Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(84)

Issue 25015005: Re-export Skia defines from content_browser.gypi. (Closed)

Created:
7 years, 2 months ago by bungeman-chromium
Modified:
7 years, 2 months ago
CC:
chromium-reviews, joi+watch-content_chromium.org, darin-cc_chromium.org
Visibility:
Public.

Description

Re-export Skia defines from content_browser.gypi. Doing a code search for "file:src/content/public/browser.*\.h skia" shows six public headers which are including Skia headers. These Skia headers require that the defines exported from skia.gyp be available. Currently the build is getting away with this by accident, but some proposed changes to some Skia headers will cause this to fail. Correct this by re-exporting the skia.gyp settings.

Patch Set 1 #

Unified diffs Side-by-side diffs Delta from patch set Stats (+3 lines, -0 lines) Patch
M content/content_browser.gypi View 1 chunk +3 lines, -0 lines 0 comments Download

Messages

Total messages: 7 (0 generated)
bungeman-chromium
7 years, 2 months ago (2013-09-27 22:21:30 UTC) #1
bungeman-chromium
7 years, 2 months ago (2013-10-09 19:45:30 UTC) #2
bungeman-chromium
Any owner willing to take a look? I'd like to avoid having skia.gyp use all_dependent_settings ...
7 years, 2 months ago (2013-10-11 14:40:18 UTC) #3
jam
On 2013/10/11 14:40:18, bungeman2 wrote: > Any owner willing to take a look? I'd like ...
7 years, 2 months ago (2013-10-11 14:58:52 UTC) #4
bungeman-chromium
I am rather frustrated by the lack of interest in having clean build files. I ...
7 years, 2 months ago (2013-10-11 15:07:02 UTC) #5
jam
On 2013/10/11 15:07:02, bungeman2 wrote: > I am rather frustrated by the lack of interest ...
7 years, 2 months ago (2013-10-14 20:38:32 UTC) #6
brettw
7 years, 2 months ago (2013-10-22 17:16:42 UTC) #7
Message was sent while issue was closed.
I noticed this go by so I thought I would comment since I've been doing a lot of
thinking about this as part of the new build.

There are two schools of thought about whether this kind of thing should be
"all" or "direct." The direct school is basically like private inheritance in
C++ and saying that you shouldn't have any more flags than necessary. The all
school says that this is too difficult to maintain. We do both randomly in
Chrome.

My opinion is that the "direct" is better, BUT that it's not practical until we
have better dependency management. Currently you can get really weird and
difficult-to-track-down compilation errors, and there's not a lot of
infrastructure to help. This is what John is mainly concerned with, I think.

Ideally our dependency management system would have some notion of "public"
headers and we could check that if you include a header that you have the proper
permissions and dependencies to do it. Google3 has some things along these
lines. Until we have something like this, I don't have a strong opinion about
what the best way to do this.

Powered by Google App Engine
This is Rietveld 408576698