[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: rfc822 date handling in imap4d/util.c
From: |
Sergey Poznyakoff |
Subject: |
Re: rfc822 date handling in imap4d/util.c |
Date: |
Fri, 15 Jun 2001 11:28:54 +0300 |
Bonjour,
> This appears to be becomeing the official GNU mailutils greeting, so
> what do you guys think? Fight the unilingualism of POP3 servers!
<..snip..>
> + fprintf (ofile, "+OK Bonjour, POP3 Ready %s\r\n", md5shared);
Je suis d'accord. Votre avis, Alain? Mieux vaudrait mettre:
> + fprintf (ofile, "+OK Bonjour, POP3 est pret %s\r\n", md5shared)
mais ici nous manquons le circonflexe...
> Pulled the code out so it could be linked by imap4d, since imap4d's
> implementation of this had sscanf bugs, and was essentially identical.
>
> It should be prototyped, but I don't know what the plan for sharing
> this kind of stuff is.
It would be very useful to have extern imap_envelope_date() function,
but it seems you have forgotten the attachment :(
> Also implemented the timezone, I think.
There are some problems with it, though. struct tm does not have
members tm_gmtoff and tm_zone on GNU/Linux (with libc 5) and
on Solaris. They are present on FreeBSD, though. GNU/Linux's struct tm
has two fields (long int __tm_gmtoff__ and __const char *__tm_zone__)
which are not used now. So we are facing a portability problem. We
could test for existence of these fields in configure.in, but what
to do when they are absent? I believe we need a workaround. A possible
solution would be to have our version of mktime, say mu_mktime(),
which whould take care of TZ properly. What do you think?
> -util_parse_rfc822_date (char *date, time_t *timep)
> +util_parse_ctime_date (char *date, time_t *timep)
Mea culpa! The name was actually a misnomer. Thanks a lot!
I have fixed sscanf() format bugs in util.c. Thanks for noticing.
The patches for pacifying gcc -Wall were also applied. I have tested
the stuff, it is working fine. The only thing I was not able to
apply was
> + if(imap_parse_date_time(datep, &tm))
since the function is not defined anywhere.
Au revoir,
Sergey