- The scheduler's is_path_absolute() code could cause a DoS (STR #1042)
- The scheduler's device loading code used the wrong size limits for the
make/model and info parameters (STR #1035)
- The PNG loading code did not use a "long unsigned integer" format
specifier for the width and height (STR #1032)
- The web interface only showed the first 4 or 8 characters of
"{variable-name}" for undefined template variables (STR #1031)
- The hpgltops filter did not handle a common PCL command to enter
HP-GL/2 mode (STR #1037)
- The scheduler no longer sends the page-set option when printing banner
pages (STR #995)
- The hpgltops filter contained two buffer overflows that could
potentially allow remote access to the "lp" account (STR #1024)
- The lppasswd command did not protect against file descriptor or ulimit
attacks (STR #1023)
- The "lpc status" command used the wrong resource path when querying
the list of printers and jobs, causing unnecessary authentication
requests (STR #1018)
- The httpWait() function did not handle signal interruptions (STR #1020)
- The USB backend used the wrong size status variable when checking the
printer status (STR #1017)
- The scheduler did not delete classes from other classes or implicit
classes, which could cause a crash (STR #1015)
- The IPP backend now logs the remote print job ID at log level NOTICE
instead of INFO (so it shows up in the error_log file...)
3dd399c2ae