RESOLVED FIXED Bug 38334
Allow other ports to compile ATSUI and CoreText functions in SimpleFontData for Mac. 
https://bugs.webkit.org/show_bug.cgi?id=38334
Summary Allow other ports to compile ATSUI and CoreText functions in SimpleFontData f...
Kevin Ollivier
Reported 2010-04-29 10:53:54 PDT
Allow other ports to compile ATSUI and CoreText functions in SimpleFontData for Mac.
Attachments
Patch (16.67 KB, patch)
2010-04-29 11:06 PDT, Kevin Ollivier 		no flags
DetailsFormatted DiffDiff
Patch (88.48 KB, patch)
2010-04-29 16:00 PDT, Kevin Ollivier 		no flags
DetailsFormatted DiffDiff
Patch (90.33 KB, patch)
2010-04-29 16:07 PDT, Kevin Ollivier 		no flags
DetailsFormatted DiffDiff
Patch (90.59 KB, patch)
2010-04-29 16:11 PDT, Kevin Ollivier 		mitz: review+
DetailsFormatted DiffDiff
Show Obsolete (3) View All Add attachment proposed patch, testcase, etc.
Kevin Ollivier
Comment 1 2010-04-29 11:06:24 PDT
Created attachment 54719 [details] Patch
Kevin Ollivier
Comment 2 2010-04-29 11:25:02 PDT
This is needed so that wx can compile and use ComplexTextController, BTW.
mitz
Comment 3 2010-04-29 11:28:47 PDT
Comment on attachment 54719 [details] Patch This should be done by svn cp’ing SimpleFontDataMac.mm to the two new files and removing everything but the needed methods, includes and copyright headers. The #if USE()s should go outside the namespace and the second paragraph of #includes.
Kevin Ollivier
Comment 4 2010-04-29 16:00:54 PDT
Created attachment 54747 [details] Patch
WebKit Review Bot
Comment 5 2010-04-29 16:03:42 PDT
Attachment 54747 [details] did not pass style-queue: Failed to run "['WebKitTools/Scripts/check-webkit-style', '--no-squash']" exit_code: 1 WebCore/platform/graphics/mac/SimpleFontDataCoreText.cpp:48: Use 0 instead of NULL. [readability/null] [5] WebCore/platform/graphics/mac/SimpleFontDataCoreText.cpp:74: Use 0 instead of NULL. [readability/null] [5] WebCore/platform/graphics/mac/SimpleFontDataCoreText.cpp:81: Use 0 instead of NULL. [readability/null] [5] Total errors found: 3 in 4 files If any of these errors are false positives, please file a bug against check-webkit-style.
mitz
Comment 6 2010-04-29 16:05:33 PDT
Comment on attachment 54747 [details] Patch Is there a reason why you didn’t address my second feedback item, regarding the placement of the #if directives?
Kevin Ollivier
Comment 7 2010-04-29 16:07:45 PDT
Created attachment 54749 [details] Patch
Kevin Ollivier
Comment 8 2010-04-29 16:08:51 PDT
(In reply to comment #6) > (From update of attachment 54747 [details]) > Is there a reason why you didn’t address my second feedback item, regarding the > placement of the #if directives? Oops, sorry, no, I just got caught up in redoing the patch and forgot about it. I'll fix and re-upload.
Kevin Ollivier
Comment 9 2010-04-29 16:11:02 PDT
Created attachment 54751 [details] Patch
Kent Tamura
Comment 10 2010-04-29 23:48:07 PDT
The patch was committed as r58557, and rolled out by r58560 because of compile errors on Tiger, Leopard, and Chromium/Mac.
Kent Tamura
Comment 11 2010-04-29 23:49:49 PDT
(In reply to comment #10) > The patch was committed as r58557, and rolled out by r58560 because of compile > errors on Tiger, Leopard, and Chromium/Mac. Correction: Tiger was ok. Leopard and Chromium/Mac failed.
Jeremy Moskovich
Comment 12 2010-04-30 04:44:39 PDT
I haven't looked into the compile failures but FYI Chromium compiles with both the ATSUI & Core text functionality enabled and switches the APIs used at runtime. This, as opposed to other ports that make the decision at compile time.
WebKit Review Bot
Comment 13 2010-04-30 09:51:13 PDT
http://trac.webkit.org/changeset/58581 might have broken Chromium Mac Release
Kevin Ollivier
Comment 14 2010-04-30 11:47:11 PDT
Landed in r58581, with Chromium build fix in r58582.
Note You need to log in before you can comment on or make changes to this bug.