Closed Bug 643460 Opened 13 years ago Closed 13 years ago

Update OTS version to r62 or later

Categories

(Core :: Layout: Text and Fonts, defect)

x86
Windows 7
defect
Not set
normal

Tracking

()

RESOLVED FIXED

People

(Reporter: emk, Assigned: emk)

References

Details

Attachments

(3 files, 3 obsolete files)

Latest OTS has added a support for OTL tables (GDEF, GPOS and GSUB).
We can remove the most part of our modification by updating the library.
I'm afraid that OTS doesn't support vhea/vmtx tables yet. So we need to update a  bug 637481 patch to latest OTS (and submit it to upstream).
Adding dependency to track patches which are not yet merged to upstream.
bug 527276 part 1a is also required.
http://hg.mozilla.org/mozilla-central/rev/88ff5f0a9681
Depends on: 602558, 628252, 643015, 637481
Summary: Update OTS version to r57 or later → Update OTS version to r60 or later
vhea/vmtx support was added in upstream.
No longer depends on: 637481
Summary: Update OTS version to r60 or later → Update OTS version to r61 or later
No longer depends on: 643015
Summary: Update OTS version to r61 or later → Update OTS version to r62 or later
Assignee: nobody → VYV03354
Status: NEW → ASSIGNED
Attachment #524378 - Flags: review?(jfkthame)
This pref is no longer required because OTS validates OTL tables now.
Attachment #524380 - Flags: review?(jfkthame)
Attachment #524379 - Attachment description: Part 1: apply local patches to fix build errors on Mozilla tree → Part 2: apply local patches to fix build errors on Mozilla tree
Attachment #524380 - Attachment is patch: true
Attachment #524380 - Attachment mime type: application/octet-stream → text/plain
Forgotten to add a countermeasure against comma-at-end-of-enumerator errors.
Attachment #524379 - Attachment is obsolete: true
Attachment #524379 - Flags: review?(jfkthame)
Attachment #524624 - Flags: review?(jfkthame)
I think one more local fix is needed in part 2, see bug 643137 (which gets reverted by the import in part 1 here).
Also, upstream OTS is now up to r64, with some Solaris fixes from Ginn Chen. So we should take that version. But let's wait until after the merge to Aurora, as there's no compelling need to fix this before the Fx5 train leaves.
Depends on: 643137
updated to tip
Attachment #524380 - Attachment is obsolete: true
Attachment #524380 - Flags: review?(jfkthame)
(In reply to comment #8)
> Also, upstream OTS is now up to r64, with some Solaris fixes from Ginn Chen. So
> we should take that version. But let's wait until after the merge to Aurora, as
> there's no compelling need to fix this before the Fx5 train leaves.
I chose r62 because Chromium did'nt roll into newer OTS yet.
http://src.chromium.org/viewvc/chrome/trunk/src/DEPS?view=markup
I'll update to a latest rolled revision at the time of aurora merge.
The latest rolled revision is still r62, so I didn't update Part 1.
Attachment #524624 - Attachment is obsolete: true
Attachment #524624 - Flags: review?(jfkthame)
Attachment #527390 - Flags: review?(jfkthame)
Attachment #525166 - Flags: review?(jfkthame)
Attachment #524378 - Flags: review?(jfkthame) → review+
Attachment #527390 - Flags: review?(jfkthame) → review+
Comment on attachment 525166 [details] [diff] [review]
Part 3: remove gfx.downloadable_fonts.sanitize.preserve_otl_tables

Looks great, thanks. (I didn't attempt to actually review all the new OTS code, but the integration into Gecko looks correct.)

This should be tested on tryserver to make sure it doesn't cause any reftest surprises, etc; then let's get it landed in m-c.
Attachment #525166 - Flags: review?(jfkthame) → review+
I do not have a commit access yet. Could you vouch me?
I've files bug 652685.
(In reply to comment #13)
> I do not have a commit access yet. Could you vouch me?
> I've files bug 652685.

I could, but I see Gavin did already. :)

Meanwhile, I've pushed these patches to try, and plan to land them in a while once the tests confirm all is good.
Blocks: 653100
You need to log in before you can comment on or make changes to this bug.