PATCH: Re: _netscape

PATCH: Re: _netscape

[Sven Wischnowsky]

Oliver Kiddle wrote:

> ...
> 
> Where things go wrong is when I don't use a quote:
> 
> netscape -remote open<tab> completes '\(', subsequent tabs insert more, I would
>     expect to see openFile and openURL listed
> netscape -remote openU<tab> works - completes to openURL\(
> netscape -remote openURL\(<tab> also works - it completes URLs.
> 
> The first of these is where it isn't workling. The thing which is missing is that I never get the list of matches, just the inserted brackets.

So why do we use `-s', then.

This'll do? 

Bye
 Sven

Index: Completion/User/_netscape
===================================================================
RCS file: /cvsroot/zsh/zsh/Completion/User/_netscape,v
retrieving revision 1.4
diff -u -r1.4 _netscape
--- Completion/User/_netscape	2000/04/11 07:57:57	1.4
+++ Completion/User/_netscape	2000/05/18 11:02:38
@@ -56,15 +56,9 @@
       fi
     ;;
     *)
-      if [[ -z "$QIPREFIX" ]]; then
-	_wanted commands expl 'remote commands' \
-  	    compadd -s'(' -S '' -M 'm:{a-zA-Z}={A-Za-z}' - \
-                    $remote_commands && ret=0
-      else
-	_wanted commands expl 'remote commands' \
-            compadd -qS '(' -M 'm:{a-zA-Z}={A-Za-z}' - \
-                    $remote_commands && ret=0
-      fi
+      _wanted commands expl 'remote commands' \
+  	  compadd -qS "${${QIPREFIX:+(}:-\(}" -M 'm:{a-zA-Z}={A-Za-z}' - \
+                  $remote_commands && ret=0
     ;;
   esac
 fi

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

Re: PATCH: Re: _netscape

[Oliver Kiddle]

Sven Wischnowsky wrote:
> 
> So why do we use `-s', then.

I'm not sure but it did used to work like that ages ago.

> This'll do?

That's great, thanks.

I've gone on to try to fix the handling of the closing bracket suffix in
netscape remote commands which has resulted in the following thoughts on
suffix handling in the new completion.

There are a few places where the contents of a remote command comes to a
definite finish: a file/ftp/http URL that ends with a file or empty
directory, a saveAs command after the file type, about: URLS. In these
cases, it would be nice if completing the last stage added both the
closing bracket (back-slash quoted if necessary) to terminate the remote
command and any closing quote.

Adding a closing quote sometimes happens, for example:
netscape -remote 'openURL(about:c<tab>
this completes to the about:cache URL with the closing quote appended.
Where does the quote come from: is the default suffix actually
"$compstate[quote] " instead of just a space?

I can basically get this to work by using
-S "${${QIPREFIX:+)}:-\)}$compstate[quote] "

One of the problems which this highlighted is that _urls didn't handle
being passed a -S option very well. Unlike many of the types completed
by helper functions, urls contain several components which are completed
separately (connection type, hostname, filename etc), all of which are
completed separately. _urls passes "$@" to _next_label in several places
so the -S option was passed at these points. This means that if you
complete urls with _urls -S ')', ht<tab> will complete to 'http://)',
i.e. the suffix is added yet the URL isn't finished. In the patch below,
what I have done is removed the "$@" for all the places where we are not
completing the final component of a URL. This is of course not ideal
because -S is not the only compadd option which might get passed around.
I think we need to work out a slightly better system for the handling of
suffixes like this. Basically, what a helper function needs to do is
take the suffix passed to it and when it is completing a final component
of itself, it should pass any suffix it wants with the one passed to it
appended. Pulling out -S options from "$@" is going to look messy
without some special handling at a lower level somewhere. compadd seems
to only use the first -S option it is passed but this isn't documented
so I didn't want to rely on it. If we do make it a documented feature
then we still need to handle the situation where you are completing the
final part of something and you want to use both your own suffix and the
one passed to the function.

Note that _files can no longer complete subdirectories if you give it a
-S option which is something other than a slash. _files should only add
the suffix after a file (and possible after an empty directory).

