gmane.comp.gnu.findutils.patches

[PATCH] updatedb: add further entries toPRUNEFS defaults

Bernhard Voelker — 2012-12-05T11:16:05

From f18c966940c7769f4be1acea5567430ee290158c Mon Sep 17 00:00:00 2001
From: Bernhard Voelker <mail< at >bernhard-voelker.de>
Date: Wed, 5 Dec 2012 12:09:41 +0100
Subject: [PATCH] updatedb: add further entries to PRUNEFS defaults

Merge new entries from OpenSuSE project, e.g. there were hangs
on nfs4 mounts.

* locate/updatedb.sh (PRUNEFS): Add 9P, cifs, nfs4, and unknown.
Sort entries alphabetically.
---
 locate/updatedb.sh |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/locate/updatedb.sh b/locate/updatedb.sh
index a9ded7b..cb3f39a 100644
--- a/locate/updatedb.sh
+++ b/locate/updatedb.sh
< at >< at > -246,7 +246,7 < at >< at > do
 done


-: ${PRUNEFS="nfs NFS proc afs smbfs autofs iso9660 ncpfs coda devpts ftpfs devfs mfs sysfs shfs"}
+: ${PRUNEFS="9P NFS afs autofs cifs coda devfs devpts ftpfs iso9660 mfs ncpfs nfs nfs4 proc shfs smbfs sysfs unknown"}

 if test -n "$PRUNEFS"; then
 prunefs_exp=`echo $PRUNEFS |sed -e 's/\([^ ][^ ]*\)/-o -fstype \1/g' \

[PATCH] fix comment/changelog typos

Jim Meyering — 2012-06-04T11:29:24

Here's a patch to fix some typos.

I made this change manually, because the tool proposed three
alternatives:

diff --git a/m4/nullsort.m4 b/m4/nullsort.m4
index 5f592f7..10a1a93 100644
--- a/m4/nullsort.m4
+++ b/m4/nullsort.m4
< at >< at > -13,7 +13,7 < at >< at > else
         # sort thinks there is more than one input line there.   We have
         # to make sort -c do its thing with the input lines in the wrong
         # order to determine that (we can't do it in one shot because
-        # if sort returns nonzero we cant tell that it wasn't just
+        # if sort returns nonzero we can't tell that it wasn't just
         # complaining about this unknown -z option.
         if $SORT -z -c < "${srcdir:-.}/m4/order-bad.bin" 2>/dev/null >/dev/null
         then

Then noticed "dum" being changed to "dumb", but only in two
exception lists.  There, just removing the non-word is better.

Finally, using http://github.com/lyda/misspell-check and
some perl to massage its diagnostics into sed commands:

git ls-files|grep -v '\.po$'|misspellings -f -|perl -nl \
  -e '/^(.*?)\[(\d+)\]: (\w+) -> "(.*?)"$/ or next;' \
  -e '($file,$n,$l,$r)=($1,$2,$3,$4); $q="'\''"; $r=~s/$q/$q\\$q$q/g;'\
  -e 'print "sed -i $q${n}s!$l!$r!$q $file"'|bash


From edf76262937c363e6d329693570e250a24338018 Mon Sep 17 00:00:00 2001
From: Jim Meyering <meyering< at >redhat.com>
Date: Mon, 4 Jun 2012 12:30:52 +0200
Subject: [PATCH] fix comment/changelog typos

First, change "cant" to "can't" manually in m4/nullsort.m4,
since "cant" had 3 replacement options.  Then remove "dum"
from the LocalWords lists in doc/*.texi.  It was not used.
Finally, perform the remaining changes mechanically using
http://github.com/lyda/misspell-check and
some perl to massage its diagnostics into sed -i commands:

git ls-files|grep -v '\.po$'|misspellings -f -|perl -nl \
  -e '/^(.*?)\[(\d+)\]: (\w+) -> "(.*?)"$/ or next;' \
  -e '($file,$n,$l,$r)=($1,$2,$3,$4); $q="'\''"; $r=~s/$q/$q\\$q$q/g;'\
  -e 'print "sed -i $q${n}s!$l!$r!$q $file"'|bash
---
 ChangeLog                                     | 28 +++++++++++++--------------
 NEWS                                          |  2 +-
 TODO                                          |  2 +-
 doc/find-maint.texi                           |  4 ++--
 doc/find.texi                                 |  2 +-
 find/defs.h                                   |  2 +-
 find/print.c                                  |  2 +-
 find/testsuite/find.gnu/exec-one-rtn-fail.exp |  2 +-
 locate/locate.c                               |  2 +-
 m4/nullsort.m4                                |  2 +-
 10 files changed, 24 insertions(+), 24 deletions(-)

diff --git a/ChangeLog b/ChangeLog
index 8455352..7f240a6 100644
--- a/ChangeLog
+++ b/ChangeLog
< at >< at > -56,7 +56,7 < at >< at >
 * find/print.c (parse_octal_escape): Instead of updating a const
 char* pointer to indicate how many characters from the input we
 consumed, update a size_t value.
-(insert_fprintf): Change function defintion to match updated
+(insert_fprintf): Change function definition to match updated
 prototype.  Eliminate fmt_inpos.  Introduce a `readpos' offset
 variable which takes the previous role of fmt_inpos.

< at >< at > -271,7 +271,7 < at >< at >
 %(, %{ and %[, which are all rejected.
 * find/testsuite/Makefile.am (EXTRA_DIST_EXP): Add
 find.gnu/printf-reserved.exp.
-* doc/find.texi (Reserved and Unkown Directives): Document this.
+* doc/find.texi (Reserved and Unknown Directives): Document this.
 * find/find.1: Document this.
 * NEWS: Mention this change.

