Gnus development mailing list
 help / color / mirror / Atom feed
* New DNS tld “might be bogus”.
@ 2018-10-08 15:53 Garreau, Alexandre
  2018-10-08 17:12 ` Eric Abrahamsen
  0 siblings, 1 reply; 8+ messages in thread
From: Garreau, Alexandre @ 2018-10-08 15:53 UTC (permalink / raw)
  To: ding

It seems Gnus is getting issues with new fancy ICANN tlds, like (from my
*Messages*):

”Address ‘van@scratch.space’ might be bogus.  Continue? (y or n) y”

PS: please take care of putting me in copy as since some times I appear
unable to re-subscribe to this ding mailing-list (or maybe I should use
help-gnu-emacs instead as I’m subscribed and they got archives and such?
it’s still unclear to me where is it more appropriated to ask)…



^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: New DNS tld “might be bogus”.
  2018-10-08 15:53 New DNS tld “might be bogus” Garreau, Alexandre
@ 2018-10-08 17:12 ` Eric Abrahamsen
  2018-10-08 17:26   ` Garreau, Alexandre
  0 siblings, 1 reply; 8+ messages in thread
From: Eric Abrahamsen @ 2018-10-08 17:12 UTC (permalink / raw)
  To: ding

"Garreau, Alexandre" <galex-713@galex-713.eu> writes:

> It seems Gnus is getting issues with new fancy ICANN tlds, like (from my
> *Messages*):
>
> ”Address ‘van@scratch.space’ might be bogus.  Continue? (y or n) y”
>
> PS: please take care of putting me in copy as since some times I appear
> unable to re-subscribe to this ding mailing-list (or maybe I should use
> help-gnu-emacs instead as I’m subscribed and they got archives and such?
> it’s still unclear to me where is it more appropriated to ask)…

What's your value of `message-bogus-addresses'? My setup doesn't
complain about that message:

(message-bogus-recipient-p "van@scratch.space") => nil

Eric




^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: New DNS tld “might be bogus”.
  2018-10-08 17:12 ` Eric Abrahamsen
@ 2018-10-08 17:26   ` Garreau, Alexandre
  2018-10-08 18:21     ` Eric Abrahamsen
  0 siblings, 1 reply; 8+ messages in thread
From: Garreau, Alexandre @ 2018-10-08 17:26 UTC (permalink / raw)
  To: Eric Abrahamsen; +Cc: ding

Le 08/10/2018 à 10h12, Eric Abrahamsen a écrit :
> What's your value of `message-bogus-addresses'? My setup doesn't
> complain

message-bogus-addresses
=> ("noreply" "nospam" "invalid" "@@" "[^[:ascii:]].*@" "[ 	]")

(message-bogus-recipient-p "van@scratch.space") => ("van@scratch.space")

Perhaps a version problem? I’m using “Gnus v5.13”, under “GNU Emacs
25.1.1 (i686-pc-linux-gnu, GTK+ Version 3.22.11) of 2017-09-15, modified
by Debian”.  What’s yours? Mine is built-in in my debian emacs25 I guess.

PS(bis): I’m subscribed now it seems ^^



^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: New DNS tld “might be bogus”.
  2018-10-08 17:26   ` Garreau, Alexandre
@ 2018-10-08 18:21     ` Eric Abrahamsen
  2018-10-08 18:56       ` Garreau, Alexandre
  0 siblings, 1 reply; 8+ messages in thread
From: Eric Abrahamsen @ 2018-10-08 18:21 UTC (permalink / raw)
  To: ding

"Garreau, Alexandre" <galex-713@galex-713.eu> writes:

> Le 08/10/2018 à 10h12, Eric Abrahamsen a écrit :
>> What's your value of `message-bogus-addresses'? My setup doesn't
>> complain
>
> message-bogus-addresses
> => ("noreply" "nospam" "invalid" "@@" "[^[:ascii:]].*@" "[ 	]")

Huh, that's pretty close to mine.

> (message-bogus-recipient-p "van@scratch.space") => ("van@scratch.space")

I don't know why it would flag that message when it doesn't match
`message-bogus-addresses'.

> Perhaps a version problem? I’m using “Gnus v5.13”, under “GNU Emacs
> 25.1.1 (i686-pc-linux-gnu, GTK+ Version 3.22.11) of 2017-09-15, modified
> by Debian”.  What’s yours? Mine is built-in in my debian emacs25 I guess.

I'm running Emacs master -- Gnus is now developed within the Emacs code
base, as of... I can't remember, a year or two?

I don't know what `message-bogus-recipient-p' looks like in your
version, or why it's matching that address. If you haven't tried edebug
before, it's a useful tool. Find the function definition, hit "C-u
C-M-x" somewhere on it, then run the function. Keep whacking <SPC> and
see what it does. Do "C-M-x" again, without the prefix, to un-instrument
it.

