* bug#39682: 26.3; Help window for list-packages disabled
@ 2020-02-19 22:23 Tim Johnson
2020-02-20 0:47 ` Noam Postavsky
0 siblings, 1 reply; 19+ messages in thread
From: Tim Johnson @ 2020-02-19 22:23 UTC (permalink / raw)
To: 39682
Alternate email address is tim@akwebsoft.com
;; -------------------------------------------------------------
Problem:
When the following occurs
1) dired+ is loaded
2) package is loaded
3) package is initialized
No *Help* window is opened and the error message: "No Package
Specified" is displayed in the minibuffer.
Test Case:
loading both dired+ and packages AND initializing packages seems to
cause this problem when an abbreviated init.el is used.
If packages is not initialized
OR
dired+ is not loaded
the list-packages *Help* window will be displayed for package info
Code for the test init.el follows:
;; code begins
(add-to-list 'load-path "~/.emacs.d/elisp") ;; edit for your path to dired+
(require 'package)
;; if the following is executed and dired+ is also loaded
;; the list-packages *Help* window will be disabled
(package-initialize)
;; if dired+ is loaded and (package-initialize) is also executed
;; the list-packages *Help* window will be disabled
(require 'dired+)
;; code ends
;; It appears to me that if either of the two cases occur
;; 1) dired+ is NOT loaded
;; 2) package is NOT initialized
;; Then the *Help* window WILL be displayed
;; -------------------------------------------------------------
In GNU Emacs 26.3 (build 1, x86_64-pc-linux-gnu, GTK+ Version 2.24.30)
of 2020-02-17 built on linus
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.
scroll-up-command: End of buffer
Configured features:
XPM JPEG TIFF GIF PNG SOUND GSETTINGS GLIB NOTIFY ACL GNUTLS LIBXML2
FREETYPE XFT ZLIB TOOLKIT_SCROLL_BARS GTK2 X11 XDBE XIM THREADS
Important settings:
value of $LANG: en_US.UTF-8
locale-coding-system: utf-8-unix
Major mode: Fundamental
Minor modes in effect:
tooltip-mode: t
global-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
blink-cursor-mode: t
auto-composition-mode: t
auto-encryption-mode: t
auto-compression-mode: t
buffer-read-only: t
line-number-mode: t
transient-mark-mode: t
Load-path shadows:
/home/tim/.emacs.d/elpa/bm-20190807.1217/bm hides ~/.emacs.d/elisp/bm
~/.emacs.d/elisp/newcomment hides
/usr/local/share/emacs/26.3/lisp/newcomment
Features:
(shadow sort mail-extr emacsbug message rmc puny rfc822 mml mml-sec epa
epg gnus-util rmail rmail-loaddefs mm-decode mm-bodies mm-encode
mail-parse rfc2231 mailabbrev gmm-utils mailheader sendmail rfc2047
rfc2045 ietf-drums mm-util mail-prsvr mail-utils elec-pair edmacro
kmacro easy-mmode derived advice dired+ image-dired image-mode
format-spec image-file help-fns+ wid-edit help-fns radix-tree help-mode
dired-x dired-aux dired dired-loaddefs cl finder-inf
dired-single-autoloads idomenu-autoloads iodine-theme-autoloads info
mark-tools-autoloads molokai-theme-autoloads show-marks-autoloads
fm-autoloads simpleclip-autoloads visible-mark-autoloads package
easymenu epg-config url-handlers url-parse auth-source cl-seq eieio
eieio-core cl-macs eieio-loaddefs password-cache url-vars seq byte-opt
gv bytecomp byte-compile cconv cl-loaddefs cl-lib time-date 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 inotify dynamic-setting system-font-setting
font-render-setting move-toolbar gtk x-toolkit x multi-tty
make-network-process emacs)
Memory information:
((conses 16 174874 7573)
(symbols 48 25419 1)
(miscs 40 95 146)
(strings 32 45594 1867)
(string-bytes 1 1401206)
(vectors 16 18756)
