* Changed layout
* Removed text, split up 3.1.x->3.2.x upgrade info to Unreal31_to_32.html
(actually that's just .RELEASE.NOTES, need a bit more explanation and
stuff about converting configs etc).
* Improved Me, Class, Allow, Listen and Set block text/layout/etc.
* Removed typo's
at windows and not at *NIX? Also it's incompatible with running as a service.
AFAIK it doesn't increase security very much: if you own the box you can easily
snif the keyboard, read the certificate from memory, etc..
- Fixed bug in +q/+a list (/mode #chan q) <= lol this was some historic
bug by DrBin or something ;). In case you wanna know.. the whole loop logic
was wrong... initalisation vs null pointer check... ;).
which is set if dead_link is called. You will now see "Write error",
"Max SendQ exceeded" etc error messages in the quit reason instead of just
the "Dead socket" message. Changed "notice" parameter of dead_link, now just
the reason and not a format string, maybe rename that var.
instead of using a 2nd flag here for the special case of "the first zip msg"
like in hybrid/etc I use cptr->zip->first to flag that. Except for the many
#ifdef ZIP_LINKS blocks added I also had to do some stuff outside it, like
crc32->our_crc32 because zlib defines it, made a READBUF define (8192),
added a msg var to parsing/send functions.. blah.. etc ;P.
I've also put the patch online at www.vulnscan.org/tmp/zip_links.diff so you
can easily look what I've changed.
TODO: ask in ./Config if ziplinks should be enabled and let ./configure check
for zlib + add the library to IRCDLIBS in Makefile if enabled.
TODO: some little code cleanups
then does a -x virthost gets freed and the user can join again/circumvent the ban.
virhost should probably be moved in to the User struct like char virthost[HOSTLEN+1]
but in the meantime I've just removed the free() when someone does -x.
I noticed we already do make_userhost at connect even if mode +x isn't set, and also
is_banned checks if virthost exists.. if so -> check if that's banned... so this
patch won't decrease performance much.
We were doing: T_AAAA, if that fails then T_A. But in that case when a host
has both T_A and T_AAAA entries, the reverse lookup will fail for ipv4.
Now using T_ANY to fetch both T_A and T_AAAA addresses at once.
We were doing "deny known bad characters" instead of "only allow known good characters", this was REALLY bad...
This patch limits hostnames to alphanumeric, '-', '_' and '.'.
for (i = 0; &HE(cp)->h_addr_list[i]; i++) was used in ipv4 mode, but should be
without the &... this caused some stalls (like 11 seconds) at my machine because the
loop was executing >1 million times (I don't understand why it didnt crash however).
Bram Matthys