Stop reporting redundant fragment parser error in foreign content #9809
Conversation
not-my-profile
force-pushed
the
fix-redundant-fragment-parser-error-in-foreign-content
branch
2 times, most recently
from
d119c56 to
d04ee3d
Compare
|
|
||
| <li><p>Otherwise, process the token according to the rules given in the section corresponding | ||
| to the current <span>insertion mode</span> in HTML content.</li> | ||
| <li><p>Return to the step labeled <i>loop</i>.</p></li> |
There was a problem hiding this comment.
(Rewriting this set of steps as a https://infra.spec.whatwg.org/#iteration-while "While true" could be nice.)
There was a problem hiding this comment.
Personally I don't really find "While true" any better than "loop" ... it's just as undescriptive regarding when the loop will terminate. In any case I think such a change should be done in a followup PR ... so that the diff of this clarification is minimal.
There was a problem hiding this comment.
While true with nested steps is much better than goto imo.
There was a problem hiding this comment.
Ah yes that's true ... there however aren't any nested steps here.
There was a problem hiding this comment.
Right, @annevk suggests switching from goto without nested steps to while true with nested steps.
There was a problem hiding this comment.
Right ... I think Anne meant that the real advantage of while is if you have multiple nested loops that is a loop within a loop, which as I pointed out isn't the case here.
In any case as I said I'd like to keep the changes in this commit minimal (since it fixes something that's wrong) and keep editorial changes to future PRs.
I have tested this change against the test cases from html5lib-tests. In document parsing this doesn't change any test results. In fragment parsing this fixes 7 test cases by reporting one parser error less. (That is all existing HTML parsers that pass html5lib-tests must have already implemented these steps in the order they are changed to by this commit.) The test cases where one error is reported less fall in two categories: * fragment parsing e.g. `<svg></table>` with a `tbody` HTML element as the context element => one "unexpected end tag" error is reported less (previously it was reported twice) * fragment parsing e.g. `<g></path>` with a `path` SVG element as the context element => one "found special tag while closing generic tag" error is no longer reported Fixes html5lib/html5lib-tests#173.
not-my-profile
force-pushed
the
fix-redundant-fragment-parser-error-in-foreign-content
branch
from
d04ee3d to
1d479e3
Compare
I have tested this change against the test cases from html5lib-tests.
In document parsing this doesn't change any test results.
In fragment parsing this fixes 7 test cases by reporting one parser error less. (That is all existing HTML parsers that pass html5lib-tests must have already implemented these steps in the order they are changed to by this commit.)
The test cases where one error is reported less fall in two categories:
fragment parsing e.g.
<svg></table>with atbodyHTML element as the context element => one "unexpected end tag" error is reported lessfragment parsing e.g.
<g></path>with apathSVG element as the context element => one "found special tag while closing generic tag" error is no longer reportedFixes html5lib/html5lib-tests#173.
/parsing.html ( diff )