Returning to _netscape, the next issue was how to handle the closing
bracket/quote already being there.
I used compset -S '(|\\)\)*'. But this again comes up with a problem
when we are not completing the final part of a URL. The suffix is only
moved to ISUFFIX so for example:
netscape -remote 'openURL(a<tab>)    i.e. cursor is at position of tab
will complete the url to about:)' - advancing the cursor past the
suffix. I'm able to prevent this by not using the suffix with compadd
if the compset succeeded.

This example shows how and leads me on to my next point:

_foo() {
  local lsuf="" nsuf=""
  compset -S '.*' || lsuf=". "
  if compset -P "*:"; then
    compadd -S "$lsuf" aaa bbb ccc
  else
    compset -S ':*' || nsuf=":"
    compadd -S "$rsuf" one two three
  fi
}

foo o<tab>.  completes the one but moves the cursor past the '.'. I
would prefer if it was only moved to the end of the :. We started off
with just 'o.' so there was no :. this indicates to me that the part
which comes after a colon is necessary so it is likely that the user
will want to complete it. In the case of:
foo o<tab>:ddd.
it is fair enough in a way that the completion moves past the final '.'
because there was a colon there to start with although even in this case
I'd prefer it didn't. What I would actually want it to do in this case
is move to the end of the 'ddd' but before the '.'. If 'ddd' was
incomplete, I could finish it otherwise, another tab would advance me on
to the next thing. I also looked at how
_sep_parts -S "." '(one two three)' ':' '(aaa bbb ccc)'
works as a comparison but it won't complete one in the case of o<tab>.
which I suppose has its uses in different situations.

Handling suffixes composed of more than one character is also less than
ideal. For example, in the case of closing the netscape remote command,
there is the closing bracket and the closing quote. To work out the
appropriate suffix, you have to do:

if compset -S '")*"; then
  suf=""
elif compset -S "'*"; then
  suf=")"
else
  suf=")' "
fi

This is without checking compstate[quote]. Of course there may be some
cases where later letters in the suffix could be part of a match so
can't be used to delimit the end of it. I haven't bothered to do this
but if changes are made to suffix handling, it might be worth
considering. For an example of why this is useful, try completing a
remote command of 'openURL(about:c<tab>' . With the above example, it
will also not complete the closing quote if there is a closing bracket
to the right of the cursor (before pressing tab) but not a closing
quote.

I have added a patch to _netscape and _urls. Note that I removed the
compset -q which I think (as Sven said earlier), is not needed anymore.
The removal of the "$@"s from _urls isn't particularly ideal though so
that situation with suffixes needs some thought.

Oliver Kiddle

Index: Completion/User/_netscape
===================================================================
RCS file: /cvsroot/zsh/zsh/Completion/User/_netscape,v
retrieving revision 1.5
diff -u -r1.5 _netscape
--- Completion/User/_netscape	2000/05/18 11:05:20	1.5
+++ Completion/User/_netscape	2000/05/19 18:27:35
@@ -1,6 +1,6 @@
 #compdef netscape
 
-local curcontext="$curcontext" state line ret=1
+local curcontext="$curcontext" state line ret=1 suf
 typeset -A opt_args
 
 _x_arguments -C \
@@ -27,24 +27,27 @@
 [[ "$state" = "urls" ]] &&
   _files "$@" && return 0
 
-
 # Handle netscape remote commands
 if [[ "$state" = "remote" ]]; then  
   local -a remote_commands
   remote_commands=(openURL openFile saveAs mailto addBookmark)
 
