Michael Hirohama writes:
>
> I've had much frustration in compiling the htdig3 CVS source using
> the 1999-09-27 snapshot of egcs-g++ on a powerpc-ibm-aix4.1.4.0.
> My desire to use the 2.7.7 Version of Berkeley DB and difficulties
> in getting a working g++ added to the difficulty of using the
> buggy AIX <sys/time.h> and varargs files.
Could you send the full error messages from the compiler ? Using
appropriate #if and autoconf, we should be able to get it right. Of
course I could just review all files including time.h but your messages
will help.
> I uncovered and reported an EGCS compiler bug generated by the
> mystrcasecmp(sorttypes[i].typest, st) expression in
> int ResultMatch::setSortType(const String& sorttype) in
> htnet/HtHTTP.cc.
If this version contains multiple mystrcasecmp (one const, another not const)
that's possible. I've changed back to only one const mystrcasecmp + cast
where appropriate. It should help.
> I needed to change some references to expressions like
> SetBodyReadingController(&HtHTTP::ReadBody); /* htnet/HtHTTP.cc:141 */
>
> G++ reports many errors of the form:
> In file included from ../../htlib/String_fmt.cc:17:
> ../../htlib/htString.h: In method `char String::operator [] (int) const':
> ../../htlib/htString.h:225: warning: sorry: semantics of inline function static data
>`char null' are wrong (you'll wind up with multiple copies)
> ../../htlib/htString.h:225: warning: you can work around this by removing the
>initializer
I've fixed this by moving the static char null outside
the inline function. The header has a static char null = '\0' in it and
therefore will have as many copies of null as there are source files using
htString. We loose at most 100 bytes. Tolerable ?-)
> I recommend decoupling the Berkeley DB distribution in the CVS
> tree. The version in CVS is old. If you need to build compiled
> packages for distribution which include DB, do this at build
> time. Developers should not be forced to use a particular DB
> release.
I advocate for keeping db in htdig. Otherwise we will have to support
multiple Berkeley DB versions. The 2.7.5 is not *that* old (July). Why
do you need to use 2.7.7 ? A specific bug fix in it ?
Cheers,
--
Loic Dachary
ECILA
100 av. du Gal Leclerc
93500 Pantin - France
Tel: 33 1 56 96 09 80, Fax: 33 1 56 96 09 61
e-mail: [EMAIL PROTECTED] URL: http://www.senga.org/
------------------------------------
To unsubscribe from the htdig3-dev mailing list, send a message to
[EMAIL PROTECTED] containing the single word "unsubscribe" in
the SUBJECT of the message.