(vector-slots 8 554228 11294)
(floats 8 67 93)
(intervals 56 265 0)
(buffers 992 13)
(heap 1024 23260 1613))
--
Tim Johnson
http://www.tj49.com
--
Tim
tj49.com
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-19 22:23 bug#39682: 26.3; Help window for list-packages disabled Tim Johnson
@ 2020-02-20 0:47 ` Noam Postavsky
2020-02-20 1:50 ` Drew Adams
0 siblings, 1 reply; 19+ messages in thread
From: Noam Postavsky @ 2020-02-20 0:47 UTC (permalink / raw)
To: Tim Johnson; +Cc: 39682
retitle 39682 26.3; Help window for list-packages disabled when dired+ is loaded
tags 39682 + unreproducible
quit
Tim Johnson <tim@tj49.com> writes:
> ;; code begins
> (add-to-list 'load-path "~/.emacs.d/elisp") ;; edit for your path to dired+
> (require 'package)
> ;; if the following is executed and dired+ is also loaded
> ;; the list-packages *Help* window will be disabled
> (package-initialize)
> ;; if dired+ is loaded and (package-initialize) is also executed
> ;; the list-packages *Help* window will be disabled
> (require 'dired+)
> ;; code ends
>
> ;; It appears to me that if either of the two cases occur
> ;; 1) dired+ is NOT loaded
> ;; 2) package is NOT initialized
> ;; Then the *Help* window WILL be displayed
You mean the one that pops up when pressing '?' in the *Packages*
buffer? Seems to work for me regardless of whether or not dired+ is
loaded.
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-20 0:47 ` Noam Postavsky
@ 2020-02-20 1:50 ` Drew Adams
2020-02-20 2:01 ` Noam Postavsky
0 siblings, 1 reply; 19+ messages in thread
From: Drew Adams @ 2020-02-20 1:50 UTC (permalink / raw)
To: Noam Postavsky, Tim Johnson; +Cc: 39682
> retitle 39682 26.3; Help window for list-packages disabled when dired+ is
> loaded
> tags 39682 + unreproducible
> quit
Give it any title you like, of course.
But I think this title may be misleading.
IIUC, the problem arises also without Dired+
- with el-get.el instead, for example.
https://lists.gnu.org/archive/html/help-gnu-emacs/2020-02/msg00101.html
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-20 1:50 ` Drew Adams
@ 2020-02-20 2:01 ` Noam Postavsky
2020-02-20 16:52 ` Tim Johnson
0 siblings, 1 reply; 19+ messages in thread
From: Noam Postavsky @ 2020-02-20 2:01 UTC (permalink / raw)
To: Drew Adams; +Cc: 39682, Tim Johnson
Drew Adams <drew.adams@oracle.com> writes:
>> retitle 39682 26.3; Help window for list-packages disabled when dired+ is
>
> Give it any title you like, of course.
>
> But I think this title may be misleading.
> IIUC, the problem arises also without Dired+
> - with el-get.el instead, for example.
>
> https://lists.gnu.org/archive/html/help-gnu-emacs/2020-02/msg00101.html
As far as I understood that thread, it was el-get in addition to
dired+. This bug report doesn't mention el-get though. And so far I
haven't been able to reproduce problem anyway.
Tim, could you post what M-: (tabulated-list-get-id) RET gives when this
problem happens? (That seems to be important part of
package-menu-describe-package, which is what '?' calls.)
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-20 2:01 ` Noam Postavsky
@ 2020-02-20 16:52 ` Tim Johnson
2020-02-20 17:12 ` Drew Adams
2020-02-20 17:53 ` Noam Postavsky
0 siblings, 2 replies; 19+ messages in thread
From: Tim Johnson @ 2020-02-20 16:52 UTC (permalink / raw)
To: Noam Postavsky, Drew Adams; +Cc: 39682
On 2/19/20 5:01 PM, Noam Postavsky wrote:
> Drew Adams <drew.adams@oracle.com> writes:
>
>>> retitle 39682 26.3; Help window for list-packages disabled when dired+ is
>> Give it any title you like, of course.