-  [[ $compstate[quoting] = (double|single) ]] && compset -q
   compset -P '*\('
+  if compset -S '(|\\)\)*'; then
+    set - -S "" "$@"
+  else
+    set - -S"${${QIPREFIX:+)}:-\)}$compstate[quote] " "$@"
+  fi
   case $IPREFIX in
     openURL*|addBookmark*) state=urls;;
-    openFile*) _files -W ~;;
+    openFile*) _files "$@" -W ~;;
     saveAs*) 
       if compset -P "*,"; then
         _wanted types expl 'data type' \
-            compadd -s")" -M 'm:{a-zA-Z}={A-Za-z}' HTML Text PostScript &&
-                ret=0
+            compadd "$@" -M 'm:{a-zA-Z}={A-Za-z}' HTML Text PostScript && ret=0
       else
-        _files -W ~ && ret=0
+        compset -S ",*" || suf=","
+        _files -qS "$suf" -W ~ && ret=0
       fi
     ;;
     mailto*)
@@ -52,12 +55,14 @@
       if compset -P '*@'; then
         _wanted hosts expl 'remote host name' _hosts -q -S, && ret=0
       else
-        _wanted users expl 'login name' _users -q -S@ && ret=0
+        compset -S "@*" || suf="@"
+        _wanted users expl 'login name' _users -q -S "$suf" && ret=0
       fi
     ;;
     *)
+      compset -S '(|\\)\(*' || suf="${${QIPREFIX:+(}:-\(}"
       _wanted commands expl 'remote commands' \
-  	  compadd -qS "${${QIPREFIX:+(}:-\(}" -M 'm:{a-zA-Z}={A-Za-z}' - \
+          compadd -qS "$suf" -M 'm:{a-zA-Z}={A-Za-z}' - \
                   $remote_commands && ret=0
     ;;
   esac
@@ -67,14 +72,15 @@
   # Complete netscape urls
   if compset -P about: ; then
     _wanted values expl 'about what' \
-        compadd authors blank cache document fonts global hype image-cache \
+        compadd "$@" authors blank cache document fonts global hype image-cache \
             license logo memory-cache mozilla plugins && ret=0
   else
     _tags prefixes
     while _tags; do
       while _next_label prefixes expl 'URL prefix'; do
-        compadd "$expl[@]" -S '' about: mocha: javascript: && ret=0
         _urls "$@" && ret=0
+	compset -S '[^:]*'
+        compadd "$expl[@]" -S '' about: mocha: javascript: && ret=0
       done
       (( ret )) || return 0
     done
Index: Completion/User/_urls
===================================================================
RCS file: /cvsroot/zsh/zsh/Completion/User/_urls,v
retrieving revision 1.7
diff -u -r1.7 _urls
--- Completion/User/_urls	2000/05/19 08:26:48	1.7
+++ Completion/User/_urls	2000/05/19 18:27:35
@@ -57,7 +57,7 @@
 if ! compset -P '(#b)([-+.a-z0-9]#):'; then
   _tags -C argument prefixes
   while _tags; do
-    while _next_label prefixes expl 'URL prefix' "$@"; do
+    while _next_label prefixes expl 'URL prefix'; do
       [[ -d $urls_path/bookmark ]] &&
         compadd "$expl[@]" -S '' bookmark: && ret=0
       compadd "$expl[@]" -S '' file: ftp:// gopher:// http:// && ret=0
@@ -79,7 +79,7 @@
     if ! compset -P //; then
       _tags -C file files
       while _tags; do
-        while _next_label files expl 'local file' "$@"; do
+        while _next_label files expl 'local file'; do
           if [[ -prefix / ]]; then
 	    _path_files "$expl[@]" -S '' -g '*(^/)' && ret=0
 	    _path_files "$expl[@]" -S/ -r '/' -/ && ret=0
@@ -119,7 +119,7 @@
 
   _tags hosts
   while _tags; do
-    while _next_label hosts expl host "$@"; do
+    while _next_label hosts expl host; do
       (( $#uhosts )) || _hosts -S/ && ret=0
       [[ "$scheme" = http ]] && uhosts=($uhosts $localhttp_servername)
       compadd "$expl[@]" -S/ - $uhosts && ret=0
@@ -143,7 +143,7 @@
     user="$match[1]"
     while _tags; do
       while _next_label files expl 'local file'; do
-        _path_files "$expl[@]" -W ~$user/$localhttp_userdir -g '*(^/)' && ret=0
+        _path_files "$expl[@]" "$@" -W ~$user/$localhttp_userdir -g '*(^/)' && ret=0
         _path_files "$expl[@]" -W ~$user/$localhttp_userdir -S/ -r '/' -/ && ret=0
       done
       (( ret )) || return 0
@@ -151,7 +151,7 @@
   else
     while _tags; do
       while _next_label files expl 'local file'; do
-        _path_files "$expl[@]" -W $localhttp_documentroot -g '*(^/)' && ret=0
+        _path_files "$expl[@]" "$@" -W $localhttp_documentroot -g '*(^/)' && ret=0
         _path_files "$expl[@]" -W $localhttp_documentroot -S/ -r '/' -/ && ret=0
       done
       (( ret )) || return 0
@@ -160,7 +160,7 @@
 else
   while _tags; do
     while _next_label files expl 'local file'; do
-      _path_files "$expl[@]" -W $urls_path/$scheme/$host -g '*(^/)' && ret=0
+      _path_files "$expl[@]" "$@" -W $urls_path/$scheme/$host -g '*(^/)' && ret=0
       _path_files "$expl[@]" -W $urls_path/$scheme/$host -S/ -r '/' -/ && ret=0
     done
     (( ret )) || return 0