< at >< at > -2892,7 +2892,7 < at >< at >

 2009-04-13  Andreas Metzler <gnu< at >downhill.at.eu.org>

-Fix Savannah bug# 26092: find.1 still marks -iwholename as prefered
+Fix Savannah bug# 26092: find.1 still marks -iwholename as preferred
 * find/find.1 (-iwholename): Indicate that this option is no
 longer preferred over -ipath.

< at >< at > -3059,7 +3059,7 < at >< at >
 * po/sl.po: Updated Slovenian translation.
 * po/sv.po: Updated Swedish translation.
 * po/tr.po: Updated Turkish translation.
-* po/uk.po: Updated Ukranian translation.
+* po/uk.po: Updated Ukrainian translation.
 * po/vi.po: Updated Vietnamese translation.
 * NEWS: Mention these updates.

< at >< at > -3394,7 +3394,7 < at >< at >
 (UNUSUAL FILENAMES): Describe how LC_CTYPE affects -printf %p,
 -print, -fprint.
 (STANDARDS CONFORMANCE): -ok and -okdir now obey the system's
-defintion of yes/no if POSIXLY_CORRECT is set.  In any case,
+definition of yes/no if POSIXLY_CORRECT is set.  In any case,
 LC_CTYPE and LC_COLLATE may have some effect.  Point out that
 the setting of LC_MESSAGES deternmines what pattern is used to
 interpret the user's response to -ok/-okdir.
< at >< at > -3937,7 +3937,7 < at >< at >
 2007-08-18  Eric Blake  <ebb9< at >byu.net>

 Fix Savannah bug #20751.
-* lib/listfile.c (list_file): Accomodate gnulib change of 3 Jul
+* lib/listfile.c (list_file): Accommodate gnulib change of 3 Jul
 2006.
 * NEWS: Document this.
 Reported by Nigel Stepp.
< at >< at > -4885,7 +4885,7 < at >< at >

 * find/defs.h (pred_open) Rename to pred_openparen to avoid
 problems with the macrtos which build the parser tabnles on
-platforms where 'open' is in fact a macro whcih expands to
+platforms where 'open' is in fact a macro which expands to
 open64.   The problem was that token pasting put pred_open64 into
 the parser table, but the function was still defined as
 parse_open.   This fixes Savannah bug #19371.
< at >< at > -5035,7 +5035,7 < at >< at >

 2007-04-20  Maxim V. Dziumanenko <dziumanenko< at >gmail.com>

-* po/uk.po: New Ukranian translation.
+* po/uk.po: New Ukrainian translation.
 * configure.in: Added "uk" for Ukranian.

 2007-04-19  Peter Breitenlohner <peb< at >mppmu.mpg.de> (tiny change)
< at >< at > -5225,7 +5225,7 < at >< at >
 2007-04-09  James Youngman  <jay< at >gnu.org>

 * doc/find.texi: Change fully-specified < at >node directives to
-single-argument < at >node directives in order to accomodate the
+single-argument < at >node directives in order to accommodate the
 inclusion of getdate.texi.

 * doc/find.texi: Document -newerXY.
< at >< at > -5589,7 +5589,7 < at >< at >
   find/testsuite/find.gnu/perm000.xo: -perm /000 is now quivalent
   to -perm -000, as dscussed in Savannah bug #14748.  The warning
   message we issue when the user does this now explains that the
-  defintion changed as we promised it would in 2005.
+  definition changed as we promised it would in 2005.

 * xargs/xargs.c (read_line): Correctly handle quoted empty
 arguments occurring first or last on a line (by adding an empty
< at >< at > -6122,7 +6122,7 < at >< at >

 * find/fstype.c:
 Savannah bug #14921: when a Linux bind filesystem is in use, find
-would expand '-printf %F' to 'none' if a bind mount targetted the same
+would expand '-printf %F' to 'none' if a bind mount targeted the same
 filesystem as the one containing the file to be described.

 * NEWS: Updated to describe recent changes
< at >< at > -6604,7 +6604,7 < at >< at >
 from Andreas Metzler)

 * INSTALL, depcomp, doc/texinfo.tex, install-sh, missing, mkinstalldirs:
-Updated auxilliary files from the automake-1.9 distribution
+Updated auxiliary files from the automake-1.9 distribution

 * configure.in:
 Introduce new Automake conditional, CROSS_COMPILING, true when we are corss compiling
< at >< at > -8763,7 +8763,7 < at >< at >
 Debian bug #176201.

 * xargs/xargs.c:
-Fixed Debian bug #176201, "xargs enviroment size limited to 20k", by
+Fixed Debian bug #176201, "xargs environment size limited to 20k", by
 reading a patch offered by Bob Proulx and implementing something
 substantially similar myself.

< at >< at > -10404,7 +10404,7 < at >< at >
 * wait.h: New file taken from xargs.c.

 * xargs.c: Make limits.h vs. sys/param.h conditional on
-LIMITS_MISSING instead of USG, to accomodate SVR2.
+LIMITS_MISSING instead of USG, to accommodate SVR2.

 1990-11-08  David J. MacKenzie  <djm< at >gnu.org>

diff --git a/NEWS b/NEWS
index d8435fd..5b4881b 100644
--- a/NEWS
+++ b/NEWS
< at >< at > -790,7 +790,7 < at >< at > with a space

 ** Translations

-Findutils 4.3.4 includes a translation for the Ukranian language.
+Findutils 4.3.4 includes a translation for the Ukrainian language.


