DragonFly BSD
DragonFly kernel List (threaded) for 2004-07
[Date Prev][Date Next]  [Thread Prev][Thread Next]  [Date Index][Thread Index]

(followup): ANSI-fy of ranlib, ruptime and rdist [patches]


From: Matthew Dillon <dillon@xxxxxxxxxxxxxxxxxxxx>
Date: Fri, 23 Jul 2004 15:16:44 -0700 (PDT)

    I didn't post the complete list!  Here is a summary of all of
    Douwe's email to date.  His patches are at:

    http://virtus.ath.cx/dfbsd

    Again, I'm looking for a committer who can start/continue running
    through these for review, feedback, and commit.

    Followups To: submit@

						-Matt

:Hello,
:
:/usr/bin/who performs no error checking on the stat() call, while
:it should error out when a utmp file can not be accessed. Here is
:a simple patch:
:
:http://virtus.ath.cx/dfbsd/who.c.diff
:
:Regards,
:Douwe


:Hello,
:
:/usr/bin/who performs no error checking on the stat() call, while
:it should error out when a utmp file can not be accessed. Here is
:a simple patch:
:
:http://virtus.ath.cx/dfbsd/who.c.diff
:
:Regards,
:Douwe

:Hello,
:
:the DragonFlyBSD source code contains several typographical errors,
:this specific patch is directed at the typing of 'sucess', while it is
:ofcourse
:'success'. The patch can be found here:
:
:http://virtus.ath.cx/dfbsd/success-typo.diff
:
:Or the patches seperate per file can be found in the 'old' directory
:on http://virtus.ath.cx/dfbsd
:
:Regards,
:Douwe

:Hello,
:
:/usr/bin/join misses the -j option in its usage() function, this should be
:added.
:A patch can be found here:
:
:http://virtus.ath.cx/dfbsd/join.c.diff
:
:Regards,
:Douwe

:Hello,
:
:/usr/bin/fetch lacks error checking in several places, it is especially
:important in
:the case of calling mkstemps() and fopen(), where it is not sufficiently
:checked.
:A patch can be found here:
:
:http://virtus.ath.cx/dfbsd/fetch.c.diff
:
:Regards,
:Douwe

:Hello,
:
:/usr/bin/dirname uses getopt(), but doesn't do anything with it, thus it can
:be removed.
:A patch is to be found here:
:
:http://virtus.ath.cx/dfbsd/dirname.c.diff
:
:Regards,
:Douwe

:Hey,
:
:/bin/kill's usage() function states the usage of -signalname
:and -signalnumber
:as if they are needed variables and not options, this is wrong and should be
:changed
:so that these optiones are within square brackets [ and ]. A patch can be
:found
:here:
:
:http://virtus.ath.cx/dfbsd/kill.c.diff
:
:Regards,
:Douwe

:Hey peeps,
:
:According to SUSv3, the -p option should behave exactly the same as rmdir
:without any option, only that it recursively removes all directories
:mentioned
:as arguments in the command line. In the current rmdir, you can't remove
:multiple directories and meanwhile recursively remove all it's
:subdirectories.
:
:This doesn't work:
:
:daedalus# rmdir -p ./a ./z
:rmdir: ./a: Directory not empty
:rmdir: ./z: Directory not empty
:
:while these directories are completely empty, except for other empty
:subdirectories.
:SUSv3 states, aswell as the manual page:
:
:-p
:Remove all directories in a pathname
:
:This is due to an error in the rmdir source code in the loop that handles
:command
:line arguments. A patch can be found here:
:
:http://virtus.ath.cx/dfbsd/rmdir.c.diff
:
:Regards,
:Douwe

:Hey people,
:I found a small error in share/examples/cvsup; the DragonFly-* files contain
:a line saying how the execute cvsup, I am used to just copying that line
:after
:I've edited it, pasting it to command-line, and pressing enter. I now got an
:error,
:so I guess these files should be edited. Atleast change it in the
:DragonFly-* files,
:we don't really have to care about the other files.. do we?
:
:Here is a small patch:
:
:http://virtus.ath.cx/dfbsd/supfile.diff
:
:Regards,
:Douwe

:Hey guys,
:
:bin/sh/mksyntax.c contains call to the perror() function, while
:sh has its own error function, that is used everywhere in the code,
:and is define in error.c, namely error(). For code consistency mksyntax.c
:should also be made to use this function, here is a patch:
:
:http://virtus.ath.cx/dfbsd/mksyntax.c.diff
:
:Regards,
:Douwe

:Hello again,
:
:usr.bin/make uses the perror() error reporting function, while this should
:ofcourse be something
:of the warn() and/or err() family. A patch can be found here, with the
:perror() function changed
:to what they have been changed into on FreeBSD-CURRENT:
:
:http://virtus.ath.cx/dfbsd/make.diff

:Hey,
:I was browsing around the source code and discovered that a lot of stuff is
:still
:not ANSI. I was bored, so I decided to patch it, here are the patches:
:
:http://virtus.ath.cx/dfbsd/ruptime.c.diff
:http://virtus.ath.cx/dfbsd/ranlib.diff
:http://virtus.ath.cx/dfbsd/rdist.diff
:
:While ANSI-fying rdist, I discovered some possible security hazards, which
:I have decided to fix aswell, these can be found here:
:
:http://virtus.ath.cx/dfbsd/rdist2.diff
:
:This depends on rdist.diff, which has to be applied first and then
:rdist2.diff over it.
:
:Kind regards,
:Douwe




[Date Prev][Date Next]  [Thread Prev][Thread Next]  [Date Index][Thread Index]