Eric




^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: New DNS tld “might be bogus”.
  2018-10-08 18:21     ` Eric Abrahamsen
@ 2018-10-08 18:56       ` Garreau, Alexandre
  2018-10-08 19:33         ` Eric Abrahamsen
  0 siblings, 1 reply; 8+ messages in thread
From: Garreau, Alexandre @ 2018-10-08 18:56 UTC (permalink / raw)
  To: Eric Abrahamsen; +Cc: ding

On 2018/10/08 at 11:21, Eric Abrahamsen wrote:
> If you haven't tried edebug before, it's a useful tool. Find the
> function definition, hit "C-u C-M-x" somewhere on it, then run the
> function. Keep whacking <SPC> and see what it does. Do "C-M-x" again,
> without the prefix, to un-instrument it.

I knew edebug-defun, however I didn’t know it was automatically called
by C-M-x with a C-u prefix! thanks a lot!

> "Garreau, Alexandre" <galex-713@galex-713.eu> writes:
>> Le 08/10/2018 à 10h12, Eric Abrahamsen a écrit :
>>> What's your value of `message-bogus-addresses'? My setup doesn't
>>> complain
>> (message-bogus-recipient-p "van@scratch.space") => ("van@scratch.space")
>
> I don't know why it would flag that message when it doesn't match
> `message-bogus-addresses'.

> I don't know what `message-bogus-recipient-p' looks like in your
> version, or why it's matching that address.

No need to go until source or even debugging, my documentation’s last
paragraph says “An address might be bogus if the domain part is not
fully qualified, see ‘message-valid-fqdn-regexp’, or if there’s a
matching entry in ‘message-bogus-addresses’”, and in fact
`message-valid-fqdn-regexp' (content afterwards [0]) doesn’t contain
“space” as a TLD in my version.

So maybe the object of this discussion should have been since the
beginning “is it a good idea to just check with a manually-updated
variable a such quickly-changing concept such as valid TLD”? especially
as now, because it makes money I guess, there will be easily more and
more new TLDs more often than a new emacs release comes out.

So do we check for gTLDs + traditional generic TLDs and warn user about
“new shiny commercial TLDs”, or should we retrieve a TLD list from elpa,
or should we have a list of TLD updated at each release, lazily updated
for each TLD found in address that would appear to really exist (because
resolving the address with this TLD would work for instance), or should
we do one of the former options but by, instead of whitelisting,
blacklisting forbidden TLDs, or should we stop checking it?

Usually I would have been more incline to the last option, as all the
examples I saw who tried, failed to check for valid mail addresses
(since there are special syntaxes for embedding “@” or “ ” in address),
valid domains (unless it pre-translate to IDN, Gnus apparently fails at
this too), or, more importantly, failed at recognizing correctly a valid
TLD, as gTLD are regularly being added, and I see no problems in this.

However, since the new trend of popularizing all sort of whatever TLD,
and since we also regularely gets lying DNS resolvers (some ISP do that,
some questionable-legislations constrains in doing that too), and
potentially that’d help cracking I guess, and some new *forbidden* TLDs,
for questions of security, sometimes appear (.onion, .local, .pkey,
(.gnu?), etc.), a such functionality such as the firsts options could be
interesting.

In we go by blacklisting we go into interesting questions such as how to
deal with more decentralized things such as trying to send mail to a
.onion, .local, etc. domain.

> I'm running Emacs master -- Gnus is now developed within the Emacs code
> base, as of... I can't remember, a year or two?

Then yours version is more up-to-date and maybe “right” than mine, what
does `message-bogus-recipient-p'’s documentation says at yours? doesn’t
it mention `message-valid-fqdn-regexp' so to suggest debugging is the
best option?  or do you just have a more up-to-date variable content
that mentions even newer TLDs?

