Re: `install -d -m MODE dir' doesn't honor MODE [Re: Bug#37150...

Jim Meyering <[EMAIL PROTECTED]> writes: > Or maybe the documentation should simply admit that when using -d, certain > directory attributes (e.g., for ACLs, too) may be inherited. One argument > in favor of this approach is that Solaris 9's /usr/sbin/install also > works this way. But when I tr

Re: on "dd" using "seek" vs. "read" & >32-bit sizes

Linda Walsh <[EMAIL PROTECTED]> writes: > I'll have to recheck this with more current versions. > It wasn't that "dd" didn't copy >32-bit files, I just had > problems with specifying offsets >32bits. My goodness you go a way back. The bug you mention was fixed a bit later, on 1997-11-29 accordin

Re: proposed tweaks for openat in coreutils

Jim Meyering <[EMAIL PROTECTED]> writes: > please do at least simulate it once, e.g., on Linux by tweaking > lib/openat.c so that its open always fails and it falls through to > the save_cwd/restore_cwd code. Then, run rm from a write-only > directory (forcing save_cwd to use getcwd) with two rel

Re: "sleep" command generates hardisk activity

Jochen Baier <[EMAIL PROTECTED]> writes: > nanosleep({3, 0}, NULL) = 0 > even a self made program with uses the "sleep() or usleep()" C > function show the same behavier. That's strange, but it indicates that the problem is not with coreutils or with the C library, but with your

Re: Bug#340236: coreutils: cp -p to nfs destination fails to preserve timestamps

The Wanderer <[EMAIL PROTECTED]> writes in : > I copied a directory hierarchy across a network to a shellfs mount, What's a shellfs mount? Sorry, I've never heard of shellfs. > When they arrived at the far end, all of the on

Re: Bug#340236: coreutils: cp -p to nfs destination fails to preserve timestamps

Mikko Rasa <[EMAIL PROTECTED]> writes: > 1. cp executes utimes before closing the file. This might be a problem, > although it seems to work on local destinations. I don't observe the problem on my Debian host, when running coreutils 5.93 (which I compiled). This is with Linux kernel 2.4.18-686

Re: bug in chdir-safer

2005-12-25 Paul Eggert <[EMAIL PROTECTED]> * chdir-safer.h (FCHMOD_SAFER_H): Remove: it was misnamed, and wasn't needed anyay. * chdir-safer.c (chdir_no_follow): Don't include stdio.h, assert.h, fcntl--.h; not needed. (O_DIRECTORY): D

Re: bug in chdir-safer

Jim Meyering <[EMAIL PROTECTED]> writes: > I recognize the value of the added open-with-O_WRONLY, > but note that it's wasted on modern Linux systems. Come to think of it, it's wasted on any system that conforms to POSIX, since POSIX requires that open-with-write-access must fail on directories.

Re: bug in chdir-safer

Paul Eggert <[EMAIL PROTECTED]> writes: > I don't think it's worth worrying about non-POSIX systems, so I'll > prepare a patch that undoes that stuff (both here and elsewhere). > There are some related things that need normalizing, too, e.g., > whether to

some other problems with chmod-safer.c, chown.c, etc.

I noticed a problem with chown.c (when CHOWN_MODIFIES_SYMLINK) and chmod-safer.c. If the file being chmod'ed or chown'ed is a special file, opening it can cause undesirable side effects. For example, it might cause a tape drive to be rewound. Also, the current implementations of mkdir, mkfifo, a

Re: some other problems with chmod-safer.c, chown.c, etc.

Jim Meyering <[EMAIL PROTECTED]> writes: > As for the use of chmod_safer in mknod.c, bearing in mind that it is > called only for a device that has just been created, does it matter > if there might be such a side effect? Yes, I think it matters. Someone may be using mknod, for example, to creat

Re: Use of bison

Eric Blake <[EMAIL PROTECTED]> writes: > Maybe bison.m4 should be updated to better check for a new enough > bison That might be nice but it's not essential, since the problem is detected at build time anyway. ___ Bug-coreutils mailing list Bug-coreut

Re: sort -b also ignores tabs

Thanks, I installed this: 2005-12-29 Paul Eggert <[EMAIL PROTECTED]> * doc/coreutils.texi (sort invocation): Clarify that a blank is a space or a tab. --- doc/coreutils.texi 17 Dec 2005 10:50:00 - 1.302 +++ doc/coreutils.texi 29 Dec 2005 21:45:32 -

Re: some other problems with chmod-safer.c, chown.c, etc.

