We create and destroy ScaledFonts for every piece of text we write.
That causes a huge amount of duplicated data within the recording.
This splits out the recording of the font data itself from the ScaledFont.
The key generated to determine uniqueness could probably be fairly easily faked, but for our purposes that doesn't matter.
These are mainly changes to make sure we have recorded relevant dependencies to each draw operation.
Where we can't record them on the fly like this, it makes sure the object has originated from our DrawTarget.
We create and destroy ScaledFonts for every piece of text we write.
That causes a huge amount of duplicated data within the recording.
This splits out the recording of the font data itself from the ScaledFont.
The key generated to determine uniqueness could probably be fairly easily faked, but for our purposes that doesn't matter.
These are mainly changes to make sure we have recorded relevant dependencies to each draw operation.
Where we can't record them on the fly like this, it makes sure the object has originated from our DrawTarget.
The bulk of this commit was generated by running:
run-clang-tidy.py \
-checks='-*,llvm-namespace-comment' \
-header-filter=^/.../mozilla-central/.* \
-fix