I believe that there is a hidden "default branch" setting in Azure
DevOps, and ours is set to "master". CG only attempts to automatically
inject to builds off the default branch... and because we're a GitHub
repo running builds in AzDO we _can't change what the default branch
is_. Oops.
A redo of #6290. That PR was overkill. In that one, we'd toss focus back to the active control any time that the tab view item got focus. That's maybe not the _best_ solution.
Instead, this PR is precision strikes. We're re-using a lot of what we already have from #9260.
* When the context menu is closed, yeet focus to the control.
* When the renamer is dismissed, yeet focus to the control.
* When the TabViewItem is tapped (meaning no one else handled it), yeet focus to the control.
### checklist
* [x] I work here
* [ ] This is UI so it doesn't have tests
* [x] Closes#3609
* [x] Closes#5750
* [x] Closes#6680
### scenarios:
* [x] focus the window by clicking on the tab -> Control is focused.
* [x] Open the color picker with the context menu, can move the focus inside the picker with the arrow keys.
* [x] Dismiss the picker with esc -> Control is focused.
* [x] Dismiss the picker with enter -> Control is focused.
* [x] Dismiss the renamer with esc -> Control is focused.
* [x] Dismiss the renamer with enter -> Control is focused.
* [x] Dismiss the context menu with esc -> Control is focused.
* [x] Start renaming, then click on the tab -> Rename is committed, Control is focused.
* [x] Start renaming, then click on the text box -> focus is still in the text box
This test has some load bearing specifics in it.
In short, it must trigger multiple boost::small_vector growths. The
vector is reused for the lifetime of the console (!) so it must be
forced to act the way the repro requires.
Oh, and the test needs to be run with Application Verifier +Heap.
This validates the fix in c0ab9cb5b5.
## Summary of the Pull Request
ControlCore::AttachUiaEngine receives a IRenderEngine as a raw pointer,
which TermControl owns. We must ensure that we first destroy the
ControlCore before the UiaEngine instance (both owned by TermControl).
Otherwise a deallocated IRenderEngine is accessed when
ControlCore calls Renderer::TriggerTeardown.
## References
This crash was introduced in #10031.
## PR Checklist
* [x] I work here
* [x] Tests added/passed
## Validation Steps Performed
* Run accevent.exe to cause a UiaEngine to be attached to a TermControl.
* Close the current tab
* Ensured no crashes occur
Until there's a "Wait", there's usually only one API message inflight at
a time.
In our quest for performance, we put that single API message in charge
of its own buffer management: instead of allocating buffers on the heap
and deleting them later (storing pointers to them at the far corners of
the earth), it would instead allocate them from small internal pools (if
possible) and only heap allocate (transparently) if necessary. The
pointers flung to the corners of the earth would be pointers (1) back
into the API_MSG or (2) to a heap block owned by boost::small_vector.
It took us months to realize that those bare pointers were being held by
COOKED_READ and RAW_READ and not actually being updated when the API
message was _copied_ as it was shuffled off to the background to become
a "Wait" message.
It turns out that it's trivially possible to crash the console by
sending two API calls--one that waits and one that completes
immediately--when the waiting message or the "wait completer" has a
bunch of dangling pointers in it. It further turns out that some
accessibility software (like JAWS) attaches directly to the console
session, much like winpty and ConEmu and friends. They're trying to read
out the buffer (API call!) and sometimes there's a shell waiting for
input (API call!). Oops.
In this commit, we fix up the message's internal pointers (in lieu of
giving it a proper copy constructor; see GH-10076) and then tell the
wait completion routine (which is going to be a COOKED_READ, RAW_READ,
DirectRead or WriteData) about the new buffer location.
This is a scoped fix that should be replaced (TODO GH-10076) with a
final one after Ask mode.
Retrieved from https://microsoft.visualstudio.com os.2020 OS official/rs_wdx_dxp_windev eca0875950fd3a9735662474613405e2dc06f485
References GH-10076
Fixes MSFT-33127449
Fixes GH-9692
The ES folks removed support for implicit binary dependencies in test projects, which required our testmd to change to include its dependency on testnetv5.
Retrieved from https://microsoft.visualstudio.com os.2020 OS official/rs_wdx_dxp_windev a2a74747feda018c347ba787e166e8385bb17e47
Related work items: MSFT-31480819
## PR Checklist
* [x] Closes#9920
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
## Validation Steps Performed
- When opening a profile page and selecting a random pivot, when then navigating away from the profile page to another item and navigating back to the same profile the last selected pivot was still active.
- When opening a profile page and selecting a random pivot, when then navigating away from the profile page to a second profile the last selected pivot was immediately active on the second profile.
- When selecting random pivots and navigating only between other profile pages the last selected pivot was always active on the opened profile page.
What felt a bit strange at first sight was:
- open a profile page, select a random pivot
- start adding a new profile
- the last selected pivot is immediately active and not the "General" pivot what might be the obvious pivot to show when a user wants to add a new profile.
After playing a bit more with this this started to feel ok. Since as a user you might go to "Add new profile" immediately anyway instead of opening other profiles first.
Just explicitly highlighting this last point since maybe someone prefers that we need to make sure to always start with the 'General' pivot when adding a new profile and then this fix is not correct.
<!-- 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
When the client application sets the console input mode with the `ENABLE_MOUSE_INPUT` flag, we send along the appropriate VT sequences to the connected terminal (if there is one).
<!-- Other than the issue solved, is this relevant to any other issues/existing PRs? -->
## References
#376#6859
<!-- 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.
* [x] I work here
<!-- Describe how you validated the behavior. Add automated tests wherever possible, but list manual validation steps taken as well -->
## Validation Steps Performed
Far manager works
