|
|
Chromium Code Reviews
Descriptioncontent: Make record_whole_document available for all platforms.
Move record_whole_document in WebPreferences from OS_ANDROID to all platforms. This will be used on the blimp engine to record the whole page rather than just the interest rect.
BUG=591575
Committed: https://crrev.com/2da2b2322423c358333c95f017539d835f677884
Cr-Commit-Position: refs/heads/master@{#382937}
Patch Set 1 #Patch Set 2 : Missed renderer change. #Patch Set 3 : Add build flag for blimp web prefs #
Total comments: 2
Patch Set 4 : Use public_configs instead of all_dependent_configs. #Patch Set 5 : Back to patch #1. #
Messages
Total messages: 31 (13 generated)
khushalsagar@chromium.org changed reviewers: + sievers@chromium.org
khushalsagar@chromium.org changed reviewers: + palmer@chromium.org
lgtm
Description was changed from ========== content: Make record_whole_document available for all platforms This will be used for benchmarking in Blimp to record the whole page rather than just the interest rect. BUG= 591575 ========== to ========== content: Add build flag for Web Preferences for the Blimp Engine This will be used for benchmarking in Blimp to record the whole page rather than just the interest rect. BUG= 591575 ==========
Description was changed from ========== content: Add build flag for Web Preferences for the Blimp Engine This will be used for benchmarking in Blimp to record the whole page rather than just the interest rect. BUG= 591575 ========== to ========== content: Add build flag for Web Preferences for the Blimp Engine - Add a build flag to enable web preferences specific to the Blimp Engine. - Move record_whole_document to Android and the Blimp Engine. This will be used for benchmarking in Blimp to record the whole page rather than just the interest rect. BUG= 591575 ==========
palmer@, I've put the settings being a build flag for web preferences used on the blimp engine. PTAL. +wez
wez@chromium.org changed reviewers: + wez@chromium.org
https://codereview.chromium.org/1808363002/diff/40001/content/public/common/B... File content/public/common/BUILD.gn (right): https://codereview.chromium.org/1808363002/diff/40001/content/public/common/B... content/public/common/BUILD.gn:78: defines = [ "ENABLE_BLIMP_WEB_PREFS" ] Could we instead have this be enable_record_whole_document or something? Seems strange to bake any understanding of Blimp in at this level.
https://codereview.chromium.org/1808363002/diff/40001/content/public/common/B... File content/public/common/BUILD.gn (right): https://codereview.chromium.org/1808363002/diff/40001/content/public/common/B... content/public/common/BUILD.gn:78: defines = [ "ENABLE_BLIMP_WEB_PREFS" ] On 2016/03/22 00:18:14, Wez wrote: > Could we instead have this be enable_record_whole_document or something? Seems > strange to bake any understanding of Blimp in at this level. I assume we would need to add more web prefs for Blimp going forward, for instance to disable custom fonts. So might be better to put all such changes for web prefs behind a single flag.
khushalsagar@chromium.org changed reviewers: + agrieve@chromium.org
+agrieve, changed it to public_configs like you suggested.
So generally a blimp specific compile time config makes sense to me especially since we are already expecting that a lot of things that are currently behind the vague notion of the platform don't work here (because we have two platforms involved, one that is the host and the other one which is the content target). But I'm also wondering if having #ifdef all over the place defends the purpose of having generic configuration structures a bit (such as WebPreferences or LayerTreeSettings) that ideally the embedder would configure. And finally it would also be nice if the notion of the blimp engine build config was a higher level meta-config of specific features somehow, instead of being visible throughout the codebase in many places.
