76412 - View Source does not do error-reporting correctly if VIEW_SOURCE_HTML is set

Status: RESOLVED FIXED

(Core :: DOM: HTML Parser, defect, P4)

Description

[Boris Zbarsky [:bzbarsky]]

Comments from bug 52154:

------- Additional Comments From [e-mail] 2001-04-17 14:37 -------

It just occurred to me that something was left out when jce2 initially
implemented viewsource in html. I am mentioning it here for the record in case
someone wants to fix it. It has to do with error reporting. The XML version
outputs <error>...</error> [see CViewSourceHTML::WriteTagWithError()]. In the
HTML version, this has to become <span class="error">...</span>.

Comment 1

[rbs]

The rule for that was in viewsouce.css for the error (others prefer blink :-)
.error {
  background-color: #FFFFCC; 
  font-weight: bold;
} 

Comment 2

[Boris Zbarsky [:bzbarsky]]

There is another issue here.  Currently mValidator is always null as far as I
can tell.  So we end up not doing any of the error reporting stuff anyway... 
I'll investigate this a bit.

Comment 3

[Boris Zbarsky [:bzbarsky]]

Does the removal of the strict DTD have anything to do with this?  I imagine we
used that as the validator.... See bug 50070

Comment 4

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

The strict DTD wouldn't always report the errors correctly, so I don't think we
should use it.

Comment 5

[Boris Zbarsky [:bzbarsky]]

OK.  Well, does the quirks DTD report errors?  If so, I would be willing to try
and hook up this error reporting again...

Comment 6

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

I hope not, because I strongly suspect it's concept of an error is different
from what's an error according to the spec's.  If we have error reporting it
should be correct, and I doubt we could get anything close to that with any of
the existing DTDs, and especially CNavDTD.

Comment 7

[Andreas M. "Clarence" Schneider]

This is completely broken. Setting mValidator is easy, but does not make it work.
I think the obsolete code can be savely removed.

My fix for bug 57724 will include some syntax highlighting at the tokenizer
level, but that is quite different from the old error reporting (will mark
things such as unescaped ampersands, forgotten semicolons and attributes in end
tags).

Comment 8

[Boris Zbarsky [:bzbarsky]]

Andreas, do you want to morph this bug into "remove the error-reporting code"? 
And take it?

Comment 9

[Andreas M. "Clarence" Schneider]

Do we all agree on removing this code?

Comment 10

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

I certainly agree, but Harish should have his say.

Comment 11

[harishd]

Since we don't have a DTD that could be used to report errors correctly ( unless
someone is willing to implement an nsErrorsDTD...which I really doubt ) it makes
sense to remove the unused code.

Comment 12

[Asa Dotzler [:asa]]

Bugs targeted at mozilla1.0 without the mozilla1.0 keyword moved to mozilla1.0.1 
(you can query for this string to delete spam or retrieve the list of bugs I've 
moved)

Comment 13

[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 14

[Blake Kaplan (:mrbkap) (inactive)]

It looks like there's very little left to rip out, might as well get this done.

Comment 15

[Blake Kaplan (:mrbkap) (inactive)]

Attachment: Cleanup in aisle 5!

This rips out some more unnecessary garbage.

Comment 16

[Blake Kaplan (:mrbkap) (inactive)]

Fix checked in.