I believe I have addressed all of gstein's issues about the date
parser, as well as some issues I found myself today.
http://subversion.tigris.org/nonav/issues/showattachment.cgi/270/diff
contains the updated patch.
To gstein: Please verify that the issues you presented have been
addressed, and consider withdrawing your veto.
To fitz: I am told that chapter 3 of the book is your chapter.
http://subversion.tigris.org/nonav/issues/showattachment.cgi/271/doc-diff
contains a suggested minimal change to the book for the new date
parser, but you may wish to make more extensive changes. The main
patch includes an addition to doc/book/REVIEW which should help you
document the new date parser.
(Chapter 9 also needs a small tweak; doc-diff has that as well. I
don't know whose chapter that is.)
To all who reviewed #408: your +1 votes are out of date. Please
review the new diff. Changes vs. the old diff include:
* Tests for svn_parse_date() have been added.
* A subtle daylight savings time issue has been addressed. As a
consequence, the new svn_parse_date() API no longer takes a gmtoff
parameter, since the local time zone cannot be expressed as a
single number. (The GMT offset for the parsed time may not be the
same as the GMT offset for the current time, since one may be in
DST while the other is not.)
* The date formats have been changed to require minutes when there
are hours, and to require a time when there is a time zone,
following the lead of:
http://www.w3.org/TR/NOTE-datetime
* The date templates have been fixed to process time zone offsets
correctly.
* The template parser verifies that explicit time zone offsets are
within bounds, since that cannot be done once the offset has been
combined into a single number.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Mon Jan 12 00:08:30 2004