Re: completion: highlight matching part

Re: completion: highlight matching part

[Bart Schaefer]

[> workers]

On Aug 3, 10:31am, Bart Schaefer wrote:
} Subject: Re: completion: highlight matching part
}
} The complist module installs some defaults if $LS_COLORS is empty.  The
} problem is that the form beginning with an equal or a star is supposed
} to take precedence over those defaults, but it does not.  Instead (if
} I'm reading the code correctly) it takes precedence only over explicit
} settings of all the possible $LS_COLORS colorings.

I've tracked this (by a much more roundabout route than necessary) to
PWS's patch 25006, in which he asserts (for compatibility with GNU ls)
that "File type tests from stat should come before extension tests."

The documentation still says that extension tests win and file type
tests come last, so this should have been changed at the same time
that complist.c was modified.  But clearly in this instance we want
pattern tests to take precedence over file type tests.  I don't think
there's any equivalent in GNU ls to this particular zsh-ism, so we
have several choices:

(1) Change complist.c:putfilecol() so that pattern tests come first,
    then mode tests, and finally extension tests.

(2) Do something convoluted where we check for an extension match,
    but if we find one, try the mode tests before returning the
    extension color.  We end up with a sort of rock-paper-scissors
    behavior, where modes beat extensions beat patterns beat modes.

(3) Simply update the doc and make it impossible to do what Tomasz
    originally asked for, except in old versions of the shell that
    are not GNU ls compatible.

(4) Revert to the pre-25006 behavior.

Obviously the doc needs updating no matter what (in case 4, to point
out that we're not compatible with ls).

I have the impression that prior to this discussion the backref-using
patterns for completion coloring were not particularly widespread, so
it probably won't cause any major upheaval whichever way we go.

Re: completion: highlight matching part

[Peter Stephenson]

On Sun, 03 Aug 2008 17:33:21 -0700
Bart Schaefer <schaefer@brasslantern.com> wrote:
> (1) Change complist.c:putfilecol() so that pattern tests come first,
>     then mode tests, and finally extension tests.

I should think this is the right thing to do, isn't it?

-- 
Peter Stephenson <p.w.stephenson@ntlworld.com>
Web page now at http://homepage.ntlworld.com/p.w.stephenson/

Re: completion: highlight matching part

[Bart Schaefer]

On Aug 4,  8:44pm, Peter Stephenson wrote:
} Subject: Re: completion: highlight matching part
}
} On Sun, 03 Aug 2008 17:33:21 -0700
} Bart Schaefer <schaefer@brasslantern.com> wrote:
} > (1) Change complist.c:putfilecol() so that pattern tests come first,
} >     then mode tests, and finally extension tests.
} 
} I should think this is the right thing to do, isn't it?

I agree, but wanted to mention the other options because this is arguably
the most distant from what zsh used to do (before workers/25006).

Re: completion: highlight matching part

[Bart Schaefer]

On Aug 4,  4:54pm, Bart Schaefer wrote:
} Subject: Re: completion: highlight matching part
}
} On Aug 4,  8:44pm, Peter Stephenson wrote:
} } Subject: Re: completion: highlight matching part
} }
} } On Sun, 03 Aug 2008 17:33:21 -0700
} } Bart Schaefer <schaefer@brasslantern.com> wrote:
} } > (1) Change complist.c:putfilecol() so that pattern tests come first,
} } >     then mode tests, and finally extension tests.
} } 
} } I should think this is the right thing to do, isn't it?
} 
} I agree

Here's the patch, mostly so you can proof my doc change.

Index: Doc/Zsh/mod_complist.yo
===================================================================
RCS file: /cvsroot/zsh/zsh/Doc/Zsh/mod_complist.yo,v
retrieving revision 1.24
diff -u -r1.24 mod_complist.yo
--- Doc/Zsh/mod_complist.yo	5 Jul 2008 19:55:29 -0000	1.24
+++ Doc/Zsh/mod_complist.yo	5 Aug 2008 00:15:00 -0000
@@ -103,9 +103,9 @@
 pattern; the tt(EXTENDED_GLOB) option will be turned on for evaluation
 of the pattern.  The var(value) given for this pattern will be used for all
 matches (not just filenames) whose display string are matched by
-the pattern.  Definitions for both of these take precedence over the
-values defined for file types and the form with the leading asterisk 
-takes precedence over the form with the leading equal sign.
+the pattern.  Definitions for the form with the leading equal sign take
+precedence over the values defined for file types, which in turn take
+precedence over the form with the leading asterisk (file extensions).
 