 * Major changes in release 4.3.3, 2007-04-15
diff --git a/TODO b/TODO
index aff56d4..1f694c3 100644
--- a/TODO
+++ b/TODO
< at >< at > -17,7 +17,7 < at >< at > Perhaps a better description in texi pages as well.
 * Add option for find to sort output in lexical order for use for updatedb
 olarsac< at >airfrance.fr (Olivier) made the following suggestion:

-As I was running thru the code looking for the bug I wondered why the updatedb
+As I was running through the code looking for the bug I wondered why the updatedb
 has to use sort...
 why not add an option to find that sorts the output in lexical order?
 my point is:
diff --git a/doc/find-maint.texi b/doc/find-maint.texi
index 0ee7e72..8fa6c1e 100644
--- a/doc/find-maint.texi
+++ b/doc/find-maint.texi
< at >< at > -839,7 +839,7 < at >< at > traditional ("old") format that locate uses on other Unix systems.

 When locate reads filenames from a LOCATE02 database (the default
 format), the buffer into which data is read is automatically extended
-to accomodate the length of the filenames.
+to accommodate the length of the filenames.

 This automatic buffer extension does not happen for old-format
 databases.  Instead a 1026-byte buffer is used.  When a longer
< at >< at > -1178,7 +1178,7 < at >< at > Release} field of these bugs appropriately and make sure the
 < at >comment LocalWords: POSIXLY fls fprintf strftime locale's EDT GMT AP
 < at >comment LocalWords: EST diff perl backquotes sprintf Falstad Oct cron
 < at >comment LocalWords: eg vmunix mkdir afs allexec allwrite ARG bigram
-< at >comment LocalWords: bigrams cd chmod comp crc CVS dbfile dum eof
+< at >comment LocalWords: bigrams cd chmod comp crc CVS dbfile eof
 < at >comment LocalWords: fileserver filesystem fn frcode Ghazi Hnewc iXX
 < at >comment LocalWords: joeuser Kaveh localpaths localuser LOGNAME
 < at >comment LocalWords: Meyering mv netpaths netuser nonblank nonblanks
diff --git a/doc/find.texi b/doc/find.texi
index 0be7292..cc02822 100644
--- a/doc/find.texi
+++ b/doc/find.texi
< at >< at > -5649,7 +5649,7 < at >< at > Expressions}, for more information on expressions.
 < at >comment LocalWords: POSIXLY fls fprintf strftime locale's EDT GMT AP
 < at >comment LocalWords: EST diff perl backquotes sprintf Falstad Oct cron
 < at >comment LocalWords: eg vmunix mkdir afs allexec allwrite ARG bigram
-< at >comment LocalWords: bigrams cd chmod comp crc CVS dbfile dum eof
+< at >comment LocalWords: bigrams cd chmod comp crc CVS dbfile eof
 < at >comment LocalWords: fileserver filesystem fn frcode Ghazi Hnewc iXX
 < at >comment LocalWords: joeuser Kaveh localpaths localuser LOGNAME
 < at >comment LocalWords: Meyering mv netpaths netuser nonblank nonblanks
diff --git a/find/defs.h b/find/defs.h
index 6e11174..ca7d604 100644
--- a/find/defs.h
+++ b/find/defs.h
< at >< at > -637,7 +637,7 < at >< at > struct state
   /* The file being operated on, relative to the current directory.
      Used for stat, readlink, remove, and opendir.  */
   char *rel_pathname;
-  /* The directory fd to which rel_pathname is relative.  Thsi is relevant
+  /* The directory fd to which rel_pathname is relative.  This is relevant
    * when we're navigating the hierarchy with fts() and using FTS_CWDFD.
    */
   int cwd_dir_fd;
diff --git a/find/print.c b/find/print.c
index e307d95..025cd7d 100644
--- a/find/print.c
+++ b/find/print.c
< at >< at > -608,7 +608,7 < at >< at > format_date (struct timespec ts, int kind)

   /* human_readable() assumes we pass a buffer which is at least as
    * long as LONGEST_HUMAN_READABLE.  We use an assertion here to
-   * ensure that no nasty unsigned overflow happend in our calculation
+   * ensure that no nasty unsigned overflow happened in our calculation
    * of the size of buf.  Do the assertion here rather than in the
    * code for %< at > so that we find the problem quickly if it exists.  If
    * you want to submit a patch to move this into the if statement, go
diff --git a/find/testsuite/find.gnu/exec-one-rtn-fail.exp b/find/testsuite/find.gnu/exec-one-rtn-fail.exp
index 91ec226..b761d1d 100644
--- a/find/testsuite/find.gnu/exec-one-rtn-fail.exp
+++ b/find/testsuite/find.gnu/exec-one-rtn-fail.exp
< at >< at > -1,4 +1,4 < at >< at >
 # If -exec \; fails, it should return false.   The return
-# value of find shoudl stil; be 0, unless another error has
+# value of find should stil; be 0, unless another error has
 # happened.
 find_start p { /tmp -exec false \; -o \( -printf "yep\n" -quit \) }
diff --git a/locate/locate.c b/locate/locate.c
index db90ce2..64ad4fa 100644
--- a/locate/locate.c
+++ b/locate/locate.c
< at >< at > -144,7 +144,7 < at >< at > static const char warn_name_units[] = N_("days");
 enum visit_result
   {
     VISIT_CONTINUE = 1,  /* please call the next visitor */
-    VISIT_ACCEPTED = 2,  /* accepted, call no futher callbacks for this file */
+    VISIT_ACCEPTED = 2,  /* accepted, call no further callbacks for this file */
     VISIT_REJECTED = 4,  /* rejected, process next file. */
     VISIT_ABORT    = 8   /* rejected, process no more files. */
   };
diff --git a/m4/nullsort.m4 b/m4/nullsort.m4
index 5f592f7..10a1a93 100644
--- a/m4/nullsort.m4
+++ b/m4/nullsort.m4
< at >< at > -13,7 +13,7 < at >< at > else
         # sort thinks there is more than one input line there.   We have
         # to make sort -c do its thing with the input lines in the wrong
         # order to determine that (we can't do it in one shot because
-        # if sort returns nonzero we cant tell that it wasn't just
+        # if sort returns nonzero we can't tell that it wasn't just
         # complaining about this unknown -z option.
         if $SORT -z -c < "${srcdir:-.}/m4/order-bad.bin" 2>/dev/null >/dev/null
         then
--
1.7.11.rc1

