UseCounter pre-loading features might have a higher count than pagevisits count |
||||||
Issue descriptionUseCounter applies multiple filters to the pages (e.g., NTP, none http:// https://, etc) it measures. But those filters are applied upon load commit. There are pre-loading features (e.g. ContentSecurityPolicy) that are logged before load commit. The filters are not applied correctly to these features. This causes a higher count on the pre-loading features than pagevisits, which breaks UseCounter.
,
Feb 22 2018
Other features that have higher count than PageVisists: V8SloppyMode, V8StrictMode, SecureContextCheck*, CleanScriptElementWithNonce, HasIDClassTagAttribute
,
Feb 22 2018
A quick fix to this problem would be not count feature usage until load commit. I am not so sure how doable it is for the features listed above, especially v8 stuff. I am also trying to 1. move the UseCounter to the browser side; 2. buffer pre-loading features and send them to the browser side upon load commit, which will fix this problem eventually.
,
Mar 9 2018
Pri-0 bugs are critical regressions or serious emergencies, and this bug has not been updated in three days. Could you please provide an update, or adjust the priority to a more appropriate level if applicable? If a fix is in active development, please set the status to Started. Thanks for your time! To disable nags, add the Disable-Nags label. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Mar 9 2018
Not on the Chrome team anymore. removing from cc
,
Mar 22 2018
,
Mar 22 2018
,
Mar 22 2018
There should no longer be feature usage higher than page visits as the features usecounter has beeen moved to the browser side. However there is still incorrect behaviour in terms of dropping counts for pre-loading features. But that bug is being tracked in another place. Closing this bug now. |
||||||
►
Sign in to add a comment |
||||||
Comment 1 by loonyb...@chromium.org
, Feb 22 2018