X-Face and depth

X-Face and depth

[Lars Magne Ingebrigtsen]

I just had scary thought -- what if we don't limit X-Faces to a single
bit plane?

I tried running a few images through various bit levels, and it's
pretty amazing how much more recognizable images are when there are
three or even two bits per pixel.  What looks totally unrecognizable
with 1 bit, looks OK with 2 bits and beautiful with 3 bits.

That's when I had that scary thought -- we don't have to limit X-Faces
to a single bit.  We can just output several X-Face headers, where
each X-Face header describes one bit plane.  Then the viewing agent
adds them all up, and displays them.  It's backwards-compatible.  :-)

However, I'm not sure this is a good idea.  Do we want to encourage
people to add even more data to the headers?  On the other hand, this
is 2002, and bandwidth and disk space is so much cheaper than it was
when X-Faces were dreamt up.  On the third hand, why not define a new
format altogether?  X-Gif-Face, or something.  Or even referring to an
outside source via HTTP.

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

Re: X-Face and depth

[Didier Verna]

Lars Magne Ingebrigtsen wrote:

> That's when I had that scary thought -- we don't have to limit X-Faces
> to a single bit.  We can just output several X-Face headers, where
> each X-Face header describes one bit plane.  Then the viewing agent
> adds them all up, and displays them.  It's backwards-compatible.  :-)

        It's not. I currently have two different X-Face headers that display 2
images next to each other. What you could do is understand X-Face-[1,2,3...]
headers that would add bitplanes to the previous X-Face one. *That* is
backward compatible, and I could still have two faces.

-- 
Didier Verna, didier@lrde.epita.fr, http://www.lrde.epita.fr/~didier

EPITA / LRDE, 14-16 rue Voltaire   Tel.+33 (1) 53 14 59 47
94276 Le Kremlin-Bicêtre, France   Fax.+33 (1) 53 14 59 22   didier@xemacs.org

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Didier Verna <didier@lrde.epita.fr> writes:

>         It's not. I currently have two different X-Face headers that
> display 2 images next to each other. What you could do is understand
> X-Face-[1,2,3...]  headers that would add bitplanes to the previous
> X-Face one. *That* is backward compatible, and I could still have
> two faces.

Good idea.

But you don't really need two X-Faces -- your face already shows up
due to picons.  :-)

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

Re: X-Face and depth

[Didier Verna]

Lars Magne Ingebrigtsen wrote:

> But you don't really need two X-Faces

        Yes I do because ...

> your face already shows up due to picons. :-)

        ... I can't have 2 picons :-)

-- 
Didier Verna, didier@lrde.epita.fr, http://www.lrde.epita.fr/~didier

EPITA / LRDE, 14-16 rue Voltaire   Tel.+33 (1) 53 14 59 47
94276 Le Kremlin-Bicêtre, France   Fax.+33 (1) 53 14 59 22   didier@xemacs.org

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Didier Verna <didier@lrde.epita.fr> writes:

>> But you don't really need two X-Faces
>
>         Yes I do because ...
>
>> your face already shows up due to picons. :-)
>
>         ... I can't have 2 picons :-)

But in your From line I see:

From: [Face][Playing Guitar] Didier Verna <[face]@lrde.epita.[tri-color]>

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

Re: X-Face and depth

[Didier Verna]

Lars Magne Ingebrigtsen wrote:

> But in your From line I see:
>
> From: [Face][Playing Guitar] Didier Verna <[face]@lrde.epita.[tri-color]>

        That's because my picon is the same as one of my X-Face ... which I
intend to change soon (2 pictures is not enough; 3 is ok :-).

-- 
Didier Verna, didier@lrde.epita.fr, http://www.lrde.epita.fr/~didier

EPITA / LRDE, 14-16 rue Voltaire   Tel.+33 (1) 53 14 59 47
94276 Le Kremlin-Bicêtre, France   Fax.+33 (1) 53 14 59 22   didier@xemacs.org

X-Face and depth

[luis fernandes]

    larsi> However, I'm not sure this is a good idea.  

I was thinking the same thing so it must be a good  idea. 

    larsi> Do we want to encourage people to add even more data to
    larsi> the headers?

