gnus-ignored-from-addresses

gnus-ignored-from-addresses

[Peter Münster]

Hi,

The format specification character %f in the summary line should display
the To: header, if gnus-ignored-from-addresses matches the From:
address.

I use gnus-ignored-from-addresses = "Peter Münster", because it would
match any of my addresses (Peter Münster <any@email.address>).

I'm quite sure, that this used to work, but with the latest git-version,
it does not.

Is this a new feature of a bug, and who could I get back the old
behaviour please?

TIA for any help,
-- 
           Peter

Re: gnus-ignored-from-addresses

[Katsumi Yamaoka]

Peter Münster wrote:
> The format specification character %f in the summary line should display
> the To: header, if gnus-ignored-from-addresses matches the From:
> address.

> I use gnus-ignored-from-addresses = "Peter Münster", because it would
> match any of my addresses (Peter Münster <any@email.address>).

> I'm quite sure, that this used to work, but with the latest git-version,
> it does not.

> Is this a new feature of a bug, and who could I get back the old
> behaviour please?

The code in question was changed (2013-01-14) so as to compare it
against only addresses, not the ones containing names.  I think it
is necessary to get rid of unexpected matching.  Now you can use
a regexp that matches the addresses you're using or a list of your
mail addresses.

Re: gnus-ignored-from-addresses

[Reiner Steib]

On Tue, Jun 25 2013, Katsumi Yamaoka wrote:

> Peter Münster wrote:
[...]
>> I use gnus-ignored-from-addresses = "Peter Münster", because it would
>> match any of my addresses (Peter Münster <any@email.address>).
>
>> I'm quite sure, that this used to work, but with the latest git-version,
>> it does not.
>
>> Is this a new feature of a bug, and who could I get back the old
>> behaviour please?
>
> The code in question was changed (2013-01-14) so as to compare it
> against only addresses, not the ones containing names.  I think it
> is necessary to get rid of unexpected matching.  Now you can use
> a regexp that matches the addresses you're using or a list of your
> mail addresses.

Please make the old behavior the default again.  You can add an option
to match only addresses or surround addresses with <...> to get the
new behavior.

The doc string talks about "From headers" so I'd expect it to match
the Fron header and not only the address. 

I rely on this as well since ages: My `gnus-ignored-from-addresses'
contains \\<Reiner Steib\\>.

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/

Re: gnus-ignored-from-addresses

[Peter Münster]

On Tue, Jun 25 2013, Reiner Steib wrote:

> Please make the old behavior the default again.  You can add an option
> to match only addresses or surround addresses with <...> to get the
> new behavior.

+1

-- 
           Peter

Re: gnus-ignored-from-addresses

[Katsumi Yamaoka]

Peter Münster wrote:
> On Tue, Jun 25 2013, Reiner Steib wrote:
>> Please make the old behavior the default again.  You can add an option
>> to match only addresses or surround addresses with <...> to get the
>> new behavior.

> +1

±0

But it might be good to *always* surround a string with \<...\>
under the `standard-syntax-table'.  I can only imagine that it
will help a FB employee whose name is "Mark Zucker", though.
OTOH, the current way is absolutely safe since an email address
is absolutely unique.  Even if it is hard to beginners to design
a regexp, it can be a start of learning Emacs.

Re: gnus-ignored-from-addresses

[David Engster]

Katsumi Yamaoka writes:
> Peter Münster wrote:
>> On Tue, Jun 25 2013, Reiner Steib wrote:
>>> Please make the old behavior the default again.  You can add an option
>>> to match only addresses or surround addresses with <...> to get the
>>> new behavior.
>
>> +1
>
> ±0
>
> But it might be good to *always* surround a string with \<...\>
> under the `standard-syntax-table'.  I can only imagine that it
> will help a FB employee whose name is "Mark Zucker", though.
> OTOH, the current way is absolutely safe since an email address
> is absolutely unique.  Even if it is hard to beginners to design
> a regexp, it can be a start of learning Emacs.

