This is an attempt to simplify the SGR (Select Graphic Rendition)
implementation in conhost, to cut down on the number of methods required
in the `ConGetSet` interface, and pave the way for future improvements
and bug fixes. It already fixes one bug that prevented SGR 0 from being
correctly applied when combined with meta attributes.
* This a first step towards fixing the conpty narrowing bugs in issue
#2661
* I'm hoping the simplification of `ConGetSet` will also help with
#3849.
* Some of the `TextAttribute` refactoring in this PR overlaps with
similar work in PR #1978.
## Detailed Description of the Pull Request / Additional comments
The main point of this PR was to simplify the
`AdaptDispatch::SetGraphicsRendition` implementation. So instead of
having it call a half a dozen methods in the `ConGetSet` API, depending
on what kinds of attributes needed to be set, there is now just one call
to get current attributes, and another call to set the new value. All
adjustments to the attributes are made in the `AdaptDispatch` class, in
a simple switch statement.
To help with this refactoring, I also made some change to the
`TextAttribute` class to make it easier to work with. This included
adding a set of methods for setting (and getting) the individual
attribute flags, instead of having the calling code being exposed to the
internal attribute structures and messing with bit manipulation. I've
tried to get rid of any methods that were directly setting legacy, meta,
and extended attributes.
Other than the fix to the `SGR 0` bug, the `AdaptDispatch` refactoring
mostly follows the behaviour of the original code. In particular, it
still maps the `SGR 38/48` indexed colors to RGB instead of retaining
the index, which is what we ultimately need it to do. Fixing that will
first require the color tables to be unified (issue #1223), which I'm
hoping to address in a followup PR.
But for now, mapping the indexed colors to RGB values required adding an
an additional `ConGetSet` API to lookup the color table entries. In the
future that won't be necessary, but the API will still be useful for
other color reporting operations that we may want to support. I've made
this API, and the existing setter, standardise on index values being in
the "Xterm" order, since that'll be essential for unifying the code with
the terminal adapter one day.
I should also point out one minor change to the `SGR 38/48` behavior,
which is that out-of-range RGB colors are now ignored rather than being
clamped, since that matches the way Xterm works.
## Validation Steps Performed
This refactoring has obviously required corresponding changes to the
unit tests, but most were just minor updates to use the new
`TextAttribute` methods without any real change in behavior. However,
the adapter tests did require significant changes to accommodate the new
`ConGetSet` API. The basic structure of the tests remain the same, but
the simpler API has meant fewer values needed to be checked in each test
case. I think they are all still covering the areas there were intended
to, though, and they are all still passing.
Other than getting the unit tests to work, I've also done a bunch of
manual testing of my own. I've made sure the color tests in Vttest all
still work as well as they used to. And I've confirmed that the test
case from issue #5341 is now working correctly.
Closes#5341
cmatrix is somewhat of a pathological case for our infrastructure: it
prints out a bunch of green and white characters and then updates them a
million times a second.
It also maintains a column of space between every green character. When
it prints this column, it prints it in "default" or "white". This ends
up making runs of text that look like this:
(def: G=green B=bright white W=white *=matrix char =space)
G W G W G W G W G W G W G W G W
G W G W G W G W G W G W G W G W
G W G W G W G W G W G W G W G W
G W G W G W G W G W G W G W G W
G W G W G W G W G W G W G W G W
G W G W G W G W G W G W G W G W
G W G W G W G W G W G W G W G W
G W G W G W G W G W G W G W G W
As characters trickle in:
G*W G*W G*W G*W G*W G*W G*W B*W
G*W G*W G*W G*W G*W G*W G*W G W
G*W G*W G*W B*W G*W G*W G*W G W
G*W B*W G*W G W G*W G*W G*W G*W
G*W G W G*W G W G*W B*W G*W G*W
B*W G W G*W G W G*W G W B*W G*W
G W G W G*W G W G*W G W G W B*W
G W G W B*W G W G*W G W G W G W
Every one of those color transitions causes us to break up the run of
text and start rendering it again. This impacts GDI, Direct2D *and*
ConPTY. In the example above, there are 120 runs.
The problem is, printing a space doesn't **use** the foreground color!
This commit introduces an optimization. When we're about to break a text
cluster because its attributes changed, we make sure that it's not just
filled with spaces and doesn't differ in any visually-meaningful way
(like underline or strikethrough, considering global invert state).
This lets us optimize both the rendering _and_ the PTY output to look
like this:
G* * * * * * * B*G
G* * * * * * *
G* * * B*G * * *
G* B*G * * * * *
G* * * B*G * *
B*G * * B*G *
G * * B*G
G B*G *
Text will be printed at best line-by-line and at worst only when the
visible properties of the screen actually change. In the example
above, there are only 21 runs.
This speeds up cmatrix remarkably.
Refs #1064
## Summary of the Pull Request
Operations that erase areas of the screen are typically meant to do so using the current color attributes, but with the rendition attributes reset (what we refer to as meta attributes). This also includes scroll operations that have to clear the area of the screen that has scrolled into view. The only exception is the _Erase Scrollback_ operation, which needs to reset the buffer with the default attributes. This PR updates all of these cases to apply the correct attributes when scrolling and erasing.
## PR Checklist
* [x] Closes#2553
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [x] Tests added/passed
* [ ] Requires documentation to be updated
* [ ] I've not really discussed this with core contributors. 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
My initial plan was to use a special case legacy attribute value to indicate the "standard erase attribute" which could safely be passed through the legacy APIs. But this wouldn't cover the cases that required default attributes to be used. And then with the changes in PR #2668 and #2987, it became clear that our requirements could be better achieved with a couple of new private APIs that wouldn't have to depend on legacy attribute hacks at all.