As opposed to adding it to the body-- i.e. spam? 

Keep the X-Face as-is or there'll have to be RFDs and RFCs before
modifying it. 

Another idea: what if there were various image formats of Face
headers:

X-Gif-Face:
X-Jpg-Face:
X-Png-Face: 

Gnus could then decode appropriately depending on what converters it
had built-in.

The adventureous could always add X-XXX-Face...that would have to be
password protected, of course

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

luis fernandes <elf@ee.ryerson.ca> writes:

> I was thinking the same thing so it must be a good  idea. 

There you go.  :-)

I've now implemented a set of functions in gnus-fun.el that basically
splits up and puts back together again bit planes.  I couldn't find
any ppm command for doing anything like that, so I wrote it in Lisp.

>     larsi> Do we want to encourage people to add even more data to
>     larsi> the headers?
>
> As opposed to adding it to the body-- i.e. spam? 

But sensible people don't do that.  Sensible people might very well
put something like this in the headers.

> Another idea: what if there were various image formats of Face
> headers:
>
> X-Gif-Face:
> X-Jpg-Face:
> X-Png-Face: 

The nice thing about X-Face headers is that they are so constrained.
If people start posting 1024x768 GIF images in their headers, I think
people will be annoyed.

I've had a quick peek at the size of a 48x48x2 "extended" X-Face and
the equivalent GIF file, and the "extended" X-Face was shorter.  Which
is nice.

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

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Here we go...  There should be three X-Face parts up there.

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

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

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

> Here we go...  There should be three X-Face parts up there.

And with the latest from CVS, you can view it as well.

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

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

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

> And with the latest from CVS, you can view it as well.

Indeed.

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

Re: X-Face and depth

[Matt Pharr]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
> Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
>> And with the latest from CVS, you can view it as well.
>
> Indeed.

Err, I'm getting the following (with xemacs running in a tty, which may be
what's tickling this...) The only x-face related thing in my .gnus.el file
is (setq gnus-treat-display-xface 'head) .

-matt

Signaling: (wrong-type-argument number-char-or-marker-p nil)
  gnus-convert-gray-x-face-to-xpm(("%xgMM\\zp%q}gak7U|f]Ko;YL6G=]+6k5{OFZ}}?^Bm&DPLtD;JxaQP&s:4])e*m55t^4BFz\n :l%y~#NeF_S'8qSSp6<>33(go,MuE*h=>vs$Yq5b28B9<(QDiewrc>|:O-Uk4tYLG&%1wn32IH.)\"O\n WB3/~m:`O$$+kzL=,}i[a=+(ilS-3nch*ji4WvBN{Cs)N-lU&+@h)v~D((wuG&K3_<Y-^WCCTj^51V\n ?z!lj9B`h5h5sk(_Ja#Y(0^8o+sZMqE/{;|EH*ljQR!Wpb=[85)xI'A2b]3d_XhN^hzCW~{a^w:2,_\n f\n" "$qH|!b,FpnFCj0`?$U(D5W~cb$)xY]Iw$U>CHfU:/HX/MyjMT.qm%!k[tSWw0;Qp|5HW!9W\n z3]Xc5-mu0Irp/(ozBB1%rxR:fGjfrL<dspq6]\\)kV60<}Q`r6%fgflUi-:s1=@bjI52~x@`+PE}SF\n .:\"5/bw'2lWc:\"Fwgizg-+k@\\LXYlzo|aJ'ekT(DoFj2][mt[G-zmpq4ScP5m^6\\dW&AB2Ng~OM4>C\n =:hR|Y]WJ(A\n" "oipAKEI5J'!,n,7/<\\/`*G/&oY.G]x<Ql6rrU~doaN(O>af6;G7gG;f$_*,:Gxff54W??52\n 3s;d0_-C|2b^=\\=:zpvHGogw6,p>&}srUBGk>SJt0iGWN7#3\\\n"))
  article-display-x-face()
  apply(article-display-x-face nil)
  (if interactive (call-interactively (quote article-display-x-face)) (apply (quote article-display-x-face) args))
)
  (save-excursion (set-buffer gnus-article-buffer) (if interactive (call-interactively ...) (apply ... args)))
)
  gnus-article-display-x-face()
  gnus-treat-article(head)
  gnus-display-mime()
  gnus-article-prepare-display()
  gnus-article-prepare(30020 nil)
  gnus-summary-display-article(30020)
  gnus-summary-next-page(nil)
  call-interactively(gnus-summary-next-page)

