Findutils 4.2.0 is released on alpha.gnu.org

[James Youngman]

I'm happy to announce the availability of findutils version 4.2.0 at
ftp://alpha.gnu.org/gnu/findutils.  The NEWS file for this release
says :-

* Major changes in release 4.2.0
** Functionality Changes
*** xargs now works better when the environment variables are very extensive.
    The xargs command used to run into difficulties if the environment 
    data contained more than 2480 bytes.
*** New options -wholename and -iwholename 
    As per the GNU Projecvt coding standard, These are preferred over 
    the -path and -ipath options.  Using -ipath now generates a warning,
    though -path does not (since HPUX also offers -path).   
*** The environment variable FIND_BLOCK_SIZE is now ignored.
*** New option "-ignore_readdir_race"
    silences an error messages which would otherwise occur if a file is removed
    after find has read it from the directory using readdir(), but before 
    find stats the file.  There is also an option
    -noignore_readdir_race which has the opposite effect.   
**  Documentation improvements
*** The -size option of find is now documented in more detail
*** POSIX compliance and GNU extensions
    The find manual page also now includes a section 
    which describes the relationship between the features of GNU 
    find and the POSIX standard.  Some other small improvements 
    to the find and xargs manual pages have been made.
*** The argument to the -fprintf directive is now better documented.
    The escape code '\0' for the `-printf' predicate of find is now
    documented, and the documentation for the %k and %b specifiers
    has been improved.  
*** xargs -i is now more clearly documented.
** Bug Fixes
*** locate 'pa*d' will now find /etc/passwd (if it exists, of course)
*** xargs standard input is not inherited by child processes
    If the command invoked by xargs reads from its standard input, 
    it now gets nothing, as opposed to stealing data from the 
    list of files that xargs is trying to read.
*** Better support for 64-bit systems.
*** The command "xargs -i -n1" now works as one might expect, 
    I think this is a strange thing to want to do.  
*** Arguments to find -mtime that are too large are now diagnosed 
    Previously, this just used to cause find just to do the wrong thing.
*** updatedb is now somewhat more robust
    The updatedb shell script now does not generate an empty 
    database if it fails.  
*** Sanity-check on some data read from locatedb
    Locate now detects some types of file corruption in the 
    locate database.
*** The %k format specifier for -fprintf now works
    This was broken in 4.1.20.

Many of you will recall that over the last few months there have been
a number of reports of defects that were already fixed in the
development code.  Well, all those fixes are present in this release.
The notable ones are :-

 #3992  xargs leaves stdin open and unchanged 
 #4295  Outdated config.guess and config.sub 
 #4298  find -regex is either not working properly or not documented properly 
 #4380  updatedb: produces invalid db if external commands fail. 
 #4391  Race condition in find causes spurious error messages in updatedb
 #5034  Documentation for '-printf %k' insufficient 
 #5087  Documentation for find -size does not explain use of +/- 
 #8623  locate fails to check input data with getline 
 #9494  xargs -s123 is broken in CVS. 
 #9833  updatedb should change to / before running. 
 #9923  get_short returns incorrect values for -ve inputs on Solaris 8 
 #10533 Change from whoami to id -u breaks on Solaris 


Regards,
James Youngman