 The last form also allows different parts of the displayed
 strings to be colored differently.  For this, the pattern has to use the
Index: Src/Zle/complist.c
===================================================================
RCS file: /cvsroot/zsh/zsh/Src/Zle/complist.c,v
retrieving revision 1.117
diff -u -r1.117 complist.c
--- Src/Zle/complist.c	5 Jul 2008 19:55:29 -0000	1.117
+++ Src/Zle/complist.c	5 Aug 2008 00:15:00 -0000
@@ -878,6 +878,21 @@
     Extcol ec;
     Patcol pc;
 
+    nrefs = MAX_POS - 1;
+
+    for (pc = mcolors.pats; pc; pc = pc->next)
+	if ((!pc->prog || !group || pattry(pc->prog, group)) &&
+	    pattryrefs(pc->pat, n, -1, -1, 0, &nrefs, begpos, endpos)) {
+	    if (pc->cols[1]) {
+		patcols = pc->cols;
+
+		return 1;
+	    }
+	    zlrputs(pc->cols[0]);
+
+	    return 0;
+	}
+
     if (special != -1) {
 	colour = special;
     } else if (S_ISDIR(m)) {
@@ -920,21 +935,6 @@
 	    return 0;
 	}
 
-    nrefs = MAX_POS - 1;
-
-    for (pc = mcolors.pats; pc; pc = pc->next)
-	if ((!pc->prog || !group || pattry(pc->prog, group)) &&
-	    pattryrefs(pc->pat, n, -1, -1, 0, &nrefs, begpos, endpos)) {
-	    if (pc->cols[1]) {
-		patcols = pc->cols;
-
-		return 1;
-	    }
-	    zlrputs(pc->cols[0]);
-
-	    return 0;
-	}
-
     zcputs(group, COL_FI);
 
     return 0;

Re: completion: highlight matching part

[Peter Stephenson]

Bart Schaefer wrote:
> On Aug 4,  4:54pm, Bart Schaefer wrote:
> } Subject: Re: completion: highlight matching part
> }
> } On Aug 4,  8:44pm, Peter Stephenson wrote:
> } } Subject: Re: completion: highlight matching part
> } }
> } } On Sun, 03 Aug 2008 17:33:21 -0700
> } } Bart Schaefer <schaefer@brasslantern.com> wrote:
> } } > (1) Change complist.c:putfilecol() so that pattern tests come first,
> } } >     then mode tests, and finally extension tests.
> } } 
> } } I should think this is the right thing to do, isn't it?
> } 
> } I agree
> 
> Here's the patch, mostly so you can proof my doc change.

I can't see any problem.  I think it's one of those things (like
DEBUG_BEFORE_CMD) where nobody knew it wasn't the best way of doing it
until they tried it.

-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR PLC, Churchill House, Cambridge Business Park, Cowley Road
Cambridge, CB4 0WZ, UK                          Tel: +44 (0)1223 692070

Re: completion: highlight matching part

[Rocky Bernstein]

On Tue, Aug 5, 2008 at 4:13 AM, Peter Stephenson <pws@csr.com> wrote:
> Bart Schaefer wrote:
>> On Aug 4,  4:54pm, Bart Schaefer wrote:
>> } Subject: Re: completion: highlight matching part
>> }
>> } On Aug 4,  8:44pm, Peter Stephenson wrote:
>> } } Subject: Re: completion: highlight matching part
>> } }
>> } } On Sun, 03 Aug 2008 17:33:21 -0700
>> } } Bart Schaefer <schaefer@brasslantern.com> wrote:
>> } } > (1) Change complist.c:putfilecol() so that pattern tests come first,
>> } } >     then mode tests, and finally extension tests.
>> } }
>> } } I should think this is the right thing to do, isn't it?
>> }
>> } I agree
>>
>> Here's the patch, mostly so you can proof my doc change.
>
> I can't see any problem.  I think it's one of those things (like
> DEBUG_BEFORE_CMD) where nobody knew it wasn't the best way of doing it
> until they tried it.

Interesting. You mean there isn't going to be an option to keep
compatibility with prior versions of zsh?  ;-)
>
> --
> Peter Stephenson <pws@csr.com>                  Software Engineer
> CSR PLC, Churchill House, Cambridge Business Park, Cowley Road
> Cambridge, CB4 0WZ, UK                          Tel: +44 (0)1223 692070
>

Re: completion: highlight matching part

[Peter Stephenson]

"Rocky Bernstein" wrote:
> Interesting. You mean there isn't going to be an option to keep
> compatibility with prior versions of zsh?  ;-)

This is buried right down in the details of completion where teams of
scientists have worked for years to determine what's actually going on
and where no one else dares to change anything.  It's hard enough
keeping it compatible with itself.  I certainly don't put this in the
same class as features directly visible to the normal shell programmer.

-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR PLC, Churchill House, Cambridge Business Park, Cowley Road
Cambridge, CB4 0WZ, UK                          Tel: +44 (0)1223 692070