-- 
Matt Pharr    |    Exluna, Inc.    |    <URL:http://graphics.stanford.edu/~mmp>
===============================================================================
In a cruel and evil world, being cynical can allow you to get some
entertainment out of it. --Daniel Waters

Re: X-Face and depth

[Steve Youngs]

|--==> "MP" == Matt Pharr <mmp@exluna.com> writes:

  MP> Err, I'm getting the following (with xemacs running in a tty, which may be
  MP> what's tickling this...) 

No, I'm seeing it in X as well.

-- 
|---<Steve Youngs>---------------<GnuPG KeyID: 10D5C9C5>---|
|            XEmacs - It's not just an editor.             |
|                    It's a way of life.                   |
|------------------------------------<youngs@xemacs.org>---|

Re: X-Face and depth

[Jason R. Mastaler]

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

>> Here we go...  There should be three X-Face parts up there.
>
> And with the latest from CVS, you can view it as well.

Is there something special I have to set to be able to view
the multi-depth X-Faces?  Nothing shows up at all.  I'm using Oort
from CVS circa 10 minutes ago.

-- 
(TMDA (http://tmda.sourceforge.net/))
(user-level UCE intrusion prevention)

Re: X-Face and depth

[Vincent Bernat]

OoO Lors de la soirée naissante du mercredi 02 janvier 2002, vers
18:02, Lars Magne Ingebrigtsen <larsi@gnus.org> disait:

> Here we go...  There should be three X-Face parts up there.

I don't understand what is behind this feature. There is already some
people which have several X-Face by just using two or more fields
X-Face. May someone confirm this ?
-- 
die_if_kernel("Kernel gets FloatingPenguinUnit disabled trap", regs);
	2.2.16 /usr/src/linux/arch/sparc/kernel/traps.c

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Vincent Bernat <bernat@scientist.com> writes:

> I don't understand what is behind this feature. There is already some
> people which have several X-Face by just using two or more fields
> X-Face. May someone confirm this ?

Sure.  But they aren't gray-scale.

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

Re: X-Face and depth

[Raymond Scholz]

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

> Here we go...  There should be three X-Face parts up there.

Erm, a "junk" buffer keeps popping up when I select your article and
it massively (3 secs) delays the article display. 

,----[ junk ]
| \f
| ppmtoxpm: (Computing colormap...
| ppmtoxpm: ...Done.  8 colors found.)
`----

How do I suppress this?

Cheers, Ray'what about X-Sound?'mond
-- 
Emacs is a fine operating system, but I still prefer UNIX.  
                                        (Tom Christiansen)

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Raymond Scholz <ray-2002@zonix.de> writes:

> Erm, a "junk" buffer keeps popping up when I select your article and
> it massively (3 secs) delays the article display. 

I've now fixed the junk thing, but the implementation is absolutely
pessimal.  I'm not sure whether the whole thing is a good idea or
not.  I we decide that it is, I'll optimize the implementation.

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

Re: X-Face and depth

[Fabien Penso]

[-- Attachment #1: Type: text/plain, Size: 263 bytes --]


Lars on Wed, 02 Jan 2002 18:02:14 +0100 wrote:

 > Here we go...  There should be three X-Face parts up there.

Nice !

-- 
Fabien Penso <penso@linuxfr.org> | LinuxFr a toujours besoin de :
http://perso.LinuxFR.org/penso/  | http://linuxfr.org/dons/


[-- Attachment #2: Type: application/pgp-signature, Size: 239 bytes --]

Re: X-Face and depth

[Fabien Penso]

[-- Attachment #1: Type: text/plain, Size: 426 bytes --]


Fabien on Thu, 03 Jan 2002 11:02:38 +0100 wrote:

 > Lars on Wed, 02 Jan 2002 18:02:14 +0100 wrote:

 >> Here we go...  There should be three X-Face parts up there.