Jim Meyering <[EMAIL PROTECTED]> writes: > Paul Eggert <[EMAIL PROTECTED]> wrote: >> For mkdir, mknod, and mkfifo, how about this idea instead? If -m is >> used, use only the umask to set the file permission bits; do not use >> chmod (or fchmod) at all.

Re: some other problems with chmod-safer.c, chown.c, etc.

ular files, falling back on lchmod for everything else; but this still suffers from the same race conditions that lchmod suffers from, and I'm not sure it's worth the hassle. 2006-01-01 Paul Eggert <[EMAIL PROTECTED]> * NEWS: Document that mkfifo and mknod -m

chown regression from coreutils 5.2.1 for execute-only files

The chown problem I mentioned in the last paragraph of still remains. My initial reaction is to do something for chown that is similar to what I just installed fro chmod; this would prefer lchown to chown. lchown is not quite

Re: env (GNU coreutils) 5.93 patch

[EMAIL PROTECTED] (Bob Proulx) writes: > On what systems is env located in /bin/env? The normal location is in > /usr/bin/env. POSIX doesn't specify the location for env, so either location conforms to POSIX. On Solaris 10, the standard location is /usr/xpg4/bin/env. (/usr/bin/env doesn't confo

Re: problem with split

John Joseph Bachir <[EMAIL PROTECTED]> writes: > split -b4700m myFile.tar.gz myFile.tar.gz_ ; > > Which should be about DVD sized chunks. But the files come out a > little over 600 megs each. Perhaps you're running an old version of split? "split --version" should tell you. If it's before versi

Re: date bug

case). You probably want %Y instead of %G. I installed this patch to the documentation to try to clarify things a bit better. 2006-01-02 Paul Eggert <[EMAIL PROTECTED]> * doc/coreutils.texi: Use @acronym around "ISO" uniformly. (Date conversion specifiers):

Re: date -d and the leapsecond

onds. On hosts that don't one could argue either way, but at least the behavior should be documented. The other point you mentioned I think has already been addressed in CVS coreutils. Thanks for mentioning the problems. I installed this patch. 2006-01-02 Paul Eggert <[EMAIL PROTEC

Re: chown regression from coreutils 5.2.1 for execute-only files

an fix the bug reported in that email without resorting to lchown, as follows, so I installed this. I included the idea of the bug report in a new test case. 2006-01-02 Paul Eggert <[EMAIL PROTECTED]> * src/chown-core.c (RC_do_ordinary_chown): New enum value. (restricte

Re: env (GNU coreutils) 5.93 patch

>> On Solaris 10, the standard location is /usr/xpg4/bin/env. >> (/usr/bin/env doesn't conform to POSIX, I guess) > > I have a hard time imagining in what way /usr/bin/env would not > conform to POSIX. On Solaris 10, /usr/bin/env invokes /usr/bin/sh, which does not conform to POSIX in several

Re: date -d and the leapsecond

"Dr. David Alan Gilbert" <[EMAIL PROTECTED]> writes: > Are you sure that whatever change made date stop taking > the 60 only stopped it on hosts that don't support leap second? I did verify that it worked on my host, when I configured it for leap seconds (which I normally don't do): 521-pengu

Re: build failure on sparc64-sun-solaris2.9

Aleksandar Milivojevic <[EMAIL PROTECTED]> writes: > if gcc -DHAVE_CONFIG_H -DLIBDIR=\"/opt/pbl/lib/sparcv9\" -I. -I../../lib -I.. > -I.. -I../../lib -g -O2 -MT canon-host.o -MD -MP -MF ".deps/canon-host.Tpo" > -c -o canon-host.o ../../lib/canon-host.c; \ > then mv -f ".deps/canon-host.Tpo" ".d

Re: build failure on sparc64-sun-solaris2.9

Aleksandar Milivojevic <[EMAIL PROTECTED]> writes: > Sparc64 gcc when built with NLS support (default) has a problem that > it core dumps from time to time if there are any warnings or errors > during compilation. It's a known bug. Thanks for tracking it down. I think we can chalk this up to "p

Re: build failure on sparc64-sun-solaris2.9

27;t know any reason why we can't simply remove all the various LDADD variables, and put everything into plain LDADD, but perhaps other linker experts can correct me. If you want to use this idea on coreutils 5.93 you can do this: ./configure LDFLAGS='-Xlinker -zignore' on your pl

Re: XML-like date/time support ?

