Re: Maybe a bug after df89d6d6dc429e638e425536d1c201b5373f2abd

2019-12-27 Thread Garjola Dindi
On Sat 28-Dec-2019 at 02:29:12 +01, Lars Ingebrigtsen  wrote: 
> Adam Sjøgren via info-gnus-english  writes:
>
>> Is this a fix for:
>
> [...]
>
>>   * lisp/gnus/gnus-start.el: Use lexical-binding
>
> Yup.

Thank you for the quick answer and fix!

I lost my .newrc.eld, but this is my fault for forgetting to add it to
my backup script.

-- 


___
info-gnus-english mailing list
info-gnus-english@gnu.org
https://lists.gnu.org/mailman/listinfo/info-gnus-english


Re: Maybe a bug after df89d6d6dc429e638e425536d1c201b5373f2abd

2019-12-27 Thread Lars Ingebrigtsen
Adam Sjøgren via info-gnus-english  writes:

> Is this a fix for:

[...]

>   * lisp/gnus/gnus-start.el: Use lexical-binding

Yup.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

___
info-gnus-english mailing list
info-gnus-english@gnu.org
https://lists.gnu.org/mailman/listinfo/info-gnus-english


From and Reply-To mangling?!

2019-12-27 Thread Adam Sjøgren via info-gnus-english
So this mailing list now changes the From: and adds Reply-To: ?!?

Only for some users/posts. Hm.

-- 
 "The Commodore 64. Under $600. You can't buy a Adam Sjøgren
  better computer at twice the price." a...@koldfront.dk


___
info-gnus-english mailing list
info-gnus-english@gnu.org
https://lists.gnu.org/mailman/listinfo/info-gnus-english


Re: Maybe a bug after df89d6d6dc429e638e425536d1c201b5373f2abd

2019-12-27 Thread Adam Sjøgren via info-gnus-english
Adam Sjøgren writes:

>> I've now pushed a fix for this to both emacs-27 and the trunk.
>
> Is this a fix for:

Your commit e1fd6b1107 indeed fixes the issue for me.

Thanks!


  Best regards,

Adam

-- 
 "Vilken sanning, Måns, är sann?"   Adam Sjøgren
   a...@koldfront.dk


___
info-gnus-english mailing list
info-gnus-english@gnu.org
https://lists.gnu.org/mailman/listinfo/info-gnus-english


Re: Maybe a bug after df89d6d6dc429e638e425536d1c201b5373f2abd

2019-12-27 Thread Adam Sjøgren via info-gnus-english
Lars writes:

> Garjola Dindi  writes:
>
>> I have a strange behaviour with recent versions of gnus (in emacs
>> master and branch emacs-27).
>>
>> When I try to subscribe to a group in the *Browse server buffer*, all
>> groups appear with the ’K’ mark and the subscribe command ’u’ does not
>> subscribe (the ’K’ still appears and the group is not subscribed).
>
> This was due to a recent change -- whenever a group changed level, it
> would put a nil onto gnus-newsrc-alist, thereby making it invalid.
>
> I've now pushed a fix for this to both emacs-27 and the trunk.

Is this a fix for:

  7b3b27eef1325b41187f2e9cc898644d5eae9ff3 is the first bad commit
  commit 7b3b27eef1325b41187f2e9cc898644d5eae9ff3
  Author: Stefan Monnier 
  Date:   Sun Dec 22 16:09:31 2019 -0500

  * lisp/gnus/gnus-start.el: Use lexical-binding

  (gnus-group-change-level, gnus-make-hashtable-from-newsrc-alist):
  Use gnus-info-make.
  (gnus-make-hashtable-from-newsrc-alist): Prefer `gnus-info-group`
  to `car` when applied to a `gnus-info` object.
  (gnus-make-hashtable-from-killed): Remove unused vars `lists` and `list`.
  (gnus-gnus-to-quick-newsrc-format): Extract common code from if branches.

  * lisp/gnus/gnus.el (gnus-info-make): New constructor.

   lisp/gnus/gnus-start.el | 59 
++---
   lisp/gnus/gnus.el   |  4 
   2 files changed, 31 insertions(+), 32 deletions(-)

?

