Re: [Bug-readline] libreadline.so has rl_sort_completion_matche instead of rl_sort_completion_matches

[Chet Ramey]

On 7/25/13 3:11 PM, Sébastien Boisvert wrote:
> Hello,
> 
> This is a bug report for readline.
> 
> 
> (1) The version number and release status of Readline (e.g., 4.2-release)
> 
>   6.2 from ftp://ftp.cwru.edu/pub/bash/readline-6.2.tar.gz
> 
> 
> (2) The machine and OS that it is running on
> 
>   machine: Lenovo Thinkpad X230
>   OS: Fedora 18 with Linux 3.9.9-201.fc18.x86_64
> 
> 
> (3) A list of the compilation flags or the contents of `config.h', if
> appropriate
> 
>   command: ./configure --prefix=$(pwd)/1; make; make install
> 
> 
> (4) a description of the bug
> 
>   libreadline.so has rl_sort_completion_matche instead of
> rl_sort_completion_matches
> 
> $ readelf -s 1/lib/libreadline.so|grep rl_sort_completion_matche
>    511: 000000000023fca8     4 OBJECT  GLOBAL DEFAULT   23
> rl_sort_completion_matche
>    488: 000000000023fca8     4 OBJECT  GLOBAL DEFAULT   23
> rl_sort_completion_matche
> 
> The header has the correct name (with a s):

This is not reflected by the source code:

$ pwd
/usr/src/local/readline/readline-6.2-patched
$ grep rl_sort_completion_matche *.[ch]
complete.c:int rl_sort_completion_matches = 1;
complete.c:  if (i && rl_sort_completion_matches)
complete.c:  if (rl_ignore_completion_duplicates == 0 &&
rl_sort_completion_matches)
readline.h:extern int rl_sort_completion_matches;

It's not clear what the problem is, but it appears to have been introduced
by the library creation process.

Chet
-- 
``The lyf so short, the craft so long to lerne.'' - Chaucer
                 ``Ars longa, vita brevis'' - Hippocrates
Chet Ramey, ITS, CWRU    address@hidden    http://cnswww.cns.cwru.edu/~chet/