bug#36459: 27.0.50; octave-mode ElDoc support bug

bug#36459: 27.0.50; octave-mode ElDoc support bug

[Mauro Aranda]

[-- Attachment #1: Type: text/plain, Size: 3979 bytes --]

Hello

Recipe to reproduce the bug:
1) emacs -Q
2) M-x run-octave
3) Type something that doesn't involve a octave function call, like the
following matrix. Type it slowly, to give octave-eldoc-function time to run:
A = [10 1;2 3]

(Of course, if you don't want to type slowly, set eldoc-idle-delay to 0).

4) When point is past a whitespace, such as in:
A |=
(with | being where point is)
the echo area displays: 'A' not found
Similar situation when point is at:
A = [10 |1;2 3]
Echo area here displays: '10' not found.

So, for every symbol that is not a function, 'not found' might be
displayed.  I think it is a bug, since as a a user I would expect that
if I'm not about to call a function, ElDoc shouldn't be displaying
anything.

Best regards,
Mauro.



In GNU Emacs 27.0.50 (build 1, i686-pc-linux-gnu, GTK+ Version 3.18.9)
 of 2019-06-30 built on the-blackbeard
Repository revision: fd9ea1e511ba3b1f5d338c3fc8faf0b85df951f9
Repository branch: master
Windowing system distributor 'The X.Org Foundation', version 11.0.11906000
System Description: Ubuntu 16.04.6 LTS

Recent messages:
For information about GNU Emacs and the GNU system, type C-h C-a.
delete-backward-char: Text is read-only [2 times]
Quit
You can run the command ‘run-octave’ with M-x ru-o RET

Configured using:
 'configure CFLAGS=-O3'

Configured features:
XPM JPEG TIFF GIF PNG RSVG SOUND GPM DBUS GSETTINGS GLIB NOTIFY INOTIFY
ACL LIBSELINUX GNUTLS LIBXML2 FREETYPE HARFBUZZ M17N_FLT LIBOTF XFT ZLIB
TOOLKIT_SCROLL_BARS GTK3 X11 XDBE XIM THREADS PDUMPER LCMS2 GMP

Important settings:
  value of $LANG: en_US.utf8
  value of $XMODIFIERS:
  locale-coding-system: utf-8-unix

Major mode: Lisp Interaction

Minor modes in effect:
  tooltip-mode: t
  global-eldoc-mode: t
  eldoc-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

Load-path shadows:
None found.

Features:
(shadow sort mail-extr emacsbug message rmc puny dired dired-loaddefs
format-spec rfc822 mml mml-sec password-cache epa derived epg epg-config
gnus-util rmail rmail-loaddefs text-property-search seq byte-opt gv
bytecomp byte-compile cconv mm-decode mm-bodies mm-encode mail-parse
rfc2231 mailabbrev gmm-utils mailheader cl-loaddefs cl-lib sendmail
rfc2047 rfc2045 ietf-drums mm-util mail-prsvr mail-utils thingatpt
compile octave smie easymenu comint ansi-color ring time-date elec-pair
mule-util tooltip eldoc electric uniquify ediff-hook vc-hooks
lisp-float-type mwheel term/x-win x-win term/common-win x-dnd tool-bar
dnd fontset image regexp-opt fringe tabulated-list replace newcomment
text-mode elisp-mode lisp-mode prog-mode register page menu-bar
rfn-eshadow isearch timer select scroll-bar mouse jit-lock font-lock
syntax facemenu font-core term/tty-colors frame cl-generic cham georgian
utf-8-lang misc-lang vietnamese tibetan thai tai-viet lao korean
japanese eucjp-ms cp51932 hebrew greek romanian slovak czech european
ethiopic indian cyrillic chinese composite charscript charprop
case-table epa-hook jka-cmpr-hook help simple abbrev obarray minibuffer
cl-preloaded 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 threads dbusbind
inotify lcms2 dynamic-setting system-font-setting font-render-setting
move-toolbar gtk x-toolkit x multi-tty make-network-process emacs)

Memory information:
((conses 8 53509 7386)
 (symbols 24 6753 1)
 (strings 16 18539 2692)
 (string-bytes 1 611047)
 (vectors 8 11206)
 (vector-slots 4 137069 15278)
 (floats 8 17 36)
 (intervals 28 230 0)
 (buffers 564 12)
 (heap 1024 7670 759))

[-- Attachment #2: Type: text/html, Size: 4317 bytes --]

bug#36459: 27.0.50; octave-mode ElDoc support bug

[Dmitry Gutov]

Paging the original author.

Leo, are you still using Octave? What do you think about the issue?

