Sender: rich AT phekda DOT freeserve DOT co DOT uk Message-ID: <3E2AF372.45EC0230@phekda.freeserve.co.uk> Date: Sun, 19 Jan 2003 18:50:26 +0000 From: Richard Dawe X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.2.23 i586) X-Accept-Language: de,fr MIME-Version: 1.0 To: djgpp-workers AT delorie DOT com Subject: Re: perl 5.8.0 from w2k test site problems References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Reply-To: djgpp-workers AT delorie DOT com Hello. Eli Zaretskii wrote: > > On Sun, 19 Jan 2003 lnobody AT delorie DOT com wrote: > > > VitalModuleBar.3 > > VitalModuleBaz.3 > > 12345678.123 > > -> > > VitalMod.3 for both. > > Yes, but are those real-life examples? How many Perl man pages actually > clash like that? Using the scheme above (collapsing ::) and just by looking at file and module names: ExtUtils::* ExtUtils::*::* File::Spec::* Locale::* Math::Big[FI]* Memoize::[ES]* Pod::Parse* Test::Harness::* Time::[Ll]ocal* io::Socket::* ~20, I think. > In general, man pages whose file names exceed 8+3 limits are nothing new; > that's why I made `man' support truncated file names. The relly > important question is: how many conflicts do we have to resolve in a > given package. In Groff, I had maybe 3 such clashes, so I renamed the > clashing pages to something close. The problem then is that the users have to know about the mappings. Really it should just work out-of-the-box. I'm starting to think a map file would be the best solution. The scheme I proposed of mapping '::' is a too Perl-man-page-specific. Each package could have its own map file, perhaps under man/map. 'man ExtUtils::MM_Win95' would be looked up normally. Once that fails man would go through man/map in the order it normally does for , loading the map file and looking for matches. Then we can store the man pages in whichever format is convenient - collapsing :: or :: -> slashes - whatever. Bye, Rich =] -- Richard Dawe [ http://www.phekda.freeserve.co.uk/richdawe/ ]