|
|
Created:
7 years, 8 months ago by Raman Kakilate Modified:
7 years, 8 months ago CC:
chromium-reviews, benquan, dhollowa+watch_chromium.org, ahutter, browser-components-watch_chromium.org, dbeam+watch-autofill_chromium.org, Dane Wallinga, dyu1, estade+watch_chromium.org, Albert Bodenhamer, Ilya Sherman, Alexei Svitkine (slow) Base URL:
svn://svn.chromium.org/chrome/trunk/src Visibility:
Public. |
DescriptionAutofill:Autocheckout: In addition to "Enable Experimental Form filling" flag, support second method of enabling Autocheckout: using finch experiment. Autocheckout will be enabled if either of the conditions is true. Will remove "Enable experimental form filling" flag, once finch experiment is setup completely.
BUG=230026
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=194767
Patch Set 1 #
Total comments: 3
Patch Set 2 : Fix indentation #
Messages
Total messages: 13 (0 generated)
Albert/Ilya - PTAL. Thanks.
Please attach a more detailed description. It looks like the intent is to be able to turn this on either with the existing switch or finch experiment. That should be in the description. Do we still need the switch in chrome://flags? It seems just just having access to it from the command line should be enough.
On 2013/04/12 17:54:14, Albert Bodenhamer wrote: > Please attach a more detailed description. It looks like the intent is to be > able to turn this on either with the existing switch or finch experiment. That > should be in the description. > > Do we still need the switch in chrome://flags? It seems just just having access > to it from the command line should be enough. Wanted to test the finch experiment before removing it from chrome://flags. Will be doing it in future CL.
LGTM On 2013/04/12 18:05:05, Raman Kakilate wrote: > On 2013/04/12 17:54:14, Albert Bodenhamer wrote: > > Please attach a more detailed description. It looks like the intent is to be > > able to turn this on either with the existing switch or finch experiment. > That > > should be in the description. > > > > Do we still need the switch in chrome://flags? It seems just just having > access > > to it from the command line should be enough. > > Wanted to test the finch experiment before removing it from chrome://flags. Will > be doing it in future CL. In general, the recommended approach is to have a tri-state flag, with states "Default", "Enable", and "Disable". The field trial should only take effect when the flag is in the "Default" state. But, if we don't want external developers to be able to mess around with this feature for now, then removing the flag is ok. https://codereview.chromium.org/13811045/diff/1/components/autofill/renderer/... File components/autofill/renderer/form_autofill_util.cc (right): https://codereview.chromium.org/13811045/diff/1/components/autofill/renderer/... components/autofill/renderer/form_autofill_util.cc:85: return base::FieldTrialList::FindFullName("Autocheckout") == "Yes" || I'm not sure that all field trials are automatically sent to the renderer process. You might want to double-check that this works correctly.
On 2013/04/13 01:12:09, Ilya Sherman wrote: > LGTM > > On 2013/04/12 18:05:05, Raman Kakilate wrote: > > On 2013/04/12 17:54:14, Albert Bodenhamer wrote: > > > Please attach a more detailed description. It looks like the intent is to > be > > > able to turn this on either with the existing switch or finch experiment. > > That > > > should be in the description. > > > > > > Do we still need the switch in chrome://flags? It seems just just having > > access > > > to it from the command line should be enough. > > > > Wanted to test the finch experiment before removing it from chrome://flags. > Will > > be doing it in future CL. > > In general, the recommended approach is to have a tri-state flag, with states > "Default", "Enable", and "Disable". The field trial should only take effect > when the flag is in the "Default" state. > > But, if we don't want external developers to be able to mess around with this > feature for now, then removing the flag is ok. > > https://codereview.chromium.org/13811045/diff/1/components/autofill/renderer/... > File components/autofill/renderer/form_autofill_util.cc (right): > > https://codereview.chromium.org/13811045/diff/1/components/autofill/renderer/... > components/autofill/renderer/form_autofill_util.cc:85: return > base::FieldTrialList::FindFullName("Autocheckout") == "Yes" || > I'm not sure that all field trials are automatically sent to the renderer > process. You might want to double-check that this works correctly. I am not aware of testing this locally. Will submit it for now along with server config and will fix if needed.