Nicolas Mailhot <[EMAIL PROTECTED]> writes: > The XML folks have defined a single common international date/time > format I just looked at and it appears that (1) There are 9 formats, not 1. (2) There is no way to represent a leap second. This is not normal

Re: XML-like date/time support ?

Nicolas Mailhot <[EMAIL PROTECTED]> writes: >> > Yep, the xml schema is the official specification from hell everyone > wants to forget But if it's the official one, it's the one that "date" should conform to, no? > it'll get superseded by something more sane

Re: bug on chmod command

Jim Meyering <[EMAIL PROTECTED]> writes: > However, I'm not able to reproduce that with 5.93 and a small > hierarchy created like this (on a RHEL 2.4.28-sparc system): I can reproduce the problem as follows on a Debian GNU/Linux stable x86 system: 526-penguin $ mkdir d 527-penguin $ cd d 528-pen

Re: error compiling regex gnulib module with pgcc compiler

o regex_internal.h. Thanks. > > Yes OK, thanks, I installed the following patch into both gnulib and coreutils. A pgcc maintainer wrote me that this should be fixed when version 6.1-2 comes out, so I put that into the comment. 2006-01-10 Paul Eggert <[EMAIL PROTECTED]> * rege

Re: XML-like date/time support ?

Nicolas Mailhot <[EMAIL PROTECTED]> writes: > http://www.oasis-open.org/committees/relax-ng/spec-20011203.html This doesn't seem to say anything about dates or times. Is there some spec based on Relax NG that does? Or am I looking in the wrong place? > You're quickly moving beyond my XML schem

one more sync from gnulib

I installed this to catch up with today's gnulib changes. It doesn't really affect coreutils as coreutils already does gl_FUNC_MEMCMP independently. 2006-01-10 Paul Eggert <[EMAIL PROTECTED]> * m4/argmatch.m4 (gl_ARGMATCH): Don't require gl_FUNC_MEMCMP, for the

Re: bugs in test

Eric Blake <[EMAIL PROTECTED]> writes: > First, the 'extra argument' and 'missing argument' error messages are > missing a trailing newline. Thanks. That's a bug, and I installed the obvious patch. > Second, it is confusing that -a is silently accepted, but that non-unary > operators that look

md5, sha1, sha256, sha512 minor cleanups

odule from gnulib. 2006-01-11 Paul Eggert <[EMAIL PROTECTED]> * lib/.cvsignore: Add stdint.h. * lib/Makefile.am (BUILT_SOURCES, EXTRA_DIST, stdint.h, MOSTLYCLEANFILES): Add gnulib snippet. * lib/md5.c: Fix commentary typos. (alignof, UNALIGNED_P): N

Re: a real fts.c bug + fix

Jim Meyering <[EMAIL PROTECTED]> writes: > If any of you know of a system with file name resolution code that > doesn't fail for a chain of 400 symlinks, or for which you get a > different diagnostic than `Too many levels of symbolic links' (ELOOP), > please provide details. On Solaris 8 through

Re: Localization based problem with sort

Dirk Stoecker <[EMAIL PROTECTED]> writes: > when using the "sort" utility in German language the two options > > -d, --dictionary-order > -f, --ignore-case > > are activated by default. It is impossible to have other sorting methods > then. > > Would you please add negative forms of t

Re: Localization based problem with sort

Dirk Stoecker <[EMAIL PROTECTED]> writes: > What would be the problem when there is an option "--no-dictionary-order" > and "--no-ignore-case"? The problem is implementing those options, not specifying them. I don't know how to implement them. If you could supply a patch to implement them, tha