[0] "[a-z0-9][-.a-z0-9]+\\.\\([a-z][a-z]\\|aero\\|arpa\\|asia\\|bitnet\\|biz\\|bofh\\|cat\\|com\\|coop\\|edu\\|gov\\|info\\|int\\|jobs\\|mil\\|mobi\\|museum\\|name\\|net\\|org\\|pro\\|tel\\|travel\\|uucp\\|academy\\|actor\\|agency\\|airforce\\|archi\\|associates\\|axa\\|bar\\|bargains\\|bayern\\|beer\\|berlin\\|best\\|bid\\|bike\\|biz\\|black\\|blackfriday\\|blue\\|boutique\\|build\\|builders\\|buzz\\|cab\\|camera\\|camp\\|capital\\|cards\\|care\\|career\\|careers\\|cash\\|catering\\|center\\|ceo\\|cheap\\|christmas\\|church\\|citic\\|cleaning\\|clinic\\|clothing\\|club\\|codes\\|coffee\\|college\\|cologne\\|com\\|community\\|company\\|computer\\|construction\\|contractors\\|cooking\\|cool\\|country\\|creditcard\\|cruises\\|dance\\|dating\\|democrat\\|dental\\|desi\\|design\\|diamonds\\|directory\\|discount\\|domains\\|education\\|email\\|engineering\\|enterprises\\|equipment\\|estate\\|eus\\|events\\|exchange\\|expert\\|exposed\\|fail\\|farm\\|feedback\\|finance\\|financial\\|fish\\|fishing\\|fitness\\|flights\\|florist\\|foo\\|foundation\\|frogans\\|fund\\|furniture\\|futbol\\|gal\\|gallery\\|gift\\|glass\\|globo\\|gmo\\|gop\\|graphics\\|gratis\\|gripe\\|guide\\|guitars\\|guru\\|hamburg\\|haus\\|hiphop\\|holdings\\|holiday\\|homes\\|horse\\|house\\|immobilien\\|industries\\|info\\|ink\\|institute\\|insure\\|international\\|investments\\|jetzt\\|juegos\\|kaufen\\|kim\\|kitchen\\|kiwi\\|koeln\\|kred\\|land\\|lat\\|latino\\|lease\\|life\\|lighting\\|limited\\|limo\\|link\\|loans\\|london\\|luxe\\|luxury\\|management\\|mango\\|marketing\\|media\\|meet\\|menu\\|miami\\|moda\\|moe\\|monash\\|moscow\\|motorcycles\\|nagoya\\|name\\|net\\|neustar\\|ninja\\|nyc\\|okinawa\\|onl\\|org\\|paris\\|partners\\|parts\\|photo\\|photography\\|photos\\|pics\\|pictures\\|pink\\|plumbing\\|pro\\|productions\\|properties\\|pub\\|qpon\\|quebec\\|recipes\\|red\\|reisen\\|ren\\|rentals\\|repair\\|report\\|rest\\|reviews\\|rich\\|rocks\\|rodeo\\|ruhr\\|ryukyu\\|saarland\\|schule\\|scot\\|services\\|sexy\\|shiksha\\|shoes\\|singles\\|social\\|sohu\\|solar\\|solutions\\|soy\\|supplies\\|supply\\|support\\|surgery\\|systems\\|tattoo\\|tax\\|technology\\|tienda\\|tips\\|today\\|tokyo\\|tools\\|town\\|toys\\|trade\\|training\\|university\\|uno\\|vacations\\|vegas\\|ventures\\|viajes\\|villas\\|vision\\|vodka\\|vote\\|voting\\|voto\\|voyage\\|wang\\|watch\\|webcam\\|wed\\|wien\\|wiki\\|works\\|wtc\\|wtf\\|xyz\\|yachts\\|yokohama\\|you\\|zone\\)"



^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: New DNS tld “might be bogus”.
  2018-10-08 18:56       ` Garreau, Alexandre
@ 2018-10-08 19:33         ` Eric Abrahamsen
  2018-10-08 19:43           ` Adam Sjøgren
  2018-10-09 10:59           ` Bjørn Mork
  0 siblings, 2 replies; 8+ messages in thread
From: Eric Abrahamsen @ 2018-10-08 19:33 UTC (permalink / raw)
  To: ding

"Garreau, Alexandre" <galex-713@galex-713.eu> writes:
>> "Garreau, Alexandre" <galex-713@galex-713.eu> writes:
>>> Le 08/10/2018 à 10h12, Eric Abrahamsen a écrit :
>>>> What's your value of `message-bogus-addresses'? My setup doesn't
>>>> complain
>>> (message-bogus-recipient-p "van@scratch.space") => ("van@scratch.space")
>>
>> I don't know why it would flag that message when it doesn't match
>> `message-bogus-addresses'.
>
>> I don't know what `message-bogus-recipient-p' looks like in your
>> version, or why it's matching that address.
>
> No need to go until source or even debugging, my documentation’s last
> paragraph says “An address might be bogus if the domain part is not
> fully qualified, see ‘message-valid-fqdn-regexp’, or if there’s a
> matching entry in ‘message-bogus-addresses’”, and in fact
> `message-valid-fqdn-regexp' (content afterwards [0]) doesn’t contain
> “space” as a TLD in my version.
>
> So maybe the object of this discussion

I can short-circuit this and tell you that, in the interim, someone's
already had the discussion, and `message-valid-fqdn-regexp' doesn't even
exist anymore in Emacs 27. I think the conclusion was that times have
changed and it's harder to predict what the internet will do, and we
shouldn't be pedantically restrictive when it's not even clear what's
"right" anymore. So there's that!

