unofficial mirror of bug-gnu-emacs@gnu.org 
 help / color / mirror / code / Atom feed
From: Dmitry Gutov <dgutov@yandex.ru>
To: Eli Zaretskii <eliz@gnu.org>,
	Ludovic Brenta <ludovic@ludovic-brenta.org>
Cc: 32510@debbugs.gnu.org
Subject: bug#32510: acknowledged by developer (control message for bug #32510)
Date: Tue, 30 Jul 2019 03:06:18 +0300	[thread overview]
Message-ID: <77c6d179-4499-4a99-db58-b0c7c2cee2f5@yandex.ru> (raw)
In-Reply-To: <83k1clyywc.fsf@gnu.org>

On 14.07.2019 8:21, Eli Zaretskii wrote:
>> From: Ludovic Brenta <ludovic@ludovic-brenta.org>
>> Date: Sat, 13 Jul 2019 21:34:26 +0200
>>
>> This bug report states that a useful functionality of find-tag is *not*
>> provided by its official replacement, xref-find-definitions.  This is a
>> regression.  Just because you think this missing functionality should be
>> provided elsewhere is not a good reason to close this bug without
>> providing any solution.
> 
> With the patch below, you should be able to have what you want if you
> add tag-partial-file-name-match-p to the list in

Finally got around to reviewing it...

> diff --git a/lisp/progmodes/etags.el b/lisp/progmodes/etags.el
> index 7bf5753..b092c63 100644
> --- a/lisp/progmodes/etags.el
> +++ b/lisp/progmodes/etags.el
> @@ -2070,13 +2070,16 @@ etags--xref-find-definitions
>                 (beginning-of-line)
>                 (pcase-let* ((tag-info (etags-snarf-tag))
>                              (`(,hint ,line . _) tag-info))
> -                (unless (eq hint t) ; hint==t if we are in a filename line
> +                (unless (and (eq hint t) ; we are in a filename line
> +                             (not (eq order-fun
> +                                      'tag-partial-file-name-match-p)))

First, I was thinking we shouldn't check for the exact order-fun value 
(because others could be used, too) and replace it with something like

                               (save-excursion
                                (forward-line 0)
                                (forward-char -2)
                                (not (looking-at "\f\n")))

But then, I'm not sure why that check is there in the first place (the 
order functions make sure not to match the wrong like). Maybe because 
the code inside couldn't handle hint=t? So if it does now, the (unless 
...) conditional can be removed.

>                     (let* ((file (file-of-tag))
>                            (mark-key (cons file line)))
>                       (unless (gethash mark-key marks)
>                         (let ((loc (xref-make-etags-location
>                                     tag-info (expand-file-name file))))
> -                        (push (xref-make hint loc) xrefs)
> +                        (push (xref-make (if (eq hint t) pattern hint) loc)
> +                               xrefs)

I'm not sure using pattern as a hint works well. How about we say 
something like "(file name match)" instead? Or you could pick a better 
wording. The full proposed patch is below.

I see that it doesn't work exactly perfectly, e.g. moving point within 
the quotes in

   #include "composite.h"

and pressing M-. brings up three matches (composite.c, composite.h and 
composite.el), whereas only one of them is correct, but find-tag 
probably has the same problem anyway. Maybe CC Mode should set up 
find-tag-default-function to return the full file name when inside 
#include statements.

diff --git a/lisp/progmodes/etags.el b/lisp/progmodes/etags.el
index 7bf575340e..a052ad2ce5 100644
--- a/lisp/progmodes/etags.el
+++ b/lisp/progmodes/etags.el
@@ -2070,14 +2070,15 @@ etags--xref-find-definitions
                (beginning-of-line)
                (pcase-let* ((tag-info (etags-snarf-tag))
                             (`(,hint ,line . _) tag-info))
-                (unless (eq hint t) ; hint==t if we are in a filename line
-                  (let* ((file (file-of-tag))
-                         (mark-key (cons file line)))
-                    (unless (gethash mark-key marks)
-                      (let ((loc (xref-make-etags-location
-                                  tag-info (expand-file-name file))))
-                        (push (xref-make hint loc) xrefs)
-                        (puthash mark-key t marks)))))))))))
+                (let* ((file (file-of-tag))
+                       (mark-key (cons file line)))
+                  (unless (gethash mark-key marks)
+                    (let ((loc (xref-make-etags-location
+                                tag-info (expand-file-name file))))
+                      (push (xref-make (if (eq hint t) "(filename 
match)" hint)
+                                       loc)
+                            xrefs)
+                      (puthash mark-key t marks))))))))))
      (nreverse xrefs)))

  (defclass xref-etags-location (xref-location)





  reply	other threads:[~2019-07-30  0:06 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <87ftnazlz3.fsf@mouse.gnus.org>
2018-08-23 15:32 ` bug#32510: xref-find-definitions should return file names, too Ludovic Brenta
2019-07-13  2:50   ` Lars Ingebrigtsen
     [not found]   ` <handler.32510.C.15629862456126.notifdonectrl.0@debbugs.gnu.org>
2019-07-13 19:34     ` bug#32510: acknowledged by developer (control message for bug #32510) Ludovic Brenta
2019-07-13 23:25       ` Drew Adams
2019-07-14  5:21       ` Eli Zaretskii
2019-07-30  0:06         ` Dmitry Gutov [this message]
2019-07-30 14:00           ` Dmitry Gutov
2019-08-03 10:00             ` Eli Zaretskii
2019-07-15 13:54   ` bug#32510: Tags: wontfix -> patch Ludovic Brenta
2019-07-18 14:53   ` bug#32510: xref-find-definitions should return file names, too Ludovic Brenta
2019-07-18 15:16     ` Eli Zaretskii
2019-07-18 15:54       ` Ludovic Brenta
2019-07-18 16:07         ` Eli Zaretskii
2019-07-19 22:23       ` Dmitry Gutov
2019-07-20  7:17         ` Eli Zaretskii

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://www.gnu.org/software/emacs/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=77c6d179-4499-4a99-db58-b0c7c2cee2f5@yandex.ru \
    --to=dgutov@yandex.ru \
    --cc=32510@debbugs.gnu.org \
    --cc=eliz@gnu.org \
    --cc=ludovic@ludovic-brenta.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/emacs.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).