Re: compadd -f with IPREFIX.

Re: compadd -f with IPREFIX.

[Sven Wischnowsky]

Tanaka Akira wrote:

> heredoc> compadd -f -r ':/ \t\n\-' - *
> 
> is27e1u11% tst :Completion/
> 
> There is no problem in above operation.
> But, I put ":", "/" is retained as follows.
> 
> is27e1u11% tst :Completion/:
> 
> I think that "/" should be removed.

Currently -r and -R only work on user-added suffixes given with -S.
This is the way -q always worked (I didn't change anything ;-). It is
also stated explicitly in the docs.

So the question to everyone is: should we change this to make -[rR]
work even on automatically added suffixes?

Bye
 Sven


--
Sven Wischnowsky                         wischnow@informatik.hu-berlin.de

Re: compadd -f with IPREFIX.

[Tanaka Akira]

In article <199905180954.LAA16898@beta.informatik.hu-berlin.de>,
  Sven Wischnowsky <wischnow@informatik.hu-berlin.de> writes:

> Currently -r and -R only work on user-added suffixes given with -S.
> This is the way -q always worked (I didn't change anything ;-). It is
> also stated explicitly in the docs.

Hmm.

> So the question to everyone is: should we change this to make -[rR]
> work even on automatically added suffixes?

It's consistent and useful, I suppose.
-- 
Tanaka Akira

Re: compadd -f with IPREFIX.

[Sven Wischnowsky]

Bart Schaefer wrote:

> Didn't you say the documentation explicitly says this won't happen?  So
> shouldn't there be some kind of documentation patch to go with the change?

Ahem. Thanks.

Bye
 Sven

diff -u od/Zsh/compwid.yo Doc/Zsh/compwid.yo
--- od/Zsh/compwid.yo	Thu May 20 08:32:30 1999
+++ Doc/Zsh/compwid.yo	Mon May 31 09:02:10 1999
@@ -424,7 +424,8 @@
 )
 item(tt(-r) var(remove-chars))(
 This is a more versatile form of the tt(-q) option.
-The suffix given with tt(-S) will be automatically removed if
+The suffix given with tt(-S) or the slash automatically added after
+completing directories will be automatically removed if
 the next character typed inserts one of the characters given in the
 var(remove-chars).  This string is parsed as a characters class and
 understands the backslash sequences used by the tt(print) command.  For
@@ -436,8 +437,8 @@
 as `tt(-S "=" -r "= \t\n\-")'.
 )
 item(tt(-R) var(remove-func))(
-This is another form of the tt(-r) option. When a suffix given with the
-tt(-S) option has been inserted and the completion accepted, the function
+This is another form of the tt(-r) option. When a suffix 
+has been inserted and the completion accepted, the function
 var(remove-func) will be called after the next character typed.  It is
 passed the length of the suffix as an argument and can use the special
 parameters available in ordinary (non-completion) zle widgets (see

--
Sven Wischnowsky                         wischnow@informatik.hu-berlin.de

Re: compadd -f with IPREFIX.

[Bart Schaefer]

On May 21,  3:18pm, Sven Wischnowsky wrote:
} Subject: Re: compadd -f with IPREFIX.
}
} Tanaka Akira wrote:
} 
} > Sven Wischnowsky <wischnow@informatik.hu-berlin.de> writes:
} > > So the question to everyone is: should we change this to make -[rR]
} > > work even on automatically added suffixes?
} > 
} > It's consistent and useful, I suppose.
} 
} Yes, maybe.

Didn't you say the documentation explicitly says this won't happen?  So
shouldn't there be some kind of documentation patch to go with the change?

-- 
Bart Schaefer                                 Brass Lantern Enterprises
http://www.well.com/user/barts              http://www.brasslantern.com

Re: compadd -f with IPREFIX.

[Sven Wischnowsky]

Tanaka Akira wrote:

> In article <199905180954.LAA16898@beta.informatik.hu-berlin.de>,
>   Sven Wischnowsky <wischnow@informatik.hu-berlin.de> writes:
> > So the question to everyone is: should we change this to make -[rR]
> > work even on automatically added suffixes?
> 
> It's consistent and useful, I suppose.

Yes, maybe. And since this is only with explicitly given characters or 
functions to remove the suffix, it won't interfere with previous
behavior anyway.

