691187 - Prune zero-length line segments before stroking a path

Status: RESOLVED FIXED

(Core :: Graphics: Canvas2D, defect)

Description

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

http://www.w3c-test.org/html/tests/submission/PhilipTaylor/canvas/index.2d.path.stroke.prune.html

Comment 1

[Grant Skinner]

It would be nice if this was fixed. Here's the relevant blurb from the standard: "Zero-length line segments must be pruned before stroking a path. Empty subpaths must be ignored."

And a working example (since the URL above is 404):
http://jsfiddle.net/lannymcnie/wM975/2/

Note how a red dot (0 length path) is drawn in FF. This is not the case in other browsers.

Comment 3

[Andrew Osmond [:aosmond] (he/him)]

Attachment: Bug 691187 - Prune zero-length segments for canvas strokes. r=lsalzman

Comment 4

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/50707b02bc59
Prune zero-length segments for canvas strokes. r=lsalzman

Comment 5

[Cristian Tuns]

Backed out for causing mochitest failures in test_bug1567544.html

• Backout link
• Push with failures
• Failure Log
• Failure line: TEST-UNEXPECTED-FAIL | dom/canvas/test/test_bug1567544.html | pixel 82,48 is 0,0,0,0; expected 255,0,0,255 +/- 0

Wr failure log: https://treeherder.mozilla.org/logviewer?job_id=417918202&repo=autoland
Reftest failure log: https://treeherder.mozilla.org/logviewer?job_id=417918238&repo=autoland
Bc failure log: https://treeherder.mozilla.org/logviewer?job_id=417917269&repo=autoland
Another reftest failure: https://treeherder.mozilla.org/logviewer?job_id=417913639&repo=autoland
WD failures: https://treeherder.mozilla.org/logviewer?job_id=417915535&repo=autoland

Comment 6

[Andrew Osmond [:aosmond] (he/him)]

I seem to have caused some test failures since my last push in my latest update, as these were passing before.

Comment 7

[Andrew Osmond [:aosmond] (he/him)]

Silly typo when I made a change. Tests should pass now.

Comment 8

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/e6e098584f0a
Prune zero-length segments for canvas strokes. r=lsalzman

Comment 9

[Narcis Beleuzu [:NarcisB]]

https://hg.mozilla.org/mozilla-central/rev/e6e098584f0a

Comment 10

[Lee Salzman [:lsalzman]]

Attachment: Bug 691187 - Backout changeset e6e098584f0ab (ESR 115 only). r?aosmond

The initial fix in bug 691187 has caused numerous regressions that have required
later fixups to address. Since only the initial version that caused all these
regressions made it into 115, it is safer to just back out the initial change rather
than try to uplift all the regression fixes. This at least reverts rendering to
the stable, tenable state it was in prior to 115 for ESR.

Comment 11

[Lee Salzman [:lsalzman]]

Comment on attachment 9347791 [details]
Bug 691187 - Backout changeset e6e098584f0ab (ESR 115 only). r?aosmond

ESR Uplift Approval Request

• If this is not a sec:{high,crit} bug, please state case for ESR consideration: Several Canvas2D rendering regressions were introduced that did not exist prior to 115. We believe it is better to try to back out this potentially controversial change in rendering semantics for 115 rather than try to uplift all the regression fixes.
• User impact if declined:
• Fix Landed on Version: 116
• Risk to taking this patch: Low
• Why is the change risky/not risky? (and alternatives if risky): All this does is back out the other patch in this bug.

Comment 12

[Ryan VanderMeulen [:RyanVM]]

Comment on attachment 9347791 [details]
Bug 691187 - Backout changeset e6e098584f0ab (ESR 115 only). r?aosmond

This doesn't apply cleanly to ESR115.

Comment 13

[Lee Salzman [:lsalzman]]

Comment on attachment 9347791 [details]
Bug 691187 - Backout changeset e6e098584f0ab (ESR 115 only). r?aosmond

ESR Uplift Approval Request

• If this is not a sec:{high,crit} bug, please state case for ESR consideration: Several Canvas2D rendering regressions were introduced that did not exist prior to 115. We believe it is better to try to back out this potentially controversial change in rendering semantics for 115 rather than try to uplift all the regression fixes.
• User impact if declined:
• Fix Landed on Version: 116
• Risk to taking this patch: Low
• Why is the change risky/not risky? (and alternatives if risky): Backs out patches to bring us back to prior rendering status quo.

Comment 14

[Ryan VanderMeulen [:RyanVM]]

Comment on attachment 9347791 [details]
Bug 691187 - Backout changeset e6e098584f0ab (ESR 115 only). r?aosmond

Approved for 115.2esr with the updated patch.

Comment 15

[Pulsebot]

https://hg.mozilla.org/releases/mozilla-esr115/rev/f51b2c8d1ccb