debugger statement not stopping flow
Reported by
glen.lit...@gmail.com,
Sep 12
|
||||
Issue descriptionUserAgent: Mozilla/5.0 (Windows NT 6.3; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.92 Safari/537.36 Steps to reproduce the problem: 1. Open attached test HTML file in Chrome with DevTools open 2. Click on <div> as indicated What is the expected behavior? The console should show: touch and then stop in the debugger. What went wrong? The console shows: touch touch and then stops in the debugger. Did this work before? N/A Chrome version: 69.0.3497.92 Channel: stable OS Version: 6.3 Flash Version:
,
Sep 12
The same problem happens in incognito mode (no extensions active).
,
Sep 13
This test works as expected in Firefox (with touch simulation enabled).
,
Sep 13
The code in the handler function appears to run twice... and only during the second execution is the 'debugger' honored.
,
Sep 13
On a tablet (real touch), when DevTools is open (not in emulation mode), the same problem happens and "touch" is shown in the console twice before the execution stops on 'debugger'. When the DevTools is closed, there is no problem and the handler code only runs once.
,
Sep 13
If the 'debugger' statement removed from the event handler code, the problem does not appear. If the developer then sets a breakpoint in the handler code, the problem does happen again and the handler code runs twice, only stopping at the breakpoint on the second execution. So, adding a debugger or a breakpoint in the handler code appears to cause the handler to run twice, only stopping on the second pass.
,
Sep 13
Tried with and without "Eager evaluation" enabled in the DevTools console and it has no effect on this issue.
,
Sep 13
So, a workaround, for now, is to NOT put a 'debugger' statement in the event handler routine. At least now I can avoid this bug and find other ways to work on the real problem that I was actually trying to solve!
,
Sep 13
Able to reproduce the issue on Windows 10, Debian Rodete and Mac 10.13.6 with chrome Stable/Beta #69.0.3497.92, Dev #70.0.3538.16 but not on Canary #71.0.3551.0 Issue is fixed in M71 Regression Range: ================== Good Build : 71.0.3544.0, #589076 bad Build : 71.0.3543.0, #588718 For this range per revision bisect script is not launching chrome Regression Range: https://chromium.googlesource.com/chromium/src/+log/71.0.3543.0..71.0.3544.0?pretty=fuller&n=10000 Suspecting 2 CL's which may cause the fix. https://chromium.googlesource.com/chromium/src/+/085697fefeaee5047b30030016edb95d3bd2bcb7 https://chromium.googlesource.com/chromium/src/+/0e743fb703d753762a8e15ef64c9751090ff3587 Assigning it to the luoe@/einbinder@ to look into this issue and merge the fixed CL in Stable,Beta,Dev
,
Sep 13
,
Sep 13
,
Sep 24
I bisected this one and it was fixed by https://chromium-review.googlesource.com/c/chromium/src/+/1172234 . I prepared another CL that might fix it on input event side and that is a little smaller and can be merged back: https://chromium-review.googlesource.com/c/chromium/src/+/1241216
,
Oct 2
|
||||
►
Sign in to add a comment |
||||
Comment 1 by glen.lit...@gmail.com
, Sep 12