Bye
 Sven

diff -u os/Zle/zle_tricky.c Src/Zle/zle_tricky.c
--- os/Zle/zle_tricky.c	Thu May 20 08:32:47 1999
+++ Src/Zle/zle_tricky.c	Fri May 21 15:13:30 1999
@@ -7112,9 +7112,13 @@
 		menuinsc++;
 		if (menuwe)
 		    menuend++;
-		if ((!menucmp || menuwe) && isset(AUTOREMOVESLASH)) {
-		    makesuffix(1);
-		    suffixlen['/'] = 1;
+		if (!menucmp || menuwe) {
+		    if (m->remf || m->rems)
+			makesuffixstr(m->remf, m->rems, 1);
+		    else if (isset(AUTOREMOVESLASH)) {
+			makesuffix(1);
+			suffixlen['/'] = 1;
+		    }
 		}
 	    }
 	}

--
Sven Wischnowsky                         wischnow@informatik.hu-berlin.de

Re: compadd -f with IPREFIX.

[Tanaka Akira]

In article <199905180715.JAA15766@beta.informatik.hu-berlin.de>,
  Sven Wischnowsky <wischnow@informatik.hu-berlin.de> writes:

> addmatches() used the IPREFIX it got also as `ripre' (the expanded
> form of the ignored prefix -- this is used for things like `~' and
> `$foo/<TAB>'). The patch makes it use a NULL pointer there.

Thanks. But I found another problem.

Z(2):akr@is27e1u11% ./Src/zsh -f
is27e1u11% cat <<'End' > Completion/User/_tst
heredoc> #compdef tst
heredoc> compset -P :
heredoc> compadd -f -r ':/ \t\n\-' - *
heredoc> End
is27e1u11% fpath=($PWD/Completion/*(/))
is27e1u11% . Completion/Core/compinit
is27e1u11% tst :Compl<TAB>

Then, I get following.

is27e1u11% tst :Completion/

There is no problem in above operation.
But, I put ":", "/" is retained as follows.

is27e1u11% tst :Completion/:

I think that "/" should be removed.

# The aim is colon-separated directories completion.
# compset -P '*:'; _path_files -r ':/ \t\n\-' -/
-- 
Tanaka Akira

Re: compadd -f with IPREFIX.

[Sven Wischnowsky]

Tanaka Akira wrote:

> is27e1u11% cat <<'End' > Completion/User/_tst
> heredoc> #compdef tst
> heredoc> compset -P :
> heredoc> compadd -f - *
> heredoc> End
> 
> then, I get following.
> 
> is27e1u11% tst :Completion
> 
> I think zsh should complete a suffix "/".

addmatches() used the IPREFIX it got also as `ripre' (the expanded
form of the ignored prefix -- this is used for things like `~' and
`$foo/<TAB>'). The patch makes it use a NULL pointer there.

Bye
 Sven

diff -u os/Zle/zle_tricky.c Src/Zle/zle_tricky.c
--- os/Zle/zle_tricky.c	Sat May  8 14:31:49 1999
+++ Src/Zle/zle_tricky.c	Tue May 18 09:07:49 1999
@@ -3734,7 +3735,7 @@
 		    continue;
 		}
 		if (doadd) {
-		    cm = add_match_data(isalt, ms, lc, dat->ipre, dat->ipre,
+		    cm = add_match_data(isalt, ms, lc, dat->ipre, NULL,
 					dat->isuf, dat->pre, dat->prpre,
 					dat->ppre, dat->psuf, dat->suf,
 					bpl, bsl, dat->flags, isexact);

--
Sven Wischnowsky                         wischnow@informatik.hu-berlin.de

compadd -f with IPREFIX.

[Tanaka Akira]

I found a problem with zsh-3.1.5-pws-18.

Z:akr@is27e1u11% ./Src/zsh -f
is27e1u11% cat <<'End' > Completion/User/_tst
heredoc> #compdef tst
heredoc> compset -P :
heredoc> compadd -f - *
heredoc> End
is27e1u11% fpath=($PWD/Completion/*(/))
is27e1u11% . Completion/Core/compinit
is27e1u11% tst :Compl<TAB>

then, I get following.

is27e1u11% tst :Completion

I think zsh should complete a suffix "/".

Also, truss reports that zsh stat ":Completion" as following.

stat64(":Completion", 0xFFBEEF00)               Err#2 ENOENT
-- 
Tanaka Akira