>>
>> But I think this title may be misleading.
>> IIUC, the problem arises also without Dired+
>> - with el-get.el instead, for example.
>>
>> https://lists.gnu.org/archive/html/help-gnu-emacs/2020-02/msg00101.html
> As far as I understood that thread, it was el-get in addition to
> dired+. This bug report doesn't mention el-get though. And so far I
> haven't been able to reproduce problem anyway.
>
> Tim, could you post what M-: (tabulated-list-get-id) RET gives when this
> problem happens? (That seems to be important part of
> package-menu-describe-package, which is what '?' calls.)
>
Sorry about that Drew. The conclusions from the thread were as you
describe them. While following code from el-get I found that the error
manifested at the point of a sexp that loaded packages with additional
arguments. I.E. in
el-get-elpa.el -> (require 'package nil t) or something like that.
I'm operating with a bare minimum of elisp knowledge AND I've been
retired for 5 years. ... :)
Here is *Messages* and tabulated-list-get-id
For information about GNU Emacs and the GNU system, type C-h C-a.
Importing package-keyring.gpg...done
Package refresh done
No package specified
Quit
#s(package-desc ace-window (0 9 0) "Quickly switch windows." ((avy (0 2
0))) single "gnu" nil ((:keywords "window" "location") (:authors ("Oleh
Krehel" . "ohwoeowho@gmail.com")) (:maintainer "Oleh Krehel" .
"ohwoeowho@gmail.com") (:url . "https://github.com/abo-abo/ace-window"))
nil)
HTH
--
Tim
tj49.com
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-20 16:52 ` Tim Johnson
@ 2020-02-20 17:12 ` Drew Adams
2020-02-20 17:53 ` Noam Postavsky
1 sibling, 0 replies; 19+ messages in thread
From: Drew Adams @ 2020-02-20 17:12 UTC (permalink / raw)
To: Tim Johnson, Noam Postavsky; +Cc: 39682
Thanks for clarifying. And congratulations (!) on retiring.
> Sorry about that Drew. The conclusions from the thread were as you
> describe them. While following code from el-get I found that the error
> manifested at the point of a sexp that loaded packages with additional
> arguments. I.E. in
>
> el-get-elpa.el -> (require 'package nil t) or something like that.
>
> I'm operating with a bare minimum of elisp knowledge AND I've been
> retired for 5 years. ... :)
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-20 16:52 ` Tim Johnson
2020-02-20 17:12 ` Drew Adams
@ 2020-02-20 17:53 ` Noam Postavsky
2020-02-20 20:12 ` Tim Johnson
1 sibling, 1 reply; 19+ messages in thread
From: Noam Postavsky @ 2020-02-20 17:53 UTC (permalink / raw)
To: Tim Johnson; +Cc: 39682
Tim Johnson <tim@tj49.com> writes:
> Here is *Messages* and tabulated-list-get-id
>
> For information about GNU Emacs and the GNU system, type C-h C-a.
> Importing package-keyring.gpg...done
> Package refresh done
> No package specified
> Quit
> #s(package-desc ace-window (0 9 0) "Quickly switch windows." ((avy (0
> 2 0))) single "gnu" nil ((:keywords "window" "location") (:authors
> ("Oleh Krehel" . "ohwoeowho@gmail.com")) (:maintainer "Oleh Krehel"
> . "ohwoeowho@gmail.com") (:url
> . "https://github.com/abo-abo/ace-window")) nil)
Hmm, that seems like it should be okay. What about
M-: (package-desc-p (tabulated-list-get-id)) RET
Also try (setq debug-on-message "No package specified") and then trigger
the error again. Perhaps the resulting backtrace will give some hints
as to what is happening.