CQ is trying da patch. Follow status at https://chromium-status.appspot.com/cq/ramankk@chromium.org/13811045/1
On Mon, Apr 15, 2013 at 12:37 AM, <ramankk@chromium.org> wrote: > On 2013/04/13 01:12:09, Ilya Sherman wrote: > >> LGTM >> > > On 2013/04/12 18:05:05, Raman Kakilate wrote: >> > On 2013/04/12 17:54:14, Albert Bodenhamer wrote: >> > > Please attach a more detailed description. It looks like the intent >> is to >> be >> > > able to turn this on either with the existing switch or finch >> experiment. >> > That >> > > should be in the description. >> > > >> > > Do we still need the switch in chrome://flags? It seems just just >> having >> > access >> > > to it from the command line should be enough. >> > >> > Wanted to test the finch experiment before removing it from >> chrome://flags. >> Will >> > be doing it in future CL. >> > > In general, the recommended approach is to have a tri-state flag, with >> states >> "Default", "Enable", and "Disable". The field trial should only take >> effect >> when the flag is in the "Default" state. >> > > But, if we don't want external developers to be able to mess around with >> this >> feature for now, then removing the flag is ok. >> > > > https://codereview.chromium.**org/13811045/diff/1/** > components/autofill/renderer/**form_autofill_util.cc<https://codereview.chromium.org/13811045/diff/1/components/autofill/renderer/form_autofill_util.cc> > >> File components/autofill/renderer/**form_autofill_util.cc (right): >> > > > https://codereview.chromium.**org/13811045/diff/1/** > components/autofill/renderer/**form_autofill_util.cc#**newcode85<https://codereview.chromium.org/13811045/diff/1/components/autofill/renderer/form_autofill_util.cc#newcode85> > >> components/autofill/renderer/**form_autofill_util.cc:85: return >> base::FieldTrialList::**FindFullName("Autocheckout") == "Yes" || >> I'm not sure that all field trials are automatically sent to the renderer >> process. You might want to double-check that this works correctly. >> > > I am not aware of testing this locally. Will submit it for now along with > server > config and will fix if needed. > You can test this locally using by passing the command-line --force-fieldtrials=Autocheckout/Yes/ to Chrome. > > https://codereview.chromium.**org/13811045/<https://codereview.chromium.org/1... >
On 2013/04/15 04:47:44, Alexei Svitkine wrote: > On Mon, Apr 15, 2013 at 12:37 AM, <mailto:ramankk@chromium.org> wrote: > > > On 2013/04/13 01:12:09, Ilya Sherman wrote: > > > >> LGTM > >> > > > > On 2013/04/12 18:05:05, Raman Kakilate wrote: > >> > On 2013/04/12 17:54:14, Albert Bodenhamer wrote: > >> > > Please attach a more detailed description. It looks like the intent > >> is to > >> be > >> > > able to turn this on either with the existing switch or finch > >> experiment. > >> > That > >> > > should be in the description. > >> > > > >> > > Do we still need the switch in chrome://flags? It seems just just > >> having > >> > access > >> > > to it from the command line should be enough. > >> > > >> > Wanted to test the finch experiment before removing it from > >> chrome://flags. > >> Will > >> > be doing it in future CL. > >> > > > > In general, the recommended approach is to have a tri-state flag, with > >> states > >> "Default", "Enable", and "Disable". The field trial should only take > >> effect > >> when the flag is in the "Default" state. > >> > > > > But, if we don't want external developers to be able to mess around with > >> this > >> feature for now, then removing the flag is ok. > >> > > > > > > https://codereview.chromium.**org/13811045/diff/1/** > > > components/autofill/renderer/**form_autofill_util.cc<https://codereview.chromium.org/13811045/diff/1/components/autofill/renderer/form_autofill_util.cc> > > > >> File components/autofill/renderer/**form_autofill_util.cc (right): > >> > > > > > > https://codereview.chromium.**org/13811045/diff/1/** > > > components/autofill/renderer/**form_autofill_util.cc#**newcode85<https://codereview.chromium.org/13811045/diff/1/components/autofill/renderer/form_autofill_util.cc#newcode85> > > > >> components/autofill/renderer/**form_autofill_util.cc:85: return > >> base::FieldTrialList::**FindFullName("Autocheckout") == "Yes" || > >> I'm not sure that all field trials are automatically sent to the renderer > >> process. You might want to double-check that this works correctly. > >> > > > > I am not aware of testing this locally. Will submit it for now along with > > server > > config and will fix if needed. > > > > You can test this locally using by passing the command-line > --force-fieldtrials=Autocheckout/Yes/ to Chrome. > > Thanks for the tip Alexei. Having trouble verifying over NX. removed the commit bit for now. Will test it and submit. > > > > > https://codereview.chromium.**org/13811045/%3Chttps://codereview.chromium.org...> > >
https://codereview.chromium.org/13811045/diff/1/components/autofill/renderer/... File components/autofill/renderer/form_autofill_util.cc (right): https://codereview.chromium.org/13811045/diff/1/components/autofill/renderer/... components/autofill/renderer/form_autofill_util.cc:85: return base::FieldTrialList::FindFullName("Autocheckout") == "Yes" || On 2013/04/13 01:12:09, Ilya Sherman wrote: > I'm not sure that all field trials are automatically sent to the renderer > process. You might want to double-check that this works correctly. I believe we only tell the renderer of studies that have had their group value queried by the browser process. So to make sure its visible in the renderer, ensure that you do a base::FieldTrialList::FindFullName("Autocheckout") in the browser prior to that.
https://codereview.chromium.org/13811045/diff/1/components/autofill/renderer/... File components/autofill/renderer/form_autofill_util.cc (right): https://codereview.chromium.org/13811045/diff/1/components/autofill/renderer/... components/autofill/renderer/form_autofill_util.cc:85: return base::FieldTrialList::FindFullName("Autocheckout") == "Yes" || On 2013/04/15 15:24:13, Alexei Svitkine wrote: > On 2013/04/13 01:12:09, Ilya Sherman wrote: > > I'm not sure that all field trials are automatically sent to the renderer > > process. You might want to double-check that this works correctly. > > I believe we only tell the renderer of studies that have had their group value > queried by the browser process. So to make sure its visible in the renderer, > ensure that you do a base::FieldTrialList::FindFullName("Autocheckout") in the > browser prior to that. Done. Verified that Autocheckout field trial is in "Yes" state in renderer process.
LGTM
CQ is trying da patch. Follow status at https://chromium-status.appspot.com/cq/ramankk@chromium.org/13811045/19001
Message was sent while issue was closed.
Change committed as 194767 |