find.texi: uk and other consistency

Karl Berry — 2011-12-29T21:56:15

Hi James,

I'm not sure it's really viable to try to write Texinfo in UK English
(there's a hyphenation problem at least, without running TeX in some
special way), but since evidently find.texi is mostly written that way
now, here is a small patch to make things a little more consistent.
(I don't doubt plenty of infelicities remain, this is just what I
noticed in a quick look.)

Not included in this patch: all those < at >subsections that don't have
corresponding should really be < at >subheadings.  Or, probably better, they
should get nodes, since now the nodes end up being hundreds of lines
long.

Sorry I can't easily deal with git and give you the patch that way,
hope this is ok.

Thanks,
k

 Karl Berry  <karl< at >gnu.org>
* find.texi (< at >copying): add 2011, period outside quotes.
(thoughout): use UK spelling and " -- " convention.
(section names): use < at >code where needed.

--- ORIG/find.texi2011-12-29 11:09:38.000000000 -0800
+++ find.texi2011-12-29 11:23:41.000000000 -0800
< at >< at > -33,4 +33,4 < at >< at >
 
-Copyright (C) 1994, 1996, 1998, 2000, 2001, 2003, 2004, 2005, 2006,
-2007, 2008, 2009, 2010 Free Software Foundation, Inc.
+Copyright < at >copyright{} 1994, 1996, 1998, 2000, 2001, 2003, 2004, 2005, 2006,
+2007, 2008, 2009, 2010, 2011 Free Software Foundation, Inc.
 
< at >< at > -41,3 +41,3 < at >< at >
 Texts.  A copy of the license is included in the section entitled
-``GNU Free Documentation License.''
+``GNU Free Documentation License''.
 < at >end copying
< at >< at > -536,4 +536,4 < at >< at >
 
-If a pattern is a plain string---it contains no
-metacharacters---< at >code{locate} displays all file names in the database
+If a pattern is a plain string -- it contains no
+metacharacters -- < at >code{locate} displays all file names in the database
 that contain that string.  If a pattern contains
< at >< at > -727,3 +727,3 < at >< at >
 
-The following differences in behavior occur when the < at >samp{-L} option
+The following differences in behaviour occur when the < at >samp{-L} option
 is used:
< at >< at > -2893,3 +2893,3 < at >< at >
 predecessor.  (The counts can be negative.)  Following the count is a
-null-terminated ASCII remainder---the part of the name that follows
+null-terminated ASCII remainder -- the part of the name that follows
 the shared prefix.
< at >< at > -4035,3 +4035,3 < at >< at >
 archiving program.  GNU < at >code{tar} and < at >code{cpio} can both read lists
-of file names from the standard input---either delimited by nulls (the
+of file names from the standard input -- either delimited by nulls (the
 safe way) or by blanks (the lazy, risky default way).  To use
< at >< at > -4354,3 +4354,3 < at >< at >
 
-< at >subsection Making Use of xargs
+< at >subsection Making Use of < at >code{xargs}
 
< at >< at > -4393,3 +4393,3 < at >< at >
 list of files that < at >code{xargs} reads could in fact contain white space
-characters --- spaces, tabs and newline characters.  Since by default,
+characters -- spaces, tabs and newline characters.  Since by default,
 < at >code{xargs} assumes that the list of files it is reading uses white
< at >< at > -4427,3 +4427,3 < at >< at >
 
-< at >subsection Going back to -exec
+< at >subsection Going back to < at >code{-exec}
 
< at >< at > -4452,3 +4452,3 < at >< at >
 
-< at >subsection A more secure version of -exec
+< at >subsection A more secure version of < at >code{-exec}
 
< at >< at > -4532,3 +4532,3 < at >< at >
 
-< at >subsection Using the -delete action
+< at >subsection Using the < at >code{-delete} action
 
< at >< at > -4740,3 +4740,3 < at >< at >
 
-< at >subsection Using -printf and sort to compare timestamps
+< at >subsection Using < at >code{-printf} and < at >code{sort} to compare timestamps
 
< at >< at > -4771,3 +4771,3 < at >< at >
 
-< at >subsection Solving the problem with make
+< at >subsection Solving the problem with < at >code{make}
 
< at >< at > -5027,3 +5027,3 < at >< at >
 < at >node Problems with -exec and filenames
-< at >subsection Problems with -exec and filenames
+< at >subsection Problems with < at >code{-exec} and filenames
 
