Closed Bug 673759 (namedspaceOverriding) Opened 8 years ago Closed 7 years ago

[MathML3] Remove namedspace value overriding

Categories

(Core :: MathML, defect)

defect
Not set

Tracking

()

RESOLVED FIXED
mozilla15

People

(Reporter: fredw, Assigned: fredw)

References

(Blocks 1 open bug)

Details

(Keywords: dev-doc-complete)

Attachments

(2 files, 8 obsolete files)

3.34 KB, patch
karlt
: review+
emorley
: checkin+
Details | Diff | Splinter Review
5.49 KB, patch
karlt
: review+
Details | Diff | Splinter Review
The REC says:

"MathML2 allowed the binding of namedspaces to new values. It appears that this capability was never implemented, and is now deprecated; namedspaces are now considered constants. For backwards compatibility, the following attributes are accepted on the mstyle element, but are expected to have no effect."

We implement namedspace value overriding in nsMathMLFrame::ParseNamedSpaceValue. Removing this will allow to get rid of many almost never used nsGkAtoms for namespaces. This will also simplify the parsing code: we could move the parsing for namespaces into nsMathMLElement::ParseNumericValue (and add a flag PARSE_ALLOW_NAMEDSPACE).
Assignee: nobody → hage.jonathan
Status: NEW → ASSIGNED
(In reply to comment #5)
> Created attachment 550043 [details] [diff] [review] [diff] [details] [review]
> Moving nsMathMLFrame::ParseNamedSpaceValue into
> nsMathMLElement::ParseNumericValue + Removing of namedspace value overriding

Compilation fails with this patch:

/src-obj/mozilla/src/content/base/src/nsTreeSanitizer.cpp:952: error: ‘negativemediummathspace_’ is not a member of ‘nsGkAtoms’
...

The nsGkAtoms for spaces should be removed everywhere.
Blocks: 677036
Comment on attachment 549073 [details] [diff] [review]
Reftest for positive namedspaces

We can probably already take this reftest.
Attachment #549073 - Flags: review?(karlt)
Attachment #549073 - Flags: review?(karlt) → review+
Keywords: checkin-needed
Whiteboard: [good first bug] → [good first bug][please push attachment 549073]
I'm presuming this needs to be left open for the other patch? Should a review be set on that patch?

https://hg.mozilla.org/integration/mozilla-inbound/rev/f5ce0b70c2e1
Flags: in-testsuite+
Keywords: checkin-needed
Whiteboard: [good first bug][please push attachment 549073] → [good first bug] [leave open for remaining patch]
Target Milestone: --- → mozilla9
Attachment #549073 - Flags: checkin+
Remaining patch isn't going to make it for mozilla9, so removing target milestone for now.
Target Milestone: mozilla9 → ---
Assignee: hage.jonathan → fred.wang
Alias: namedspaceOverriding
Depends on: 696647
Attached patch Patch V3 - part 1 (obsolete) — Splinter Review
Attachment #551414 - Attachment is obsolete: true
Part 2, removing the corresponding MathML Atoms. To apply over patch from bug 696647.

I'm not sure this patch is good idea, though. Bug 324472 comment 5 says that atoms are needed for clipboard operations.
Comment on attachment 586105 [details] [diff] [review]
Patch V3 - part 1

Note: I have not changed the flags passed to ParseNumericValue, as I expect this job to be done in bug 411227.
Attachment #586105 - Flags: review?(karlt)
Attachment #586105 - Flags: review?(karlt)
No longer blocks: 677036
Depends on: 677036
Attachment #586107 - Attachment is obsolete: true
Attachment #586105 - Attachment is obsolete: true
Attachment #586107 - Attachment is obsolete: false
Whiteboard: [good first bug] [leave open for remaining patch] → [leave open for remaining patch]
Keywords: helpwanted
Comment on attachment 586107 [details] [diff] [review]
Patch V3 - part 2

So now bug 677036 is fixed, it should not be possible to override namedspace. Karl, do you think we should take attachment 586107 [details] [diff] [review]? Bug 324472 comment 5 indicates that atoms are needed for clipboard operations. However, I don't think people have ever used namedspace overriding, so that should not be a problem?
Attachment #586107 - Flags: review?(karlt)
Comment on attachment 586107 [details] [diff] [review]
Patch V3 - part 2

Yes, makes sense to remove these now, thanks.
Attachment #586107 - Flags: review?(karlt) → review+
https://tbpl.mozilla.org/?tree=Try&rev=5adfc5ff68c7

I'm not sure how the build results can be red (with empty summaries) but the tests are executed and seem successful.
No longer depends on: 677036
Depends on: 677036
(In reply to Frédéric Wang (:fredw) from comment #21)
> https://tbpl.mozilla.org/?tree=Try&rev=5adfc5ff68c7
> 
> I'm not sure how the build results can be red (with empty summaries) but the
> tests are executed and seem successful.

Karl, do you have any idea about this problem? I can build opt/debug builds without problems with this patch applied. I tried the patch twice on the Try server and always got this build results in red, with empty summaries.
I don't know what happened there but some others had similar problems
e.g. https://tbpl.mozilla.org/?tree=Try&rev=be00392bbc1f

I'd try pulling and rebasing against the latest known good changeset.
OK, that looks better:
https://tbpl.mozilla.org/?tree=Try&rev=3dc56ac6f8c3
Keywords: checkin-needed
Whiteboard: [leave open for remaining patch]
https://hg.mozilla.org/mozilla-central/rev/a048bf8506f7
Status: ASSIGNED → RESOLVED
Closed: 7 years ago
Resolution: --- → FIXED
Keywords: dev-doc-needed
You need to log in before you can comment on or make changes to this bug.