summaryrefslogtreecommitdiff
path: root/usr.bin/file
AgeCommit message (Collapse)Author
2019-11-30After fork(2) the pledge(2) in the parent proc can be reduced toRicardo Mestre
"stdio rpath sendfd" so that it can call {l,}stat/open and sendfd for imsg_* in order to send fds to the child proc which is already pledged by recvfd to receive them OK brynet@ deraadt@
2019-06-28asprintf returns -1, not an arbitrary value < 0. Also upon error theTheo de Raadt
(very sloppy specification) leaves an undefined value in *ret, so it is wrong to inspect it, the error condition is enough. discussed a little with nicm, and then much more with millert until we were exasperated
2019-02-05dev_t is signed to permit passing -1 as an invalid condition, but theTheo de Raadt
decomposition into major and minor is unsigned, so we should print them with %u instead of %d. ok guenther
2019-01-15Regex flags should include REG_NEWLINE and REG_ICASE should actually beNicholas Marriott
used if specified.
2019-01-06the parent process doesn't do much but open files, but just the sameTed Unangst
there's no reason it can't use pledge to enforce that. ok brynet deraadt
2018-10-04Update qcow file definitionsccardenas
Taken from netbsd "go for it" deraadt@
2018-06-14Return from magic_test does not need to be copied because the callerNicholas Marriott
will do so anyway and it was just being leaked. ok brynet deraadt
2018-01-15Revert my changes in r1.60 back to nicm@'s latest for now.Bryan Steele
Fixes "file *|grep" breakage reported by espie@ ok nicm, deraadt
2017-11-30Default unknowns to application/octet-stream instead of x-not-regular-file.Anthony J. Bentley
This makes more sense and matches what the latest "other" file(1) now does. ok nicm@
2017-09-20add missing types; from klemens nanniJason McIntyre
ok nicm
2017-07-02Replace fgetln with POSIX getline, even though file was using fgetln inBryan Steele
the idiomatic safe way. Also, check for stream errors with ferror and err(1, "") in that case. with feedback from joerg@, schwarze@, nicm@ ok nicm "code is shorter anyway", and joerg@ for an earlier diff
2017-07-01close open fd after testing files, ensuring that fd isn't stdin.Bryan Steele
ok nicm@
2017-07-01magicfp is fclosed where it was opened, remove fclose from magic_load()Bryan Steele
ok nicm deraadt
2017-06-28Fix stdin file read support, accidentally broken in my last commit.Bryan Steele
2017-06-28remove excess #includeTheo de Raadt
2017-06-28perform an initial pledge very early on, and drop tzset to later.Theo de Raadt
ok nicm brynet
2017-06-28Simplify file(1) by removing the no longer necessary parent/child separationBryan Steele
and just drop privileges in the main process. Also allows for a tighter "stdio" pledge. passing regress tests still pass ok nicm@ with helpful feedback
2017-04-18Style nits; no binary change.Nicholas Marriott
2017-02-19Recognise AArch64 binaries and core dumps.Jonathan Gray
2016-08-26Pull in <time.h> for ctime_r, gmtime, etcPhilip Guenther
ok deraadt@
2016-06-12Remove old unused prototype, from Ilya Kaliman.Nicholas Marriott
2016-05-01Tidy up some #include lines.Nicholas Marriott
2016-05-01Exact match use and name with strcmp rather than prefix with strncmp.Nicholas Marriott
2016-05-01Use the right size (include the length byte) when working out if aNicholas Marriott
pstring is too big.
2016-05-01Add support for 'name' and 'use' which allows more of the latest magicNicholas Marriott
files to work unchanged. (We are still missing 'indirect' and a few other bits.)
2016-05-01Trim = prefix from regex (it a noop).Nicholas Marriott
2016-05-01Remove __unused that are now lies.Nicholas Marriott
2016-05-01Add support for 'clear' test, and fix 'default' to expand the resultNicholas Marriott
string if any (used by, for example, rtf).
2016-04-30magic_test_type_default parameter is no longer __unused.Nicholas Marriott
2016-04-30Fix the default type to work properly, that is to only match if noNicholas Marriott
previous test at the same level has matched.
2016-04-30Replace regex match string in results (some new magic entries use this).Nicholas Marriott
2016-04-28Partially sync magdir/sniffer from the file distribution on astron.com,Stuart Henderson
adds pcap-ng files and some more link-layer types for pcap. Not an exact merge because they're using "use" which we don't currently support. ok nicm@ jasper@
2016-04-24typos;Jason McIntyre
2016-03-06Sync 'archive' magic from file 5.25. Various improvements includingStuart Henderson
epub detection which jirib at devio dot us reported as not working. nicm says "The only thing I see that we do not support is the indirect on the last line, I suggest just commenting it" (done).
2016-03-04fix broken listDmitrij Czarkoff
2016-03-04fix typoDmitrij Czarkoff
2016-02-20Fix the C type used for 32-bit float/lefloat/befloat, it should be floatNicholas Marriott
not double (they are not used in any magic files anyway at the moment though). Spotted by and ok jsg.
2016-01-29Therefor -> Therefore (where appropriate)tb
from ray@, ok jmc@
2016-01-02Sync magdir/animation with file 5.25, amongst other updates this fixes anStuart Henderson
incorrect detection of mp3 sample frequency/audio channels with non-mp3 files tracked down by Brandon Vincent. ok nicm@
2015-12-24Add --brief and --dereference, used by xdg-open/xdg-mime.Jeremie Courreges-Anglas
From Ralf Horstmann, ok tb@ deraadt@ nicm@ sthen@
2015-12-05EAGAIN handling for imsg_read. OK henning@ benno@Claudio Jeker
2015-11-17Merge xmalloc.[ch] files across base, skipping OpenSSH for now.Tobias Stoeckmann
ok nicm
2015-11-15When validating a char by calling strchr() with a string of allowed chars,Tobias Stoeckmann
check for '\0' first, because strchr() would return non-NULL. ok nicm
2015-11-13Break the message preparation bit of the main loop into its own functionNicholas Marriott
for less excessive level of indentation.
2015-11-13Call stat not lstat with -L, makes links actually be followed. ReportedNicholas Marriott
by and ok semarie@.
2015-11-05palmpilot should have gone when palm was added, it is redundant.Nicholas Marriott
2015-10-31Update Matroska magic file.Nicholas Marriott
2015-10-17The file(1) magic-parsing process was using pledge "stdio getpw proc recvfd"Theo de Raadt
early on, then a set of getpwnam/setresuid/... before quickly dropping to "stdio recvfd". It receives fd's and runs the magic code on them in a chroot'd "stdio" jail. We can do better than that. Before the recent change, "proc" contained both the concepts of "forking" and "setuid". "id" is now split out as a seperate request, and it is exactly what this process needs momentarily. So this loses another window of opportunity, in case we have a major bug in .... hmm, it'd have to be in getpwnam.... ok tedu doug semarie gilles
2015-10-09Change all tame callers to namechange to pledge(2).Theo de Raadt
2015-10-06Move from tame "cmsg" to tame "sendfd" or "recvfd", depending on whichTheo de Raadt
way the process moves fd's.