Zsh Mailing List Archive
Messages sorted by:
Reverse Date,
Date,
Thread,
Author
PATCH: Re: Bugs thrown up by _perforce
- X-seq: zsh-workers 18299
- From: Oliver Kiddle <okiddle@xxxxxxxxxxx>
- To: zsh-workers@xxxxxxxxxx
- Subject: PATCH: Re: Bugs thrown up by _perforce
- Date: Wed, 26 Feb 2003 09:47:43 +0100
- In-reply-to: <22979.1046178445@xxxxxxx>
- Mailing-list: contact zsh-workers-help@xxxxxxxxxx; run by ezmlm
- References: <22979.1046178445@xxxxxxx>
- Sender: kiddleo@xxxxxxxxxx
Peter wrote:
> Oliver Kiddle wrote:
> > [_next_tags always inserts an unambiguous completion]
> >
> > You can fix this by changing the ins=unambiguous line to just ins= so
> > that compstate[insert] remains empty but this changes _next_tags
> > behaviour in other cases.
>
> I'll test this out.
That would be useful.
> > behaviour would be. Does anyone want _next_tags to actually complete
> > stuff ever?
>
> I can't see why --- certainly not treat it as an unambiguous insertion,
> fait accompli. You can't even see what it's going to complete next.
Okay, I think it can be fixed then.
> > [an _arguments --- or any inner tags loop --- screws up _next_tags]
> Um. I'm still a bit vague as to how all this works. Can we prevent
> this behaviour happening with argument-rest? After all, it's being
> smuggled in behind our back anyway --- it's not specified either in the
> styles, nor in the list of tags I'm explicitly generating.
You could but a more generic solution would be much better. The
argument- tags can be useful.
> Can we make the system remember we're in a nested loop? And then inside
> it can be more lenient? Or some combination of the tow? Or something?
This patch should do the job. It only puts tags from the inner most tag
loops into $_lastcomp[tags] and hence $_next_tags_not. Not blocking an
outer loop tag does not matter because all inner tags will be blocked
for it so the effect is the same.
Does anyone see a problem with the change to what $_lastcomp[tags]
contains? Note that it isn't used by anything other than _next_tags.
Potential problems remain if the same tag names are used across
different branches (the inner loops of separate outer loops). Doing
that is otherwise not very sensible anyway.
Oliver
Index: Completion/Base/Core/_all_labels
===================================================================
RCS file: /cvsroot/zsh/zsh/Completion/Base/Core/_all_labels,v
retrieving revision 1.2
diff -u -r1.2 _all_labels
--- Completion/Base/Core/_all_labels 12 Feb 2002 13:37:03 -0000 1.2
+++ Completion/Base/Core/_all_labels 26 Feb 2003 08:33:31 -0000
@@ -24,6 +24,8 @@
fi
while comptags "-A$__prev" "$1" curtag __spec; do
+ (( $#funcstack > _tags_level )) && _comp_tags="${_comp_tags% * }"
+ _tags_level=$#funcstack
_comp_tags="$_comp_tags $__spec "
if [[ "$curtag" = *[^\\]:* ]]; then
zformat -f __descr "${curtag#*:}" "d:$3"
Index: Completion/Base/Core/_main_complete
===================================================================
RCS file: /cvsroot/zsh/zsh/Completion/Base/Core/_main_complete,v
retrieving revision 1.6
diff -u -r1.6 _main_complete
--- Completion/Base/Core/_main_complete 26 Jun 2002 11:07:46 -0000 1.6
+++ Completion/Base/Core/_main_complete 26 Feb 2003 08:33:31 -0000
@@ -23,6 +23,7 @@
_matchers _matcher _c_matcher _matcher_num _comp_tags _comp_mesg \
mesg str context state line opt_args val_args curcontext="$curcontext" \
_last_nmatches=-1 _last_menu_style _def_menu_style _menu_style sel \
+ _tags_level=0 \
_saved_exact="${compstate[exact]}" \
_saved_lastprompt="${compstate[last_prompt]}" \
_saved_list="${compstate[list]}" \
Index: Completion/Base/Core/_next_label
===================================================================
RCS file: /cvsroot/zsh/zsh/Completion/Base/Core/_next_label,v
retrieving revision 1.2
diff -u -r1.2 _next_label
--- Completion/Base/Core/_next_label 12 Feb 2002 13:37:04 -0000 1.2
+++ Completion/Base/Core/_next_label 26 Feb 2003 08:33:31 -0000
@@ -6,6 +6,8 @@
zparseopts -D -a __gopt 1 2 V J x
if comptags -A "$1" curtag __spec; then
+ (( $#funcstack > _tags_level )) && _comp_tags="${_comp_tags% * }"
+ _tags_level=$#funcstack
_comp_tags="$_comp_tags $__spec "
if [[ "$curtag" = *[^\\]:* ]]; then
zformat -f __descr "${curtag#*:}" "d:$3"
Index: Completion/Base/Widget/_next_tags
===================================================================
RCS file: /cvsroot/zsh/zsh/Completion/Base/Widget/_next_tags,v
retrieving revision 1.3
diff -u -r1.3 _next_tags
--- Completion/Base/Widget/_next_tags 25 Feb 2003 18:43:51 -0000 1.3
+++ Completion/Base/Widget/_next_tags 26 Feb 2003 08:33:31 -0000
@@ -34,6 +34,8 @@
fi
while comptags "-A$__prev" "$1" curtag __spec; do
+ (( $#funcstack > _tags_level )) && _comp_tags="${_comp_tags% * }"
+ _tags_level=$#funcstack
[[ "$_next_tags_not" = *\ ${__spec}\ * ]] && continue
_comp_tags="$_comp_tags $__spec "
if [[ "$curtag" = *[^\\]:* ]]; then
@@ -59,6 +61,8 @@
zparseopts -D -a __gopt 1 2 V J x
if comptags -A "$1" curtag __spec; then
+ (( $#funcstack > _tags_level )) && _comp_tags="${_comp_tags% * }"
+ _tags_level=$#funcstack
[[ "$_next_tags_not" = *\ ${__spec}\ * ]] && continue
_comp_tags="$_comp_tags $__spec "
if [[ "$curtag" = *[^\\]:* ]]; then
This e-mail and any attachment is for authorised use by the intended recipient(s) only. It may contain proprietary material, confidential information and/or be subject to legal privilege. It should not be copied, disclosed to, retained or used by, any other party. If you are not an intended recipient then please promptly delete this e-mail and any attachment and all copies and inform the sender. Thank you.
Messages sorted by:
Reverse Date,
Date,
Thread,
Author