Re: PATCH: Re: _netscape

[Sven Wischnowsky]

Oliver Kiddle wrote:

> ...
> 
> > > Note that _files can no longer complete subdirectories if you give it a
> > > -S option which is something other than a slash. _files should only add
> > > the suffix after a file (and possible after an empty directory).
> 
> > Sometimes you want it, sometimes not...
> 
> When might you not want it (the slash)? If you don't want to decend
> directories, it would be easier to use compadd * than _files. I'd have
> thought that in the vast majority of cases, the passed suffix would be
> wanted only after a file or empty directory.

When completing only directories.

> ...
> 
> > The way suffixes are handled has mostly historical reasons and, yes,
> > it is a bit simple minded. If someone wants to help to improve it, I'd
> > like to have a short and comprehensive list of things one might want
> > to do, so that we can discuss it easliy and find the (hopefully few)
> > basic things we need for that. With that we could then think about
> > changing the code. Ok? (I don't think I'll have much time to think
> > about this...)
> 
> I'll try to come up with such a list but it may be a little while before
> I have time. Does anyone know of any parts of the completion system
> where there is an interesting sitation or currently not very good
> handling of suffixes: it might be useful if I investigate a few
> different instances first.

You have every time you want or need, I don't think we really need to
do this before 3.1.7.

And, no, off the top of my head I can't think of places where
suffix-handling caused that much trouble.

Oh, and, btw, when thinking about suffix handling, we should probably
also re-think the -R stuff (auto-removal in general). Maybe we find
something better, maybe we don't...

Bye
 Sven


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

Re: PATCH: Re: _netscape

[Oliver Kiddle]

Sven Wischnowsky wrote:
> 
> > compadd seems
> > to only use the first -S option it is passed but this isn't documented
> > so I didn't want to rely on it.
> 
> It *is* documented, in the compadd entry, below the options.
> _approximate always relied on it.

That's useful. The patch below makes use of this for url completion. It
doesn't seem to work in _path_files though.

> > Note that _files can no longer complete subdirectories if you give it a
> > -S option which is something other than a slash. _files should only add
> > the suffix after a file (and possible after an empty directory).

> Sometimes you want it, sometimes not...

When might you not want it (the slash)? If you don't want to decend
directories, it would be easier to use compadd * than _files. I'd have
thought that in the vast majority of cases, the passed suffix would be
wanted only after a file or empty directory.

> You may also want to have a look at $compstate[to_end]...

Thanks, that handles some situations but more control would be nice (at
least moving to the end of the match plus suffix).

> The way suffixes are handled has mostly historical reasons and, yes,
> it is a bit simple minded. If someone wants to help to improve it, I'd
> like to have a short and comprehensive list of things one might want
> to do, so that we can discuss it easliy and find the (hopefully few)
> basic things we need for that. With that we could then think about
> changing the code. Ok? (I don't think I'll have much time to think
> about this...)

I'll try to come up with such a list but it may be a little while before
I have time. Does anyone know of any parts of the completion system
where there is an interesting sitation or currently not very good
handling of suffixes: it might be useful if I investigate a few
different instances first.

