aboutsummaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* Added spb's bsd chflags support.Jason Stubbs2005-02-264-0/+201
|
* updated filter-env for another type of parsing.Brian Harring2005-02-091-6/+10
| | | | offhand, walk_command probably needs a cleanup for DOLLARED.
* change to NO_PARSING mode. allows it to handle escaped chars.Brian Harring2005-02-091-3/+3
|
* main:char *end -> main:const char *endBrian Harring2004-11-231-1/+2
|
* added a few asserts.Brian Harring2004-11-231-2/+4
|
* la la la. stupid typos suck.Brian Harring2004-11-231-9/+30
|
* I hate realloc, and I am starting to dislike reading from pipes.Brian Harring2004-11-221-10/+11
|
* hopefully a final fix for the "lets ignore part of the dumped env" bug.Brian Harring2004-11-221-24/+24
|
* null termination bug.Brian Harring2004-11-201-1/+1
|
* Added qualifiers to bmh_search parameters and return type. SlightlyJason Stubbs2004-11-203-7/+3
| | | | modified parameter type on main() to get rid of a warning.
* dumb ass realloc bug.Brian Harring2004-11-201-1/+1
| | | | I hate realloc.
* close out a leak.Brian Harring2004-11-201-0/+1
|
* autotooling fun.Brian Harring2004-11-202-0/+41
|
* oops. commited wrong base file. :)Brian Harring2004-11-202-401/+3
|
* hellllllllooooooo filter-env.Brian Harring2004-11-204-0/+948
| | | | | | | | | | | | | | | | | | | simple binary to walk the structure of a bash env/prog, and optionally mask portions of the it via specified opts fex, to read from stdin and outputing to stdout, and filtering func1,func2 and var1 filter-env -f func1,func2 -v var1 alternativelly, from file, and getting var1,var2 filter-env -i file -v var1,var2 etc. Always outputs to stdout, although that might change. Tested it against a full vdb environment dump (all files)- 100+ envs merged, walked it fine. That said, if you hit a bug, it's quite likely a HERE related structure- cat <<BLAH asdf asdf BLAH fex. If you hit oddities, use -d to turn on debugging info. currently 2 levels of it- first is func matching and env matching, second level enable debug info for HERE related structure matching. No deps aside from regex supplied by glibc.
* sandbox is now in gentoo-src/sandbox.Brian Harring2004-11-2011-3987/+0
| | | | Hopefully this doesn't hose up the stable portage branch. if it does, these files will be restored.
* files are dead, no longer used. killing 'em.Brian Harring2004-11-2010-2907/+0
|
* bug #70225, sandbox_pids_file potential overflow.Brian Harring2004-11-153-22/+20
|
* bye bye unused sandbox directory...Brian Harring2004-11-159-2006/+0
|
* bug #70351, compilation failure.Brian Harring2004-11-072-4/+9
|
* doins: Added vapier's fix for doins changing existing directoryJason Stubbs2004-11-051-1/+2
| | | | | | permissions. #69896 tbz2tool.c: Added vapier's fix for tbz2tool on ia64. #70001
* futils fix from bug #65201 via solar, and libsandbox log path checks via #69137Brian Harring2004-11-033-35/+42
|
* Lots of stuff. Digest handling enhancements. Pickle speedups. Sandbox fixes ↵Nicholas Jones2004-10-192-16/+16
| | | | for ia64/security cleanup. Revertion of the lock code to _rc7 + touchups.
* _rc8, basically.Nicholas Jones2004-10-103-136/+86
| | | | | Lots of little cleanups. Solar's patch for sandbox. Spanky's patches on various outstanding bugs.