I just bisected the problem that Garjola and I have hit, where if you
press u on an unsubscribed group, you get an error, and when quitting
Gnus you get something like this (with toggle-debug-on-error):

  Debugger entered--Lisp error: (wrong-type-argument stringp nil)
encode-coding-string(nil utf-8-emacs)
#f(compiled-function (info) #)(nil)
mapcar(#f(compiled-function (info) #) (nil 
("nntp+fb:feedbase.blog.adafruit.q..." [...cut...] ...))
gnus-gnus-to-quick-newsrc-format()
gnus-save-newsrc-file()
gnus-group-exit()
funcall-interactively(gnus-group-exit)
call-interactively(gnus-group-exit nil nil)
command-execute(gnus-group-exit)

and when quitting Emacs, it asks if .newsrc.eld should be saved (which
will save an almost empty file).


  Best regards,

Adam

-- 
 "It's a shite state of affairs to be in, and noAdam Sjøgren
  amount of fresh air is ever going to change that."   a...@koldfront.dk


___
info-gnus-english mailing list
info-gnus-english@gnu.org
https://lists.gnu.org/mailman/listinfo/info-gnus-english


Re: Maybe a bug after df89d6d6dc429e638e425536d1c201b5373f2abd

2019-12-27 Thread Lars Ingebrigtsen
Garjola Dindi  writes:

> I have a strange behaviour with recent versions of gnus (in emacs
> master and branch emacs-27).
>
> When I try to subscribe to a group in the *Browse server buffer*, all
> groups appear with the ’K’ mark and the subscribe command ’u’ does not
> subscribe (the ’K’ still appears and the group is not subscribed).

This was due to a recent change -- whenever a group changed level, it
would put a nil onto gnus-newsrc-alist, thereby making it invalid.

I've now pushed a fix for this to both emacs-27 and the trunk.

> Also, the .newsrc.eld file does not contain anything after the line
> showing the Gnus version number. This means that all existing
> subscriptions are lost after exiting emacs.

Geez!  I didn't see that problem -- I just got some extra nil entries at
the start of the gnus-newsrc-alist variable.  I hope there's a backup of
the file...

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

___
info-gnus-english mailing list
info-gnus-english@gnu.org
https://lists.gnu.org/mailman/listinfo/info-gnus-english


Re: Maybe a bug after df89d6d6dc429e638e425536d1c201b5373f2abd

2019-12-27 Thread Adam Sjøgren via info-gnus-english
Garjola writes:

> When I try to subscribe to a group in the *Browse server buffer*, all
> groups appear with the ’K’ mark and the subscribe command ’u’ does not
> subscribe (the ’K’ still appears and the group is not subscribed).
>
> Also, the .newsrc.eld file does not contain anything after the line
> showing the Gnus version number. This means that all existing
> subscriptions are lost after exiting emacs.

Thanks for reporting, I have seen this problem as well - I thought it
was just me.


  Best regards,

Adam

-- 
 "For alle os andre betyder det vel ikke andet end at   Adam Sjøgren
  et par tosser for lang tid siden blev enige om ata...@koldfront.dk
  tælle. Og nu er de altså nået til 2000. Hurra.
  Mægtigt."


___
info-gnus-english mailing list
info-gnus-english@gnu.org
https://lists.gnu.org/mailman/listinfo/info-gnus-english


Re: Incorrect character encoding in received messages

2019-12-27 Thread Lars Ingebrigtsen
Garjola Dindi  writes:

> I have recently been having trouble with Gnus decoding some e-mails as
> ASCII when actually they should be decoded as unicode.

Sounds like there's no Content-Type header in the message that says what
the charset it.

> For instance, in French, the “à” char gets displayed as “\340”.

Then the message isn't encoded as utf-8, but is probably latin-1.

`C-u W M c' in the summary buffer should allow you to decode the message
in the proper charset.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

___
info-gnus-english mailing list
info-gnus-english@gnu.org
https://lists.gnu.org/mailman/listinfo/info-gnus-english