On Jan 22, 2008 7:16 PM, Nick Pilon <[EMAIL PROTECTED]> wrote:
> On Jan 21, 2008 7:34 PM, Axel Liljencrantz <[EMAIL PROTECTED]> wrote:
> > One simple solution is to disable with the X clipboard by sending
> > --without-xsel to configure. This means fish won't share the clipboard
> > with X, butmaybe that isn't such a big deal? Does X share the
> > clipboard with the rest of the GUI?
>
> Yes, it does. I'm not particularly worried about the clipboard
> functionality. If I want to copy something from a terminal, I'll grab
> the mouse and copy it from the terminal. I just don't want X starting
> every time I clear a line of text. --without-xsel sounds like it'll do
> the trick, thanks!
>

Happy to help.

> > That's weird. According to darcs, the last change to that file is from
> > 2007-01-07, which is before the previos fish release...
> >
> > Perhaps the output of apropos is what has changed?
>
> After poking around at it some more, you're definitely right. The
> problem is that there's more broken apropos formats that aren't parsed
> right. The two that I've managed to track down are:
>
> 1) Man pages for commands that are in subsections. This is easy enough
> to fix. For example:
> lastwords(1m)            - print syscalls before exit. Uses DTrace
>
> This is clearly a valid command, so I think the solution is to match
> the section with ([18][[:alpha:]]*) instead of just ([18]).

Agreed. Patch? :-)

>
> 2) Man pages with alternative names with just plain weird formatting.
> Perl seems to be the biggest culprit here:
> perllexwarn(1), Xref "warning(1), lexical warnings warning"
> perllexwarn(1) - Perl Lexical Warnings
>
> I'm not even sure how to fix this. The easiest answer is to be a lot
> more strict about what we accept as a "command name" (IE,
> alphanumerics plus dashes and underscores only, no spaces or symbols,
> or even just no spaces), but I'm not sure if that excludes actual
> in-use command names. Is this a sane restriction?

I don't have a problem with simply ignoring lines that are too weirdly formated.

>
> Ideally, we wouldn't have to parse apropos output to get this
> information, since it varies widely from system to system and even
> program to program. It's particularly funky on Solaris. I can't find
> any other way to get at this information, so I guess we're stuck with
> it.

Yup, the underlying database for whatis and apropos varies between
systems. Apropos is pretty much the state of the art in portability in
this area, sad as it might seem.

>
> > Actually, it's a variation on a general bug. PWD might have the wrong
> > value under any platform. :-/
>
> Well, it seems to be fixed in darcs now, so I guess it's no longer a problem?

I hope so!

>
> Except prompt_pwd was broken in darcs and was printing every path
> element twice. IE, /usr/usr. So I've fixed that instead and sent the
> patch to the list.

Oops. Thanks.

Axel

>
> --
> -Nick Pilon
>

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Fish-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/fish-users

Reply via email to