Emacs 27 isn't released yet, but I'd recommend switching to 26 if
possible. It looks like this option was removed in Feb 2016, so Emacs 26
probably has that change.

Eric




^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: New DNS tld “might be bogus”.
  2018-10-08 19:33         ` Eric Abrahamsen
@ 2018-10-08 19:43           ` Adam Sjøgren
  2018-10-09 10:59           ` Bjørn Mork
  1 sibling, 0 replies; 8+ messages in thread
From: Adam Sjøgren @ 2018-10-08 19:43 UTC (permalink / raw)
  To: ding

Eric writes:

> I can short-circuit this and tell you that, in the interim, someone's
> already had the discussion, and `message-valid-fqdn-regexp' doesn't even
> exist anymore in Emacs 27. I think the conclusion was that times have
> changed and it's harder to predict what the internet will do, and we
> shouldn't be pedantically restrictive when it's not even clear what's
> "right" anymore. So there's that!

Yes, it was removed a couple of years back:

  commit 762986c9d3671138f8fd4b248dbf536453f231dd
  Author: Lars Ingebrigtsen <larsi@gnus.org>
  Date:   Sun Feb 7 14:19:59 2016 +1100

      Remove message-valid-fqdn-regexp, since it changes too much now

      * lisp/gnus/message.el (message-valid-fqdn-regexp): Remove.
      (message-bogus-recipient-p): Don't use it any more.
      (message-make-fqdn): Ditto.  Suggested by Lars-Johan Liman.

   - http://git.savannah.gnu.org/cgit/emacs.git/commit/?id=762986c9d3


  commit 3a2b6aa33109dc40b2c1bcc621a624d38fe814fc
  Author: Lars Ingebrigtsen <larsi@gnus.org>
  Date:   Sun Feb 7 14:30:16 2016 +1100

      message-valid-fqdn-regexp no longer exists

      * lisp/gnus/gnus-art.el (gnus-button-valid-fqdn-regexp): Don't use
      the no-longer-existing message-valid-fqdn-regexp variable.

   - http://git.savannah.gnu.org/cgit/emacs.git/commit/?id=3a2b6aa331


  Best regards,

    Adam

-- 
 "The key to making programs fast is to make them do          Adam Sjøgren
  practically nothing. ;-)"                              asjo@koldfront.dk




^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: New DNS tld “might be bogus”.
  2018-10-08 19:33         ` Eric Abrahamsen
  2018-10-08 19:43           ` Adam Sjøgren
@ 2018-10-09 10:59           ` Bjørn Mork
  1 sibling, 0 replies; 8+ messages in thread
From: Bjørn Mork @ 2018-10-09 10:59 UTC (permalink / raw)
  To: Eric Abrahamsen; +Cc: ding

Eric Abrahamsen <eric@ericabrahamsen.net> writes:
> "Garreau, Alexandre" <galex-713@galex-713.eu> writes:
>
>> No need to go until source or even debugging, my documentation’s last
>> paragraph says “An address might be bogus if the domain part is not
>> fully qualified, see ‘message-valid-fqdn-regexp’, or if there’s a
>> matching entry in ‘message-bogus-addresses’”, and in fact
>> `message-valid-fqdn-regexp' (content afterwards [0]) doesn’t contain
>> “space” as a TLD in my version.
>>
>> So maybe the object of this discussion
>
> I can short-circuit this and tell you that, in the interim, someone's
> already had the discussion, and `message-valid-fqdn-regexp' doesn't even
> exist anymore in Emacs 27. I think the conclusion was that times have
> changed and it's harder to predict what the internet will do, and we
> shouldn't be pedantically restrictive when it's not even clear what's
> "right" anymore. So there's that!
>
> Emacs 27 isn't released yet, but I'd recommend switching to 26 if
> possible. It looks like this option was removed in Feb 2016, so Emacs 26
> probably has that change.


Or simply set it to something like

 (setq message-valid-fqdn-regexp "[a-z0-9][-.a-z0-9]+\\.[a-z0-9][-.a-z0-9]+")

until you can upgrade emacs.  Works For Me (tm).




Bjørn



^ permalink raw reply	[flat|nested] 8+ messages in thread

end of thread, other threads:[~2018-10-09 10:59 UTC | newest]

Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-10-08 15:53 New DNS tld “might be bogus” Garreau, Alexandre
2018-10-08 17:12 ` Eric Abrahamsen
2018-10-08 17:26   ` Garreau, Alexandre
2018-10-08 18:21     ` Eric Abrahamsen
2018-10-08 18:56       ` Garreau, Alexandre
2018-10-08 19:33         ` Eric Abrahamsen
2018-10-08 19:43           ` Adam Sjøgren
2018-10-09 10:59           ` Bjørn Mork

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 NNTP newsgroup(s).