Bug 149092 – document.lastModified should use the user's local time zone

• Component: DOM
• Status: RESOLVED FIXED

[Chris Dumez]

document.lastModified should use the user's local time zone:
https://html.spec.whatwg.org/multipage/dom.html#dom-document-lastmodified

Chrome and Firefox comply with the specification. However, it looks like WebKit returns UTC (7 hours difference on my machine).

[Chris Dumez]

rdar://problem/22567705

[Chris Dumez]

Created attachment 261062 [details]
Patch

[Alexey Proskuryakov]

Comment on attachment 261062 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=261062&action=review

> LayoutTests/http/tests/misc/last-modified-parsing-expected.txt:6
> +11/20/2008 17:03:33

What will happen with this test if my time zone is not PST?

[Chris Dumez]

Comment on attachment 261062 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=261062&action=review

>> LayoutTests/http/tests/misc/last-modified-parsing-expected.txt:6
>> +11/20/2008 17:03:33
> 
> What will happen with this test if my time zone is not PST?

Good point. I should probably edit the test so that it prints UTC time.

[Chris Dumez]

Created attachment 261075 [details]
Patch

[Chris Dumez]

(In reply to comment #3)
> Comment on attachment 261062 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=261062&action=review
> 
> > LayoutTests/http/tests/misc/last-modified-parsing-expected.txt:6
> > +11/20/2008 17:03:33
> 
> What will happen with this test if my time zone is not PST?

This is addressed in the latest iteration of the patch.

[Daniel Bates]

Comment on attachment 261075 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=261075&action=review

> Source/WebCore/ChangeLog:17
> +        No new tests, already covered by existing tests.
> +

Do we have any tests/infrastructure to check the correctness of document.lastModified with respect to different time zones? If not, then I suggest that we add infrastructure to support changing to a time zone t at runtime and tests to ensure that document.lastModified is correct with respect t.

[Daniel Bates]

Comment on attachment 261075 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=261075&action=review

> Source/WebCore/ChangeLog:17
> +        No new tests, already covered by existing tests.
> +

Do we have any tests/infrastructure to check the correctness of document.lastModified with respect to different time zones? If not, then I suggest that we add infrastructure to support changing to a time zone t at runtime and tests to ensure that document.lastModified is correct with respect t.

[Chris Dumez]

(In reply to comment #8)
> Comment on attachment 261075 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=261075&action=review
> 
> > Source/WebCore/ChangeLog:17
> > +        No new tests, already covered by existing tests.
> > +
> 
> Do we have any tests/infrastructure to check the correctness of
> document.lastModified with respect to different time zones? If not, then I
> suggest that we add infrastructure to support changing to a time zone t at
> runtime and tests to ensure that document.lastModified is correct with
> respect t.

I don't believe we currently have test infrastructure for changing the system's timezone. While this would potentially be a nice thing to have, I don't believe this is a pre-requirement for this patch though. The existing W3C tests check that document.lastModified is using the current system timezone. They used to fail for every timezone other than UTC and now they pass.

[Ryosuke Niwa]

Comment on attachment 261075 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=261075&action=review

>>>> Source/WebCore/ChangeLog:17
>>>> +
>>> 
>>> Do we have any tests/infrastructure to check the correctness of document.lastModified with respect to different time zones? If not, then I suggest that we add infrastructure to support changing to a time zone t at runtime and tests to ensure that document.lastModified is correct with respect t.
>> 
>> Do we have any tests/infrastructure to check the correctness of document.lastModified with respect to different time zones? If not, then I suggest that we add infrastructure to support changing to a time zone t at runtime and tests to ensure that document.lastModified is correct with respect t.
> 
> I don't believe we currently have test infrastructure for changing the system's timezone. While this would potentially be a nice thing to have, I don't believe this is a pre-requirement for this patch though. The existing W3C tests check that document.lastModified is using the current system timezone. They used to fail for every timezone other than UTC and now they pass.

I don't think changing the system clock's timezone is a realistic solution since we parallelize tests.
In any case, I don't think we should block this patch for adding such a test infrastructure.

> LayoutTests/ChangeLog:15
> [phone]  Chris Dumez  <[e-mail]>

Duplicated change log entries.

[Chris Dumez]

Created attachment 261098 [details]
Patch

[Chris Dumez]

Comment on attachment 261098 [details]
Patch

Clearing flags on attachment: 261098

Committed r189679: <http://trac.webkit.org/changeset/189679>

[Chris Dumez]

All reviewed patches have been landed.  Closing bug.

[Alexey Proskuryakov]

*** Bug 4363 has been marked as a duplicate of this bug. ***