message-make-fqdn generating strange Message-ID

message-make-fqdn generating strange Message-ID

[Adam Sjøgren]

Just having switched to Gnus from GNU Emacs master, Gnus complained:

  The Message-ID looks strange: "<87lgt59kxb.fsf@tullinup>".  Really post? (y or n)

This because gethostname() on my machine returns "tullinup" and
previously message-make-fqdn caught this by matching against
message-valid-fqdn-regexp, which was removed in 762986c9d36:

  -     ((and (string-match message-valid-fqdn-regexp sysname)
  -          (not (string-match message-bogus-system-names sysname)))
  +     ((and (string-match message-bogus-system-names sysname))

(larsi fixed the "((and" to "((not" later).

Should message-make-fqdn grow a simpler check ("Is there a '.' in the
sysname"), should the order of checks in it be shuffled around, or is my
machine broken when gethostname()¹ returns just the hostname
(hostname(1) returns the fqdn when given the -f option)?

For now I have set message-user-fqdn.


  Best regards,

    Adam


¹: #include <unistd.h>
#include <stdio.h>
#include <stdlib.h>

int main(void) {
  char hostname[1001];
  int ret;
  
  ret=gethostname(hostname, 1000);
  printf("%d gethostname: %s\n", ret, hostname);
  exit(0);
}

prints: 0 gethostname: tullinup

-- 
 "Angels can fly because they take themselves lightly."       Adam Sjøgren
                                                         asjo@koldfront.dk

Re: message-make-fqdn generating strange Message-ID

[Rasmus]

asjo@koldfront.dk (Adam Sjøgren) writes:

> Just having switched to Gnus from GNU Emacs master, Gnus complained:
>
>   The Message-ID looks strange: "<87lgt59kxb.fsf@tullinup>".  Really post? (y or n)

It's the same issue as #25702, right?

     https://debbugs.gnu.org/cgi/bugreport.cgi?bug=25702

Rasmus

-- 
Hooray!

Re: message-make-fqdn generating strange Message-ID

[Adam Sjøgren]

Rasmus writes:

> asjo@koldfront.dk (Adam Sjøgren) writes:
>
>> Just having switched to Gnus from GNU Emacs master, Gnus complained:
>>
>>   The Message-ID looks strange: "<87lgt59kxb.fsf@tullinup>".  Really post? (y or n)
>
> It's the same issue as #25702, right?
>
>      https://debbugs.gnu.org/cgi/bugreport.cgi?bug=25702

Yes, looks like the same thing.

Maybe I should start looking at the bug tracker before posting here. Hm.


  :-),

   Adam

-- 
 "The soap bubble experiment reveals something                Adam Sjøgren
  fundamental about the universe: it is lazy."           asjo@koldfront.dk