Re: Localization based problem with sort

Dirk Stoecker <[EMAIL PROTECTED]> writes: > What about the following solution? It is not exactly implementing the > above options, but nevertheless fixes the problem directly: But isn't this equivalent to setting LC_COLLATE="C" and LC_TIME="C" in the environment? Why bother to have two differen

Re: new module lib-ignore; new section build_lib in MODULES.html

#x27;-z ignore' option if this works. That should be enough in practice. This way, we don't need to worry about ldd portability. And perhaps the ldd module itself isn't necessary any more, at least for now. I installed the following into both gnulib and coreutils. This fixes anot

Re: new module lib-ignore; new section build_lib in MODULES.html

"Mark D. Baushke" <[EMAIL PROTECTED]> writes: > I know that gcc-2.7.2.1/gcc.c supports both -Wl and -Xlinker. Thanks for checking that. That's old enough for me. I installed this, in both gnulib and coreutils: 2006-01-20 Paul Eggert <[EMAIL PROTECTED

quotearg minor fix to pacify gcc -Wswitch-default

I installed this into gnulib (and coreutils) after a Bison installer noted the issue. I'm not a big fan of -Wswitch-default in general but here the workaround is harmless. 2006-01-22 Paul Eggert <[EMAIL PROTECTED]> * quotearg.c (quotearg_buffer_restyled): Add &quo

Re: date -I now undocumented

Dan Jacobson <[EMAIL PROTECTED]> writes: > If deprecated then still say so. Well, I dunno, is that really needed? Often we don't document what's deprecated. After all, we don't want to make it easy to use stuff that people shouldn't be using. If you really want to know the details you can look

Re: portability fix for bison-1.75

ld fix the problem for Bison 2.2, when it comes out. I'll also propagate these patches to coreutils. 2006-01-23 Paul Eggert <[EMAIL PROTECTED]> Work around porting bugs reported by Dieter in <http://lists.gnu.org/archive/html/bug-bison/2006-01/msg00049.html>.

Re: coreutils-5.2.1 : df show negative number for an cifs - mounted dir

4 -480512 2673216 - /mnt/GSA I've installed the following patch to CVS coreutils, so that it generates this output. 2006-01-24 Paul Eggert <[EMAIL PROTECTED]> * src/df.c (show_dev): If the file system claims to have more available than total blocks, report the nu

merging stdbool changes from gnulib

I installed this to help merge stdbool with gnulib: 2006-01-25 Paul Eggert <[EMAIL PROTECTED]> Merge from gnulib; we still don't quite match exactly, but we're getting closer. * lib/stdbool_.h (true, false) [defined __BEOS__]: undef, as before.

Re: date -I now undocumented

Dan Jacobson <[EMAIL PROTECTED]> writes: > Well I don't suppose we will ever see the word "deprecated" in any of > your Info or man pages, No, actually we do see it. For example, CVS coreutils.texi says that cp --reply is deprecated and is scheduled to be removed in 2008. > Contrast this with p

Re: proposed stdbool fixes for AIX, HP-UX, and now IRIX

Bruno Haible <[EMAIL PROTECTED]> writes: > I remember that for several gcc versions, the problems were in the > stdbool.h file, not in the built-in type. So I propose to continue using > _Bool in that case, unlike your patch. Yes, that sounds even nicer. I copied this into coreutils, so now the

Re: strange behavior with fread, fseek

I didn't have time to read and understand that puzzle, but for further enlightenment perhaps you could look at the POSIX spec for how file descriptors and streams interact and then, if glibc seems to violate t

Re: problem with multibyte constants under Mac OS X 10.3.9 in lib/fnmatch_loop.c

cause of the very same issue. I tested the patch myself on a couple of non-MacOS hosts, found and fixed a bug in it, and installed the following instead, into both gnulib and coreutils. 2006-01-27 Paul Eggert <[EMAIL PROTECTED]> * fnmatch.c (L_): Renamed from L, to work around a

Re: changes (bugs?) in date --date=

"Gus Michel" <[EMAIL PROTECTED]> writes: > (Besides the source) is there any documentation to reliably predict > what is meant by date strings which mix [partial] absolute time > references and relative ones? Sorry, no. Perhaps some day we can all come to a consensus on what that sort of string

