libtool 1.4.3 is scary as hell and caused just about every build on all sorts
of platforms to stop working, thanks to the fact that it ruquires a SED
variables somehow set by the configure script. It works fine on my linux
running autoconf 2.57 and automake 1.7 but others seem not to do as fine.

Reverting back to the ltmain.sh we had previously, which I believe is 1.4.2
including handmade patches for FreeBSD.

ALERT ALERT ALERT before we try 1.4.3 or similar versions again, check the
${SED} stuff and similar carefully.

AIX when this fix is applied and I cannot see how this will break any
systems.

and two makefiles fixed.

automatic disable won't cause a warning

*_r() functions as they're not needed (and if fact mess things up for us).
Brought to our attention by the friendly Troels Walsted Hansen in bug report
#696217.

code as this makes compiler warnings. We *should* fix the code to deal with
the return codes instead...

lingering POST-stuff that confuses GET requests. Juan F. Codagnone reported
this problem in bug report #653859.

compiler, mentioned by Andi Jahja

the recent libcurl changes.

looks like.

when SIZE is used on the file while being in BINARY mode. To work
around that (stupid) behavior, we attempt to parse the RETR response
even if the SIZE returned size zero.

Debugging help from Salvatore Sorrentino on February 26, 2003.

include size in the 150-reply

function, as this could easily end up looping for a very long time (more or
less until the whole transfer was done) and no library-using app would want
that.

Found thanks to a report by Kyle Sallee.

header is used, we must wait for a 100-code (or timeout), before we send the
data. The timeout is merely 1000 ms at this point. We may have reason to set
a longer timeout in the future.

than 5 could cause a segfault.