< at >< at > -5103,3 +5103,3 < at >< at >
 < at >node O_NOFOLLOW
-< at >subsubsection O_NOFOLLOW
+< at >subsubsection < at >code{O_NOFOLLOW}
 
< at >< at > -5143,3 +5143,3 < at >< at >
 < at >node Systems without O_NOFOLLOW
-< at >subsubsection Systems without O_NOFOLLOW
+< at >subsubsection Systems without < at >code{O_NOFOLLOW}
 
< at >< at > -5184,3 +5184,3 < at >< at >
 < at >node Race Conditions with -exec
-< at >subsection Race Conditions with -exec
+< at >subsection Race Conditions with < at >code{-exec}
 
< at >< at > -5238,3 +5238,3 < at >< at >
 < at >node Race Conditions with -print and -print0
-< at >subsection Race Conditions with -print and -print0
+< at >subsection Race Conditions with < at >code{-print} and < at >code{-print0}
 
< at >< at > -5470,3 +5470,3 < at >< at >
 you move into them (that is how it knows you want to use the
-filesystem).  So, if you do see this message, be wary ---
+filesystem).  So, if you do see this message, be wary --
 < at >code{automount} may not have been responsible.  Consider the
< at >< at > -5532,3 +5532,3 < at >< at >
 < at >node Error Messages From xargs
-< at >section Error Messages From xargs
+< at >section Error Messages From < at >code{xargs}
 
< at >< at > -5606,3 +5606,3 < at >< at >
 < at >node Error Messages From updatedb
-< at >section Error Messages From updatedb
+< at >section Error Messages From < at >code{updatedb}
 

Diff finished at Thu Dec 29 11:23:48

[PATCH 1/9] File system type names should beuntranslated strings.

James Youngman — 2011-07-09T22:15:43

* find/fstype.c (file_system_type_uncached): When the file system
type is not known, return the literal string "unknown", instead of
a translation of that.  This allows a find command line like "find
. -fstype unknown" to work in a variety of locales, instead of
just the C locale.  Similarly, for the AFS file system, always
return "afs".
* NEWS: Mention this change.
---
 ChangeLog     |   11 +++++++++++
 NEWS          |    7 +++++++
 find/fstype.c |    4 ++--
 3 files changed, 20 insertions(+), 2 deletions(-)

diff --git a/ChangeLog b/ChangeLog
index 412d219..581de09 100644
--- a/ChangeLog
+++ b/ChangeLog
< at >< at > -1,3 +1,14 < at >< at >
+2011-07-09  James Youngman  <jay< at >gnu.org>
+
+File system type names should be untranslated strings.
+* find/fstype.c (file_system_type_uncached): When the file system
+type is not known, return the literal string "unknown", instead of
+a translation of that.  This allows a find command line like "find
+. -fstype unknown" to work in a variety of locales, instead of
+just the C locale.  Similarly, for the AFS file system, always
+return "afs".
+* NEWS: Mention this change.
+
 2011-06-29  James Youngman  <jay< at >gnu.org>
 
 Fix some constness warnings when dealing with -printf formats.
diff --git a/NEWS b/NEWS
index 2f031d4..2a019d8 100644
--- a/NEWS
+++ b/NEWS
< at >< at > -51,6 +51,13 < at >< at > last match found in this file, rather than the first, to better deal
 with implementations which have duplicate entries (for example
 /proc/mounts on systems running the Linux kernel).
 
+** Translations
+
+The file system type names "afs" and "unknown" are now always
+untranslated strings.  This means that a find command lines like "find
+. -fstype unknown" and "find . -fstype afs" will work as-is in a
+variety of locales, instead of just the C locale.
+
 * Major changes in release 4.5.10, 2011-05-11
 
 ** Documentation Changes
diff --git a/find/fstype.c b/find/fstype.c
index a47b6d3..f8bb793 100644
--- a/find/fstype.c
+++ b/find/fstype.c
< at >< at > -206,7 +206,7 < at >< at > file_system_type_uncached (const struct stat *statp, const char *path)
   if (in_afs (path))
     {
       fstype_known = 1;
-      return xstrdup ("afs");
+      return xstrdup (N_("afs"));
     }
 #endif
 
< at >< at > -241,7 +241,7 < at >< at > file_system_type_uncached (const struct stat *statp, const char *path)
   /* Don't cache unknown values. */
   fstype_known = (type != NULL);
 
-  return type ? type : xstrdup (_("unknown"));
+  return type ? type : xstrdup (N_("unknown"));
 }

[PATCH 1/7] Clarify variable naming ininsert_fprintf.

James Youngman — 2011-06-29T09:16:50

* find/print.c (insert_fprintf): Rename some variables for greater
clarity:
format is now segstart, since it points to the start of the part
of the format string we're next going to pass to make_segment.
scan2 is now fmt_inpos (which is a const char*) and is the
scanning position beyond segstart which we're reading to locate
the end of this segment.
scan is now fmt_editpos and is the scanning position at which we
are modifying the format string (for example to change \n in the
format to the value of the '\n' character).
---
 ChangeLog    |   14 +++++++
 find/print.c |  122 ++++++++++++++++++++++++++++++---------------------------
 2 files changed, 78 insertions(+), 58 deletions(-)

diff --git a/ChangeLog b/ChangeLog
index c003e0f..fc6f945 100644
--- a/ChangeLog
+++ b/ChangeLog
< at >< at > -1,3 +1,17 < at >< at >
+2011-06-25  James Youngman  <jay< at >gnu.org>
+
+Clarify variable naming in insert_fprintf.
+* find/print.c (insert_fprintf): Rename some variables for greater
+clarity:
+format is now segstart, since it points to the start of the part
+of the format string we're next going to pass to make_segment.
+scan2 is now fmt_inpos (which is a const char*) and is the
+scanning position beyond segstart which we're reading to locate
+the end of this segment.
+scan is now fmt_editpos and is the scanning position at which we
+are modifying the format string (for example to change \n in the
+format to the value of the '\n' character).
+
 2011-06-19  James Youngman  <jay< at >gnu.org>
 
