61107 - bad link interpretation in utf-8 document (Build ID: 2000101014)

Status: VERIFIED INVALID

(Core :: DOM: HTML Parser, defect, P3)

Description

[Stanislav Pecha]

Page: http://gw.internet.pb.cz/noconv/spoluzaci/index2.php?lang=cs
two links in menu:
1.http://gw.internet.pb.cz/noconv/spoluzaci/vyber.php?did=[some number]%E2%8C%A9=en
2.http://gw.internet.pb.cz/noconv/spoluzaci/nova.php?lang=en&did=[some number]
Both links are same but 1st link has switch did first and lang 2nd.
Second link has lang 1st and did 2nd.
When is lang 2nd mozilla change string '&lang' to '%E2%8C%A9'
It do only with switch named 'lang' if it is not used like first switch in anchor
All the page is in UTF-8 charcode.
When i look on page source, all is fine. It is critical for my web (I must
rebuild it) but not for Mozilla. It should be something worse in Mozilla code
and it's better to check it.
IE 5.5 has no problem with this.

Comment 1

[Keyser Sose]

Reporter is this still a problem in the latest nightlies?

Comment 2

[Keyser Sose]

WORKSFORME
Platform: PC
OS: Linux 2.2.16
Mozilla Build: 20002808 M18 Trunk Build

Marking as such.

Comment 3

[Keyser Sose]

Really marking WORKSFORME.

Comment 4

[Stanislav Pecha]

Build ID: 2000122820 still do it.
http://gw.internet.pb.cz/spoluzaci/index2.php?lang=en

Comment 5

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Confirming in Linux build 2001-01-02-11-Mtrunk.  I'm not sure that we follow the
links incorrectly but they certainly show up wrong in the toolbar.

Comment 6

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Attachment: reduced testcase

Comment 7

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Oops.  I see the problem now.  This has nothing to do with Layout's mucking
about with URI character encodings...  This bug is probably invalid, but if it's
not it's a parser issue.  The problem is that you need to escape & in URLs
within HTML to &.  Otherwise we interpret &lang as the HTML entity for a
left angle bracket.

Harish - you should probably check what the correct set of termination
characters for entities is.  I'm not sure we implement that correctly.

Comment 8

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

er, entity references, not entities

Comment 9

[harishd]

David, the bug ( not sure if it's a bug! ) is confined only to the strict 
documents. Removing the DOCTYPE should exhibit the same behavior as in Nav 4.x 
and IE.

Btw, entity parsing, for attributes, happens in the content sink.

Comment 10

[harishd]

This bug has been marked "future" because the original netscape engineer working 
on this is over-burdened. If you feel this is an error, that you or another
known resource will be working on this bug,or if it blocks your work in some way 
-- please attach your concern to the bug for reconsideration -----

Comment 11

[bsharma]

updated qa contact.

Comment 12

[Christopher Hoess (gone)]

Per SGML, the part of the general entity reference between the "&" and the 
reference end must be name characters (the actual production is a bit more 
complicated, but this is all we have to deal with).  The reference end is 
normally ; or a record end, but it may be omitted "if the reference is not 
followed by a character that could occur in the reference".  This is indeed the 
case (since "=" cannot appear in names), the entity is terminated, and our 
parsing is correct.  INVALID.

Comment 13

[Moied]

Verified invalid