Re: making coreutils depend on c99

Albert Chin <[EMAIL PROTECTED]> writes: > On Tue, Nov 22, 2005 at 07:51:50PM +0100, Jim Meyering wrote: >> So, people building coreutils will have a choice: apply the >> c99->c89 patch or install a modern compiler and use that >> instead of the vendor-supplied one. > > There are two issues with C9

Re: 5.93 tail -N foo bar invalid option

[EMAIL PROTECTED] (Karl Berry) writes: > In coreutils 5.93, with _POSIX2_VERSION=0 (and POSIXLY_CORRECT unset), > touch foo bar > tail -3 foo bar > yields the error > tail: invalid option -- 3 > Both `tail -3 foo' and `head -3 foo bar' are ok. Well, that's a long story. 7th Edition Unix "

Re: making coreutils depend on c99

Albert Chin <[EMAIL PROTECTED]> writes: > Substitute the code in the c99.c file with any > other C99 idiom and the results should be the same. That hasn't been my experience. For example, many C89 compilers support "long long" in some form, even though it's a C99 idiom. Similarly for compound li

Re: 5.93 tail -N foo bar invalid option

t usage "tail -DIGITS ARG ARG [ARG]", where none of the ARGs start with "-". That's no way to run a railroad. > The error `invalid option -- 3' just confused me more, since it's not > the -3 that is invalid. _That_ we can fix. I installed the follow

Re: od -t a to warn for non-ASCII

; If you don't want to do that, please update the help text and info > documentation to make it clear that the 'a' format maps high-bit-set > characters onto the same output as ASCII Good point, and thanks for the suggestion. I installed this patch: 2006-02-01 Paul Eggert

Re: join bug?

german rigau <[EMAIL PROTECTED]> writes: > [EMAIL PROTECTED] BaseConcepts]$ more kk2.sort > icecream%1:13:00:: 07510835 1 0 > ice_cream%1:13:00:: 07510835 1 1 > life_style%1:07:00:: 04875322 1 2 > part-time%3:00:00:: 01131371 1 21 kk2.sort isn't sorted correctly, at least not for the default C lo

extensions.m4 patch for Solaris 10 Sun C 5.7 c89 and some -D options

SIONS__. With that combination, you cannot compile a simple file containing only "#include " because types like uint64_t are used without being defined. I installed this patch to work around the problem: it causes "configure" to not defined __EXTENSIONS__ in this case. Thi

Re: Slash appears twice using ls -ldF

[EMAIL PROTECTED] (Eric Blake) writes: > What do other systems do in this case? Solaris 10 "ls" (both /bin/ls and /usr/xpg4/bin/ls) appends the slash. 1001-otter $ /usr/xpg4/bin/ls -d -F / /usr /usr/ // /usr/ /usr// 1002-otter $ /bin/ls -d -F / /usr /usr/ // /usr/ /usr// 1003-otter

Re: join bug?

german rigau <[EMAIL PROTECTED]> writes: > Obviously, the problem is in the sort command. With C locale > runs perfectly. However, I use LANG=en_US.UTF-8 ... > And then it seems that the "sort" command have different behaviour ... I don't see any bug in the examples that you gave. Getting back t

Re: join bug?

german rigau <[EMAIL PROTECTED]> writes: > If you see carefully the last example I sent, we obtain two different > sortings with locale en_US.UTF-8 ... with "sort kk2" we obtain "icecream" > before "ice_cream" and with "sort -k 1,2 kk2" we obtain "ice_cream" before > "icecream"! That is because i

Re: configure breakage under Interix

Jerker Bäck <[EMAIL PROTECTED]> writes: > > > > Is there a solution for this yet? Some of the problems described in those links have been fixed, in coreu

Re: SV: configure breakage under Interix

Jerker Bäck <[EMAIL PROTECTED]> writes: > sed: Cannot allocate memory > ... > > It seems to me (but now I'm only guessing) that the script is suffering from > buffer overruns. It's more likely that your 'sed' implementation is buggy. >> Some of the problems described in those links have been fi

Re: how does O_NOCTTY help? [Re: bug in chdir-safer

Jim Meyering <[EMAIL PROTECTED]> writes: > how can setting the controlling terminal cause trouble if we're > guaranteed never to read from or write to the corresponding file > descriptor. Once the terminal controls you, someone typing Control-C on that terminal can send you a signal, even if you'