Oliver

Index: Completion/User/_netscape
===================================================================
RCS file: /cvsroot/zsh/zsh/Completion/User/_netscape,v
retrieving revision 1.6
diff -u -r1.6 _netscape
--- Completion/User/_netscape	2000/05/19 18:44:22	1.6
+++ Completion/User/_netscape	2000/05/23 15:44:11
@@ -77,10 +77,10 @@
   else
     _tags prefixes
     while _tags; do
-      while _next_label prefixes expl 'URL prefix'; do
-        _urls "$@" && ret=0
+      while _next_label prefixes expl 'URL prefix' "$@"; do
+        _urls "$expl[@]" && ret=0
 	compset -S '[^:]*'
-        compadd "$expl[@]" -S '' about: mocha: javascript: && ret=0
+        compadd -S '' "$expl[@]" about: mocha: javascript: && ret=0
       done
       (( ret )) || return 0
     done
Index: Completion/User/_urls
===================================================================
RCS file: /cvsroot/zsh/zsh/Completion/User/_urls,v
retrieving revision 1.8
diff -u -r1.8 _urls
--- Completion/User/_urls	2000/05/19 18:44:22	1.8
+++ Completion/User/_urls	2000/05/23 15:44:11
@@ -57,10 +57,11 @@
 if ! compset -P '(#b)([-+.a-z0-9]#):'; then
   _tags -C argument prefixes
   while _tags; do
-    while _next_label prefixes expl 'URL prefix'; do
+    while _next_label prefixes expl 'URL prefix' -S '' "$@"; do
+      compset -S '[^:/]*' && compstate[to_end]=''
       [[ -d $urls_path/bookmark ]] &&
-        compadd "$expl[@]" -S '' bookmark: && ret=0
-      compadd "$expl[@]" -S '' file: ftp:// gopher:// http:// && ret=0
+        compadd "$expl[@]" bookmark: && ret=0
+      compadd "$expl[@]" file: ftp:// gopher:// http:// && ret=0
     done
     (( ret )) || return 0
   done
@@ -71,7 +72,7 @@
 case "$scheme" in
   http|ftp|gopher)
     if ! compset -P //; then
-      _wanted -C "$scheme" prefixes expl 'end of prefix' compadd "$@" -S '' //
+      _wanted -C "$scheme" prefixes expl 'end of prefix' compadd -S '' "$@" //
       return
     fi
   ;;
@@ -84,7 +85,7 @@
 	    _path_files "$expl[@]" -S '' -g '*(^/)' && ret=0
 	    _path_files "$expl[@]" -S/ -r '/' -/ && ret=0
           elif [[ -z "$PREFIX" ]]; then
-	    compadd "$expl[@]" -S '/' -r '/' - "${PWD%/}" && ret=0
+	    compadd -S '/' -r '/' "$expl[@]" "$@" - "${PWD%/}" && ret=0
           fi
         done
 	(( ret )) || return 0
@@ -104,7 +105,7 @@
         while _next_label files expl 'bookmark'; do
           _path_files -W "$urls_path/$scheme" "$expl[@]" -S '' -g '*(^/)' && 
               ret=0
-          _path_files -W "$urls_path/$scheme" "$expl[@]" -S/ -r '/' -/ && ret=0
+          _path_files -W "$urls_path/$scheme" -S/ -r '/' "$expl[@]" -/ && ret=0
         done
 	(( ret )) || return 0
       done
@@ -119,10 +120,11 @@
 
   _tags hosts
   while _tags; do
-    while _next_label hosts expl host; do
-      (( $#uhosts )) || _hosts -S/ && ret=0
+    while _next_label hosts expl host "$@"; do
+      compset -S '/*' || suf="/"
+      (( $#uhosts )) || _hosts -S "$suf" "$expl[@]" && ret=0
       [[ "$scheme" = http ]] && uhosts=($uhosts $localhttp_servername)
-      compadd "$expl[@]" -S/ - $uhosts && ret=0
+      compadd -S "$suf" "$expl[@]" - $uhosts && ret=0
     done
     (( ret )) || return 0
   done
@@ -137,14 +139,14 @@
 if [[ "$localhttp_servername" = "$host" ]]; then
   if compset -P \~; then
     if ! compset -P '(#b)([^/]#)/'; then
-      _users -S/
+      _users -S/ "$@"
       return
     fi
     user="$match[1]"
     while _tags; do
       while _next_label files expl 'local file'; do
         _path_files "$expl[@]" "$@" -W ~$user/$localhttp_userdir -g '*(^/)' && ret=0
-        _path_files "$expl[@]" -W ~$user/$localhttp_userdir -S/ -r '/' -/ && ret=0
+        _path_files -S/ -r '/'  "$expl[@]" -W ~$user/$localhttp_userdir-/ && ret=0
       done
       (( ret )) || return 0
     done
@@ -152,7 +154,7 @@
     while _tags; do
       while _next_label files expl 'local file'; do
         _path_files "$expl[@]" "$@" -W $localhttp_documentroot -g '*(^/)' && ret=0
-        _path_files "$expl[@]" -W $localhttp_documentroot -S/ -r '/' -/ && ret=0
+        _path_files -S/ -r '/' "$expl[@]" -W $localhttp_documentroot -/ && ret=0
       done
       (( ret )) || return 0
     done
@@ -161,7 +163,7 @@
   while _tags; do
     while _next_label files expl 'local file'; do
       _path_files "$expl[@]" "$@" -W $urls_path/$scheme/$host -g '*(^/)' && ret=0
-      _path_files "$expl[@]" -W $urls_path/$scheme/$host -S/ -r '/' -/ && ret=0
+      _path_files -S/ -r '/' "$expl[@]" -W $urls_path/$scheme/$host -/ && ret=0
     done
     (( ret )) || return 0
   done