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-20 17:53 ` Noam Postavsky
@ 2020-02-20 20:12 ` Tim Johnson
2020-02-20 20:22 ` Noam Postavsky
0 siblings, 1 reply; 19+ messages in thread
From: Tim Johnson @ 2020-02-20 20:12 UTC (permalink / raw)
To: Noam Postavsky; +Cc: 39682
Following Noam's instructions, here is a dump of *Messages*
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
For information about GNU Emacs and the GNU system, type C-h C-a.
Importing package-keyring.gpg...done
Package refresh done
No package specified
t
No package specified
previous-line: Beginning of buffer
t
"No package specified"
No package specified
Entering debugger...
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
I note that
(package-desc-p (tabulated-list-get-id))
just returns t
below is dump of *Backtrace* using different methods
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
Debugger entered--Lisp error: "No package specified"
message("No package specified")
describe-package(#s(package-desc :name ack :version (1 8) :summary "interface to ack-like tools" :reqs nil :kind tar :archive "gnu" :dir nil :extras ((:url . "https://github.com/leoliu/ack-el") (:maintainer "João Távora" . "joaotavora@gmail.com") (:authors ("Leo Liu" . "sdl.web@gmail.com")) (:keywords "tools" "processes" "convenience")) :signed nil))
package-menu-describe-package()
funcall-interactively(package-menu-describe-package)
call-interactively(package-menu-describe-package nil nil)
command-execute(package-menu-describe-package)
recursive-edit()
debug(error "No package specified")
message("No package specified")
describe-package(#s(package-desc :name ack :version (1 8) :summary "interface to ack-like tools" :reqs nil :kind tar :archive "gnu" :dir nil :extras ((:url . "https://github.com/leoliu/ack-el") (:maintainer "João Távora" . "joaotavora@gmail.com") (:authors ("Leo Liu" . "sdl.web@gmail.com")) (:keywords "tools" "processes" "convenience")) :signed nil))
package-menu-describe-package()
funcall-interactively(package-menu-describe-package)
call-interactively(package-menu-describe-package nil nil)
command-execute(package-menu-describe-package)
recursive-edit()
debug(error "No package specified")
message("No package specified")
describe-package(#s(package-desc :name ace-window :version (0 9 0) :summary "Quickly switch windows." :reqs ((avy (0 2 0))) :kind single :archive "gnu" :dir nil :extras ((:keywords "window" "location") (:authors ("Oleh Krehel" . "ohwoeowho@gmail.com")) (:maintainer "Oleh Krehel" . "ohwoeowho@gmail.com") (:url . "https://github.com/abo-abo/ace-window")) :signed nil))
package-menu-describe-package(#<marker (moves after insertion) at 8 in *Packages*>)
button-activate(#<marker (moves after insertion) at 8 in *Packages*> t)
push-button(8 t)
push-button((mouse-2 (#<window 6 on *Packages*> 8 (65 . 11) 260014406 nil 8 (7 . 0) nil (2 . 11) (9 . 17))))
funcall-interactively(push-button (mouse-2 (#<window 6 on *Packages*> 8 (65 . 11) 260014406 nil 8 (7 . 0) nil (2 . 11) (9 . 17))))
call-interactively(push-button nil nil)
command-execute(push-button)
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
On 2/20/20 8:53 AM, Noam Postavsky wrote:
> (setq debug-on-message "No package specified")
Thanks
--
Tim
tj49.com
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-20 20:12 ` Tim Johnson
@ 2020-02-20 20:22 ` Noam Postavsky
2020-02-20 21:16 ` Tim Johnson
0 siblings, 1 reply; 19+ messages in thread
From: Noam Postavsky @ 2020-02-20 20:22 UTC (permalink / raw)
To: Tim Johnson; +Cc: 39682
Tim Johnson <tim@tj49.com> writes:
> (package-desc-p (tabulated-list-get-id))
> just returns t
> message("No package specified")
> describe-package(#s(package-desc :name ack :version (1 8) :summary...
