This repository was archived by the owner on Feb 25, 2025. It is now read-only.
[wip] Revise the raster cache unit test #17790
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Previously, it's sending the wrong ctm: the one before the translation. It's important because canvas.translate is applying the offset to the matrix using preTranslate, so the fractional scale factor could turn an integral offset into a final (postTranslate) fractional offset. After the revision, we no longer fail the assertion that the image rect should exactly matches the bounds. I didn't find any case in our code where we used the wrong ctm for raster cache key. Therefore, we have to find another unit test that could surface the failed check in the 1st gen Nexus7.
flar
suggested changes
Apr 17, 2020
|
|
||
| cache.Get(*picture, ctm).draw(canvas); | ||
| #ifndef SUPPORT_FRACTIONAL_TRANSLATION | ||
| canvas.setMatrix(RasterCache::GetIntegralTransCTM(canvas.getTotalMatrix())); |
Contributor
There was a problem hiding this comment.
This line defeats the purpose of this test as stated in the comment above it.
Contributor
Author
There was a problem hiding this comment.
If we don't snap, we'd fail the integral translation assert instead of of bounds assert above. This test tries to reproduce the bounds assert.
Contributor
There was a problem hiding this comment.
Given the current way that we use the raster cache, this whole test doesn't make sense then.
It will be useful when we introduce "cache at 0,0 - render at any fractional translation" later, though. In that case we will see output size differences of up to +1 from the cached image.
liyuqian
added a commit
to liyuqian/engine
that referenced
this pull request
Apr 17, 2020
This avoids the possible matrix mismatch between RasterCache::Get and RasterCacheResult::draw. See flutter#17790 for an example that tries to fix an earlier mismatch.
liyuqian
added a commit
to liyuqian/engine
that referenced
this pull request
Apr 21, 2020
This avoids the possible matrix mismatch between RasterCache::Get and RasterCacheResult::draw. See flutter#17790 for an example that tries to fix an earlier mismatch.
liyuqian
added a commit
to liyuqian/engine
that referenced
this pull request
Apr 22, 2020
This avoids the possible matrix mismatch between RasterCache::Get and RasterCacheResult::draw. See flutter#17790 for an example that tries to fix an earlier mismatch.
liyuqian
added a commit
that referenced
this pull request
Apr 23, 2020
This avoids the possible matrix mismatch between RasterCache::Get and RasterCacheResult::draw. See #17790 for an example that tries to fix an earlier mismatch.
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Previously, it's sending the wrong ctm: the one before the translation.
It's important because canvas.translate is applying the offset to the
matrix using preTranslate, so the fractional scale factor could turn an
integral offset into a final (postTranslate) fractional offset.
After the revision, we no longer fail the assertion that the image rect
should exactly matches the bounds.
I didn't find any case in our code where we used the wrong ctm for
raster cache key. Therefore, we have to find another unit test that
could surface the failed check in the 1st gen Nexus7.