Re: PATCH: Re: _netscape

[Sven Wischnowsky]

Oliver Kiddle wrote:

> Sven Wischnowsky wrote:
> 
> > > > Basically, what a helper function needs to do is
> > > > take the suffix passed to it and when it is completing a final component
> > > > of itself, it should pass any suffix it wants with the one passed to it
> > > > appended. Pulling out -S options from "$@" is going to look messy
> > > > without some special handling at a lower level somewhere.
> 
> > This is so simple to write that I think it's worth adding. So, this
> > adds the -E option to zparseopts that can be used to extract options
> > from the positional parameters. When combined with -D, the options
> > described are actually removed from $*.
> 
> On the basis that compadd does only use the first -S option that it is
> passed, being able to remove an option is not what is wanted: what is
> wanted is a way to separate the first -S option and have it available in
> a parameter.

Ever had a look at zparseopts?

  foo() {
    local suffix

    zparseopts -D -E 'S:=suffix'

    print "user-supplied suffix: ${suffix:-<none>}"
    print "remaining arguments:  $*"
  }

  % foo -J group -S test -X format other arguments
  user-supplied suffix: -S test
  remaining arguments:  -J group -X format other arguments
  % foo -J group -X format no suffix here
  user-supplied suffix: <none>
  remaining arguments:  -J group -X format no suffix here

What else do you need?

Bye
 Sven


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

Re: PATCH: Re: _netscape

[Oliver Kiddle]

Sven Wischnowsky wrote:

> > > Basically, what a helper function needs to do is
> > > take the suffix passed to it and when it is completing a final component
> > > of itself, it should pass any suffix it wants with the one passed to it
> > > appended. Pulling out -S options from "$@" is going to look messy
> > > without some special handling at a lower level somewhere.

> This is so simple to write that I think it's worth adding. So, this
> adds the -E option to zparseopts that can be used to extract options
> from the positional parameters. When combined with -D, the options
> described are actually removed from $*.

On the basis that compadd does only use the first -S option that it is
passed, being able to remove an option is not what is wanted: what is
wanted is a way to separate the first -S option and have it available in
a parameter.

If a helper function is completing only an initial portion of whatever
it completes, it is likely that what it needs to do is use its own
suffix, ignoring any passed to it - this we can do by inserting a new -S
option before the passed args. But, if a helper function is completing
the final portion of whatever it completes, it might want to use as a
suffix, the concatenation of its own suffix with the one which was
passed to it. For this, it would need to extract any suffix passed to it
as opposed to remove it. I'll also need to think about what will happen
with -r options aswell. This -E option well be very useful as it is but
it is not quite what I meant by 'pull out -S options from "$@"'.

Oliver

Re: PATCH: Re: _netscape

[Sven Wischnowsky]

I wrote:

> Oliver Kiddle wrote:
> 
> > ...
> > 
> > Basically, what a helper function needs to do is
> > take the suffix passed to it and when it is completing a final component
> > of itself, it should pass any suffix it wants with the one passed to it
> > appended. Pulling out -S options from "$@" is going to look messy
> > without some special handling at a lower level somewhere.
> 
> It's mostly a bit of [(i)...] stuff, but yes, it's probably common
> enough... One idea would be to add an option to zparseopts to allow it 
> to extract options. Actually, I've been wishing for something like
> that, too.

This is so simple to write that I think it's worth adding. So, this
adds the -E option to zparseopts that can be used to extract options
from the positional parameters. When combined with -D, the options
described are actually removed from $*.