> package-menu-describe-package()
Hmm, very strange. Could you step through describe-package with edebug,
and see how this condition is getting taken, even though package-desc-p
should apparently be returning t?
(if (not (or (package-desc-p package) (and package (symbolp package))))
(message "No package specified")
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-20 20:22 ` Noam Postavsky
@ 2020-02-20 21:16 ` Tim Johnson
2020-02-20 21:30 ` Noam Postavsky
0 siblings, 1 reply; 19+ messages in thread
From: Tim Johnson @ 2020-02-20 21:16 UTC (permalink / raw)
To: Noam Postavsky; +Cc: 39682
Noam, I am not familiar with edebug.
Can you give me some more code examples?
On 2/20/20 11:22 AM, Noam Postavsky wrote:
> Tim Johnson <tim@tj49.com> writes:
>
>> (package-desc-p (tabulated-list-get-id))
>> just returns t
>> message("No package specified")
>> describe-package(#s(package-desc :name ack :version (1 8) :summary...
>> package-menu-describe-package()
> Hmm, very strange. Could you step through describe-package with edebug,
> and see how this condition is getting taken, even though package-desc-p
> should apparently be returning t?
>
> (if (not (or (package-desc-p package) (and package (symbolp package))))
> (message "No package specified")
>
>
--
Tim
tj49.com
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-20 21:16 ` Tim Johnson
@ 2020-02-20 21:30 ` Noam Postavsky
2020-02-20 23:14 ` Tim Johnson
2020-02-20 23:38 ` Tim Johnson
0 siblings, 2 replies; 19+ messages in thread
From: Noam Postavsky @ 2020-02-20 21:30 UTC (permalink / raw)
To: Tim Johnson; +Cc: 39682
Tim Johnson <tim@tj49.com> writes:
> Noam, I am not familiar with edebug.
>
> Can you give me some more code examples?
Basically just put point on the source code of describe-package, and hit
C-u C-M-x, that will instrument the function. Then run
package-menu-describe-package with "?" or however, and you will stop in
describe-package. Press space to step through each expression.
> On 2/20/20 11:22 AM, Noam Postavsky wrote:
>> Tim Johnson <tim@tj49.com> writes:
>>
>>> (package-desc-p (tabulated-list-get-id))
>>> just returns t
>>> message("No package specified")
>>> describe-package(#s(package-desc :name ack :version (1 8) :summary...
>>> package-menu-describe-package()
>> Hmm, very strange. Could you step through describe-package with edebug,
>> and see how this condition is getting taken, even though package-desc-p
>> should apparently be returning t?
>>
>> (if (not (or (package-desc-p package) (and package (symbolp package))))
>> (message "No package specified")
>>
>>
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-20 21:30 ` Noam Postavsky
@ 2020-02-20 23:14 ` Tim Johnson
2020-02-20 23:38 ` Tim Johnson
1 sibling, 0 replies; 19+ messages in thread
From: Tim Johnson @ 2020-02-20 23:14 UTC (permalink / raw)
To: Noam Postavsky; +Cc: 39682
I'm sorry, but I am not seeing anything that edifies me. Pressing SPC
does take me through the function, but the edebug instrumentation ends
at the end of defun without showing me anything that I could share.
But, there is something else that may be helpful:
The source code I am directed to is
.emacs.d/elpa/help-fns+-20170223.733/help-fns+.el
Does it not appear that I am executing an override function perhaps from
an older version of the helps-fns+ package?
Before I look for an upgrade, I wanted all concerned to read this in
case you have other analysis in mind.
On 2/20/20 12:30 PM, Noam Postavsky wrote:
> Tim Johnson <tim@tj49.com> writes:
>
>> Noam, I am not familiar with edebug.
>>
>> Can you give me some more code examples?