 Test find -printf %i.
diff --git a/find/print.c b/find/print.c
index 0e15123..e0a1331 100644
--- a/find/print.c
+++ b/find/print.c
< at >< at > -217,10 +217,10 < at >< at > insert_fprintf (struct format_val *vec,
 const struct parser_table *entry,
 const char *format_const)
 {
-  char *format = (char*)format_const; /* XXX: casting away constness */
-  register char *scan;/* Current address in scanning `format'. */
-  register char *scan2;/* Address inside of element being scanned. */
-  struct segment **segmentp;/* Address of current segment. */
+  char *segstart = (char*)format_const; /* XXX: casting away constness */
+  register char *fmt_editpos; /* Current address in scanning `format_const'. */
+  register const char *fmt_inpos; /* Address inside of element being scanned. */
+  struct segment **segmentp;  /* Address of current segment. */
   struct predicate *our_pred;
 
   our_pred = insert_primary_withpred (entry, pred_fprintf, format_const);
< at >< at > -233,121 +233,126 < at >< at > insert_fprintf (struct format_val *vec,
   segmentp = &our_pred->args.printf_vec.segment;
   *segmentp = NULL;
 
-  for (scan = format; *scan; scan++)
+  for (fmt_editpos = segstart; *fmt_editpos; fmt_editpos++)
     {
-      if (*scan == '\\')
+      if (*fmt_editpos == '\\')
 {
-  scan2 = scan + 1;
-  if (*scan2 >= '0' && *scan2 <= '7')
+  fmt_inpos = fmt_editpos + 1;
+  if (*fmt_inpos >= '0' && *fmt_inpos <= '7')
     {
       register int n, i;
 
-      for (i = n = 0; i < 3 && (*scan2 >= '0' && *scan2 <= '7');
-   i++, scan2++)
-n = 8 * n + *scan2 - '0';
-      scan2--;
-      *scan = n;
+      for (i = n = 0; i < 3 && (*fmt_inpos >= '0' && *fmt_inpos <= '7');
+   i++, fmt_inpos++)
+n = 8 * n + *fmt_inpos - '0';
+      fmt_inpos--;
+      *fmt_editpos = n;
     }
   else
     {
-      switch (*scan2)
+      switch (*fmt_inpos)
 {
 case 'a':
-  *scan = 7;
+  *fmt_editpos = 7;
   break;
 case 'b':
-  *scan = '\b';
+  *fmt_editpos = '\b';
   break;
 case 'c':
-  make_segment (segmentp, format, scan - format,
+  make_segment (segmentp, segstart, fmt_editpos - segstart,
 KIND_STOP, 0, 0,
 our_pred);
   if (our_pred->need_stat && (our_pred->p_cost < NeedsStatInfo))
     our_pred->p_cost = NeedsStatInfo;
   return true;
 case 'f':
-  *scan = '\f';
+  *fmt_editpos = '\f';
   break;
 case 'n':
-  *scan = '\n';
+  *fmt_editpos = '\n';
   break;
 case 'r':
-  *scan = '\r';
+  *fmt_editpos = '\r';
   break;
 case 't':
-  *scan = '\t';
+  *fmt_editpos = '\t';
   break;
 case 'v':
-  *scan = '\v';
+  *fmt_editpos = '\v';
   break;
 case '\\':
-  /* *scan = '\\'; * it already is */
+  /* *fmt_editpos = '\\'; * it already is */
   break;
 default:
   error (0, 0,
- _("warning: unrecognized escape `\\%c'"), *scan2);
-  scan++;
+ _("warning: unrecognized escape `\\%c'"), *fmt_inpos);
+  fmt_editpos++;
   continue;
 }
     }
-  segmentp = make_segment (segmentp, format, scan - format + 1,
+  segmentp = make_segment (segmentp,
+   segstart, fmt_editpos - segstart + 1,
    KIND_PLAIN, 0, 0,
    our_pred);
-  format = scan2 + 1;/* Move past the escape. */
-  scan = scan2;/* Incremented immediately by `for'. */
+  segstart = fmt_inpos + 1; /* Move past the escape. */
+  fmt_editpos = fmt_inpos;  /* Incremented immediately by `for'. */
 }
-      else if (*scan == '%')
+      else if (*fmt_editpos == '%')
 {
-  if (scan[1] == 0)
+  if (fmt_editpos[1] == 0)
     {
       /* Trailing %.  We don't like those. */
       error (EXIT_FAILURE, 0,
-     _("error: %s at end of format string"), scan);
+     _("error: %s at end of format string"), fmt_editpos);
     }
-  else if (scan[1] == '%')
+  else if (fmt_editpos[1] == '%')
     {
-      segmentp = make_segment (segmentp, format, scan - format + 1,
+      segmentp = make_segment (segmentp,
+       segstart, fmt_editpos - segstart + 1,
        KIND_PLAIN, 0, 0,
        our_pred);
-      scan++;
-      format = scan + 1;
+      fmt_editpos++;
+      segstart = fmt_editpos + 1;
       continue;
     }
   /* Scan past flags, width and precision, to verify kind. */
-  for (scan2 = scan; *++scan2 && strchr ("-+ #", *scan2);)
-    /* Do nothing. */ ;
-  while (ISDIGIT (*scan2))
-    scan2++;
-  if (*scan2 == '.')
-    for (scan2++; ISDIGIT (*scan2); scan2++)
+  for (fmt_inpos = fmt_editpos;
+       *++fmt_inpos && strchr ("-+ #", *fmt_inpos);)
+    {
+      /* Do nothing. */
+    }
+  while (ISDIGIT (*fmt_inpos))
+    fmt_inpos++;
+  if (*fmt_inpos == '.')
+    for (fmt_inpos++; ISDIGIT (*fmt_inpos); fmt_inpos++)
       /* Do nothing. */ ;
-  if (strchr ("abcdDfFgGhHiklmMnpPsStuUyYZ", *scan2))
+  if (strchr ("abcdDfFgGhHiklmMnpPsStuUyYZ", *fmt_inpos))
     {
-      segmentp = make_segment (segmentp, format, scan2 - format,
-       KIND_FORMAT, *scan2, 0,
+      segmentp = make_segment (segmentp, segstart, fmt_inpos - segstart,
+       KIND_FORMAT, *fmt_inpos, 0,
        our_pred);
-      scan = scan2;
-      format = scan + 1;
+      fmt_editpos = fmt_inpos;
+      segstart = fmt_editpos + 1;
     }
-  else if (strchr ("ABCT", *scan2) && scan2[1])
+  else if (strchr ("ABCT", *fmt_inpos) && fmt_inpos[1])
     {
-      segmentp = make_segment (segmentp, format, scan2 - format,
-       KIND_FORMAT, scan2[0], scan2[1],
+      segmentp = make_segment (segmentp, segstart, fmt_inpos - segstart,
+       KIND_FORMAT, fmt_inpos[0], fmt_inpos[1],
        our_pred);
-      scan = scan2 + 1;
-      format = scan + 1;
+      fmt_editpos = fmt_inpos + 1;
+      segstart = fmt_editpos + 1;
       continue;
     }
   else
     {
-      switch (*scan2)
+      switch (*fmt_inpos)
 {
 case '{':
 case '[':
 case '(':
   error (EXIT_FAILURE, 0,
  _("error: the format directive `%%%c' is reserved for future use"),
- (int)*scan2);
+ (int)*fmt_inpos);
   /*NOTREACHED*/
   break;
 
