[Webkit-unassigned] [Bug 48459] Glyphs in vertical text tests are rotated 90 degrees clockwise on Windows

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Mon Mar 5 13:03:05 PST 2012


Koji Ishii <kojiishi at gmail.com> changed:

           What    |Removed                     |Added
 Attachment #130022|0                           |1
        is obsolete|                            |

--- Comment #47 from Koji Ishii <kojiishi at gmail.com>  2012-03-05 13:03:04 PST ---
Created an attachment (id=130188)
 --> (https://bugs.webkit.org/attachment.cgi?id=130188&action=review)
130022: New approach without using @-font API (in progress)

Revised the patch I posted at comment #44.

Changes from the previous patch:
1. Fixed styles.
2. Vertical origin using VORG/vmtx, which is required for CFF OpenType fonts.
3. GSUB coverage format 2 support for some Adobe fonts.
4. Support !useGDI()

There are still several work items to do:
1. Several FIXME comments are in the new code.
2. Currently OpenTypeVerticalData loads multiple times for the same font. Having it in FontPlatformData might help some. Also some of OpenTypeVerticalData are size-independent, so I might try to share among multiple SimpleFontData/FontPlatformData.
3. Since it's reading OpenType tables directly, there's a risk of buffer-overrun if the font is broken. I added checks in the last minutes, but I think I need more and the code needs careful reviews.
4. By looking at the results, there seems to be some rounding errors in small font sizes which doesn't look nice. I'd like to look into that further to see if there were any good way to avoid it.
5. Non-OpenType font isn't supported very well. I'd like to have a fallback, 2nd-level support for such fonts, but not sure.
6. I'm still new to webkit (this is 2nd patch,) I'll review coding style guidelines further. Note that I did read arrays are preferred than iterator for readability reasons, but I'm using iterator since pointers are needed to check buffer overruns.
7. OpenTypeVerticalData loads information for all glyphs in its constructor. I might want to move some of them to on-demand, my thinking is to use GlyphMetricsMap class for per-page loading and caching.
8. Complex path doesn't work because it reads advance from Uniscribe. I'm not sure how I can make it work at this point.

Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

More information about the webkit-unassigned mailing list