> Basically just put point on the source code of describe-package, and hit
> C-u C-M-x, that will instrument the function. Then run
> package-menu-describe-package with "?" or however, and you will stop in
> describe-package. Press space to step through each expression.
>
>> On 2/20/20 11:22 AM, Noam Postavsky wrote:
>>> Tim Johnson <tim@tj49.com> writes:
>>>
>>>> (package-desc-p (tabulated-list-get-id))
>>>> just returns t
>>>> message("No package specified")
>>>> describe-package(#s(package-desc :name ack :version (1 8) :summary...
>>>> package-menu-describe-package()
>>> Hmm, very strange. Could you step through describe-package with edebug,
>>> and see how this condition is getting taken, even though package-desc-p
>>> should apparently be returning t?
>>>
>>> (if (not (or (package-desc-p package) (and package (symbolp package))))
>>> (message "No package specified")
>>>
>>>
--
Tim
tj49.com
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-20 21:30 ` Noam Postavsky
2020-02-20 23:14 ` Tim Johnson
@ 2020-02-20 23:38 ` Tim Johnson
2020-02-21 0:02 ` Drew Adams
1 sibling, 1 reply; 19+ messages in thread
From: Tim Johnson @ 2020-02-20 23:38 UTC (permalink / raw)
To: Noam Postavsky; +Cc: 39682
Possible solution!
I loaded emacs -q, ran list-packages, look at the info for help-fns+.el
and understood that it would need to be re-compiled with a newer version
of emacs.
I just re-compiled help-fns+.el as per instructions in the info file and
the problem is solved.
I have much snow to shovel here - may not check email for several hours.
Hopefully further testing won't turn up further hiccups.
cheers
tim
On 2/20/20 12:30 PM, Noam Postavsky wrote:
> Tim Johnson <tim@tj49.com> writes:
>
>> Noam, I am not familiar with edebug.
>>
>> Can you give me some more code examples?
> Basically just put point on the source code of describe-package, and hit
> C-u C-M-x, that will instrument the function. Then run
> package-menu-describe-package with "?" or however, and you will stop in
> describe-package. Press space to step through each expression.
>
>> On 2/20/20 11:22 AM, Noam Postavsky wrote:
>>> Tim Johnson <tim@tj49.com> writes:
>>>
>>>> (package-desc-p (tabulated-list-get-id))
>>>> just returns t
>>>> message("No package specified")
>>>> describe-package(#s(package-desc :name ack :version (1 8) :summary...
>>>> package-menu-describe-package()
>>> Hmm, very strange. Could you step through describe-package with edebug,
>>> and see how this condition is getting taken, even though package-desc-p
>>> should apparently be returning t?
>>>
>>> (if (not (or (package-desc-p package) (and package (symbolp package))))
>>> (message "No package specified")
>>>
>>>
--
Tim
tj49.com
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-20 23:38 ` Tim Johnson
@ 2020-02-21 0:02 ` Drew Adams
2020-02-21 1:25 ` Noam Postavsky
[not found] ` <1c501888-2bf4-8c45-ae49-4d1e64ea114f@tj49.com>
0 siblings, 2 replies; 19+ messages in thread
From: Drew Adams @ 2020-02-21 0:02 UTC (permalink / raw)
To: Tim Johnson, Noam Postavsky; +Cc: 39682
> Possible solution!
>
> I loaded emacs -q, ran list-packages, look at the info for help-fns+.el
> and understood that it would need to be re-compiled with a newer version
> of emacs.
>
> I just re-compiled help-fns+.el as per instructions in the info file and
> the problem is solved.
I'm glad you seem to have solved the problem.
But I don't understand how you got help-fns+.el[c] loaded if you started Emacs using `emacs -q', which doesn't load your init file.
The original problem really needs to have been narrowed down to only the library (or libraries) really necessary to repro the problem.
To be clear, what happens if you ensure that you load _only_ el-get.el, and not help-fns+.el or dired+.el or anything else? E.g., use just `M-x load-file' for elget.el (or elget.elc).
