develooper Front page | perl.perl5.porters | Postings from January 2009

Re: [PATCH] Silence time64.c build warning

Thread Previous
Glenn Linderman
January 13, 2009 10:53
Re: [PATCH] Silence time64.c build warning
Message ID:
On approximately 1/13/2009 9:50 AM, came the following characters from 
the keyboard of Craig A. Berry:
> On Tue, Jan 13, 2009 at 11:25 AM, Glenn Linderman <> wrote:
>> On approximately 1/13/2009 8:25 AM, came the following characters from the
>> keyboard of Steve Hay:
>>> Jerry D. Hedden wrote:
>>>> Silence the following build warning:
>>>>   time64.c:169: warning: 'check_tm' defined but not used
>>> Thanks, applied:
>>> 070471542c0ebe
>> The messages show up with broken links, because Exchange is chopping lines
>> to ~70 characters.  Then text-reading clients see the http: and parse and
>> create a link... of course because the line was split, it doesn't see that
>> the next line was supposed to have been part of it.
>> This is hilarious: back in about 1999, the company I then worked for
>> attempted to force everyone to use Outlook 2000 and Exchange server. And I
>> used this bug as proof that was stupid, and was able to continue using tools
>> that worked... and it is still there!
>> Oh, to keep this on-topic, it points out another deficiency of git having
>> long patch numbers, vs. perforce's shorter ones.  I'm by no means suggesting
>> that we should return to perforce, but people that post long URLs should
>> figure out a way to do it without using stupid software that inserts white
>> space behind your back.
> Luckily gitweb does just fine with chopped URLs.  This one, as chopped
> by somebody's mailer:
> or this one, manually chopped to the first 7 digits of the SHA1:
> get you to the same place.  If you're seeing a broken link, it's
> probably something else being inserted, not something missing.
> I always include only the first 7 digits when posting one of these
> URLs to avoid visual clutter as well as line wrap.  I think 7 is the
> convention followed by some of git's own tools for how much of the
> hash is very likely to be (though of course not absolutely guaranteed
> to be) unique.

Interesting!  I noticed your short one just after sending this chopped 
one.  I don't yet have git installed, and so didn't try the link, but 
trying it now I see that it does work, even though truncated, and that I 
don't need git installed to operate with it.  Handy.

On the other hand, I could see the intended link was broken, and I'm 
still amazed that people are still using M$' broken email client and/or 
server, as well as amazed that they haven't fixed it by now.  It cracks 
me up that people have put up with that all these years!

And I agree that it is white space (a newline) inserted in a run of 
non-white characters of the link, not that anything is missing.

Glenn --
A protocol is complete when there is nothing left to remove.
-- Stuart Cheshire, Apple Computer, regarding Zero Configuration Networking

Thread Previous Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at | Group listing | About