Some MIME suggestions...

Some MIME suggestions...

[Hrvoje Niksic]

Some suggestions and bug reports:

1)
When Gnus prompts me for file name when saving a binary attachment,
I'd really like to *not* be switched to "*mm*".  The binary content
looks really ugly in Emacs.

2)
There should be a way to force inlining an attachment.  For example,
many people send patches as application/octet-stream; I'd like to be
able to press a button to force inlining the patch text.  Perhaps `v'
over button should offer "inline" as a viewer.  Or something.

3)
When viewing attachments in a buffer via `c', the buffer's name and
major mode should be determined by the specified file name, if any.
Just calling it "decoded" is lack of imagination.  :-)

4)
There really, really should be a third-button context-sensitive button 
with the commands also available on the keyboard.  I can try to work
on it, if you will accept the patch.

5)
When editing an article via `e', one should be presented the raw
article.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
Don't get even, get odd!

Re: Some MIME suggestions...

[Shenghuo ZHU]

>>>>> "HN" == Hrvoje Niksic <hniksic@srce.hr> writes:

HN> 2)
HN> There should be a way to force inlining an attachment.  For example,
HN> many people send patches as application/octet-stream; I'd like to be
HN> able to press a button to force inlining the patch text.  Perhaps `v'
HN> over button should offer "inline" as a viewer.  Or something.

Is `i' over button the way to force inlining an attachment? 

Suggestion: `i' with argument inlines part with selected charset.

HN> 5)
HN> When editing an article via `e', one should be presented the raw
HN> article.

Suggestion: `e' with argument edits article in decoded text (or raw
article). Then, `C-c C-c' with argument encodes (does not encode) and
saves the article.

-- 
Shenghuo

Re: Some MIME suggestions...

[Steinar Bang]

>>>>> Hrvoje Niksic <hniksic@srce.hr>:

> 2)
> There should be a way to force inlining an attachment.  For example,
> many people send patches as application/octet-stream; I'd like to be
> able to press a button to force inlining the patch text.  Perhaps `v'
> over button should offer "inline" as a viewer.  Or something.

You would probably also like to change the MIME type to something
suitable for viewing, such as image/jpeg or image/gif, or whatever's
appropriate. 

Older versions of Outlook, and Exchange, as well as Lotus Notes, sends 
everything as application/octet-stream, even when there is a suitable
MIME type available ("File name extensions was good enough for my
grandfather!  No reasons why they shouldn't be good enough today!").

Re: Some MIME suggestions...

[Hrvoje Niksic]

Shenghuo ZHU <zsh@cs.rochester.edu> writes:

> >>>>> "HN" == Hrvoje Niksic <hniksic@srce.hr> writes:
> 
> HN> 2) There should be a way to force inlining an attachment.  For
> HN> example, many people send patches as application/octet-stream;
> HN> I'd like to be able to press a button to force inlining the
> HN> patch text.  Perhaps `v' over button should offer "inline" as a
> HN> viewer.  Or something.
> 
> Is `i' over button the way to force inlining an attachment?

Yes; that's exactly what I wanted, thanks.

> Suggestion: `i' with argument inlines part with selected charset.

No.  `i' with an argument prompts you for MIME type.  When you enter a 
textual type, you can easily add `;charset=CHARSET'.

> HN> 5)
> HN> When editing an article via `e', one should be presented the raw
> HN> article.
> 
> Suggestion: `e' with argument edits article in decoded text (or raw
> article). Then, `C-c C-c' with argument encodes (does not encode) and
> saves the article.

This assumes a MIME-aware message, which we don't have right now.  I
think the e-with-argument thing can wait.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
You have an unusual magnetic personality.  Don't walk too close to
metal objects which are not fastened down.

Re: Some MIME suggestions...

[Steinar Bang]

>>>>> Hrvoje Niksic <hniksic@srce.hr>:

> Shenghuo ZHU <zsh@cs.rochester.edu> writes:
>> >>>>> "HN" == Hrvoje Niksic <hniksic@srce.hr> writes:
>> 
HN> 2) There should be a way to force inlining an attachment.  For
HN> example, many people send patches as application/octet-stream;
HN> I'd like to be able to press a button to force inlining the
HN> patch text.  Perhaps `v' over button should offer "inline" as a
HN> viewer.  Or something.
>> 
>> Is `i' over button the way to force inlining an attachment?

> Yes; that's exactly what I wanted, thanks.

>> Suggestion: `i' with argument inlines part with selected charset.

> No.  `i' with an argument prompts you for MIME type.  When you enter a 
> textual type, you can easily add `;charset=CHARSET'.

Should the message be permanently changed to the new MIME type?

My opinion is that it should.  *If* the back end allows editing, that
is. 

I haven't gotten around to installing pgnus yet, so I can't test it
myself.  Maybe today's the day?  Hm... where did I put the article
with the receipe for using tm-edit.el in message mode, without using
TM?  Hm... I don't seem to have kept it around... oh well...