Re: how does O_NOCTTY help? [Re: bug in chdir-safer

Jim Meyering <[EMAIL PROTECTED]> writes: > So O_NOCTTY is definitely worthwhile, if only to avoid > exposure for the time between an open and close of /dev/tty, > assuming the process in question already had no controlling terminal. > Right? That's the basic idea, but I'm afraid it's a bit more c

Re: sort error

Bill Mills-Curran <[EMAIL PROTECTED]> writes: > There's a sort error Nope, no error. "sort --help" will tell you what's going on. Look at the last 5 lines of the help output. ___ Bug-coreutils mailing list Bug-coreutils@gnu.org http://lists.gnu.org/m

Re: sort error

Bill Mills-Curran <[EMAIL PROTECTED]> writes: > is it possible that the "." is simply not sorted? Yes, you're getting it. The details depend on how your locale is set up. Set LC_ALL="C" in your environment to shut all this stuff off. ___ Bug-coreuti

Re: rules, rules, and more (code policy) rules

as what I think Simon (or I :-) would like, but it addresses a few of the issues he raised anyway. 2006-02-10 Paul Eggert <[EMAIL PROTECTED]> * Version 6.0-cvs. * Makefile.maint (CVS_LIST): Don't assume cvsu is available. (CVS_LIST_EXCEPT): New macro, to sim

Re: od swap support?

"Paul A. Clarke" <[EMAIL PROTECTED]> writes: > Is there any interest in adding a flag to od to support byteswapping? Would "dd conv=swab | od" suffice? ___ Bug-coreutils mailing list Bug-coreutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-core

Re: minor bug in rm: couldn't remove empty, inaccessible directory

Jim Meyering <[EMAIL PROTECTED]> writes: > + /* CAUTION: this test and diagnostic are identical those > +following the other use of fd_to_subdirp. */ > ... > + /* CAUTION: this test and diagnostic are identical those > + following the other use of fd_to_subdi

Re: how does O_NOCTTY help? [Re: bug in chdir-safer

>> That's the basic idea, but I'm afraid it's a bit more complicated than >> that. The process is exposed even after it closes the terminal, since >> it doesn't relinquish the controlling terminal even after it closes >> the corresponding file descriptor (certainly if other processes have >> the t

Re: Maybe a bug

"Jorge Bastos - Decimal" <[EMAIL PROTECTED]> writes: > I used "du" in 2 formats, and the number os bytes doesn't correspond to the > number os megabytes. As "du --help" explains, "du -b" is not the same thing as "du --block-size=1". It also sets the --apparent-size option, which explains the di

Re: Minor patch

Charles Longeau <[EMAIL PROTECTED]> writes: > Here's a little patch about some memory leaks found with valgrind, on > sort, tail and uptime. The changes that you propose will slow the programs down slightly and make them slightly bigger. The changes won't save any memory, since the programs in q

Re: how does O_NOCTTY help? [Re: bug in chdir-safer

Jim Meyering <[EMAIL PROTECTED]> writes: > It sounds like you're explaining why it was important to use O_NOCTTY > on ancient systems. Do you really think it is important now? I don't think it's _important_, no; it's a minor issue. The scenarios that I'm thinking of are fairly unlikely and don'

sync from gnulib for install-sh, vasnprintf.c, getaddrinfo, socklen

I installed this patch to sync from gnulib. 2006-02-14 Paul Eggert <[EMAIL PROTECTED]> * build-aux/install-sh: Update from gnulib. * lib/vasnprintf.c (VASNPRINTF): Rewrite the computation so that we need not use xsum. 2006-02-14 Bruno Haible <[EMAIL

Re: Suggestion for 'sort' command

Eric Blake <[EMAIL PROTECTED]> writes: > --seed=STRING seed random hash function with STRING That '--seed' part has got to go, or be changed. The string typically isn't long enough. The arg should be a file instead, defaulting to /dev/urandom. I'll try to look into this. __

Re: canon-host problems in cygwin

Thanks; I installed that sync from gnulib. ___ Bug-coreutils mailing list Bug-coreutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-coreutils

Re: bug in yesno()?

