439293 - A <div> inside an overflow'd <pre> on an RTL page is truncating text (no scrollbar)

Status: RESOLVED INVALID

(Core :: Layout, defect)

Description

[shimi]

User-Agent:       Mozilla/5.0 (compatible; Konqueror/3.5; Linux) KHTML/3.5.7 (like Gecko)
Build Identifier: Mozilla/5.0 (X11; U; Linux i686 (x86_64); en-US; rv:1.9) Gecko/2008052912 Firefox/3.0

See attached smaller testcase (thanks for mzz on #firefox @ irc.firefox.org for making it :))

Reproducible: Always

Steps to Reproduce:
1. See testcase

Actual Results:  
div text is truncated, no scrollbar appears. did not happen in any other browser, including FF2!

Comment 1

[shimi]

Attachment: example testcase

Comment 2

[Paul O'Shannessy [:zpao] (not bugmail, email directly)]

I would say this is a regression as it worked in FF2.

Comment 3

[Simon Montagu :smontagu]

This regressed between 2006-03-15-04 and 2006-03-16-04, probably from bug 192767 http://bonsai.mozilla.org/cvsquery.cgi?treeid=default&module=all&branch=HEAD&branchtype=match&dir=&file=&filetype=match&who=&whotype=match&sortby=Date&hours=2&date=explicit&mindate=2006-03-15+04%3A00&maxdate=2006-03-16+04%3A00&cvsroot=%2Fcvsroot

Comment 4

[Uri Bernstein]

Note that the two cases in the testcase aren't symmetrical: The inner DIV in both of them has dir="ltr". Changing this to dir="rtl" in the second (RTL) case, produces the expected scrollbar (and similarly, changing the inner div in the first case to "rtl" causes the scrollbar to disappear).

Also note that Safari behaves the same way.

Comment 5

[Tomer Cohen :tomer]

(In reply to comment #4)

Uri - Any chance to get it fixed for 1.9.0.1?

Comment 6

[Uri Bernstein]

(In reply to comment #5)
> (In reply to comment #4)
> 
> Uri - Any chance to get it fixed for 1.9.0.1?
> 

Sorry for the late reply. Anyway, the gist of comment #4 was that I tend to think this is INVALID. The issue should be easy to fix on the web site by setting dir="ltr" on the <pre>.

Comment 7

[shimi]

Uri,

So just to understand this right: Are you saying, that the spec says that the combination that I opened the bug about SHOULD result what it is resulting? I am not asking if there is a workaround, rather than trying to understand why you say this bug is INVALID - because the behavior doesn't look reasonable to me [OTOH, I don't know the spec...]

Thanks

Comment 8

[Samuel Sidler (old account; do not CC)]

Not wanted based on comment 6, but please renominate if this doesn't become INVALID.

Comment 9

[shimi]

(In reply to comment #8)
> Not wanted based on comment 6, but please renominate if this doesn't become
> INVALID.
> 

Samuel - as Uri ignored my comment in reply to his claim that this is INVALID, and you ignored it as well - I will ask again: Since when a workaround for a bug make the bug void ? Are we talking here about MSIE, the browser with special hacks per each version of the browser ?

Comment 10

[Mats Palmgren (inactive)]

Overflow on the start side should be ignored (i.e. overflow on the left side for
LTR and overflow on the right side for RTL).
Chrome25 and IE10 renders the testcase the same a Nightly (no horizontal scrollbar
in the RTL case).  Opera 12.11 does have a horizontal scrollbar though.

Note that the attached test isn't symmetrical; if you change the INNER <div> in
the LTR case to <div style="text-align: right;" dir="rtl"> then it's symmetrical.
Then the result is that the end of the text is visible and the start is clipped.

It is a bug that a vertical scrollbar is visible though - I have a fix for that
in my tree (bug 806284).

Comment 11

[Mats Palmgren (inactive)]

Attachment: Symmetrical test