< at >< at > -355,19 +360,20 < at >< at > insert_fprintf (struct format_val *vec,
   /* An unrecognized % escape.  Print the char after the %. */
   error (0, 0,
  _("warning: unrecognized format directive `%%%c'"),
- *scan2);
-  segmentp = make_segment (segmentp, format, scan - format,
+ *fmt_inpos);
+  segmentp = make_segment (segmentp,
+   segstart, fmt_editpos - segstart,
    KIND_PLAIN, 0, 0,
    our_pred);
-  format = scan + 1;
+  segstart = fmt_editpos + 1;
 }
       continue;
     }
 }
     }
 
-  if (scan > format)
-    make_segment (segmentp, format, scan - format, KIND_PLAIN, 0, 0,
+  if (fmt_editpos > segstart)
+    make_segment (segmentp, segstart, fmt_editpos - segstart, KIND_PLAIN, 0, 0,
   our_pred);
   return true;
 }

Change request proposal for a new Findfunctionality

Michel Lanne — 2010-11-06T18:08:47

Good morning,

This mail just to mention a simple lacking functionality in the Find command :

There is no way to disable the warning messages sent on the error
output, and especially the "Permission non granted" error message sent
for each unaccessible directory met during the research process.

The alone way proposed on various docs or sites is to redirect the
error output on /dev/null. OK, thank's, but that's far to be great.
This is even an invitation to very bad programming, especially by
avoiding the ability to choose through the find parameters the
warnings which are suppressed and those which are not.

So, I submit to the Find maintainers the idea of a change request for
this functionality, maybe through a new option.

Sincerely yours,

Michel Lanne.

new predicate

Konrad Eisele — 2010-05-27T20:04:44

I wanted to submit a patch that is quite short and 
more thought as a feature request. It adds the predicate
"-dtype <regex>" (dtype meaning datatype). The dtype
predicate uses libmagic from the "file" command to get
the *content datatype* of the file in view, then doing a regex on
it. i.e. "echo abc>f.txt; file f.txt" yealds "ASSCII text".
Therefore "file f.txt -dtype .*text.*" would do a regex ".*text.*"
on "ASCII text" (and match). 

The problem this patch addresses is like this:
I have several source project directory with serveral million
files in them. I want to make a backup, however i want 
to only backup text files, (Makefiles, shell sripts, c and
h files etc). Currently I do something like this:
(for f in `find <srcdir> -type f`; do if (file $f | cut -d: -f2 | grep text &> /dev/null ); then echo $f; fi; done) > file.list
Then I use file.list to create a tar.
But, this pipe is very slow (I run it over night so it works but...).
With the above patch I can do:
find <stcdir> -dtype .*text.* >file.list 
This version is a magnitude faster...

As for I'm not really familiar with patch subbmissions I
send this short and easy to understand patch so that 
a developer can integrate it himself. I guess having a 
content datatype match is quite useful...
 
