, but no solution was perfect.
- Renamed "Open tabs from a previous session" to "Open windows from a previous session". (not reflected in video below)
- Now save runtime tab color and window names
- Only enabled for non-elevated windows
- Add some change tracking to ApplicationState
<!-- Describe how you validated the behavior. Add automated tests wherever possible, but list manual validation steps taken as well -->
## Validation Steps Performed
`SettingsLoader::_parse` used to skip profiles which didn't have either a "guid"
or "name" field, due to #9962. This is however wrong for fragment loading, as
fragments can alternatively use an "updates" field instead of guid/name.
`SettingsLoader::_parse` was updated to allow profiles with this alternative
field during fragment loading.
## PR Checklist
* [x] Closes#11331
* [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": [
{
"updates": "{574e775e-4f2a-5b96-ac1e-a2962a402336}",
"background": "#FFD700"
}
]
}
```
## Summary of the Pull Request
This introduces a new TIL class that is equivalent in functionality to a `std::bitset`, but where the positions in the bitset are enum values. It also has a few additional methods allowing for setting and testing multiple positions at the same time. The idea is that this class could be used in place of the `WI_SetFlag` and `WI_IsFlagSet` macros when working with sets of flags.
## PR Checklist
* [x] Closes#10432
* [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: #10432
## Validation Steps Performed
I've added a few unit tests that verify the behaviour of all the new methods that aren't part of `std::bitset`. I've also tried it out as a replacement for the `GridLines` enum used in the renderer, and confirmed that it has all the functionality needed to replace that cleanly.
This commit fixes layering of fragment profiles without an update key.
The previous CascadiaSettings deserializer first assembled all builtin
profiles and only then parsed the user's settings.json file.
This meant that even though fragment profiles were added to `_allProfiles`
unconditionally, they did get layered properly with user profiles regardless,
as user profiles were always properly layered.
The new CascadiaSettings approach since 168d28b was a direct translation of this
approach but this is incorrect: As the new approach reads user profiles first,
all inbox profiles, including fragments, must equally use proper layering,
instead of adding profiles unconditionally.
While this commit fixes the bug it maintains a regression:
Duplicate fragment profile GUIDs will not be detected and instead fragments with
identical GUID will all be added as parents to a single user profile.
I considered to fix this regression, but felt that this new behavior is better
than the old one, since a user often can't directly control installed fragments,
and is unlikely to occur in practice. This simplifies the implementation.
## PR Checklist
* [x] Closes#11323
* [x] I work here
* [x] Tests added/passed
## Validation Steps Performed
* Fragment layering works ✔️
Ensures that command-lines constructed to invoke `wt` are escaped properly.
## PR Checklist
* [x] Closes#11273
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [x] Tests added/passed
## Detailed Description of the Pull Request / Additional comments
This was broken in two places - when constructing the command-line in the shell extension and in `NewTerminalArgs::ToCommandline()`.
Both places now invoke a shared method to escape the command-line arguments that require it.
## Validation Steps Performed
Added a test and additionally:
* Invoked the shell extension from `D:\Downloads\With;Semicolon`.
* Added a `newWindow` action to `settings.json` as below and ensured the new window opened without erroring.
```json
{
"command":
{
"action": "newWindow",
"tabTitle": "\";foo\\"
},
"keys": "ctrl+shift+s"
}
```
Adds a check before every UIA function call to ensure the terminal (specifically the buffer) is initialized before doing work. Both the `ScreenInfoUiaProvider` and the `UiaTextRange` are now covered.
## References
Closes#11135#10971 & #11042
## Detailed Description of the Pull Request / Additional comments
Originally, I tried applying this heuristic to all the `RuntimeClassInitialize` on `UiaTextRangeBase` with the philosophy of "a range pointing to an invalid buffer is invalid itself", but that caused a regression on [MSFT 33353327](https://microsoft.visualstudio.com/OS/_workitems/edit/33353327).
`IUiaData` also has `GetTextBuffer()` return a `TextBuffer&`, which cannot be checked for nullness. Instead, I decided to add a function to `IUiaData` that checks if we have a valid state. Since this is shared with Conhost and Conhost doesn't have this issue, I simply make that function say that it's always in a valid state.
## Validation Steps Performed
- [X] Narrator can detect newly created terminals
- [X] (On Windows Server 2022) Windows Terminal does not hang on launch
Implements the following keyboard selection non-configurable key bindings:
- shift+arrow --> move endpoint by character
- ctrl+shift+left/right --> move endpoint by word
- shift+home/end --> move to beginning/end of line
- ctrl+shift+home/end --> move to beginning/end of buffer
This was purposefully done in the ControlCore layer to make keyboard selection an innate part of how the terminal functions (aka a shared component across terminal consumers).
## References
#715 - Keyboard Selection
#2840 - Spec
## Detailed Description of the Pull Request / Additional comment
The most relevant section is `TerminalSelection.cpp`, where we define how each movement operates. It's basically a giant embedded switch-case statement. We leverage a lot of the work done in a11y to perform the movements.
## Validation Steps Performed
- General cases:
- test all of the key bindings added
- Corner cases:
- `char`: wide glyph support
- `word`: move towards, away, and across the selection pivot
- automatically scroll viewport
- ESC (and other key combos) are still clearing the selection properly
I've done this process enough times that I should have written a script
to do it a while ago. This one is rough, but the whole changelog process
is pretty rough.