Eric Blake <[EMAIL PROTECTED]> writes: > Should yesno() be setting the properties of stdin to be unbuffered for the > duration of the getline(), so that the rest of stdin is not consumed too > early? That would make the code less portable/reliable, I think, and would slow things down. POSIX does

Re: comparing string with regular expression using test command in unix

N Gandhi Raja <[EMAIL PROTECTED]> writes: > Can we use "test" command in UNIX to compare a *string *with the > *regular expression*? No. You might look at 'expr' or 'awk' instead. ___ Bug-coreutils mailing list Bug-coreutils@gnu.org http://lists.gnu.

Re: sort on multicolumn files

ven't found a bug. However, `-k 1b,1' isn't immediately obvious, and the documentation should be improved here. I installed the following patch to try to improve things. Thanks for reporting the problem. 2006-02-20 Paul Eggert <[EMAIL PROTECTED]> * doc/coreutils.tex

AT_FDCWD glitch on Solaris 10 with Sun C 5.7

nat.h has temporarily diverged from gnulib's). 2006-02-20 Paul Eggert <[EMAIL PROTECTED]> * lib/getcwd.c (AT_FDCWD): Work around a bug in Solaris 9 and 10, where AT_FDCWD exceeds INT_MAX. * lib/openat.h (AT_FDCWD): Likewise. --- lib/getcwd.c30

Re: Bug#353911: md5sum --check checks only if _all_ are bad

Dan Jacobson <[EMAIL PROTECTED]> writes: > $ md5sum test/*|sed 1s/./Z/|>/dev/null md5sum -c --status -w -; echo $? > md5sum: standard input: 1: improperly formatted MD5 checksum line > 0 > $ md5sum test/*|sed 1s/./Z/|>/dev/null md5sum -c -; echo $? > 0 > $ md5sum test/*|sed s/./Z/|>/dev/null md5s

Re: root-dev-ino.c limitation

[EMAIL PROTECTED] (Eric Blake) writes: > a cross-compile must be pessimistic and assume that // might be > distinct from / Since the only platforms we know of that do that are Cygwin and Apollo DomainOS, can't we fall back on a system-specific #ifdef for this? It'd be a shame to penalize every cr

Re: root-dev-ino.c limitation

Jim Meyering <[EMAIL PROTECTED]> writes: > So cygwin has are two root directories, / and //, with distinct dev/inode > pairs? That would imply that there are two distinct trees. But, as I understand it, / is a subtree of //. If so, isn't // the "real" root, and can't the algorithm simply use th

Re: ls -i inefficiency

Eric Blake <[EMAIL PROTECTED]> writes: > And here's the patch. As a nice side effect, it also optimized 'ls -L' to > avoid stat() (after all, dereferencing makes no sense when all you need is > file names, and no information from the dereference). Hmmm, won't that break the behavior on dangling

Re: date --utc adds 24 hours

Craig Lawson <[EMAIL PROTECTED]> writes: > # date +'%z %Z' > -0800 PST > # date --date='2:50 am PST' > Sun Feb 26 02:50:00 PST 2006 > # date --utc --date='2:50 am PST' > Mon Feb 27 10:50:00 UTC 2006 > > PST is 8 hours ahead of UTC, so the second result should show "Feb 26". No, the re

Re: sort -u -n error : sort v 4.5.3

[EMAIL PROTECTED] (Eric Blake) writes: > I think this does point out some bugs in POSIX > compliance I don't see any bugs. In one case it appears you're misunderstanding the POSIX requirements; in the other I don't observe the symptoms that you do. > $ sort -n blah # lines with equal numeric k

dd new iflag= oflag= flags directory, nolinks

Solaris 10 O_NOLINKS option. So I installed this: 2006-03-05 Paul Eggert <[EMAIL PROTECTED]> * doc/coreutils.texi (dd invocation): New flags directory, nolinks. Alphabetize nofollow. * src/dd.c (flags, usage): New flags directory, nolinks. * src/system.h (O_

Re: Documentation discrepancy for ls -f

anks for reporting that. I installed the patch below. > I think it would also be helpful to say something about what -f is > intended to be used for and why the option character 'f' (force) is > selected for that. Sorry, I don't know that one. > should I use bug-coreu

