gnus-article-browse-html-article: boundary, gnus-visible-headers

gnus-article-browse-html-article: boundary, gnus-visible-headers

[Reiner Steib]

Hi,

`K H' (gnus-article-browse-html-article) displays the headers.  When
`gnus-treat-body-boundary' is `head', we get kind of two boundaries,
the X-Boundary line (incl. "X-Boundary:") and `<hr>'.  Therefore, I've
set `gnus-treat-body-boundary' to nil there.

While at it, I couldn't figure out, why you need to let-bind
`gnus-visible-headers' and why you consider `standard-value'.  Could
you explain, please?

    (let ((gnus-visible-headers (or (get 'gnus-visible-headers 'standard-value)
				    gnus-visible-headers))

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

Re: gnus-article-browse-html-article: boundary, gnus-visible-headers

[Katsumi Yamaoka]

>>>>> Reiner Steib <reinersteib+gmane@imap.cc> wrote:

> `K H' (gnus-article-browse-html-article) displays the headers.  When
> `gnus-treat-body-boundary' is `head', we get kind of two boundaries,
> the X-Boundary line (incl. "X-Boundary:") and `<hr>'.  Therefore, I've
> set `gnus-treat-body-boundary' to nil there.

Thanks.

> While at it, I couldn't figure out, why you need to let-bind
> `gnus-visible-headers' and why you consider `standard-value'.  Could
> you explain, please?

>     (let ((gnus-visible-headers (or (get 'gnus-visible-headers 'standard-value)
> 				    gnus-visible-headers))

I considered the main purpose to customize `gnus-visible-headers'
on most people is to add some headers to view in the article buffer.
Those will be for keeping watch on some headers or just for fun.
I also considered that while it is meaningful in the article
buffer, headers that `gnus-visible-headers' specifies by default
are necessary and sufficient for general use.

But I don't oppose deleting of that binding, if you or someone
think that the appearance of the headers displayed in an html
browser should be the same as the one in the Gnus article buffer.

Sorry for the late response.

Re: gnus-article-browse-html-article: boundary, gnus-visible-headers

[Reiner Steib]

On Fri, Mar 21 2008, Katsumi Yamaoka wrote:

> I considered the main purpose to customize `gnus-visible-headers'
> on most people is to add some headers to view in the article buffer.
> Those will be for keeping watch on some headers or just for fun.

I agree.  Thanks for your explanations.

> I also considered that while it is meaningful in the article buffer,
> headers that `gnus-visible-headers' specifies by default are
> necessary and sufficient for general use.

Maybe it would make sense to add a new variable specifying the headers
used for commands like `gnus-article-browse-html-article' and
`gnus-summary-print-article' (do we have other, comparable commands)?
We should also bind `gnus-treat-date-lapsed' to nil for those, I
think.

The new variable (say `gnus-brief-headers'?) could have the following
values:

- (list of) regexp(s):

  Show specified headers.  The variable should be initialized from
  `gnus-saved-headers'.  The purpose seems similar.  Its default value
  is gnus-visible-headers.

  Maybe some headers present (and useful) in `gnus-visible-headers'
  should be removed from the default of `gnus-saved-headers' or only
  from the new variable, e.g. Gnus-Warning, X-Sent,
  (Mail-)?Followup-To, ...

- The symbol `current':

  Use the currently visible headers (like `gnus-summary-print-article'
  does it now)

- nil:

  Use `gnus-saved-headers'.

- t (or ...?):

  Use your implementation (but with `gnus-saved-headers'?).  (Does
  providing this still make sense if we have a dedicated variable?)

Opinions? 

> Sorry for the late response.

Late?  Less than one day?  :-)

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

Re: gnus-article-browse-html-article: boundary, gnus-visible-headers

[Katsumi Yamaoka]

>>>>> Reiner Steib wrote:
> On Fri, Mar 21 2008, Katsumi Yamaoka wrote:

>> I considered the main purpose to customize `gnus-visible-headers'
>> on most people is to add some headers to view in the article buffer.
>> Those will be for keeping watch on some headers or just for fun.

> I agree.  Thanks for your explanations.

