One patch for using random instead of rand. [1] Reordered headers and removed old comments (what is ckt_write?). umask is actually in sys/stat.h and not in unistd.h Also removed expired TODO item about dpi_service.[c|h]. [1] https://www.securecoding.cert.org/confluence/display/seccode/MSC30-C.+Do+not...
On Sat, May 26, 2012 at 11:05:28PM +0400, 123 wrote:
One patch for using random instead of rand. [1]
Ack.
Reordered headers and removed old comments (what is ckt_write?). umask is actually in sys/stat.h and not in unistd.h
Dillo compiles in several OSes/platforms, which sometimes have declarations in different files. This is the to-be-tested point.
(what is ckt_write?).
ckd_write() an EINTR-protected write() (see dpid/dpid_common.c).
Also removed expired TODO item about dpi_service.[c|h].
Ack. -- Cheers Jorge.-
On Sat, May 26, 2012 at 11:05:28PM +0400, 123 wrote:
One patch for using random instead of rand. [1]
Guessing the intention here, I would recomment to use arc4random on systems that have it. time(0) should likely be spelled as time(NULL).
Reordered headers and removed old comments (what is ckt_write?). umask is actually in sys/stat.h and not in unistd.h
umask in sys/stat.h is correct. Joerg
participants (3)
-
jcid＠dillo.org -
joerg.sonnenberger＠web.de -
p37sitdu＠lavabit.com