regex maintenance?

[Karl Berry]

rms sent me this note.  I figured this was the group of people most
likely to have someone able and willing to tackle some regex work.
Anyone want to volunteer?

(Having worked on regex for a couple years back in the late 80's, I have
zero desire to go there again, even for function prototypes. :)

Thanks,
karl


Date: Tue, 26 Dec 2006 21:59:46 -0500
From: Richard Stallman <address@hidden>
To: address@hidden
Subject: GNU regex

GNU regex needs some maintenance.  Can you find someone who would like
to do it?

Here's a message which talks about what work is needed.



Date: Sun, 09 Jul 2006 09:26:01 -0700
From: (Doc) drclue !at! drclue <dot> n e t 
MIME-Version: 1.0
To: address@hidden
Subject: Re: Well got the RegEx library to copile

While I did have to move on to a  different library, I did want to
at least give you answers to some of you questions





Richard Stallman wrote:
>     I actually got this to compile ( eventually )
>     http://ftp.gnu.org/pub/gnu/regex/regex-0.12.tar.gz
>
> On what platform?  What compiler did you use, and what version?
> (It certainly compiled when we released it.)
>
>   

OS       :Debian GNU/Linux, kernel 2.6.15-k7
COMPILER :gcc 4.0.3
SHELL    :GNU bash, version 3.1.5(1)-release (i486-pc-linux-gnu)

In the new library I'm using  "boost" , I did have to edit one of it's 
header files
as it kicked up a warning about my compiler version.
( The compiler version was newer than the header expected it to be),


>     I've not tested it to see if is logically correct,
>     but at least after a lot of work I translated all the function calls to
>     a more traditional form ,
>
> I am not sure what that means; could you show me a specific example of
> a function call that you changed, before and after your change?
>
>   
void foo(a,b,c)
    int a,
    int b,
    int c
    {
    }


void foo(int a,int b,int c)
    {
   }

>      and moved all the declarations out
>     of the case statements.
Compilers complained often enough about declarations
in case statements and goto's straddling same that I simply no longer do it.

>  I also renamed the "not" variables
>     to bNot  which solved a lot of problems.
>
> That seems really strange.  There is no problem with `not' as a
> variable name.
>
>   
By changing the variable from "not" to "bNot" the compiler complaint 
went away.


> All these things lead me to wonder if you are compiling on a really
> strange platform, with unusual requirements.
>
>     There were also some parameter order errors
>     in a couple of functions.
>
> Could you show me them?
>
>   
I've already archived the materials , so I'm speaking from memory,
but one of the calling order things seem to be the fallowing ,
which might be some fancy code technique , but seemed like it
would need a cast to be so.

re_match_2(buff,string1,size1,string2, size2,...
    vs.
re_match_2(buff,string1,string2,size1,size2...

It's probably just that the coding style is so apart from
what I've been doing for the last 20 years or so , that it went over my 
head. :