Re: Some MIME suggestions...

[Hrvoje Niksic]

Steinar Bang <sb@metis.no> writes:

> > No.  `i' with an argument prompts you for MIME type.  When you
> > enter a textual type, you can easily add `;charset=CHARSET'.
> 
> Should the message be permanently changed to the new MIME type?

No.  At least not as far as I'm concerned.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
You can only be young once, but you can be immature forever.

Re: Some MIME suggestions...

[Lars Magne Ingebrigtsen]

Hrvoje Niksic <hniksic@srce.hr> writes:

> When Gnus prompts me for file name when saving a binary attachment,
> I'd really like to *not* be switched to "*mm*".  The binary content
> looks really ugly in Emacs.

Yes.  But textual content looks beautiful in Emacs.  And it is kinda
customary to be presented with the buffer you are saving in Emacs.
Although this might be regarded as "copying" instead.  I don't know.
I'll change it to whatever y'all want.  Should the buffer be displayed
or not?  Poll time.

> When viewing attachments in a buffer via `c', the buffer's name and
> major mode should be determined by the specified file name, if any.
> Just calling it "decoded" is lack of imagination.  :-)

Yup.  Fix in Pterodactyl Gnus v0.35.

> There really, really should be a third-button context-sensitive button 
> with the commands also available on the keyboard.  I can try to work
> on it, if you will accept the patch.

Sounds nice; I'll apply the patch.

> When editing an article via `e', one should be presented the raw
> article.

Er...  Yes, one really should.  But that kinda makes all the
highlighting and everything away, which makes things a bit awkward.
Hm.  Well, after we have divided all the display functions up into
"highlight" and "wash" functions, we could re-run all the wanted
"highlights", and not do the "wash" functions.  I think that should do
the trick.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: Some MIME suggestions...

[Hrvoje Niksic]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Hrvoje Niksic <hniksic@srce.hr> writes:
> 
> > When Gnus prompts me for file name when saving a binary attachment,
> > I'd really like to *not* be switched to "*mm*".  The binary content
> > looks really ugly in Emacs.
> 
> Yes.  But textual content looks beautiful in Emacs.

True, but textual content should be marked as text/something by the
sender.

And another point is that the buffer should be placed to a nice place, 
not just random.  Currently I get it instead of the Summary buffer, so 
it just screws up my display.  And it won't remove itself after I save 
it.  :-(

> And it is kinda customary to be presented with the buffer you are
> saving in Emacs.  Although this might be regarded as "copying"
> instead.

Please do regard it as copying.

> > When editing an article via `e', one should be presented the raw
> > article.
> 
> Er...  Yes, one really should.  But that kinda makes all the
> highlighting and everything away, which makes things a bit awkward.

Why would that make it awkward?  Aren't highlights *supposed* to go
away when you're editing the article?

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
Q: What's an IBM man-year?
A: 730 people trying to get a project done before noon.

Re: Some MIME suggestions...

[Lars Magne Ingebrigtsen]

Hrvoje Niksic <hniksic@srce.hr> writes:

> Please do regard it as copying.

Yup.  Fix in Pterodactyl Gnus v0.35.

> > Er...  Yes, one really should.  But that kinda makes all the
> > highlighting and everything away, which makes things a bit awkward.
> 
> Why would that make it awkward?  Aren't highlights *supposed* to go
> away when you're editing the article?

I don't really see why.  Some people edit articles quite a lot (to
correct little things), and having the article look as similar as
possible when editing and viewing seems like a good idea.  It's also
an efficiency thing -- re-highlighting everything after every little
edit would be annoying, especially in large articles.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: Some MIME suggestions...

[Hrvoje Niksic]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> > > Er...  Yes, one really should.  But that kinda makes all the
> > > highlighting and everything away, which makes things a bit
> > > awkward.
> > 
> > Why would that make it awkward?  Aren't highlights *supposed* to go
> > away when you're editing the article?
> 
> I don't really see why.

You have a point.

However, the point stands (IMHO) only as long as we're talking about
highlighting, as opposed to changing the text.  And many currenn
display hooks do change the text, e.g. `W w'.  Somehow I think that
`e' should always present the original article, the same stuff you get 
with `C-u g'.

> It's also an efficiency thing -- re-highlighting everything after
> every little edit would be annoying, especially in large articles.

Are people really editing stuff that often to make this a problem?

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
"Psychos _do not_ explode when sunlight hits them."

Re: Some MIME suggestions...

[Lars Magne Ingebrigtsen]

Hrvoje Niksic <hniksic@srce.hr> writes:

> However, the point stands (IMHO) only as long as we're talking about
> highlighting, as opposed to changing the text.  And many currenn
> display hooks do change the text, e.g. `W w'.

But I do like that.   To permanently edit an article to be what `W w'
had done with it, you now only have to type `C-u g W w e C-c C-c'.  So
I think it's useful to be able to have things like that happen, but
not things like MIME buttons.