To make it compile you need to have libmagic (from the "file" 
command distribtuin) installed. Then I added LIBS=-lmagic to the
configure call. I guess in reality you should have a --with-magic
or so configure option etc.

Fix bug #27213: avoid failed assertions fornon-executable directories

James Youngman — 2010-04-04T16:28:05

This sequence of patches fixes bug #27213 without issuing a duplicate
error message.  Clearly a NEWS file entry will also be required.


 ChangeLog            |   64 ++++++++++++++++++++++++++++++++++++++
 build-aux/.cvsignore |    1 +
 build-aux/.gitignore |    1 +
 find/defs.h          |   12 ++++++-
 find/find.c          |    8 +++--
 find/ftsfind.c       |   53 +++++++++++++++-----------------
 find/parser.c        |   10 +++---
 find/pred.c          |    8 ++--
 find/sharefile.c     |    2 +-
 find/tree.c          |    1 +
 find/util.c          |   83 +++++++++++++++++++++++++++++++++++++++-----------
 11 files changed, 182 insertions(+), 61 deletions(-)

[PATCH] Bug #27213: avoid failed assertions fornon-executable directories.

Martin von Gagern — 2009-11-22T18:38:58

Addresses Savannah bug #27213 - https://savannah.gnu.org/bugs/?27213

This used to fail in recent releases:
1. mkdir -p foo/bar
2. chmod a-x foo
3. find foo
4. find foo -ls

Now it will print error messages for the denied permission on foo, but will
not abort completely, and in 3. will even print the name foo/bar, while it
won't do so in 4., as the -ls predicate requires stat information. For now,
4. will print two identical error message. This should get fixed some day.
---
 find/ftsfind.c |   36 ++++++++++++++----------------------
 find/util.c    |   11 +++++++----
 2 files changed, 21 insertions(+), 26 deletions(-)

diff --git a/find/ftsfind.c b/find/ftsfind.c
index 61f1079..a8efdda 100644
--- a/find/ftsfind.c
+++ b/find/ftsfind.c
< at >< at > -446,27 +446,19 < at >< at > consider_visiting(FTS *p, FTSENT *ent)
     }
   else if (ent->fts_info == FTS_NS)
     {
-      if (ent->fts_level == 0)
-{
-  /* e.g., nonexistent starting point */
-  error(0, ent->fts_errno, "%s",
-safely_quote_err_filename(0, ent->fts_path));
-  error_severity(1);/* remember problem */
-  return;
-}
-      else
-{
-  /* The following if statement fixes Savannah bug #19605
-   * (failure to diagnose a symbolic link loop)
-   */
-  if (symlink_loop(ent->fts_accpath))
-    {
-      error(0, ELOOP, "%s",
-    safely_quote_err_filename(0, ent->fts_path));
-      error_severity(1);
-      return;
-    }
-}
+      error(0, ent->fts_errno, "%s",
+    safely_quote_err_filename(0, ent->fts_path));
+      error_severity(1);
+      /* Continue despite the error, as file name without stat info
+       * might be better than not even processing the file name. This
+       * can lead to repeated error messages later on, though, if a
+       * predicate requires stat information.
+       *
+       * Not printing an error message here would be even more wrong,
+       * though, as this could cause the contents of a directory to be
+       * silently ignored, as the directory wouldn't be identified as
+       * such.
+       */
     }
 
   /* Cope with the usual cases. */
< at >< at > -474,7 +466,7 < at >< at > consider_visiting(FTS *p, FTSENT *ent)
       || ent->fts_info == FTS_NS /* e.g. symlink loop */)
     {
       assert (!state.have_stat);
-      assert (ent->fts_info == FTS_NSOK || state.type != 0);
+      assert (ent->fts_info == FTS_NSOK || state.type == 0);
       mode = state.type;
     }
   else
diff --git a/find/util.c b/find/util.c
index 4a0bb9b..0e5988f 100644
--- a/find/util.c
+++ b/find/util.c
< at >< at > -214,6 +214,9 < at >< at > get_statinfo (const char *pathname, const char *name, struct stat *p)
 {
   if (!options.ignore_readdir_race || (errno != ENOENT) )
     {
+              /* FIXME: this error message might repeat the one from
+               * the FTS_NS case in consider_visiting. How to avoid this?
+               */
       error (0, errno, "%s",
      safely_quote_err_filename(0, pathname));
       state.exit_status = 1;
< at >< at > -270,6 +273,10 < at >< at > get_info (const char *pathname,
       int result = get_statinfo(pathname, state.rel_pathname, p);
       if (result != 0)
 {
+  return -1;/* failure. */
+}
+      else
+{
   /* Verify some postconditions.  We can't check st_mode for
      non-zero-ness because of Savannah bug #16378 (which is
      that broken NFS servers can return st_mode==0). */
< at >< at > -281,10 +288,6 < at >< at > get_info (const char *pathname,
     {
       assert (p->st_ino);
     }
-  return -1;/* failure. */
-}
-      else
-{
   return 0;/* success. */
 }
     }

lib/check-regexprops.sh portability fixes forFreeBSD

James Youngman — 2009-08-16T15:18:38

I just applied and pushed the attached patch.

* lib/check-regexprops.sh: Portability fixes for FreeBSD (mktemp
requires an argument, dialect differences in sed regexes).
_______________________________________________
Findutils-patches mailing list
Findutils-patches< at >gnu.org
http://lists.gnu.org/mailman/listinfo/findutils-patches