Bye
 Sven

Index: Doc/Zsh/mod_zutil.yo
===================================================================
RCS file: /cvsroot/zsh/zsh/Doc/Zsh/mod_zutil.yo,v
retrieving revision 1.6
diff -u -r1.6 mod_zutil.yo
--- Doc/Zsh/mod_zutil.yo	2000/05/22 09:32:33	1.6
+++ Doc/Zsh/mod_zutil.yo	2000/05/23 13:14:08
@@ -131,7 +131,7 @@
 This implements the internals of the `tt(_regex_arguments)'.
 )
 findex(zparseopts)
-item(tt(zparseopts) [ tt(-D) ] [ tt(-a) var(array) ] [ tt(-A) var(assoc) ] var(specs))(
+item(tt(zparseopts) [ tt(-D) ] [ tt(-E) ] [ tt(-a) var(array) ] [ tt(-A) var(assoc) ] var(specs))(
 This builtin simplifies the parsing of options in positional
 parameters, i.e. the set of arguments given by tt($*).  Each var(spec)
 describes one option and should be of the form
@@ -166,6 +166,9 @@
 positional parameters, up to but not including any not described by the
 var(specs).  This means that any options processed by tt(zparseopts) are
 removed from the positional parameters.
+
+The tt(-E) option allows to extract the options described by the
+var(specs) from the positional parameters, ignoring all other strings.
 
 For example,
 
Index: Src/Modules/zutil.c
===================================================================
RCS file: /cvsroot/zsh/zsh/Src/Modules/zutil.c,v
retrieving revision 1.4
diff -u -r1.4 zutil.c
--- Src/Modules/zutil.c	2000/05/22 09:32:33	1.4
+++ Src/Modules/zutil.c	2000/05/23 13:14:09
@@ -1261,8 +1261,8 @@
 static int
 bin_zparseopts(char *nam, char **args, char *ops, int func)
 {
-    char *o, *p, *n, **pp, **aval, **ap, *assoc = NULL;
-    int del = 0, f;
+    char *o, *p, *n, **pp, **aval, **ap, *assoc = NULL, **cp, **np;
+    int del = 0, f, extract = 0;
     Zoptdesc sopts[256], d;
     Zoptarr a, defarr = NULL;
     Zoptval v;
@@ -1290,6 +1290,14 @@
 		}
 		del = 1;
 		break;
+	    case 'E':
+		if (o[2]) {
+		    args--;
+		    o = NULL;
+		    break;
+		}
+		extract = 1;
+		break;
 	    case 'a':
 		if (defarr) {
 		    zwarnnam(nam, "default array given more than once", NULL, 0);
@@ -1400,10 +1408,19 @@
 	if (!o[1])
 	    sopts[STOUC(*o)] = d;
     }
-    for (pp = pparams; (o = *pp); pp++) {
-	if (*o != '-')
-	    break;
+    np = cp = pp = ((extract && del) ? arrdup(pparams) : pparams);
+    for (; (o = *pp); pp++) {
+	if (*o != '-') {
+	    if (extract) {
+		if (del)
+		    *cp++ = o;
+		continue;
+	    } else
+		break;
+	}
 	if (!o[1] || (o[1] == '-' && !o[2])) {
+	    if (del && extract)
+		*cp++ = o;
 	    pp++;
 	    break;
 	}
@@ -1429,8 +1446,14 @@
 		} else
 		    add_opt_val(d, NULL);
 	    }
-	    if (!o)
-		break;
+	    if (!o) {
+		if (extract) {
+		    if (del)
+			*cp++ = *pp;
+		    continue;
+		} else
+		    break;
+	    }
 	} else {
 	    if (d->flags & ZOF_ARG) {
 		char *e = o + strlen(d->name) + 1;
@@ -1450,6 +1473,10 @@
 		add_opt_val(d, NULL);
 	}
     }
+    if (extract && del)
+	while (*pp)
+	    *cp++ = *pp++;
+
     for (a = opt_arrs; a; a = a->next) {
 	aval = (char **) zalloc((a->num + 1) * sizeof(char *));
 	for (ap = aval, v = a->vals; v; ap++, v = v->next) {
@@ -1498,9 +1525,15 @@
 	sethparam(assoc, aval);
     }
     if (del) {
-	pp = zarrdup(pp);
-	freearray(pparams);
-	pparams = pp;
+	if (extract) {
+	    *cp = NULL;
+	    freearray(pparams);
+	    pparams = zarrdup(np);
+	} else {
+	    pp = zarrdup(pp);
+	    freearray(pparams);
+	    pparams = pp;
+	}
     }
     return 0;
 }

