bug#18282: 24.3.93; Tab completion in gud-gdb at the start of a line overwrites the prompt.

bug#18282: 24.3.93; Tab completion in gud-gdb at the start of a line overwrites the prompt.

[Mat Smiglarski]

Hello,
Tab completion in gud-gdb replaces the prompt with the completed
string. As a result this string is not sent to the gdb process.

This bug does not occur in 24.3.1. It does occur on trunk and
24.3.92, as well as 24.3.93.

To reproduce:

1. emacs -Q
2. M-: (gud-gdb "gdb")
3. Type the following at the gdb prompt "hel<TAB><RET>"

What happens is the command "help" is correctly completed, but the
prompt "(gdb)" is overwritten. This completed
string has the face "comint-highlight-prompt" and is not evaluated by 
the gdb process.

For what it is worth, the bug does not occur with "M-x gdb", only "M-x
gud-gdb".

Regards,
Mat


In GNU Emacs 24.3.93.1 (x86_64-unknown-linux-gnu, GTK+ Version 3.10.8)
  of 2014-08-16 on zz
Windowing system distributor `The X.Org Foundation', version 
11.0.11501000
System Description:	Ubuntu 14.04.1 LTS

Configured using:
  `configure 'CFLAGS=-O0 -g3''

Important settings:
   value of $LC_CTYPE: en_US.UTF-8
   value of $LANG: en_US.UTF-8
   value of $XMODIFIERS: @im=ibus
   locale-coding-system: utf-8-unix

Major mode: Debugger

Minor modes in effect:
   tooltip-mode: t
   electric-indent-mode: t
   mouse-wheel-mode: t
   tool-bar-mode: t
   menu-bar-mode: t
   file-name-shadow-mode: t
   global-font-lock-mode: t
   font-lock-mode: t
   blink-cursor-mode: t
   auto-composition-mode: t
   auto-encryption-mode: t
   auto-compression-mode: t
   line-number-mode: t
   transient-mark-mode: t

Recent input:
M-x M-p g u d - g d b <return> C-a M-f C-k <return>
h e <tab> <return> M-x r e p <tab> - e - b <tab> <
return>

Recent messages:
For information about GNU Emacs and the GNU system, type C-h C-a.
user-error: Beginning of history; no preceding item
Making completion list...

Load-path shadows:
None found.

Features:
(shadow sort gnus-util mail-extr emacsbug message format-spec rfc822 mml
mml-sec mm-decode mm-bodies mm-encode mail-parse rfc2231 mailabbrev
gmm-utils mailheader sendmail rfc2047 rfc2045 ietf-drums mm-util
help-fns mail-prsvr mail-utils help-mode easymenu gud easy-mmode comint
ansi-color ring time-date tooltip electric uniquify ediff-hook vc-hooks
lisp-float-type mwheel x-win x-dnd tool-bar dnd fontset image regexp-opt
fringe tabulated-list newcomment lisp-mode prog-mode register page
menu-bar rfn-eshadow timer select scroll-bar mouse jit-lock font-lock
syntax facemenu font-core frame cham georgian utf-8-lang misc-lang
vietnamese tibetan thai tai-viet lao korean japanese hebrew greek
romanian slovak czech european ethiopic indian cyrillic chinese
case-table epa-hook jka-cmpr-hook help simple abbrev minibuffer nadvice
loaddefs button faces cus-face macroexp files text-properties overlay
sha1 md5 base64 format env code-pages mule custom widget
hashtable-print-readable backquote make-network-process dbusbind
gfilenotify dynamic-setting system-font-setting font-render-setting
move-toolbar gtk x-toolkit x multi-tty emacs)

Memory information:
((conses 16 79181 7387)
  (symbols 48 18300 0)
  (miscs 40 49 200)
  (strings 32 11682 4841)
  (string-bytes 1 317029)
  (vectors 16 10558)
  (vector-slots 8 397115 7916)
  (floats 8 66 210)
  (intervals 56 233 0)
  (buffers 960 13)
  (heap 1024 17278 783))

bug#18282: 24.3.93; Tab completion in gud-gdb at the start of a line overwrites the prompt.

[Glenn Morris]

Mat Smiglarski wrote:

> 1. emacs -Q
> 2. M-: (gud-gdb "gdb")
> 3. Type the following at the gdb prompt "hel<TAB><RET>"
>
> What happens is the command "help" is correctly completed, but the
> prompt "(gdb)" is overwritten.

Bisected to emacs-24:
  
  revno: 114162
  committer: Stefan Monnier <monnier@iro.umontreal.ca>
  timestamp: Fri 2013-09-06 18:46:44 -0400
  message:
    * lisp/minibuffer.el: Make minibuffer-complete call
    completion-in-region rather than other way around.
    [...]

bug#18282: 24.3.93; Tab completion in gud-gdb at the start of a line overwrites the prompt.

[Stefan Monnier]

>> 1. emacs -Q
>> 2. M-: (gud-gdb "gdb")
>> 3. Type the following at the gdb prompt "hel<TAB><RET>"
>> 
>> What happens is the command "help" is correctly completed, but the
>> prompt "(gdb)" is overwritten.
> Bisected to emacs-24:
>   revno: 114162
>   committer: Stefan Monnier <monnier@iro.umontreal.ca>
>   timestamp: Fri 2013-09-06 18:46:44 -0400
>   message:
>     * lisp/minibuffer.el: Make minibuffer-complete call
>     completion-in-region rather than other way around.
>     [...]

Thanks for bisecting it!  That'll be very helpful,


        Stefan

bug#18282: 24.3.93; Tab completion in gud-gdb at the start of a line overwrites the prompt.

