Glyph::get_points(phantom_only=true) is the path hb-ot-font uses for advances and vertical origins of a gvar font without HVAR/VVAR. It recursed into every USE_MY_METRICS component although each one overwrites the phantoms copied from the previous one, so only the last surviving one can influence the result; a composite that sets the flag on many components could thus cost nearly a whole HB_BUDGET_GLYPH per advance lookup, and a small font could reach the same per-character cost that previously needed a megabyte-sized one. Add a cheap pre-pass over the component records (bounded by the same count the offset-collecting loop already charges) that picks the last flagged component the decycler would accept, and descend only into it. Components that are not descended into now continue before the post-child bookkeeping; this also removes a latent wrong-advance bug where the unconditional all_points.resize (length - PHANTOM_COUNT) for a skipped component ate entries pushed by earlier siblings, making the parent's phantom copy read out of range and return advance 0. Non-phantom-only traversal (drawing, extents, subsetting) is unchanged: copy_metrics reduces to the old condition and the new continue is never taken. The one behavioural difference in phantom-only mode is that a traversal failure (nesting, edge count, budget) inside a flagged subtree that is no longer visited no longer fails the whole lookup. Reported by Anthropic's OSS Scanner as ordinary bug ANT-2026-DQGMHCLX. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Adding tests
You can test shaping of a unicode sequence against a font like this:
$ ./hb-unicode-encode 41 42 43 627 | ../../util/hb-shape font.ttf
assuming an in-tree build. The 41 42 43 627 here is a sequence of
Unicode codepoints: U+0041,0042,0043,0627. When you are happy with
the shape results, you can use the record-test.sh script to add
this to the test suite. record-test.sh requires pyftsubset to
be installed. You can get pyftsubset by installing
FontTools from https://github.com/behdad/fonttools.
To use record-test.sh, just put it right before the hb-shape invocation:
$ ./hb-unicode-encode 41 42 43 627 | ./record-test.sh ../../util/hb-shape font.ttf
hb-subset and hb-view are resolved as sibling executables next to the
provided hb-shape binary.
what this does is:
- Subset the font for the sequence of Unicode characters requested,
- Compare the
hb-shapeoutput of the original font versus the subset font for the input sequence, - If the outputs differ, perhaps it is because the font does not have
glyph names; it then compares the output of
hb-viewfor both fonts. - If the outputs differ, recording fails. Otherwise, it will move the
subset font file into
data/in-house/fontsand name it after its hash, and print out the test case input, which you can then redirect to an existing or new test file indata/in-house/testsusing-o, e.g.:
$ ./hb-unicode-encode 41 42 43 627 | ./record-test.sh -o data/in-house/tests/test-name.tests ../../util/hb-shape font.ttf
If you created a new test file, add it to data/in-house/meson.build
so it is run. Check that make check does indeed run it, and that the
test passes. When everything looks good, git add the new font as well
as the new test file if you created any. You can see what new files are
there by running git status data/in-house. And commit!
Note! Please only add tests using Open Source fonts, preferably under OFL or similar license.