summaryrefslogtreecommitdiff
path: root/usr.bin
AgeCommit message (Collapse)Author
2009-03-27-S (safe mode) would copy the file over, rename it to the target, thenPeter Hessler
do the chown/chmod dance. This created a race where the new file would be in place, but with the incorrect permissions. Fix so the rename is the last thing, instead of a middle thing. looks ok to krw@, deraadt@ OK beck@
2009-03-27make sure that when we try to send a Set-sticky to the client,Joris Vink
we only do so if the file we're checking out was not supplied on the command-line. Otherwise we'll end up setting CVS/Tag for the entire directory instead of slapping a sticky tag only on the file supplied.
2009-03-26fix the way Attic is handled in our recursion code, so weJoris Vink
do not skip files or run over them twice. also fixes -l and -r for checkout/update when a file in Attic exists with that tag that in HEAD is a directory in the normal repository like gnu/usr.bin/gcc/INSTALL. as a bonus, we do not run fstat() twice per file or dir anymore... spotted by deraadt@
2009-03-26"Logging dirname" needs to be properly shown when running log/rlog.Joris Vink
spotted by deraadt@
2009-03-26fix a few typographical errors found by spell(1).Igor Sobrado
ok dtucker@, jmc@
2009-03-25switch our file and directory lists to RB trees (see tree(3)),Joris Vink
so we can benefit from faster lookup times while recursing.
2009-03-25prevent a very difficult to reach, but possible use after free.Joris Vink
ok tobias@
2009-03-25remove unused functions, definitions and outdated commentsJoris Vink
from the stone ages.
2009-03-25-q hasnt been the default verbosity mode in ages.Joris Vink
2009-03-24have CVS_OP_TAG and CVS_OP_RTAG use RCS_WRITE flagJoris Vink
when calling rcs_open(), so it gets automaticly synced at the end of operations for that file instead of explicitly calling rcs_write() when you're in RCS_READ mode only. while here, make sure we do not try to free an uninitialized list.
2009-03-24shuffle code a bit so FILE_REMOVE_ENTRY also callsJoris Vink
cvs_checkout_file() with CO_REMOVE set otherwise our server would never inform the client the entry should be removed.
2009-03-24use size_t for size, since atomicio() both accepts and returns itJoris Vink
2009-03-23My previous commit didn't fix the problem at all, so stick at my firstTobias Stoeckmann
version of the fix presented to dtucker. Issue notified by Matthias Barkhoff (matthias dot barkhoff at gmx dot de). ok dtucker
2009-03-23kill SIZE_LEFT() macro, no longer usedJoris Vink
2009-03-23simplifyJoris Vink
2009-03-23remove debug, blehJoris Vink
2009-03-23add rlog -d support, ported from openrcs byJoris Vink
Josh Elsasser <josh@elsasser.org>
2009-03-23Fixed a possible out-of-bounds memory access if the environment variableTobias Stoeckmann
SHELL is shorter than 3 characters. with input by and ok dtucker
2009-03-23LP_NOTICE should be LP_ERR when showing directory messages.Joris Vink
2009-03-23any directory messages are, believe it or not, suppose to go to LP_ERRJoris Vink
instead of LP_NOTICE, while here modify output for non-existing files to match "? foobar" instead of "cannot find foobar". makes our output a little less alien and a little bit more gnuish.
2009-03-22styleTobias Stoeckmann
2009-03-22Avoid the usage of uninitialized value, which even might result in a memoryTobias Stoeckmann
leak if a file has no line in CVS/Entries. ok joris
2009-03-22Properly free allocated memory in error case.Tobias Stoeckmann
ok joris
2009-03-21a paragraph break (.Pp macro) is not required before or after section headers.Igor Sobrado
2009-03-21use file_rcsrev when printing out Repository revision soJoris Vink
we always receive the correct revision number instead of HEAD, otherwise running status on a file that belongs to a branch results in showing HEAD revision instead of latest branch revision.
2009-03-21kill useless return;Joris Vink
2009-03-21be smarter when checking if a directory is empty by looking atJoris Vink
CVS/Entries early on in cvs_update_leavedir()
2009-03-19for "Ciphers", just point the reader to the keyword in ssh_config(5), justJason McIntyre
as we do for "MACs": this stops us getting out of sync when the lists change; fixes documentation/6102, submitted by Peter J. Philipp alternative fix proposed by djm ok markus
2009-03-19do not diff files server-side that are uptodate but require a patch,Joris Vink
as the client does not provide these files at all because they are not modified in anyway. otherwise this results in bad mojo. found by myself and sthen@
2009-03-19kill unused varsJoris Vink
2009-03-19only pass the real $var to expand_vars() instead of the entireJoris Vink
line, this way commitinfo and stuff works again. found by myself the hard way, fix by tobias@ who's not around to commit.
2009-03-19properly register Questionable files using CVS/Entries server-sideJoris Vink
in order for them to show up as "? foobar" when opencvs is in server mode.
2009-03-18be carefull when trying to access cf->file_rcs in cvs_checkout_file()Joris Vink
as it could be NULL under certain conditions.
2009-03-17pass correct argv[0] to openlog(); ok djm@Markus Friedl
2009-03-17- rename the arg to -b "nframes" and sync usage()Jason McIntyre
- grammar fix for the server section from Thomas Pfaff ok ratchov
2009-03-17DESCRIPTION should also note aucat's ability to act as a serverAlexandre Ratchov
from jmc@
2009-03-16improve spacing in synopsis.Igor Sobrado
2009-03-15fix a couple comment typosJacob Meuser
2009-03-12in full-duplex, don't send AMSG_MOVE messages twice; instead,Alexandre Ratchov
send only the record direction one. spotted by jakemsr
2009-03-09correct .Dt; from Tim van der MolenJason McIntyre
2009-03-05Call endusershell() at the end of ok_shell(), making a copy of theTodd C. Miller
(possibly) expanded shell as needed. OK deraadt@
2009-03-05document what these functions do so I don't ever have to recuse intoDamien Miller
b64_pton/ntop to remember their return values
2009-03-05refactor the (disabled) Schnorr proof code to make it a little moreDamien Miller
generally useful
2009-03-01Don't rely on non-standard setenv() behavior.Todd C. Miller
2009-02-27say that aucat can act as audio server in the NAME sectionAlexandre Ratchov
ok jmc
2009-02-25Checking if a file is ASCII should only be done when diffing, becauseRay Lai
the non-ASCII characters could be printed to the screen. For checking in files, checking out files, merging files, and removing revisions of files, we do so regardless of whether the files are ASCII or binary. Fixes PR6031. OK joris and deraadt.
2009-02-23unused varJoris Vink
2009-02-23we no longer push out Set-sticky to the client when writingJoris Vink
our Tag file on the server or in cvs_mkpath(), this caused problems for the "fast checkout" mechanism when a tag was supplied. instead, write out any directory sticky tag when we are checking out a file in that directory, this way we can use our "fast checkout" mechanism (no /tmp required) even for -r and -d without breaking working copies because they were missing CVS/Tag files. "looks good" tobias@
2009-02-23Properly convert between gmt and localtime for unchanged files as client.Tobias Stoeckmann
Problem reported by Jesus Sanchez (zexel08 at gmail dot com). ok joris
2009-02-23matthew haub pointed out that pid may be used uninitialized in some contextsTed Unangst
ok deraadt