* copyright updatesMike Frysinger2004-10-042-4/+4
|
* GPG work, CONF_LIBDIR, and stuff. Portage-2.0.51_pre18 finializationNicholas Jones2004-08-161-1/+15
|
* Cleanups, typo fixes, extra info output, binary altroot fix, bad mirrorNicholas Jones2004-08-023-37/+42
| | | | message. portage-2.0.51_pre14
* Added USE flag based PROVIDE support. Fixed lchown sandbox bug. (#58084)Jason Stubbs2004-07-241-4/+3
|
* Lots of little updates and fixes. See the changelog...Nicholas Jones2004-04-141-2/+3
| | | | portage-2.0.51_pre3
* free() fix for env in execve call -- jstubbsNicholas Jones2004-04-122-4/+10
|
* mkdir fix from jstubbs, little touches.Nicholas Jones2004-04-113-7/+12
|
* Cleanup of the security issues from bug 21923.Nicholas Jones2004-04-043-6/+123
| | | | | file_security_check() is now referenced from file_open to ensure that the file being operated on is sane.
* Portage-2.0.50-r2portage_2_0_50_r2Nicholas Jones2004-03-225-18/+51
| | | | Laundry list of fixes and updates. See the changelog.
* Fix permissions and group of pids file and logs. Permissions should be 0664Martin Schlemmer2004-02-293-15/+35
| | | | and group should be 'portage'. Bug #34260.
* Besides a small cleanup, redo how we replace LD_PRELOAD in the environ passedMartin Schlemmer2004-02-282-30/+53
| | | | | | to the real execve (in our execve wrapper). Seems that on some arches (sparc among others) do not allow us to tamper with the readonly copy passed to execve, so pass our own copy of the environment. Bug #42290.
* Lots of modifications... Mostly cleanupsportage_2_0_50_pre22Nicholas Jones2004-02-041-3/+3
| | | | 2.0.50_pre22
* portage-2.0.50_pre20portage_2_0_50_pre20Nicholas Jones2004-01-241-1/+9
|
* Quick fixes for trivial problems. _pre14Nicholas Jones2004-01-121-6/+6
|
* Warning in configs. DB operation reorganization, counter fix, sandbox ↵Nicholas Jones2004-01-122-5/+11
| | | | | | | speedup for the glibc finding stuff. 0 . 0 \_U_/
* Slightly better check for glibc version that doesn't freak out onNicholas Jones2003-12-312-3/+7
| | | | a static /bin/sh.
* Lots of stuff. -49-r17 candidate.portage_2_0_49_r17_pre1Nicholas Jones2003-11-101-2/+2
| | | | 49-r61 tarball is available for this set.
* If 'file' passed to before_syscall(const char *func, const char *file) isMartin Schlemmer2003-11-022-4/+10
| | | | | invalid, we should set errno to ENOENT, and not EINVAL. This should close bug #32238.
* Fix a bug that occurs mainly on 64bit arch, where the file passed toMartin Schlemmer2003-10-142-2/+15
| | | | | | the functions we wrap, is invalid, and then cause canonicalize to pass garbage to before_syscall(), thanks to great detective work from Andrea Luzzardi <al@sig11.org>.
* Add a uClibc detection patch from Peter S. Mazinger <ps.m@gmx.net>.Martin Schlemmer2003-10-132-3/+6
|
* Fix a bug in libsandbox.c 's checking in the rename wrapper - it basicallyMartin Schlemmer2003-10-132-7/+13
| | | | | | only checked the destination patch, and not the source, so we could move a protected file to a unprotected directory, and then delete/modify it. Thanks to Andrea Luzzardi (scox) <al@sig11.org>, bug #30992, for this fix.
* Added comment about adding python2.3 to sandbox predicts.Nicholas Jones2003-10-121-1/+4
|
* Added python2.3 to the predict section.Nicholas Jones2003-10-121-1/+3
|
* Add support to set the pids file via SANDBOX_PIDS_FILEMartin Schlemmer2003-09-285-14/+51
|
* Fix our mkdir wrapper to check if the dir exist, and return EEXIST if so,Martin Schlemmer2003-09-272-5/+18
| | | | rather than failing with a violation, bug #29748.
* ignore empty path argument to canonicalizeMartin Schlemmer2003-07-272-5/+5
|