[Glenn Morris]

I wonder if the problem is this note from gud-gdb-completions:

    ;; `gud-gdb-run-command-fetch-lines' has some nasty side-effects on the
    ;; buffer (via `gud-delete-prompt-marker'): it removes the prompt and then
    ;; re-adds it later, thus messing up markers and overlays along the way.
    ;; This is a problem for completion-in-region which uses an overlay to
    ;; create a field.
    ;; So we restore completion-in-region's field if needed.

completion-in-region doesn't use an overlay any more (so the comment and
workaround are in any case outdated), it uses a marker, which presumably
gets moved behind its back. Dunno how you fix things up to make the
marker point back where it should.

python.el seems to have a similar need, and to solve it using
python-shell-send-string-no-output (use
comint-preoutput-filter-functions to prevent completion output getting
inserted into the buffer).

bug#18282: 24.3.93; Tab completion in gud-gdb at the start of a line overwrites the prompt.

[Glenn Morris]

PS perhaps comint-dynamic-list-filename-completions can/should be
similarly updated to us explicit start/end args rather than overlays.

bug#18282: 24.3.93; Tab completion in gud-gdb at the start of a line overwrites the prompt.

[Glenn Morris]

Her's an awful kludge that seems to work somewhat, maybe.
But it can't be the right solution.
But it can't be worse than it is now (?).

*** lisp/progmodes/gud.el	2014-02-10 01:34:22 +0000
--- lisp/progmodes/gud.el	2014-08-27 06:48:45 +0000
***************
*** 778,784 ****
    (add-hook 'completion-at-point-functions #'gud-gdb-completion-at-point
              nil 'local)
    (set (make-local-variable 'gud-gdb-completion-function) 'gud-gdb-completions)
! 
    (local-set-key "\C-i" 'completion-at-point)
    (setq comint-prompt-regexp "^(.*gdb[+]?) *")
    (setq paragraph-start comint-prompt-regexp)
--- 778,784 ----
    (add-hook 'completion-at-point-functions #'gud-gdb-completion-at-point
              nil 'local)
    (set (make-local-variable 'gud-gdb-completion-function) 'gud-gdb-completions)
!   (setq-local completion-in-region-function 'gud--completion-in-region)
    (local-set-key "\C-i" 'completion-at-point)
    (setq comint-prompt-regexp "^(.*gdb[+]?) *")
    (setq paragraph-start comint-prompt-regexp)
***************
*** 787,792 ****
--- 787,795 ----
    (setq gud-filter-pending-text nil)
    (run-hooks 'gud-gdb-mode-hook))
  
+ (defun gud--completion-in-region (start end collection &optional predicate)
+   (completion--in-region (copy-marker start t) end collection predicate))
+ 
  ;; The completion process filter indicates when it is finished.
  (defvar gud-gdb-fetch-lines-in-progress)
  

bug#18282: 24.3.93; Tab completion in gud-gdb at the start of a line overwrites the prompt.

[Stefan Monnier]

> Her's an awful kludge that seems to work somewhat, maybe.
> But it can't be the right solution.
> But it can't be worse than it is now (?).

I'll come up with a better fix,


        Stefan

bug#18282: 24.3.93; Tab completion in gud-gdb at the start of a line overwrites the prompt.

[Stefan Monnier]

>> 1. emacs -Q
>> 2. M-: (gud-gdb "gdb")
>> 3. Type the following at the gdb prompt "hel<TAB><RET>"
>> 
>> What happens is the command "help" is correctly completed, but the
>> prompt "(gdb)" is overwritten.

I installed the patch below which seems to fix it for now.


        Stefan


=== modified file 'lisp/progmodes/gud.el'
--- lisp/progmodes/gud.el	2014-02-10 01:34:22 +0000
+++ lisp/progmodes/gud.el	2014-09-02 18:14:45 +0000
@@ -809,18 +809,6 @@
 					   (current-buffer)
 					   ;; From string-match above.
 					   (length context))))
-    ;; `gud-gdb-run-command-fetch-lines' has some nasty side-effects on the
-    ;; buffer (via `gud-delete-prompt-marker'): it removes the prompt and then
-    ;; re-adds it later, thus messing up markers and overlays along the way.
-    ;; This is a problem for completion-in-region which uses an overlay to
-    ;; create a field.
-    ;; So we restore completion-in-region's field if needed.
-    ;; FIXME: change gud-gdb-run-command-fetch-lines so it doesn't modify the
-    ;; buffer at all.
-    (when (/= start (- (point) (field-beginning)))
-      (dolist (ol (overlays-at (1- (point))))
-        (when (eq (overlay-get ol 'field) 'completion)
-          (move-overlay ol (- (point) start) (overlay-end ol)))))
     ;; Protect against old versions of GDB.
     (and complete-list
 	 (string-match "^Undefined command: \"complete\"" (car complete-list))
@@ -859,7 +847,14 @@
          (save-excursion
            (skip-chars-backward "^ " (comint-line-beginning-position))
            (point))))
-    (list start end
+    ;; FIXME: `gud-gdb-run-command-fetch-lines' has some nasty side-effects on
+    ;; the buffer (via `gud-delete-prompt-marker'): it removes the prompt and
+    ;; then re-adds it later, thus messing up markers and overlays along the
+    ;; way (bug#18282).
+    ;; We use an "insert-before" marker for `start', since it's typically right
+    ;; after the prompt, which works around the problem, but is a hack (and
+    ;; comes with other downsides, e.g. if completion adds text at `start').
+    (list (copy-marker start t) end
           (completion-table-dynamic
            (apply-partially gud-gdb-completion-function
                             (buffer-substring (comint-line-beginning-position)