On 01.07.2019 17:58, Mauro Aranda wrote:
> Hello
> 
> Recipe to reproduce the bug:
> 1) emacs -Q
> 2) M-x run-octave
> 3) Type something that doesn't involve a octave function call, like the
> following matrix. Type it slowly, to give octave-eldoc-function time to run:
> A = [10 1;2 3]
> 
> (Of course, if you don't want to type slowly, set eldoc-idle-delay to 0).
> 
> 4) When point is past a whitespace, such as in:
> A |=
> (with | being where point is)
> the echo area displays: 'A' not found
> Similar situation when point is at:
> A = [10 |1;2 3]
> Echo area here displays: '10' not found.
> 
> So, for every symbol that is not a function, 'not found' might be
> displayed.  I think it is a bug, since as a a user I would expect that
> if I'm not about to call a function, ElDoc shouldn't be displaying
> anything.
> 
> Best regards,
> Mauro.

bug#36459: 27.0.50; octave-mode ElDoc support bug

[Mauro Aranda]

[-- Attachment #1: Type: text/plain, Size: 1079 bytes --]

I forgot to say that my version of GNU Octave is 4.0.0

The problem is with the regexp used in
octave-eldoc-function-signatures.  It matches the signatures of the
function, that are output with a call to the GNU Octave function
print_usage.  It also matches an old-style of outputting these function
signatures, which was:
usage:  bar = foo ()

This output was also generated by calling the GNU Octave function usage,
like this:
usage ("bar = foo ()")
But usage has been deprecated in GNU Octave, I think in major version 4.
It's still present in my version, but I don't know about latest releases
(there's a recent GNU Octave 5.1.0)

I don't know why the author chose print_usage over the GNU Octave
function help, but when print_usage fails, due to being called with a
nonexistent function in his argument, like this:
print_usage ('A')
An error message like this one is output:
error: print_usage: 'A' not found

The regexp used is wrongly matching this line and therefore
octave-eldoc-function-signatures is not returning nil, as it should in
this case.

Best regards,
Mauro.

[-- Attachment #2: Type: text/html, Size: 1322 bytes --]

bug#36459: 27.0.50; octave-mode ElDoc support bug

[Leo Liu]

On 2019-07-03 17:43 +0300, Dmitry Gutov wrote:
> Paging the original author.
>
> Leo, are you still using Octave? What do you think about the issue?

Sadly I never become a serious octave user. It is possible these
functions stop working given octave seems to evolve fast these days.

On 2019-07-04 13:15 -0300, Mauro Aranda wrote:
> I forgot to say that my version of GNU Octave is 4.0.0
>
> The problem is with the regexp used in
> octave-eldoc-function-signatures.  It matches the signatures of the
> function, that are output with a call to the GNU Octave function
> print_usage.  It also matches an old-style of outputting these function
> signatures, which was:
> usage:  bar = foo ()
>
> This output was also generated by calling the GNU Octave function usage,
> like this:
> usage ("bar = foo ()")
> But usage has been deprecated in GNU Octave, I think in major version 4.
> It's still present in my version, but I don't know about latest releases
> (there's a recent GNU Octave 5.1.0)
>
> I don't know why the author chose print_usage over the GNU Octave
> function help, but when print_usage fails, due to being called with a
> nonexistent function in his argument, like this:
> print_usage ('A')
> An error message like this one is output:
> error: print_usage: 'A' not found
>
> The regexp used is wrongly matching this line and therefore
> octave-eldoc-function-signatures is not returning nil, as it should in
> this case.

Can this bug be fixed by changing the regexp in
`octave-eldoc-function-signatures' from

        "\\s-*\\(?:--[^:]+\\|usage\\):\\s-*\\(.*\\)$"

to

        "\\s-*\\(?:--[^:]+\\|\\_<usage\\):\\s-*\\(.*\\)$"
?

BTW I can't remember why print_usage was chosen. It could be it was the
one I found that worked.

>
> Best regards,
> Mauro.

Thanks for the report.

bug#36459: 27.0.50; octave-mode ElDoc support bug

[Mauro Aranda]

[-- Attachment #1: Type: text/plain, Size: 379 bytes --]

Leo Liu <sdl.web@gmail.com> writes:

Hello Leo.  Thanks for your answer.

> Can this bug be fixed by changing the regexp in
> `octave-eldoc-function-signatures' from
>
>         "\\s-*\\(?:--[^:]+\\|usage\\):\\s-*\\(.*\\)$"
>
> to
>
>         "\\s-*\\(?:--[^:]+\\|\\_<usage\\):\\s-*\\(.*\\)$"
> ?

Yes, indeed, this change fixes it.  Will you push the fix?

Best regards,
Mauro.