 > Nice !

Btw the CVS doesn't show any x-face at all for regular x-faces. I'm sure
you already know about it.

-- 
Fabien Penso <penso@linuxfr.org> | LinuxFr a toujours besoin de :
http://perso.LinuxFR.org/penso/  | http://linuxfr.org/dons/


[-- Attachment #2: Type: application/pgp-signature, Size: 239 bytes --]

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Fabien Penso <penso@linuxfr.org> writes:

> Btw the CVS doesn't show any x-face at all for regular x-faces. I'm sure
> you already know about it.

Works for me...

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

Re: X-Face and depth

[Fabien Penso]

[-- Attachment #1: Type: text/plain, Size: 471 bytes --]


Lars on Thu, 03 Jan 2002 11:37:37 +0100 wrote:

 > Fabien Penso <penso@linuxfr.org> writes:
 >> Btw the CVS doesn't show any x-face at all for regular x-faces. I'm sure
 >> you already know about it.

 > Works for me...

Ok if I add "X-Face:" into gnus-ignored-headers the X-Face isn't
displayed anymore. It was before...

-- 
Fabien Penso <penso@linuxfr.org> | LinuxFr a toujours besoin de :
http://perso.LinuxFR.org/penso/  | http://linuxfr.org/dons/


[-- Attachment #2: Type: application/pgp-signature, Size: 239 bytes --]

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Fabien Penso <penso@linuxfr.org> writes:

> Ok if I add "X-Face:" into gnus-ignored-headers the X-Face isn't
> displayed anymore. It was before...

It's in there by default, so I'm unable to see this bug...

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

Re: X-Face and depth

[Zlatko Calusic]

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

> Here we go...  There should be three X-Face parts up there.

I see it, I see it!!! :)

Now, how do you make multilevel faces, that is my question?
-- 
Zlatko

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Zlatko Calusic <zlatko.calusic@iskon.hr> writes:

> Now, how do you make multilevel faces, that is my question?

`gnus-convert-image-to-gray-x-face'.

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

Re: X-Face and depth

[John H. Palmieri]

[-- Attachment #1: Type: text/plain, Size: 1255 bytes --]

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

> Zlatko Calusic <zlatko.calusic@iskon.hr> writes:
>
>> Now, how do you make multilevel faces, that is my question?
>
> `gnus-convert-image-to-gray-x-face'.

This doesn't work for me: I get output like

  X-Face:compface: (warning) <stdin>: excess data ignored
  compface: (warning) <stdin>: excess data ignored
  X-Face-1:compface: (warning) <stdin>: excess data ignored
  compface: (warning) <stdin>: excess data ignored
  X-Face-2:compface: (warning) <stdin>: excess data ignored
  compface: (warning) <stdin>: excess data ignored

If I change this:

	    (shell-command-on-region
	     (point-min) (point-max)
	     "pbmtoxbm | compface"
	     (current-buffer) t)

to this:

	    (shell-command-on-region
	     (point-min) (point-max)
	     "pbmtoxbm | xbmtoikon | compface"
	     (current-buffer) t)

it works.  I found xbmtoikon here:

  <http://www.cs.indiana.edu/l/www/pub/faces/picons/src/bits-2.7.tar.Z>

I'm attaching xbmtoikon.  Perhaps someone could just translate it to lisp...

-- 
J. H. Palmieri                      
Dept of Mathematics, Box 354350    mailto:palmieri@math.washington.edu
University of Washington           http://www.math.washington.edu/~palmieri/
Seattle, WA 98195-4350

[-- Attachment #2: xbmtoikon (a shell script) --]
[-- Type: text/plain, Size: 460 bytes --]

#!/bin/sh -
PATH=$PATH:/l/picons/bits:/l/netpbm/bin

# xbmtoikon - convert an XBM bitmap to a Blit ikon bitmap
# Steve Kinzler, kinzler@cs.indiana.edu, Mar 91/Jan 95

# NOTE: This program has only been tested with 48x48 and 64x64 XBM inputs.

case "$1" in
-h)	echo "usage: $0 [ file ]" 1>&2; exit 1;;
esac

xbmtopbm ${1+"$@"} |
pbmtoicon |
sed '/^[ 	]*[*\/]/d; s/[ 	]//g; s/,$//' |
tr , '\012' |
sed 's/^0x//; s/^/0x/' |
pr -l1 -t -w22 -3 -s, |
sed 's/,*$/,/'