This script takes multiple revision ranges and produces something that
looks like a rough untranslated changelog, with indicators for how many
of the provided ranges had the same change (deduplicated by title.)
I use a process like this to build the Stable and Preview release notes
out of a set of revision ranges.
This introduces a spec for keyboard selection. This enables the user to create and update a selection without the use of a mouse or stylus.
## References
Contributes to #715
For some weird reason we sometimes receive a WM_KEYDOWN
message without vkey or scanCode if a user drags a tab.
The KeyChord constructor has a debug assertion ensuring that all KeyChord
either have a valid vkey/scanCode. This is important, because this prevents
accidential insertion of invalid KeyChords into classes like ActionMap.
## PR Checklist
* [x] Closes#11076
* [x] I work here
* [x] Tests added/passed
## Validation Steps Performed
* Tab dragging doesn't produce assertion failures anymore ✔️
This fixes the issue I had in the last commit. It's a little weird, but gets
rid of the muckiness of layering. Things that are local to one elevation level
won't pollute the other, and we don't need to worry about layering or where
they came from. Just write shared state to `state.json`, and window state to
`elevated-state`/`user-state`
* [x] delete all state, open terminal elevated - generated profiles go to the correct place, approved commandlines go to the elevated one.
* [x] delete some dynamic profiles, open the terminal - they don't resurrect in either IL
* [ ] GAH saving a window layout unelevated, then trying to save one elevated will blow away the unelevated one
## Summary of the Pull Request
Fixes the 24 failing generated tests. 20 of them were fixed by enforcing the following rule: when moving backwards by word...
- a degenerate range moves to the beginning of the word, then to the word behind it.
- a non-degenerate range outright moves to the word behind it.
The fix was simple: if we're a degenerate range, check if we're at the beginning of the word. If not, move there. Otherwise, move to the word before it. See UiaTextRangeBase.cpp changes for implementation details.
Along the way, several misauthored tests were found:
- 2 generated tests:
- Cause: MS Word considers a line break a word delimiter. We don't use line-wrapping to distinguish two separate words.
- `MovementAtExclusiveEnd` backwards word movement tests:
- `end` will always be `writeTarget` because...
- [degenerate range case] both `start` and `end` are moved to the beginning of the word (`writeTarget`)
- [non-degenerate range case] from the `UiaTextRangeBase` bugfix, we should be moving to the word behind it.
- this misauthored test was explicitly found by fixing the bug first explained here.
## References
#10925 Word navigation testing
This fixes the issue I had in the last commit. It's a little weird, but gets
rid of the muckiness of layering. Things that are local to one elevation level
won't pollute the other, and we don't need to worry about layering or where
they came from. Just write shared state to `state.json`, and window state to
`elevated-state`/`user-state`
This commit reduces the code surface that interacts with raw JSON data,
reducing code complexity and improving maintainability.
Files that needed to be changed drastically were additionally
cleaned up to remove any code cruft that has accrued over time.
In order to facility this the following changes were made:
* Move JSON handling from `CascadiaSettings` into `SettingsLoader`
This allows us to use STL containers for data model instances.
For instance profiles are now added to a hashmap for O(1) lookup.
* JSON parsing within `SettingsLoader` doesn't differentiate between user,
inbox and fragment JSON data, reducing code complexity and size.
It also centralizes common concerns, like profile deduplication and
ensuring that all profiles are assigned a GUID.
* Direct JSON modification, like the insertion of dynamic profiles into
settings.json were removed. This vastly reduces code complexity,
but unfortunately removes support for comments in JSON on first start.
* `ColorScheme`s cannot be layered. As such its `LayerJson` API was replaced
with `FromJson`, allowing us to remove JSON-based color scheme validation.
* `Profile`s used to test their wish to layer using `ShouldBeLayered`, which
was replaced with a GUID-based hashmap lookup on previously parsed profiles.
Further changes were made as improvements upon the previous changes:
* Compact the JSON files embedded binary, saving 28kB
* Prevent double-initialization of the color table in `ColorScheme`
* Making `til::color` getters `constexpr`, allow better optimizations
The result is a reduction of:
* 48kB binary size for the Settings.Model.dll
* 5-10% startup duration
* 26% code for the `CascadiaSettings` class
* 1% overall code in this project
Furthermore this results in the following breaking changes:
* The long deprecated "globals" settings object will not be detected and no
warning will be created during load.
* The initial creation of a new settings.json will not produce helpful comments.
Both cases are caused by the removal of manual JSON handling and the
move to representing the settings file with model objects instead
## PR Checklist
* [x] Closes#5276
* [x] Closes#7421
* [x] I work here
* [x] Tests added/passed
## Validation Steps Performed
* Out-of-box-experience is identical to before ✔️
(Except for the settings.json file lacking comments.)
* Existing user settings load correctly ✔️
* New WSL instances are added to user settings ✔️
* New fragments are added to user settings ✔️
* All profiles are assigned GUIDs ✔️
* [x] delete all state, open terminal elevated - generated profiles go to the correct place, approved commandlines go to the elevated one.
* [x] delete some dynamic profiles, open the terminal - they don't resurrect in either IL
* [ ] GAH saving a window layout unelevated, then trying to save one elevated will blow away the unelevated one