Uhm.

Hm.  Hm.  Hm.

I don't see an obvious user interface to determine which washes should
be "edit-permanent", and which should be undone before editing.

> Are people really editing stuff that often to make this a problem?

I've had reports from people who use Gnus to edit large groups of
documents that are shared between people.  This kinda-sorta worked
in a satisfactor manner until now (if some washing functions were
avoided), but the MIME button thing made the problem more obvious.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: Some MIME suggestions...

[Paul Franklin]

>>>>> Lars Magne Ingebrigtsen writes:

 > Hrvoje Niksic <hniksic@srce.hr> writes:

 >> However, the point stands (IMHO) only as long as we're talking about
 >> highlighting, as opposed to changing the text.  And many currenn
 >> display hooks do change the text, e.g. `W w'.

 > But I do like that.  To permanently edit an article to be what `W
 > w' had done with it, you now only have to type `C-u g W w e C-c
 > C-c'.

Yup; I do this too.  Sometimes I wish I could do this to cached
articles!

 > So
 > I think it's useful to be able to have things like that happen, but
 > not things like MIME buttons.

Maybe the complete wash/highlight map should be available while
editing articles.  Then you could do `e C-c C-w w C-c C-c' to
achieve the above.

Hmm.  There's already
	C-c C-w		<< gnus-article-edit-wash-map >>
but it only has one function in it.

--Paul

Re: Some MIME suggestions...

[Lars Magne Ingebrigtsen]

Paul Franklin <paul@cs.washington.edu> writes:

> Maybe the complete wash/highlight map should be available while
> editing articles.  Then you could do `e C-c C-w w C-c C-c' to
> achieve the above.

That's definitely an idea.  Any comments from dedicated message
editors out there?

The idea would then be that `e' would fetch a new, fresh article,
re-highlight according to what is normally done, but no washes, and
then put point in the article buffer.  One would then have access to
all the article washing commands under some keymap.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: Some MIME suggestions...

[Kai Grossjohann]

>>>>> Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

  > Hrvoje Niksic <hniksic@srce.hr> writes:
  > 
  > > When editing an article via `e', one should be presented the raw
  > > article.
  > 
  > Er...  Yes, one really should.  But that kinda makes all the
  > highlighting and everything away, which makes things a bit awkward.

There's two kinds of highlights: the article display highlight thingy,
and the message composition highlight thingy.  Editing articles is
somewhere in between, I think, so it would be possible to take the
stance that `e' should invoke message-mode, including its
highlighting.

Of course, in the long run, the highlighting when composing messages
should be the same as when displaying them.  Don' you agree?

kai
-- 
In Turkey, we played well and lost.  In Moldavia, we played badly and won.
This raises the question: which is better?  (Oliver Kahn, German soccer team)

Re: Some MIME suggestions...

[Kai Grossjohann]

>>>>> Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

  > Hrvoje Niksic <hniksic@srce.hr> writes:
  > 
  > > However, the point stands (IMHO) only as long as we're talking about
  > > highlighting, as opposed to changing the text.  And many currenn
  > > display hooks do change the text, e.g. `W w'.
  > 
  > But I do like that.   To permanently edit an article to be what `W w'
  > had done with it, you now only have to type `C-u g W w e C-c C-c'.  So
  > I think it's useful to be able to have things like that happen, but
  > not things like MIME buttons.
  > 
  > [...]
  > 
  > I don't see an obvious user interface to determine which washes
  > should be "edit-permanent", and which should be undone before
  > editing.

I think solving this in a satisfactory manner means that one has also
solved the MIME message composition issue.  Thus, I propose that we
choose a quick-and-dirty solution first (make it so that `e' implies a
`C-u g' first).  After that, the MIME message composition thing should
be worked on.  And then, editing articles would be very similar to
composing MIME messages (regarding what is being displayed).  And we
can use the C-c C-w map that someone else talked about.

kai
-- 
In Turkey, we played well and lost.  In Moldavia, we played badly and won.
This raises the question: which is better?  (Oliver Kahn, German soccer team)

Re: Some MIME suggestions...

[Lars Magne Ingebrigtsen]

Kai Grossjohann <Kai.Grossjohann@CS.Uni-Dortmund.DE> writes:

> There's two kinds of highlights: the article display highlight thingy,
> and the message composition highlight thingy.  Editing articles is
> somewhere in between, I think, so it would be possible to take the
> stance that `e' should invoke message-mode, including its
> highlighting.

Especially with a MIME-aware Message, this would definitely make
sense.

Ok; I'm not doing to do anything about `e' until Message has been
MIMEtilated.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: Some MIME suggestions...

[Hrvoje Niksic]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> The idea would then be that `e' would fetch a new, fresh article,
> re-highlight according to what is normally done, but no washes, and
> then put point in the article buffer.  One would then have access to
> all the article washing commands under some keymap.

Sounds good to me.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
Those who like sausages, laws, and standards are well advised not
to learn how they are made.