Re: X-Face and depth

[Steve Youngs]

|--==> "JHP" == John H Palmieri <palmieri@math.washington.edu> writes:

  JHP> Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
  >>Zlatko Calusic <zlatko.calusic@iskon.hr> writes:
  >>
  >>>Now, how do you make multilevel faces, that is my question?
  >>
  >>`gnus-convert-image-to-gray-x-face'.

  JHP> This doesn't work for me: I get output like

  JHP> X-Face:compface: (warning) <stdin>: excess data ignored
  JHP> compface: (warning) <stdin>: excess data ignored
  JHP> X-Face-1:compface: (warning) <stdin>: excess data ignored
  JHP> compface: (warning) <stdin>: excess data ignored
  JHP> X-Face-2:compface: (warning) <stdin>: excess data ignored
  JHP> compface: (warning) <stdin>: excess data ignored

I'm seeing this too.

  JHP> If I change this:

[...]

  JHP> to this:

  JHP> (shell-command-on-region
  JHP> (point-min) (point-max)
  JHP> "pbmtoxbm | xbmtoikon | compface"
  JHP> (current-buffer) t)

Yep, that fixes it. :-)

-- 
|---<Steve Youngs>---------------<GnuPG KeyID: 10D5C9C5>---|
|            XEmacs - It's not just an editor.             |
|                    It's a way of life.                   |
|------------------------------------<youngs@xemacs.org>---|

Re: X-Face and depth

[John H. Palmieri]

[-- Attachment #1: Type: text/plain, Size: 385 bytes --]

I'm attaching an almost completely untested version of
gnus-convert-image-to-gray-x-face, which shouldn't require the
presence of xbmtoikon.  Does it work for anyone else?

-- 
J. H. Palmieri                      
Dept of Mathematics, Box 354350    mailto:palmieri@math.washington.edu
University of Washington           http://www.math.washington.edu/~palmieri/
Seattle, WA 98195-4350

[-- Attachment #2: gnus-convert-image-to-gray-x-face --]
[-- Type: application/emacs-lisp, Size: 1599 bytes --]

Re: X-Face and depth

[John H. Palmieri]

Two addenda: you can delete "tr -d '[:blank:]' |" from the long line
near the end of the function (that line is the only change from
Lars' version).  Also, while this seems to work for me, it doesn't
give the same result twice in a row; anyone know why?

For example, when I execute

  (gnus-convert-image-to-gray-x-face "palmieri.gif" 3)

I get 

  X-Face: $GbgRf%-(p'HfD?g-[@{P5sFeYQTF9+LC!k>'tW2PTynVg$~5^X*[k-4w!NXV<JFpqayE_-
   50HDA"]Dz6KPmri
  [etc.]

When I run it again, I get

  X-Face: ?t<i|=y.$,lnu(K:uh3#nT^d@~9J~TEi%CB<=1HWr~K1~hNuqca.j&sk]2hBFt'l$"tD_Xq
   ia#AXW(_xgvu0
  [etc.]

The next time, I get

  X-Face: .DK?X/SSWeRfw6}fwd>mY\:,'xe]K_;U3Fv[*}v3or<y5|b"5G&mSSJ5AcW16xCN9sKI*@H
   &WrO>E(*J1[)?)
  [etc.]


palmieri@math.washington.edu (John H. Palmieri) writes:

> I'm attaching an almost completely untested version of
> gnus-convert-image-to-gray-x-face, which shouldn't require the
> presence of xbmtoikon.  Does it work for anyone else?

-- 
J. H. Palmieri                      
Dept of Mathematics, Box 354350    mailto:palmieri@math.washington.edu
University of Washington           http://www.math.washington.edu/~palmieri/
Seattle, WA 98195-4350

Re: X-Face and depth

[Andreas Büsching]

palmieri@math.washington.edu (John H. Palmieri) wrote:

> I'm attaching an almost completely untested version of
> gnus-convert-image-to-gray-x-face, which shouldn't require the
> presence of xbmtoikon.  Does it work for anyone else?

It works very well for me. :-)

crunchy

-- 
"I don't question YOUR existence." - God

Re: X-Face and depth

[Raymond Scholz]

Andreas Büsching <crunchy@tzi.de> writes:

> It works very well for me. :-)

It doesn't for me :-(  Same with the original code from Gnus CVS.

Debugger entered--Lisp error: (wrong-type-argument number-or-marker-p nil)
  logand(nil 1)
  (zerop (logand bits mask))
  (if (zerop (logand bits mask)) "0 " "1 ")
  (insert (if (zerop ...) "0 " "1 "))
  (while --dolist-temp--56627 (setq bits (car --dolist-temp--56627)) (insert (if ... "0 " "1 ")) (setq --dolist-temp--56627 (cdr --dolist-temp--56627)))
  (let ((--dolist-temp--56627 bits-list) bits) (while --dolist-temp--56627 (setq bits ...) (insert ...) (setq --dolist-temp--56627 ...)) nil)
  (catch (quote --cl-block-nil--) (let (... bits) (while --dolist-temp--56627 ... ... ...) nil))
  (cl-block-wrapper (catch (quote --cl-block-nil--) (let ... ... nil)))
  (block nil (let (... bits) (while --dolist-temp--56627 ... ... ...) nil))
  (dolist (bits bits-list) (insert (if ... "0 " "1 ")))
  (save-current-buffer (set-buffer temp-buffer) (insert "P1\n48 48\n") (dolist (bits bits-list) (insert ...)) (shell-command-on-region (point-min) (point-max) "pbmtoicon | sed '/^[ 	]*[*\\\\/]/d; s/[ 	]//g; s/,$//' | tr , '\\012' | tr -d '[:blank:]' | sed 's/^0x//; s/^/0x/' | pr -l1 -t -w22 -3 -s, | sed 's/,*$/,/' | compface" (current-buffer) t) (push (buffer-string) x-faces))
  (with-current-buffer temp-buffer (insert "P1\n48 48\n") (dolist (bits bits-list) (insert ...)) (shell-command-on-region (point-min) (point-max) "pbmtoicon | sed '/^[ 	]*[*\\\\/]/d; s/[ 	]//g; s/,$//' | tr , '\\012' | tr -d '[:blank:]' | sed 's/^0x//; s/^/0x/' | pr -l1 -t -w22 -3 -s, | sed 's/,*$/,/' | compface" (current-buffer) t) (push (buffer-string) x-faces))
  (unwind-protect (with-current-buffer temp-buffer (insert "P1\n48 48\n") (dolist ... ...) (shell-command-on-region ... ... "pbmtoicon | sed '/^[ 	]*[*\\\\/]/d; s/[ 	]//g; s/,$//' | tr , '\\012' | tr -d '[:blank:]' | sed 's/^0x//; s/^/0x/' | pr -l1 -t -w22 -3 -s, | sed 's/,*$/,/' | compface" ... t) (push ... x-faces)) (and (buffer-name temp-buffer) (kill-buffer temp-buffer)))
  (let ((temp-buffer ...)) (unwind-protect (with-current-buffer temp-buffer ... ... ... ...) (and ... ...)))
  (with-temp-buffer (insert "P1\n48 48\n") (dolist (bits bits-list) (insert ...)) (shell-command-on-region (point-min) (point-max) "pbmtoicon | sed '/^[ 	]*[*\\\\/]/d; s/[ 	]//g; s/,$//' | tr , '\\012' | tr -d '[:blank:]' | sed 's/^0x//; s/^/0x/' | pr -l1 -t -w22 -3 -s, | sed 's/,*$/,/' | compface" (current-buffer) t) (push (buffer-string) x-faces))
  (while (< bit-number --dotimes-temp--56626) (setq mask (expt 2 bit-number)) (with-temp-buffer (insert "P1\n48 48\n") (dolist ... ...) (shell-command-on-region ... ... "pbmtoicon | sed '/^[ 	]*[*\\\\/]/d; s/[ 	]//g; s/,$//' | tr , '\\012' | tr -d '[:blank:]' | sed 's/^0x//; s/^/0x/' | pr -l1 -t -w22 -3 -s, | sed 's/,*$/,/' | compface" ... t) (push ... x-faces)) (incf bit-number))