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

Re: PATCH: Re: _netscape

[Sven Wischnowsky]

Oliver Kiddle wrote:

> ...
> 
> I've gone on to try to fix the handling of the closing bracket suffix in
> netscape remote commands which has resulted in the following thoughts on
> suffix handling in the new completion.
> 
> There are a few places where the contents of a remote command comes to a
> definite finish: a file/ftp/http URL that ends with a file or empty
> directory, a saveAs command after the file type, about: URLS. In these
> cases, it would be nice if completing the last stage added both the
> closing bracket (back-slash quoted if necessary) to terminate the remote
> command and any closing quote.
> 
> Adding a closing quote sometimes happens, for example:
> netscape -remote 'openURL(about:c<tab>
> this completes to the about:cache URL with the closing quote appended.
> Where does the quote come from: is the default suffix actually
> "$compstate[quote] " instead of just a space?

Approximately, yes. If the completion code thinks or knows that there
has to be a certain closing quote (or closing quotes) and there is no
other suffix, the quote is inserted. This was added at the last
completion-in-quotes overhaul (and in most cases gives one
automatically what one wants).

> ...
> 
> Basically, what a helper function needs to do is
> take the suffix passed to it and when it is completing a final component
> of itself, it should pass any suffix it wants with the one passed to it
> appended. Pulling out -S options from "$@" is going to look messy
> without some special handling at a lower level somewhere.

It's mostly a bit of [(i)...] stuff, but yes, it's probably common
enough... One idea would be to add an option to zparseopts to allow it 
to extract options. Actually, I've been wishing for something like
that, too.

> compadd seems
> to only use the first -S option it is passed but this isn't documented
> so I didn't want to rely on it.

It *is* documented, in the compadd entry, below the options.
_approximate always relied on it.

> If we do make it a documented feature
> then we still need to handle the situation where you are completing the
> final part of something and you want to use both your own suffix and the
> one passed to the function.

One problem I see here is that in some cases utility functions might
want to use their suffix only as a default, overridden by the one
given by a calling function (easy to solve). Another problem might be
the interaction with -[qrR].

> Note that _files can no longer complete subdirectories if you give it a
> -S option which is something other than a slash. _files should only add
> the suffix after a file (and possible after an empty directory).

Sometimes you want it, sometimes not...

> Returning to _netscape, the next issue was how to handle the closing
> bracket/quote already being there.
> I used compset -S '(|\\)\)*'. But this again comes up with a problem
> when we are not completing the final part of a URL. The suffix is only
> moved to ISUFFIX so for example:
> netscape -remote 'openURL(a<tab>)    i.e. cursor is at position of tab
> will complete the url to about:)' - advancing the cursor past the
> suffix. I'm able to prevent this by not using the suffix with compadd
> if the compset succeeded.

You may also want to have a look at $compstate[to_end]...

> This example shows how and leads me on to my next point:
> 
> _foo() {
>   local lsuf="" nsuf=""
>   compset -S '.*' || lsuf=". "
>   if compset -P "*:"; then
>     compadd -S "$lsuf" aaa bbb ccc
>   else
>     compset -S ':*' || nsuf=":"
>     compadd -S "$rsuf" one two three
>   fi
> }
> 
> foo o<tab>.  completes the one but moves the cursor past the '.'. I
> would prefer if it was only moved to the end of the :. We started off
> with just 'o.' so there was no :. this indicates to me that the part
> which comes after a colon is necessary so it is likely that the user
> will want to complete it. In the case of:
> foo o<tab>:ddd.
> it is fair enough in a way that the completion moves past the final '.'
> because there was a colon there to start with although even in this case
> I'd prefer it didn't. What I would actually want it to do in this case
> is move to the end of the 'ddd' but before the '.'. If 'ddd' was
> incomplete, I could finish it otherwise, another tab would advance me on
> to the next thing.

This, of course can't be solved even by to_end.


The way suffixes are handled has mostly historical reasons and, yes,
it is a bit simple minded. If someone wants to help to improve it, I'd 
like to have a short and comprehensive list of things one might want
to do, so that we can discuss it easliy and find the (hopefully few)
basic things we need for that. With that we could then think about
changing the code. Ok? (I don't think I'll have much time to think
about this...)


Bye
 Sven


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