> I have much snow to shovel here - may not check email for several hours.
That'll probably keep you healthy. Good shoveling.
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-21 0:02 ` Drew Adams
@ 2020-02-21 1:25 ` Noam Postavsky
[not found] ` <1c501888-2bf4-8c45-ae49-4d1e64ea114f@tj49.com>
1 sibling, 0 replies; 19+ messages in thread
From: Noam Postavsky @ 2020-02-21 1:25 UTC (permalink / raw)
To: Drew Adams; +Cc: 39682, Tim Johnson
retitle 39682 describe-package disabled by help-fns+.elc compiled in older Emacs
tags 39682 = notabug
close 39682
quit
Drew Adams <drew.adams@oracle.com> writes:
>> Possible solution!
>>
>> I loaded emacs -q, ran list-packages, look at the info for help-fns+.el
>> and understood that it would need to be re-compiled with a newer version
>> of emacs.
>>
>> I just re-compiled help-fns+.el as per instructions in the info file and
>> the problem is solved.
>
> I'm glad you seem to have solved the problem.
>
> But I don't understand how you got help-fns+.el[c] loaded if you
> started Emacs using `emacs -q', which doesn't load your init file.
>
> The original problem really needs to have been narrowed down to only
> the library (or libraries) really necessary to repro the problem.
Yeah, help-fns+ wasn't even mentioned until now, so this was all a bit
surprising. However, looking at its code I think it's clear what the
problem was: help-fns+ has a copy of the describe-package function,
which it overrides the builtin one with. The describe-package code
contains a call to package-desc-p which is an inline function. So a
help-fns+.elc produced by an older version of Emacs uses an older
version of package-desc-p that gives the wrong answer when handed a
new-style struct (i.e., a record rather than a vector). That's why it
decides that it's been handed something that doesn't specify a package
and complains.
Anyway, I don't think there's anything to fix here, so I'm closing the
report.
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
[not found] ` <1c501888-2bf4-8c45-ae49-4d1e64ea114f@tj49.com>
@ 2020-02-21 1:47 ` Drew Adams
2020-02-21 2:07 ` Noam Postavsky
0 siblings, 1 reply; 19+ messages in thread
From: Drew Adams @ 2020-02-21 1:47 UTC (permalink / raw)
To: Tim Johnson, 39682
> describe-package works fine when I invoke
> emacs -q and then load el-get only.
What's needed is a minimal recipe, which involves only the one (or perhaps two) libraries that repro the problem.
Maybe the problem is only in help-fns+.el. Or maybe it's only in dired+.el. Or maybe it was only in an old byte-compiled help-fns+.el.
The point is that as long as multiple things are loaded it's hard to know what's going on, especially at a distance.
Ideally, it would be narrowed down to a single library, and even to just some of the code in that library.
My guess now, based on the fact that help-fns+.el redefines `describe-package' (to add a cross-reference to the manuals), is that help-fns+.el alone is probably causing (or probably caused) the problem.
And if the problem was only with an older help-fns+.elc (byte-compiled) then that's probably the explanation. The library has two different definitions of `describe-package', depending on which Emacs version you have (because they changed the vanilla code for it, on which the library's code is based).
If your older Emacs version was prior to Emacs 24.4 then the older definition of `describe-package' was used in the byte-compiled file. And that likely won't work with Emacs 24.4 or later.
It would be good to know for sure that that was the problem. But if you no longer have the old help-fns+.elc (byte-compiled before Emacs 24.4) then there's probably no way to reproduce the problem anymore.
---
Anyway, I see now that Noam has replied more or less the same thing, and has closed the bug report.
Thanks for reporting the problem and trying to track it down. Sorry to all for the trouble.
> >> I have much snow to shovel here - may not check email for several hours.
> > That'll probably keep you healthy. Good shoveling.
> Starting now ...
