Closed Bug 727804 Opened 13 years ago Closed 10 years ago

height of brackets generated by mfenced doesn't match height of separators

Categories

(Core :: MathML, defect)

Product:
Core
Component:
MathML
Version:
10 Branch
Platform:
x86_64
Linux
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

()

Status:
RESOLVED FIXED
Milestone:
mozilla36

People

(Reporter: adam.slaski, Assigned: jkitch)

References

Details

(Keywords: testcase)

Attachments

(2 files, 3 obsolete files)

mathml-test.html - an ilustration for this bug
(deleted), text/html
Details
test
(deleted), patch
Details | Diff | Splinter Review
User Agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:10.0.1) Gecko/20100101 Firefox/10.0.1 Build ID: 20120210025125 Steps to reproduce: I used '&' (an ampersand) as a separator in mfenced Actual results: height of brackets match height of the highest element and disregard the height of the separator Expected results: height of brackets should match both height of the highest element and the height of the separator
Attachment #597788 - Attachment mime type: text/plain → text/html
Confirmed with 2013-08-22-03-02-04-mozilla-central-firefox-26.0a1.en-US.linux-x86_64
Status: UNCONFIRMED → NEW
Component: Untriaged → MathML
Ever confirmed: true
Keywords: testcase
Product: Firefox → Core
Can you test in a recent nightly build? (see bug 670334 comment 20)
Flags: needinfo?(adam.slaski)
Attached patch test (obsolete) (deleted) — — Splinter Review
By redacting everything except the fences, I've been able to get an automated test that passes all try platforms.
Assignee: nobody → jkitch.bug
Status: NEW → ASSIGNED
Attachment #8510007 - Flags: review?(fred.wang)
Attached patch testcase (obsolete) (deleted) — — Splinter Review
Oops. Wrongly flagged Windows as failing.
Attachment #8510007 - Attachment is obsolete: true
Attachment #8510007 - Flags: review?(fred.wang)
Attachment #8510008 - Flags: review?(fred.wang)
Comment on attachment 8510008 [details] [diff] [review] testcase Review of attachment 8510008 [details] [diff] [review]: ----------------------------------------------------------------- You can use style="visibility: none;" for the <mn>'s, so that it's clearer that we want to hide them. Actually, I wonder if you just want to use <mspace width="10px"/>, so that their heights are not involved at all in the computation. To be more reliable about the <math> position and to avoid issue with the top/left border, you can absolutely position the <math> to (10px, 10px) too. Then for the div I would do "top: 0px; height: 120px;" (height = 4 times the font-size + 10px above/below) so that we are very sure the ampersand is covered.
Attachment #8510008 - Flags: review?(fred.wang) → review+
Attached patch testcase (obsolete) (deleted) — — Splinter Review
The <mspace/> change wasn't adopted as there are positioning differences between the <mfenced> and <mrow>+<mo> cases.
Attachment #8510008 - Attachment is obsolete: true
https://treeherder.mozilla.org/ui/#/jobs?repo=try&revision=26fb9f23f6d8 Adam: Reopen if this is still a problem.
Depends on: 670334
Keywords: checkin-needed
Comment on attachment 8510997 [details] [diff] [review] testcase Review of attachment 8510997 [details] [diff] [review]: ----------------------------------------------------------------- ::: layout/reftests/mathml/mfenced-12.html @@ +1,5 @@ > +<!doctype html> > +<html class="reftest-wait"> > + > +<math style="font-size:25px; position: absolute; top: 10px; left:10px;"> > + <mfenced id="outer" separators="&"> BTW, here and in the ref, this should really &amp;
Keywords: checkin-needed
Attached patch test (deleted) — — Splinter Review
Attachment #8510997 - Attachment is obsolete: true
https://hg.mozilla.org/mozilla-central/rev/ec4245765266
Status: ASSIGNED → RESOLVED
Closed: 10 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla36
Flags: needinfo?(adam.slaski)
QA Whiteboard: [good first verify]
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: