6074016782
fix address size check
...
The rad pointer is an in_addr structure (IA), not a sockaddr structure
(SA), so make sure the memcmp call restricts itself to those many bytes.
2018-01-11 17:13:14 -05:00
98c8f686aa
fix a few harmless compiler warnings
2018-01-11 03:50:29 -05:00
fce106d40c
ignore generated files
2018-01-11 03:39:13 -05:00
e0fbe7aea4
add netcat manpage
2018-01-11 03:38:32 -05:00
af2ec70280
support named ports with dashes in them
...
Normally netcat treats dashes as port ranges (e.g. "ssh-http"), but
there are port names with dashes in them (e.g. "ftp-data"). Allow
users to escape the dashes to support the latter mode.
2018-01-11 03:24:37 -05:00
3807536449
add IPv6 patch
2018-01-10 23:57:00 -05:00
9274c7d458
add a -q option for controlling automatic exit behavior
2018-01-11 03:44:26 +00:00
3c58093fdc
add a -b option to allow sending to broadcast addresses
2018-01-11 03:38:27 +00:00
5236e1513f
change inet_addr to inet_aton
...
The former cannot differentiate between errors (-1) and 255.255.255.255.
Switch to inet_aton which does not have this problem.
2018-01-11 03:32:05 +00:00
c037092c97
drop "punt" message when using CTRL+C
...
Most distros have patched this out, so might as well do it here too.
2018-01-11 03:20:55 +00:00
84f88ce209
drop default stripping
...
If people want to strip this, they can do so via `strip`.
2018-01-11 03:14:41 +00:00
74922ad47b
Fix the size of a strncpy
...
It seems that the intent of this code was to only use the first char[24]
in `addrs` for this strncpy. While that should always happen anyway
(inet_ntoa should never produce a >= 24 char string), the mismatch
between the size of poop->addrs[0] and `sizeof(poop->addrs)` caused
issues with FORTIFY on clang.
2018-01-05 23:12:30 +00:00
31f56e021f
fix crashes with large port numbers
...
the anum field isn't used anywhere, so punt it
http://bugs.gentoo.org/370827
2011-06-12 01:42:11 +00:00
fcaf538746
add an -c option which is like -e but executes through a shell
2009-03-06 02:51:18 +00:00
bda7ada5a1
do not make the -n flag affect service name lookups
2009-03-06 02:45:46 +00:00
417a3d26a0
add an option to use sig{set,long}jmp and default it to on for linux systems
2009-03-06 02:41:40 +00:00
e670149ab9
rename the arm() func to arm_timer() as gcc arm compilers define "arm" by default
2009-03-06 02:35:22 +00:00
5d9df913c4
define HAVE_SOCKLEN_T for linux systems
2009-03-06 02:32:41 +00:00
36d74baf20
use __linux__ rather than LINUX as toolchains set that up automatically
2009-03-06 02:13:34 +00:00
b9e72ce5af
dump usage() info to stdout when asking for help
2009-03-06 02:11:45 +00:00
69a0d44d75
make sure the buffer holding the command line is always NUL terminated
2009-03-06 02:10:10 +00:00
ebf4d38912
dont delay with sleep() before calling exit()
2009-03-06 01:55:09 +00:00
9059d5da39
mmm, actually allow HAVE_BIND on all systems and use resolv.h on linux systems
2009-03-06 01:51:34 +00:00
e89203975b
declare proper return types for funcs and move helpme() up before its first usage
2009-03-06 01:46:51 +00:00
b7c6c3c10e
use socklen_t with network functions
2009-03-06 01:42:41 +00:00
89659d87d9
include unistd.h for some standard file related prototypes
2009-03-06 01:35:04 +00:00
384a321928
include time.h by default for time() prototype
2009-03-06 01:32:40 +00:00
d0ec781df1
do not define HAVE_BIND by default -- let builders do it
2009-03-06 01:28:37 +00:00
3d5a4955ad
remove embedded /* in comments
2009-03-06 01:21:49 +00:00
7ed164285c
document our role
2007-01-18 04:06:44 +00:00
5086ba1ade
*Hobbit*'s netcat 1.10
2007-01-17 09:51:02 +00:00
Mike Frysinger
George Burgess IV
Anatoly Techtonik