Just don't overdo it. ;-) Good luck.
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-21 1:47 ` Drew Adams
@ 2020-02-21 2:07 ` Noam Postavsky
2020-02-21 3:27 ` Drew Adams
0 siblings, 1 reply; 19+ messages in thread
From: Noam Postavsky @ 2020-02-21 2:07 UTC (permalink / raw)
To: Drew Adams; +Cc: 39682, Tim Johnson
Drew Adams <drew.adams@oracle.com> writes:
> It would be good to know for sure that that was the problem. But if
> you no longer have the old help-fns+.elc (byte-compiled before Emacs
> 24.4) then there's probably no way to reproduce the problem anymore.
I can reproduce it with
emacs-25.3 -Q --batch -l package -f batch-byte-compile help-fns+.el
emacs-26.3 -Q -l package -l help-fns+.elc -f list-packages -f package-menu-describe-package
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-21 2:07 ` Noam Postavsky
@ 2020-02-21 3:27 ` Drew Adams
2020-02-21 18:20 ` Tim Johnson
0 siblings, 1 reply; 19+ messages in thread
From: Drew Adams @ 2020-02-21 3:27 UTC (permalink / raw)
To: Noam Postavsky; +Cc: 39682, Tim Johnson
> I can reproduce it with
>
> emacs-25.3 -Q --batch -l package -f batch-byte-compile help-fns+.el
> emacs-26.3 -Q -l package -l help-fns+.elc -f list-packages -f package-menu-
> describe-package
Good. Then it's just what we thought. Thx.
^ permalink raw reply [flat|nested] 19+ messages in thread
* bug#39682: 26.3; Help window for list-packages disabled
2020-02-21 3:27 ` Drew Adams
@ 2020-02-21 18:20 ` Tim Johnson
0 siblings, 0 replies; 19+ messages in thread
From: Tim Johnson @ 2020-02-21 18:20 UTC (permalink / raw)
To: Drew Adams, Noam Postavsky; +Cc: 39682
On 2/20/20 6:27 PM, Drew Adams wrote:
>> I can reproduce it with
>>
>> emacs-25.3 -Q --batch -l package -f batch-byte-compile help-fns+.el
>> emacs-26.3 -Q -l package -l help-fns+.elc -f list-packages -f package-menu-
>> describe-package
> Good. Then it's just what we thought. Thx.
>
I guess from my perspective this is not so much a bug as a strategic
omission on my part.
Since yesterday, I've found a number of discussions of remedial
byte-compiling of packages following upgrades of emacs. The onus is kind
of on the user.
--
Tim
tj49.com
^ permalink raw reply [flat|nested] 19+ messages in thread
end of thread, other threads:[~2020-02-21 18:20 UTC | newest]
Thread overview: 19+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2020-02-19 22:23 bug#39682: 26.3; Help window for list-packages disabled Tim Johnson
2020-02-20 0:47 ` Noam Postavsky
2020-02-20 1:50 ` Drew Adams
2020-02-20 2:01 ` Noam Postavsky
2020-02-20 16:52 ` Tim Johnson
2020-02-20 17:12 ` Drew Adams
2020-02-20 17:53 ` Noam Postavsky
2020-02-20 20:12 ` Tim Johnson
2020-02-20 20:22 ` Noam Postavsky
2020-02-20 21:16 ` Tim Johnson
2020-02-20 21:30 ` Noam Postavsky
2020-02-20 23:14 ` Tim Johnson
2020-02-20 23:38 ` Tim Johnson
2020-02-21 0:02 ` Drew Adams
2020-02-21 1:25 ` Noam Postavsky
[not found] ` <1c501888-2bf4-8c45-ae49-4d1e64ea114f@tj49.com>
2020-02-21 1:47 ` Drew Adams
2020-02-21 2:07 ` Noam Postavsky
2020-02-21 3:27 ` Drew Adams
2020-02-21 18:20 ` Tim Johnson
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).