  (let ((--dotimes-temp--56626 depth) (bit-number 0)) (while (< bit-number --dotimes-temp--56626) (setq mask ...) (with-temp-buffer ... ... ... ...) (incf bit-number)) nil)
  (catch (quote --cl-block-nil--) (let (... ...) (while ... ... ... ...) nil))
  (cl-block-wrapper (catch (quote --cl-block-nil--) (let ... ... nil)))
  (block nil (let (... ...) (while ... ... ... ...) nil))
  (dotimes (bit-number depth) (setq mask (expt 2 bit-number)) (with-temp-buffer (insert "P1\n48 48\n") (dolist ... ...) (shell-command-on-region ... ... "pbmtoicon | sed '/^[ 	]*[*\\\\/]/d; s/[ 	]//g; s/,$//' | tr , '\\012' | tr -d '[:blank:]' | sed 's/^0x//; s/^/0x/' | pr -l1 -t -w22 -3 -s, | sed 's/,*$/,/' | compface" ... t) (push ... x-faces)))
  (save-current-buffer (set-buffer temp-buffer) (insert (shell-command-to-string ...)) (goto-char (point-min)) (forward-line 3) (while (setq pixel ...) (push ... bits-list)) (setq bits-list (nreverse bits-list)) (dotimes (bit-number depth) (setq mask ...) (with-temp-buffer ... ... ... ...)))
  (with-current-buffer temp-buffer (insert (shell-command-to-string ...)) (goto-char (point-min)) (forward-line 3) (while (setq pixel ...) (push ... bits-list)) (setq bits-list (nreverse bits-list)) (dotimes (bit-number depth) (setq mask ...) (with-temp-buffer ... ... ... ...)))
  (unwind-protect (with-current-buffer temp-buffer (insert ...) (goto-char ...) (forward-line 3) (while ... ...) (setq bits-list ...) (dotimes ... ... ...)) (and (buffer-name temp-buffer) (kill-buffer temp-buffer)))
  (let ((temp-buffer ...)) (unwind-protect (with-current-buffer temp-buffer ... ... ... ... ... ...) (and ... ...)))
  (with-temp-buffer (insert (shell-command-to-string ...)) (goto-char (point-min)) (forward-line 3) (while (setq pixel ...) (push ... bits-list)) (setq bits-list (nreverse bits-list)) (dotimes (bit-number depth) (setq mask ...) (with-temp-buffer ... ... ... ...)))
  (progn (with-temp-buffer (insert ...) (goto-char ...) (forward-line 3) (while ... ...) (setq bits-list ...) (dotimes ... ... ...)) (dotimes (i ...) (insert ... ...)))
  (if (file-exists-p file) (progn (with-temp-buffer ... ... ... ... ... ...) (dotimes ... ...)))
  (when (file-exists-p file) (with-temp-buffer (insert ...) (goto-char ...) (forward-line 3) (while ... ...) (setq bits-list ...) (dotimes ... ... ...)) (dotimes (i ...) (insert ... ...)))
  (let* ((mapfile ...) (levels ...) (step ...) color-alist bits bits-list mask pixel x-faces) (with-temp-file mapfile (insert "P3\n") (insert ...) (insert "255\n") (dotimes ... ... ...)) (when (file-exists-p file) (with-temp-buffer ... ... ... ... ... ...) (dotimes ... ...)) (delete-file mapfile))
  gnus-convert-image-to-gray-x-face("~/doomicon.gif" 1)

