Re: [rh-rpm] Case Sensitivity

[Simon J Mudd <sjmudd pobox com>]

k5tsrytx001@sneakemail.com ("Greg Wright") writes:

> >> And apart from the obvious, is there a reason why RPM does not
> keep/force
> >> all records in/to lower case ?
> >
> >both answers are the same -- Case matters in *nix, unlike less 
> >discerning operating systems. <grin>
> 
> Yes, I know it matters, but why should RPM not keep records just in lower
> case ? like who would even try having  mysql-1 and MySQL-1?, then an rpm -e
> or a SRPM build will not break when a req. is not found due to CAPs, so why
> not have RPM squash it to lower in the database on install.
> 
> For now, I was more interested in how best to handle this -- you cannot use
> {mysql,MySQL} in a PreReq as the spec is not a script.

I've been maintaining a postfix rpm which optionally includes support
for MySQL's mysql rpms and RedHat's mysql rpms.

The packages are different and the names are different too.

--- snip of part of my spec file ---
%if %{REDHATMYSQL}
Requires: mysql, mysqlclient9
BuildRequires: mysql, mysqlclient9, mysql-devel
%endif
 
%if %{MYSQL}
Requires: MySQL-shared, zlib
BuildRequires: MySQL-shared, MySQL-devel, zlib-devel
%endif
--- end ----------------------------

As you can see it may _in this case_ actually be a good thing that
there is mixed case support as I can distinguish between the two
packages, but in the general case it's much better NOT to have
multiple packages by different suppliers and then mixed-case support
would be less important.

Simon
--
Simon J Mudd,   Tel: +34-91-408 4878,  Mobile: +34-605-085 219
Madrid, Spain.  email: sjmudd@pobox.com,  Postfix RPM Packager