>> I also considered that while it is meaningful in the article buffer,
>> headers that `gnus-visible-headers' specifies by default are
>> necessary and sufficient for general use.

> Maybe it would make sense to add a new variable specifying the headers
> used for commands like `gnus-article-browse-html-article' and
> `gnus-summary-print-article' (do we have other, comparable commands)?

Although I hesitate to add a new variable still more, there might
be no choice but to do so.  It will be a user option but I think
it should be assumed that ones who really need to customize it are
few.  Maybe it's `gnus-default-headers', or `gnus-brief-headers'
that you suggested.

> We should also bind `gnus-treat-date-lapsed' to nil for those, I
> think.

Good point.  Not only the date options but also the options that
will modify the article contents might be worth considering too.
E.g.:
`gnus-treat-strip-list-identifiers', `gnus-treat-hide-signature'...

(BTW, I recalled binding of `gnus-inhibit-treatment' to t has no
effect at all.)

> The new variable (say `gnus-brief-headers'?) could have the following
> values:

> - (list of) regexp(s):

>   Show specified headers.  The variable should be initialized from
>   `gnus-saved-headers'.  The purpose seems similar.  Its default value
>   is gnus-visible-headers.

I began to think it is wrong that `gnus-saved-headers' is
initialized from `gnus-visible-headers'.  Because the purpose of
them differs; one is for saving, the other is for viewing.
However, the value of `gnus-visible-headers' that a user might
have customized for viewing is now applied to `gnus-saved-headers'
as the default value.  Though I'm not sure how people use saved
article files (since I don't use them), shouldn't those three
variables be independent each other?

>   Maybe some headers present (and useful) in `gnus-visible-headers'
>   should be removed from the default of `gnus-saved-headers' or only
>   from the new variable, e.g. Gnus-Warning, X-Sent,
>   (Mail-)?Followup-To, ...

I agree.  I'm considering that the default values of
`gnus-brief-headers' and `gnus-saved-headers' have the minimum
headers, like To, Cc, Newsgroups, From, Subject, and Date (I
think it meets the argument "It should be assumed that ones who
really need to customize it are few").  And the default value of
`gnus-visible-headers' is the same as the present one.

> - The symbol `current':

>   Use the currently visible headers (like `gnus-summary-print-article'
>   does it now)

> - nil:

>   Use `gnus-saved-headers'.

It assumes headers present in `gnus-saved-headers' are less than
`gnus-visible-headers', doesn't it?  Actually those are currently
the same in my case.

> - t (or ...?):

>   Use your implementation (but with `gnus-saved-headers'?).  (Does
>   providing this still make sense if we have a dedicated variable?)

This option seems to be indeed needless.

> Opinions?

>> Sorry for the late response.

> Late?  Less than one day?  :-)

It's really late this time.  Sorry. ;-)

Re: gnus-article-browse-html-article: boundary, gnus-visible-headers

[Katsumi Yamaoka]

[-- Attachment #1: Type: text/plain, Size: 592 bytes --]

>>>>> Katsumi Yamaoka wrote:
>>>>>> Reiner Steib wrote:

>> Maybe it would make sense to add a new variable specifying the headers
>> used for commands like `gnus-article-browse-html-article' and
>> `gnus-summary-print-article' (do we have other, comparable commands)?

> Although I hesitate to add a new variable still more, there might
> be no choice but to do so.  It will be a user option but I think
> it should be assumed that ones who really need to customize it are
> few.  Maybe it's `gnus-default-headers', or `gnus-brief-headers'
> that you suggested.

I meant the one like this:


[-- Attachment #2: Type: application/emacs-lisp, Size: 684 bytes --]

[-- Attachment #3: Type: text/plain, Size: 366 bytes --]


This may be able to be the default value of `gnus-visible-headers'
or may be able to use for printing, but cannot be the default
value of `gnus-saved-headers'.

> I began to think it is wrong that `gnus-saved-headers' is
> initialized from `gnus-visible-headers'.

Anyway we must postpone any changes after the No Gnus v0.8 release.

Have a nice weekend.
Sayonara,