238989 - [FIXr]whitespace after a </b> tag is ignored when enclosed inside <fieldset></fieldset> tags

Status: RESOLVED FIXED

(Core :: DOM: HTML Parser, defect, P1)

Description

[Seth Borg]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.6) Gecko/20040206 Firefox/0.8
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.6) Gecko/20040206 Firefox/0.8

<b>text</b> more text comes out looking like this: textmore text.
<b>text </b> more text comes out correctly.
I've been having to use <span style="font-weight: bold;">text</span> more text
to make it look right.

Reproducible: Always
Steps to Reproduce:
1. Just try putting some whitespace after a </b> tag, it won't show up.
2. <b>text</b> more text comes out looking like this: textmore text.


Actual Results:  
textmore text

Expected Results:  
text more text

Comment 1

[Bill Mason]

WFM 20040327 PC/WinXP

Testcase forthcoming

Comment 2

[Bill Mason]

Attachment: testcase

Comment 3

[Seth Borg]

Attachment: Shows how </b> tags inside <fieldset></fieldset> tags ignore whitespace

Comment 4

[Seth Borg]

Attachment: whitespace after certain tags is ignored when enclosed inside <fieldset></fieldset> tags

Comment 5

[Roland Mainz]

Confirming, Mozilla1.7b, Linux/x86.

Comment 6

[Christopher Hoess (gone)]

Same as comment #19 in bug 48376, although I'm not entirely sure it's the same
underlying issue as the original problem in bug 48376.

Comment 7

[Boris Zbarsky [:bzbarsky]]

Attachment: This should fix it

Comment 8

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 145302 [details] [diff] [review]
This should fix it

So the problem here was that <fieldset> was marked as a tag which is allowed to
discard
all direct child whitespace that's not part of a text run.  That seems pretty
bogus to me.

Comment 9

[Seth Borg]

Why is that "bogus"

Comment 10

[Boris Zbarsky [:bzbarsky]]

Because there is absolutely no reason that fieldset should be allowed to strip
out whitespace?  Perhaps a better question is, "why would that possibly be not
bogus?"

Comment 11

[Peter Van der Beken [:peterv]]

Comment on attachment 145302 [details] [diff] [review]
This should fix it

Fine by me, although I don't really understand why we'd want kOmitWS at all.

Comment 12

[Christopher Hoess (gone)]

Comment on attachment 145302 [details] [diff] [review]
This should fix it

What peterv said.

Comment 13

[Boris Zbarsky [:bzbarsky]]

It's meant to be an optimization (memory-saving, etc), I think.  If you guys
think we should remove it, I'll be up for that.  Editor will likely thank us.  ;)

Comment 14

[Boris Zbarsky [:bzbarsky]]

Actually, I _think_ that this patch should be safe enough for 1.7.  Peterv,
choess, what do you think?

Comment 15

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

for the record, removing kOmitWS sonuds like a swell idea to me.

Comment 16

[Peter Van der Beken [:peterv]]

Yeah, this should be safe for 1.7 imho.

Comment 17

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 145302 [details] [diff] [review]
This should fix it

Could this be approved for 1.7?  This is a safe fix to keep us from mis-parsing
content inside <fieldset>.

Comment 18

[Brendan Eich [:brendan]]

Comment on attachment 145302 [details] [diff] [review]
This should fix it

a=[e-mail] for 1.7 final.

/be

Comment 19

[Boris Zbarsky [:bzbarsky]]

Checked in for 1.7.  I've filed bug 240139 as a followup on the kOmitWS issue.