To that end, I've added the `PrivateFillRegion` and `PrivateScrollRegion` APIs to the `ConGetSet` interface. These are just thin wrappers around the existing `SCREEN_INFORMATION::Write` method and the `ScrollRegion` function respectively, but with a simple boolean parameter to choose between filling with default attributes or the standard erase attributes (i.e the current colors but with meta attributes reset).
With those new APIs in place, I could then update most scroll operations to use `PrivateScrollRegion`, and most erase operations to use `PrivateFillRegion`.
The functions affected by scrolling included:
* `DoSrvPrivateReverseLineFeed` (the RI command)
* `DoSrvPrivateModifyLinesImpl` (the IL and DL commands)
* `AdaptDispatch::_InsertDeleteHelper` (the ICH and DCH commands)
* `AdaptDispatch::_ScrollMovement` (the SU and SD commands)
The functions affected by erasing included:
* `AdaptDispatch::_EraseSingleLineHelper` (the EL command, and most ED variants)
* `AdaptDispatch::EraseCharacters` (the ECH command)
While updating these erase methods, I noticed that both of them also required boundary fixes similar to those in PR #2505 (i.e. the horizontal extent of the erase operation should apply to the full width of the buffer, and not just the current viewport width), so I've addressed that at the same time.
In addition to the changes above, there were also a few special cases, the first being the line feed handling, which required updating in a number of places to use the correct erase attributes:
* `SCREEN_INFORMATION::InitializeCursorRowAttributes` - this is used to initialise the rows that pan into view when the viewport is moved down the buffer.
* `TextBuffer::IncrementCircularBuffer` - this occurs when we scroll passed the very end of the buffer, and a recycled row now needs to be reinitialised.
* `AdjustCursorPosition` - when within margin boundaries, this relies on a couple of direct calls to `ScrollRegion` which needed to be passed the correct fill attributes.
The second special case was the full screen erase sequence (`ESC 2 J`), which is handled separately from the other ED sequences. This required updating the `SCREEN_INFORMATION::VtEraseAll` method to use the standard erase attributes, and also required changes to the horizontal extent of the filled area, since it should have been clearing the full buffer width (the same issue as the other erase operations mentioned above).
Finally, there was the `AdaptDispatch::_EraseScrollback` method, which uses both scroll and fill operations, which could now be handled by the new `PrivateScrollRegion` and `PrivateFillRegion` APIs. But in this case we needed to fill with the default attributes rather than the standard erase attributes. And again this implementation needed some changes to make sure the full width of the active area was retained after the erase, similar to the horizontal boundary issues with the other erase operations.
Once all these changes were made, there were a few areas of the code that could then be simplified quite a bit. The `FillConsoleOutputCharacterW`, `FillConsoleOutputAttribute`, and `ScrollConsoleScreenBufferW` were no longer needed in the `ConGetSet` interface, so all of that code could now be removed. The `_EraseSingleLineDistanceHelper` and `_EraseAreaHelper` methods in the `AdaptDispatch` class were also no longer required and could be removed.
Then there were the hacks to handle legacy default colors in the `FillConsoleOutputAttributeImpl` and `ScrollConsoleScreenBufferWImpl` implementations. Since those hacks were only needed for VT operations, and the VT code no longer calls those methods, there was no longer a need to retain that behaviour (in fact there are probably some edge cases where that behaviour might have been considered a bug when reached via the public console APIs).
## Validation Steps Performed
For most of the scrolling operations there were already existing tests in place, and those could easily be extended to check that the meta attributes were correctly reset when filling the revealed lines of the scrolling region.
In the screen buffer tests, I made updates of that sort to the `ScrollOperations` method (handling SU, SD, IL, DL, and RI), the `InsertChars` and `DeleteChars` methods (ICH and DCH), and the `VtNewlinePastViewport` method (LF). I also added a new `VtNewlinePastEndOfBuffer` test to check the case where the line feed causes the viewport to pan past the end of the buffer.
The erase operations, however, were being covered by adapter tests, and those aren't really suited for this kind of functionality (the same sort of issue came up in PR #2505). As a result I've had to reimplement those tests as screen buffer tests.
Most of the erase operations are covered by the `EraseTests` method, except the for the scrollback erase which has a dedicated `EraseScrollbackTests` method. I've also had to replace the `HardReset` adapter test, but that was already mostly covered by the `HardResetBuffer` screen buffer test, which I've now extended slightly (it could do with some more checks, but I think that can wait for a future PR when we're fixing other RIS issues).
## Summary of the Pull Request
Adds support for Italics, Blinking, Invisible, CrossedOut text, THROUGH CONPTY. This does **NOT** add support for those styles to conhost or the terminal.
We will store these "Extended Text Attributes" in a `TextAttribute`. When we go to render a line, we'll see if the state has changed from our previous state, and if so, we'll appropriately toggle that state with VT. Boldness has been moved from a `bool` to a single bit in these flags.
Technically, now that these are stored in the buffer, we only need to make changes to the renderers to be able to support them. That's not being done as a part of this PR however.
## References
See also #2915 and #2916, which are some follow-up tasks from this fix. I thought them too risky for 20H1.
## PR Checklist
* [x] Closes#2554
* [x] I work here
* [x] Tests added/passed
* [n/a] Requires documentation to be updated
<hr>
* store text with extended attributes too
* Plumb attributes through all the renderers
* parse extended attrs, though we're not renderering them right
* Render these states correctly
* Add a very extensive test
* Cleanup for PR
* a block of PR feedback
* add 512 test cases
* Fix the build
* Fix @carlos-zamora's suggestions
* @miniksa's PR feedback