I'd agree if this were a new feature. But this change needlessly breaks
people's configurations, so I agree with Reiner that it should be
reverted. That the variable mentions 'addresses' might be unfortunate,
but it's a quirk we should live with.

-David

Re: gnus-ignored-from-addresses

[Katsumi Yamaoka]

David Engster <deng@randomsample.de> wrote:
> Katsumi Yamaoka writes:
>> Peter Münster wrote:
>>> On Tue, Jun 25 2013, Reiner Steib wrote:
>>>> Please make the old behavior the default again.  You can add an option
>>>> to match only addresses or surround addresses with <...> to get the
>>>> new behavior.
>>
>>> +1
>>
>> ±0
>>
>> But it might be good to *always* surround a string with \<...\>
>> under the `standard-syntax-table'.  I can only imagine that it
>> will help a FB employee whose name is "Mark Zucker", though.
>> OTOH, the current way is absolutely safe since an email address
>> is absolutely unique.  Even if it is hard to beginners to design
>> a regexp, it can be a start of learning Emacs.

> I'd agree if this were a new feature. But this change needlessly breaks
> people's configurations, so I agree with Reiner that it should be
> reverted. That the variable mentions 'addresses' might be unfortunate,
> but it's a quirk we should live with.

Ah, I agree.  Julien?

Re: gnus-ignored-from-addresses

[Julien Danjou]

[-- Attachment #1: Type: text/plain, Size: 1470 bytes --]

On Tue, Jul 02 2013, Katsumi Yamaoka wrote:

> David Engster <deng@randomsample.de> wrote:
>> Katsumi Yamaoka writes:
>>> Peter Münster wrote:
>>>> On Tue, Jun 25 2013, Reiner Steib wrote:
>>>>> Please make the old behavior the default again.  You can add an option
>>>>> to match only addresses or surround addresses with <...> to get the
>>>>> new behavior.
>>>
>>>> +1
>>>
>>> ±0
>>>
>>> But it might be good to *always* surround a string with \<...\>
>>> under the `standard-syntax-table'.  I can only imagine that it
>>> will help a FB employee whose name is "Mark Zucker", though.
>>> OTOH, the current way is absolutely safe since an email address
>>> is absolutely unique.  Even if it is hard to beginners to design
>>> a regexp, it can be a start of learning Emacs.
>
>> I'd agree if this were a new feature. But this change needlessly breaks
>> people's configurations, so I agree with Reiner that it should be
>> reverted. That the variable mentions 'addresses' might be unfortunate,
>> but it's a quirk we should live with.
>
> Ah, I agree.  Julien?

Everything was said, things are better now, but if it breaks people
configuration, I guess that should be reverted. So whoever wants to do
it go ahead.
When I'll get annoyed again I'll add another option to handle this
correctly again rather than fixing the existing one.

-- 
Julien Danjou
// Free Software hacker / freelance consultant
// http://julien.danjou.info

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 835 bytes --]

Re: gnus-ignored-from-addresses

[Peter Münster]

On Tue, Jul 02 2013, Julien Danjou wrote:

> Everything was said, things are better now, but if it breaks people
> configuration, I guess that should be reverted. So whoever wants to do
> it go ahead.
> When I'll get annoyed again I'll add another option to handle this
> correctly again rather than fixing the existing one.

What was your initial problem that you tried to solve?

-- 
           Peter

Re: gnus-ignored-from-addresses

[Julien Danjou]

[-- Attachment #1: Type: text/plain, Size: 246 bytes --]

On Tue, Jul 02 2013, Peter Münster wrote:

> What was your initial problem that you tried to solve?

Unfortunately, I don't recall exactly.

-- 
Julien Danjou
// Free Software hacker / freelance consultant
// http://julien.danjou.info

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 835 bytes --]

Re: gnus-ignored-from-addresses

[Katsumi Yamaoka]

Julien Danjou wrote:
> On Tue, Jul 02 2013, Peter Münster wrote:
>> What was your initial problem that you tried to solve?
> Unfortunately, I don't recall exactly.

This is a good reason to revert it.  Done.