25d9c5a6c7
add helpful #define HAVE_EAGAIN_READWRITE
11 years ago
39e7ee90bb
introduce io_eagain_read and io_eagain_write (discontinue using io_eagain plz)
11 years ago
d9cbb3940c
it turns out you don't need -1/EGAIN to know edge triggering told you
...
there is no more data, it's enough to read/write less than you asked for
11 years ago
f3c0e69988
fix and document iarray
11 years ago
5baf943f0a
finish epoll move to edge triggering
11 years ago
0f6e586fc9
fix some bugs that cppcheck found
11 years ago
6d18d4ecc3
better comment
11 years ago
4ecc33dcea
catch not enough bytes in input buffer case
11 years ago
8106f8c5a6
remove superfluous &0x7f
11 years ago
8581b74a84
on FreeBSD, on a PF_INET6 socket, recvfrom and friends can actually
...
return a sockaddr with family PF_INET. WTF? Work around that.
11 years ago
63b08692d0
check value to large error for asn1derlength
11 years ago
cf1406735c
much shorter way to do the same thing
11 years ago
708c358a74
for asn1derlength: save a few bytes
...
for asn1dertag: catch too-large-value overflow
11 years ago
24d1ccb1b7
reject non-minimally encoded ASN.1 DER data
...
split scan_asn1derlength into scan_asn1derlength and scan_asn1derlengthvalue
11 years ago
f4414a2e80
remove io_starteventloop
11 years ago
b6af872d43
update Makefile
11 years ago
321b30565d
change return type to void
11 years ago
90acf965e7
add iarray_length
11 years ago
a544abc39c
switch io_fds from array to newly implemented (hopefully thread-safe) iarray
...
switch epoll from level triggering to edge triggering
11 years ago
dd436c5bf8
map mmap_read return a const char*
11 years ago
c548cce5a1
another attempt at a thread-safe array
11 years ago
c608502cd8
add scan_ulongn, scan_8longn, scan_xlongn, scan_longn, scan_netstring
11 years ago
74d676fc62
fix typo in man page
...
make use of __uint128_t in scan_ulongn and provide long long based
implementation for 32-bit platforms
11 years ago
9b6671efee
add buffer_tosa check to marshal.c
...
failed attempt to improve scan_ulongn, results in essentially same code
but the C code looks slighly better so I'm keeping it
11 years ago
cf3dc34288
actually increase stralloc.len
11 years ago
53e2af7389
never close fd 0, 1 or 2 in buffer_close
11 years ago
85d14a2703
ignore entities.h
11 years ago
9af06ac843
remove can't happen code and instead force a compile time error
11 years ago
baec005507
improved code path for typical platforms where there is a wider integer type
11 years ago
08ba483bfb
catch other potential overflow
11 years ago
4a04c40595
get rid of more -Wconversion warnings
11 years ago
f7fee036c1
SECURITY: check for integer overflow in stralloc_ready
11 years ago
539ef564ab
add unit tests for uint*
11 years ago
a9ca21ffc9
remove compiler warnings
11 years ago
324c8e0e8b
add rules regarding ent
11 years ago
d14d2536f0
get rid of -Wconversion warnings
11 years ago
e1081e4b13
remove -Wconversion warnings
11 years ago
18afec7d68
scan_*longlong now properly abort if the number would not fit
11 years ago
9a1addce2f
remove compiler warnings
11 years ago
01ffc04006
do proper early abort in the other scan_* routines if the numeric value
...
is too large
11 years ago
35942878c2
$ make WERROR=-Werror now builds with -Werror
...
add some single char escaping routines to fmt.h
pull in html5 entities from w3c and use those to do a proper scan_html decoding
fix an off-by-one in fmt_to_array
add a ton of unit tests for the fmt routines
11 years ago
dd4f86fbfa
add some unit tests for the formatting routines
11 years ago
73db80daec
get rid of some compiler warnings
11 years ago
35382a430e
if called with srclen > maxlen, return maxlen not srclen
11 years ago
21d0ea4549
if srclen>maxlen, truncate at maxlen
11 years ago
a85bfc1b81
fmt_strn aborted one byte too late
11 years ago
fb5bd841a8
| not +
11 years ago
1d2add208f
add json encoding routines to textcode
11 years ago
66cb1c1c82
fix dependencies
11 years ago
ff3a3410ec
hopefully this fixes the looping gatling processes
11 years ago
leitner