On 2016/03/22 18:55:49, sievers wrote: > So generally a blimp specific compile time config makes sense to me especially > since we are already expecting that a lot of things that are currently behind > the vague notion of the platform don't work here (because we have two platforms > involved, one that is the host and the other one which is the content target). > > But I'm also wondering if having #ifdef all over the place defends the purpose > of having generic configuration structures a bit (such as WebPreferences or > LayerTreeSettings) that ideally the embedder would configure. We have been trying to avoid defining a global define for blimp and rather use feature specific flags. However, that seemed hard to do in this case since the aim was to restrict a publicly exposed config for the embedder to the blimp engine. In this particular case, we can have the define as allow_record_whole_document, instead of a blanket flag for blimp, like Wez was also suggesting if that's preferable. +dpranke. > > And finally it would also be nice if the notion of the blimp engine build config > was a higher level meta-config of specific features somehow, instead of being > visible throughout the codebase in many places.
khushalsagar@chromium.org changed reviewers: + dpranke@chromium.org
khushalsagar@chromium.org changed reviewers: + tsepez@chromium.org
+tsepez since palmer@ is OOO.
Messages LGTM.
Description was changed from ========== content: Add build flag for Web Preferences for the Blimp Engine - Add a build flag to enable web preferences specific to the Blimp Engine. - Move record_whole_document to Android and the Blimp Engine. This will be used for benchmarking in Blimp to record the whole page rather than just the interest rect. BUG= 591575 ========== to ========== content: Make record_whole_document available for all platforms. Move record_whole_document in WebPreferences from OS_ANDROID to all platforms. This will be used on the blimp engine to record the whole page rather than just the interest rect. BUG= 591575 ==========
On 2016/03/23 20:03:25, Tom Sepez wrote: > Messages LGTM. Spoke with Tom, and he agreed that a build time flag is not necessary for this change, so back to the original patch#1.
The CQ bit was checked by khushalsagar@chromium.org
The patchset sent to the CQ was uploaded after l-g-t-m from sievers@chromium.org, tsepez@chromium.org Link to the patchset: https://codereview.chromium.org/1808363002/#ps80001 (title: "Back to patch #1.")
CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1808363002/80001 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1808363002/80001
For the record, this seems like a perfectly reasonable thing to be unconditionally available everywhere to me, so patchset #1 lgtm , too. (though we can make adding flags work also, of course).
Message was sent while issue was closed.
Description was changed from ========== content: Make record_whole_document available for all platforms. Move record_whole_document in WebPreferences from OS_ANDROID to all platforms. This will be used on the blimp engine to record the whole page rather than just the interest rect. BUG= 591575 ========== to ========== content: Make record_whole_document available for all platforms. Move record_whole_document in WebPreferences from OS_ANDROID to all platforms. This will be used on the blimp engine to record the whole page rather than just the interest rect. BUG= 591575 ==========
Message was sent while issue was closed.
Committed patchset #5 (id:80001)
Message was sent while issue was closed.
Description was changed from ========== content: Make record_whole_document available for all platforms. Move record_whole_document in WebPreferences from OS_ANDROID to all platforms. This will be used on the blimp engine to record the whole page rather than just the interest rect. BUG= 591575 ========== to ========== content: Make record_whole_document available for all platforms. Move record_whole_document in WebPreferences from OS_ANDROID to all platforms. This will be used on the blimp engine to record the whole page rather than just the interest rect. BUG= 591575 Committed: https://crrev.com/2da2b2322423c358333c95f017539d835f677884 Cr-Commit-Position: refs/heads/master@{#382937} ==========
Message was sent while issue was closed.
Patchset 5 (id:??) landed as https://crrev.com/2da2b2322423c358333c95f017539d835f677884 Cr-Commit-Position: refs/heads/master@{#382937}
Message was sent while issue was closed.
On 2016/03/23 21:43:11, Dirk Pranke wrote: > For the record, this seems like a perfectly reasonable thing to be > unconditionally available everywhere to me, so patchset #1 lgtm , too. > > (though we can make adding flags work also, of course). Agreed, let's try to stick with either runtime plumbed through layers so the embedder or so can set it. Or compile time where we can set it in one place. But both is weird esp. since the plumbing requires ifdef'ing all over the place too then. |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
