| Summary: | Allow C API clients to get the WKProcessTerminationReason in processDidCrash functions, and use in WebKitTestRunner | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Simon Fraser (smfr) <simon.fraser> | ||||||
| Component: | WebKit2 | Assignee: | Simon Fraser (smfr) <simon.fraser> | ||||||
| Status: | RESOLVED FIXED | ||||||||
| Severity: | Normal | CC: | achristensen, cdumez, kkinnunen, sam, simon.fraser, webkit-bug-importer | ||||||
| Priority: | P2 | Keywords: | InRadar | ||||||
| Version: | WebKit Nightly Build | ||||||||
| Hardware: | Unspecified | ||||||||
| OS: | Unspecified | ||||||||
| Attachments: |
|
||||||||
|
Description
Simon Fraser (smfr)
2022-04-15 13:02:35 PDT
Created attachment 457718 [details]
Patch
Created attachment 457723 [details]
Patch
Comment on attachment 457723 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=457723&action=review > Source/WebKit/UIProcess/API/C/WKContext.h:137 > + WKContextChildProcessDidCrashWithDetailsCallback networkProcessDidCrashWithDetails; How would a single one of these that gets passed a "process-kind" enum be? Comment on attachment 457723 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=457723&action=review >> Source/WebKit/UIProcess/API/C/WKContext.h:137 >> + WKContextChildProcessDidCrashWithDetailsCallback networkProcessDidCrashWithDetails; > > How would a single one of these that gets passed a "process-kind" enum be? Or better yet, since this seems to be a pattern, a single one that gets passed a new WKObjectRef (I think that is the base name) "crash info" object that has some functions to grab the pid/reason? Both good suggestions, but is it really worth spending effort polishing the legacy C SPI? Alternatively we could replace this with Objective-C SPI (In reply to Simon Fraser (smfr) from comment #5) > Both good suggestions, but is it really worth spending effort polishing the > legacy C SPI? I was thinking of it not as polish but as just more ergonomical for going forward, since we don't really have a path off of it. (In reply to Simon Fraser (smfr) from comment #6) > Alternatively we could replace this with Objective-C SPI Isn't this used by the non-cocoa ports as well? Committed r292970 (249733@main): <https://commits.webkit.org/249733@main> All reviewed patches have been landed. Closing bug and clearing flags on attachment 457723 [details]. |