[-- Attachment #2: Type: text/html, Size: 561 bytes --]

bug#36459: 27.0.50; octave-mode ElDoc support bug

[Dmitry Gutov]

On 09.07.2019 1:28, Mauro Aranda wrote:
> Leo Liu <sdl.web@gmail.com <mailto:sdl.web@gmail.com>> writes:
> 
> Hello Leo.  Thanks for your answer.
> 
>  > Can this bug be fixed by changing the regexp in
>  > `octave-eldoc-function-signatures' from
>  >
>  >         "\\s-*\\(?:--[^:]+\\|usage\\):\\s-*\\(.*\\)$"
>  >
>  > to
>  >
>  >         "\\s-*\\(?:--[^:]+\\|\\_<usage\\):\\s-*\\(.*\\)$"
>  > ?
> 
> Yes, indeed, this change fixes it.  Will you push the fix?

I've tried to verify the fix myself, and foremost to see whether the 
cases where ElDoc should be displayed still work.

But I haven't managed to find working ones, with either version of the 
regexp. My Octave version is 4.2.2.

The output of e.g. print_usage ('abs') is:

   error: Invalid call to abs.  Correct usage is:

    -- abs (Z)

   Additional help for built-in functions and operators is
   available in the online version of the manual.  Use the command
   'doc <topic>' to search the manual index.

Neither regexp seems to match the second line.

Ideas? Are there help outputs that do match? Is 4.2.2 too new/too old?

bug#36459: 27.0.50; octave-mode ElDoc support bug

[Leo Liu]

On 2019-07-30 17:37 +0300, Dmitry Gutov wrote:
> I've tried to verify the fix myself, and foremost to see whether the
> cases where ElDoc should be displayed still work.
>
> But I haven't managed to find working ones, with either version of the
> regexp. My Octave version is 4.2.2.
>
> The output of e.g. print_usage ('abs') is:
>
>   error: Invalid call to abs.  Correct usage is:
>
>    -- abs (Z)

I just tried this on a few octaves >= 4.0 and they all had this problem.
If I remember correctly the function used to work with octave 3.8. It
seems that is the new output format i.e. they all start with this
useless "error: Invalid call to XXX". The regexp might need change.

bug#36459: 27.0.50; octave-mode ElDoc support bug

[Dmitry Gutov]

On 30.07.2019 20:53, Leo Liu wrote:
> It
> seems that is the new output format i.e. they all start with this
> useless "error: Invalid call to XXX".

I wouldn't say this is a problem exactly, because the code tries to 
match the regexp against every line, not just the first one.

> The regexp might need change.

Do you have an example of the output from other versions of Octave, to 
make sure any change in regexp will keep them working? Not sure if we 
still need to support Octave 3, but 3.8 is only 6 years old. :-)

bug#36459: 27.0.50; octave-mode ElDoc support bug

[Mauro Aranda]

[-- Attachment #1: Type: text/plain, Size: 512 bytes --]

My GNU Octave version is 4.0.0
In 4.0.x versions, the output of print_usage is like this:
 -- Mapping Function: abs (Z)

That's why the regexp works.  You can see an example in:
https://octave.org/doc/v4.0.0/Raising-Errors.html#Raising-Errors

Looking at the doc sources, it looks like the output was the same in GNU
Octave 3.

Now, after 4.0.x, namely 4.2.x, it seems that the output has changed to
what you see, and an example can be seen here:
https://octave.org/doc/v4.2.0/Raising-Errors.html#Raising-Errors

[-- Attachment #2: Type: text/html, Size: 759 bytes --]

bug#36459: 27.0.50; octave-mode ElDoc support bug

[Mauro Aranda]

[-- Attachment #1.1: Type: text/plain, Size: 407 bytes --]

tags 36459 patch
quit

I attach a patch, which I think tackles the problems presented.  It
stops matching incorrectly when there is no function defined, with the
change suggested by Leo Liu.

Also, the regexp used now matches the newer help output (GNU Octave
4.2.x and newer), while still supporting older GNU Octave versions.

I've tested with GNU Octave 4.0.0 and GNU Octave 5.1.0.

Best regards,
Mauro.