Re: Join enhancements

"David G. Pickett" <[EMAIL PROTECTED]> writes: > I think we might extend the gnu join in a backwards compatible way > to have this flavor of capabilities, and make the it much more > useful. It sounds like you have a useful set of extensions to GNU join, though I admit I don't completely understa

Re: make error

Robert Tellamalla <[EMAIL PROTECTED]> writes: > LINK.EXE /subsystem:console /DLL /nologo /base:"0x4ad0" /NOENTRY > /IMPLIB:ic > udt.lib /out:icudt34.dll stubdata.o > LINK: extra operand `/nologo' > Try `LINK --help' for more information. > make[1]: *** [icudt34.dll] Error 1 > make[1]: Leaving

[bug #15971] Coreutils 5.94 need to build with -std=iso9899:199x with gcc 4.0.2

Update of bug #15971 (project coreutils): Status:None => Need Info ___ Follow-up Comment #1: I don't observe this problem on my Debian GNU/Linux stable host, in which I have installed GC

Re: ls -i inefficiency

[EMAIL PROTECTED] (Eric Blake) writes: > What does Solaris 10 do? Good point. My Solaris 10 host is down right now, but Solaris 9 does not complain: 54-pete $ ls -l foo lrwxrwxrwx 1 eggert eggert 7 Feb 26 15:22 foo -> nowhere 55-pete $ /bin/ls -L eggert.kshh foo

Re: ls -i inefficiency

[EMAIL PROTECTED] (Eric Blake) writes: > So do OpenBSD and Solaris warn when -L appears with an > explicit listing of a broken link? Solaris does. (Dunno about OpenBSD.) The distinction, I think, is partly that "ls -L" merely reads ".", whereas "ls -L foo" must dereference "foo" to see whether

Re: module for PRI*MAX macros?

Eric Blake <[EMAIL PROTECTED]> writes: > Is it time to make a gnulib module that exposes what is currently > done in coreutils' src/system.h to provide PRIuMAX and friends on > systems that lack them, so that it becomes possible to portably > print an ino_t value Ideally there'd be an inttypes mo

Re: dd new iflag= oflag= flags directory, nolinks

Phillip Susi <[EMAIL PROTECTED]> writes: > What is atomic about having dd do this? open() with O_DIRECTORY to test for > existence of a directory is exactly what test does isn't it? No, because "test -d foo && test -r foo" is _two_ invocations of "test", not one. A race condition is therefore p

tests/dd/misc patch for typo in noatime test

I installed this patch which I caught by code inspection (I don't have easy access to a kernel with the noatime feature): 2006-03-08 Paul Eggert <[EMAIL PROTECTED]> * tests/dd/misc: iflags->iflag. This fixes a typo that meant the noatime test never tested anythi

Re: gnulib regex problem

Jim Meyering <[EMAIL PROTECTED]> writes: > -AC_DEFINE([_REGEX_WIDE_OFFSETS], 1, > +AC_DEFINE([_REGEX_LARGE_OFFSETS], 1, Ouch, thanks for the quick fix. exg mentions that there must be a better way, in case glibc ever buys back the _REGEX_LARGE_OFFSETS fix. I'll see if I can do that. _

Re: controlling stdio buffering

If this is merely for debugging, can't you write a wrapper that forces line buffering for a program? The wrapper could substitute your own C library functions via LD_PRELOAD, or could run the program inside a pseudotty a la 'expect'. If it's for production use then I'm not sure it's a good idea t

[patch #4978] hexadecimal support for dd

Update of patch #4978 (project coreutils): Assigned to:None => eggert ___ Follow-up Comment #1: Alas, this is an incompatible change; look for how 'x' is treated in the very next line of s

Re: dd new iflag= oflag= flags directory, nolinks

Olivier Delhomme <[EMAIL PROTECTED]> writes: > So if this flag is linux specific does it mean that dd iflag=directory > won't run on other unices ? Yes, that's right. It's just like "dd iflag=noatime". The general rule is that dd issues a runtime diagnostic and immediately fails if you try to u

<    5   6   7   8   9   10   11   12   13   14   >