Now the code loops over system locations where libssl/libcrypto
might be located. Aside from linking, it actually tries to run
the executable before concluding that the linking "worked".
Published in <sxs4rua6zxm.fsf@florida.arsdigita.de>.
* hash.c (hash_table_map): Allow deletion and change of the
element processed by MAPFUN.
(string_hash): Use the function from glib.
* hash.c (hash_table_remove): Rewrite to actually clear deleted
entries instead of just marking them as deleted.
Published in <sxsu23tvdur.fsf@florida.arsdigita.de>.
configuration section" of top Makefile and analogous spot in others.
* po/Makefile.in.in: Previous addition of top_builddir to
po/Makefile.in was bogus -- it's generated from po/Makefile.in.in.
wrong order, causing a link failure if you're using libcrypto.a and libssl.a
rather than shared libraries. Also put in checks for -ldl, necessary since the
libcrypto shared library doesn't record its dependency on libdl.
wget.texi (Recursive Retrieval Options): Explained that you need
to use -r -l1 -p to get the two levels of requisites for a
<FRAMESET> page. Also made a few other wording improvements.
me to believe it wasn't was exposing a different bug -- URLs specified on the
commandline as opposed to being recursed to don't always get re-converted at the
end of the Wget run.
appropriate -I, -L, and -R/-rpath flags in environment variables,
manually. Automated everything, including bundling libtool so we can
successfully link with the OpenSSL shared libraries on just about any
platform.
NetWinder ARM Linux system (among others). Updated to 2001-03-16 vers.
* config.sub: Hadn't been updated since 1996 -- didn't work for
NetWinder ARM Linux system (among others). Updated to 2001-03-12 vers.
get from a file over HTTP (FTP only supports ranges ending at the end of the
file, though forcibly disconnecting from the server at the desired endpoint
might be workable).
multiple ChangeLog files (currently ./ChangeLog, doc/ChangeLog, and
src/ChangeLog), since this is unusual and people have complained their patches
hadn't been applied after checking only the top-level ChangeLog.
removed. Hopefully all the failures I was seeing were due to the fact that it
wasn't documented that non-globbing, non-recursive FTP downloads need -N to get
the remote timestamp to be preserved.