[-- Attachment #1.2: Type: text/html, Size: 498 bytes --]

[-- Attachment #2: 0001-Fix-octave-mode-ElDoc-support.patch --]
[-- Type: text/x-patch, Size: 1940 bytes --]

From 6e704f00be224371a2aab9b6bce3a792dfeb934b Mon Sep 17 00:00:00 2001
From: Mauro Aranda <maurooaranda@gmail.com>
Date: Tue, 6 Aug 2019 12:45:28 -0300
Subject: [PATCH] Fix octave-mode ElDoc support

* lisp/progmodes/octave.el (octave-eldoc-function-signatures): Fix the
regexp used, so no match happens when there is no defined function FN.
Also, tweak the regexp to support GNU Octave 4.2.x and newer.  (Bug#36459)
---
 lisp/progmodes/octave.el | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)

diff --git a/lisp/progmodes/octave.el b/lisp/progmodes/octave.el
index b770edb..51ba34f 100644
--- a/lisp/progmodes/octave.el
+++ b/lisp/progmodes/octave.el
@@ -1616,8 +1616,23 @@ octave-eldoc-function-signatures
      (list (format "print_usage ('%s');\n" fn)))
     (let (result)
       (dolist (line inferior-octave-output-list)
+        ;; The help output has changed a few times in GNU Octave.
+        ;; Earlier versions output "usage: " before the function signature.
+        ;; After deprecating the usage function, and up until GNU Octave 4.0.3,
+        ;; the output looks like this:
+        ;; -- Mapping Function: abs (Z).
+        ;; After GNU Octave 4.2.0, the output is less verbose and it looks like
+        ;; this:
+        ;; -- abs (Z)
+        ;; The following regexp matches these three formats.
+        ;; The "usage: " alternative matches the symbol, because a call to
+        ;; print_usage with a non-existent function (e.g., print_usage ('A'))
+        ;; would output:
+        ;; error: print_usage: 'A' not found
+        ;; and we wouldn't like to match anything in this case.
+        ;; See bug #36459.
         (when (string-match
-               "\\s-*\\(?:--[^:]+\\|usage\\):\\s-*\\(.*\\)$"
+               "\\s-*\\(?:--[^:]+:\\|\\_<usage:\\|--\\)\\s-*\\(.*\\)$"
                line)
           (push (match-string 1 line) result)))
       (setq octave-eldoc-cache
-- 
2.7.4

bug#36459: 27.0.50; octave-mode ElDoc support bug

[Leo Liu]

On 2019-08-06 13:07 -0300, Mauro Aranda wrote:
> Also, the regexp used now matches the newer help output (GNU Octave
> 4.2.x and newer), while still supporting older GNU Octave versions.
>
> I've tested with GNU Octave 4.0.0 and GNU Octave 5.1.0.

Thank you for the patch. What is the recommended way to get function
signatures these days if print_usage is deprecated?

Leo

bug#36459: 27.0.50; octave-mode ElDoc support bug

[Mauro Aranda]

[-- Attachment #1: Type: text/plain, Size: 666 bytes --]

> Thank you for the patch. What is the recommended way to get function
> signatures these days if print_usage is deprecated?
>
> Leo

Hello Leo.

The deprecated function is usage, not print_usage.  The usage function
was deprecated in GNU Octave 4.0, and in the 4.4 version it
has been removed.

Here is the NEWS file that says usage is deprecated, and that its
replacement is print_usage:
https://www.gnu.org/software/octave/NEWS-4.0.html

Since it says print_usage is the replacement function, I think calling
print_usage is reliable.

Here is the NEWS file that says usage has been removed:
https://www.gnu.org/software/octave/NEWS-4.4.html

Best regards,
Mauro.

[-- Attachment #2: Type: text/html, Size: 899 bytes --]

bug#36459: 27.0.50; octave-mode ElDoc support bug

[Leo Liu]

On 2019-08-08 08:09 -0300, Mauro Aranda wrote:
> Hello Leo.

Hi Mauro,

> The deprecated function is usage, not print_usage.  The usage function
> was deprecated in GNU Octave 4.0, and in the 4.4 version it
> has been removed.

Thanks for the clarification. I misread the comment in the patch to mean
print_usage is deprecated.

Hi Eli,

octave-eldoc-function-signatures is broken for too long. I wonder if it
is possible to merge Mauro's patch for the upcoming 26.3 release so that
there is at least one working version in 26.x releases? Thanks.

Leo

bug#36459: 27.0.50; octave-mode ElDoc support bug

[Eli Zaretskii]

> From:  Leo Liu <sdl.web@gmail.com>
> Cc: Eli Zaretskii <eliz@gnu.org>, 36459@debbugs.gnu.org,  Dmitry Gutov <dgutov@yandex.ru>
> Date: Thu, 08 Aug 2019 20:39:49 +0800
> 
> octave-eldoc-function-signatures is broken for too long. I wonder if it
> is possible to merge Mauro's patch for the upcoming 26.3 release so that
> there is at least one working version in 26.x releases?

Yes, please go ahead.

bug#36459: 27.0.50; octave-mode ElDoc support bug

[Leo Liu]

version: 26.3

On 2019-08-08 16:50 +0300, Eli Zaretskii wrote:
> Yes, please go ahead.

Excellent. Thanks to everyone for help and testing.