Open
Bug 1330643
Opened 8 years ago
Updated 2 years ago
Develop a printing test plan oriented around sandboxing related changes
Categories
(Core :: Security: Process Sandboxing, defect, P3)
Core
Security: Process Sandboxing
Tracking
()
NEW
People
(Reporter: jimm, Unassigned)
References
(Blocks 3 open bugs, )
Details
(Whiteboard: sb+)
We're trying to improve the way we test features through test plans. (See attached template.) Sandboxing work has had a large impact on our printing code, so lets start by putting together a plan oriented around sandboxing changes.
Reporter | ||
Comment 1•8 years ago
|
||
Bob/Haik - please help Tracy out here by hooking up various print related bugs that you know of that make for good test cases. Also Bob, is there a bug on that automated print test work you wanted to do, I'd like that linked here as well.
Flags: needinfo?(bobowencode)
Reporter | ||
Updated•8 years ago
|
Flags: needinfo?(haftandilian)
Reporter | ||
Updated•8 years ago
|
status-firefox53:
affected → ---
Reporter | ||
Updated•8 years ago
|
Whiteboard: sb+
Comment 2•8 years ago
|
||
Bug 1303051 wasn't sandboxing related, but the kinds of things that were fixed would be good to test--page sizes, color vs black and white, print scaling.
With Bug 1324610, just opening the print dialog triggered the bug.
Comment 3•8 years ago
|
||
Didn't want to cause a load of bugmail spam by adding dependencies.
Tried to pick out ones that highlight different test scenarios, both in what is being printed and different print settings / config.
Which remind me, we should test with and without acceleration as well.
bug 1152353 - not directly to do with e10s, but it made it worse. This also highlights the need to manually test the Print dialog, because on Windows (at least) it is a native dialog.
bug 1324908 - adding this just as a note that layers.shared-buffer-provider.enabled=true currently causes crash on Mac when print.print_via_parent=false
bug 1326642 - not sure if this one is a new bug (haven't had time to look into it) and possibly only non-e10s.
bug 1328337 - mistake I made in one of my original patches. Adding it to remind us to have test pages with decimal font sizes.
bug 1328072 - reminder to test when no printers are installed and that print to PDF/PS still works.
bug 1316783 - something that was missing from the Moz2D recording code that caused box-shadow to print incorrectly in many cases.
bug 1310165 - make sure print to PDF renders actual text not just images.
bug 1189846 - reminder that we might want to test with Print Edit, which seems to be the most popular Printing addon.
bug 1306001 - not sandboxing/e10s related, but it would be useful to test printing from an iframe sandbox, with and without correct allow-modals keyword.
bug 1278537 - need to test with pref print.always_print_silent=true.
bug 1288194 - couple of issues with Moz2D recording Arc and PushLayer incorrectly (one existing and one I introduced :-( ).
bug 1287446 - issue with print progress dialog when document has a long title.
bug 1285943 - not sandboxing, but print preview used to be broken in e10s-multi.
bug 1238964 - initial problem with my print via parent changes where it caused a hang when not printing with Letter paper size.
bug 1271900 - reminder that people print labels using Firefox, with very small page sizes.
bug 1271348 - this one would probably require a dot matrix printer to test or something similar that had differing X and Y pixel counts.
bug 1184936 - this was an e10s issue I think, with a non-English system language on Ubuntu.
bug 1093774 - again not directly related, but to do with a native "Save the file as" dialog, so probably something we can only do in manual testing.
Flags: needinfo?(bobowencode)
Updated•7 years ago
|
Assignee: twalker → nobody
Updated•2 years ago
|
Severity: normal → S3
You need to log in
before you can comment on or make changes to this bug.
Description
•