Cheers, Ray
-- 
It's clear that whoever set up the font colorings for most programming
modes has seen too many Peter Max posters, or did more acid than I did
in the 60's.                     (Charles R. Martin in gnu.emacs.help)

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Raymond Scholz <ray-2002@zonix.de> writes:

> It doesn't for me :-(  Same with the original code from Gnus CVS.
>
> Debugger entered--Lisp error: (wrong-type-argument number-or-marker-p nil)
>   logand(nil 1)
>   (zerop (logand bits mask))
>   (if (zerop (logand bits mask)) "0 " "1 ")

Do you still see this, or has it gone away by now?

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

Re: X-Face and depth

[Raymond Scholz]

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

> Do you still see this, or has it gone away by now?

I'm on a different machine right now (SuSE 7.1) and that error is
gone, however a new one occured.  This is what the function inserts
into the Message Buffer:

X-Face:pbmtoxbm: EOF / read error
compface: <stdin>: insufficient or invalid data
X-Face-1:pbmtoxbm: EOF / read error
compface: <stdin>: insufficient or invalid data
X-Face-2:pbmtoxbm: EOF / read error
compface: <stdin>: insufficient or invalid data
X-Face-3:pbmtoxbm: EOF / read error
compface: <stdin>: insufficient or invalid data

(no backtrace)

Cheers, Ray
-- 
Gnus is already huge.  Maybe it doesn't need all of this kind of stuff. -
I think you mean "Gnus is huge, so adding more stuff won't make it
noticeably bigger".               (Chris Beggy and LMI on gnu.emacs.gnus)

Re: X-Face and depth

[Raymond Scholz]

Raymond Scholz <ray-2002@zonix.de> writes:

> It doesn't for me :-(  Same with the original code from Gnus CVS.

It now works, but I dunno why...

Cheers, Ray
-- 
I have a local archive newsserver with groups with more than 50k messages in it. 
When I try to enter those groups with Gnus I can have a cigarette before Gnus 
is ready. -- Oh no, I can see the lawsuits ... "Gnus causes cancer".
                                   (Frank Schmitt and Simon Josefsson on *ding*)

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

palmieri@math.washington.edu (John H. Palmieri) writes:

> I'm attaching an almost completely untested version of
> gnus-convert-image-to-gray-x-face, which shouldn't require the
> presence of xbmtoikon. 

Could you mail me a patch?

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

Re: X-Face and depth

[Thomas Kosch]

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

> Here we go...  There should be three X-Face parts up there.

:(
When I access (only) your article with multiple X-Faces: I get:

Symbol's function definition is void: create-image

I want also to see it

-- 
Klein bottle for rent -- inquire within.

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Thomas Kosch <t.kosch@cyvermail.de> writes:

> When I access (only) your article with multiple X-Faces: I get:
>
> Symbol's function definition is void: create-image

Is this with the current CVS?

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

Re: X-Face and depth

[Thomas Kosch]

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

> Is this with the current CVS?

>From 15 minutes ago

-- 
Do what comes naturally now.  Seethe and fume and throw a tantrum.

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Thomas Kosch <t.kosch@cyvermail.de> writes:

> When I access (only) your article with multiple X-Faces: I get:
>
> Symbol's function definition is void: create-image

`(setq debug-on-error t)', repeat the bug and mail me the resulting
backtrace.

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

Re: X-Face and depth

[Karl Kleinpaste]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
> I've now implemented a set of functions in gnus-fun.el that basically
> splits up and puts back together again bit planes.

Compiling under XEmacs 21.4.6:

| Compiling /home/karl/Emacs/gnus/lisp/gnus-fun.el...
| While compiling gnus-convert-gray-x-face-to-xpm in file /home/karl/Emacs/gnus/lisp/gnus-fun.el:
|   ** shell-command-on-region called with 6 arguments, but accepts only 3-5
| Wrote /home/karl/Emacs/gnus/lisp/gnus-fun.elc

XEmacs' doc for this function observes its calling structure:

(shell-command-on-region START END COMMAND &optional OUTPUT-BUFFER REPLACE)

Now, I looked at gnus-fun.el to see how this function might be
mis-called, and the code in use is...

       (shell-command-on-region
	 (point-min) (point-max)
	 "uncompface -X | xbmtopbm | pnmtoplainpnm"
	 (current-buffer) t)

...which looks a lot like 5 arguments.  I'm confused.

Also, for the record, I'm running Linux RH7.1 with most things
installed, but I haven't got any program "pnmtoplainpnm".  What does
it do?  Is this an excessively recent addition to the PBM package, and
could the same result be achieved with something else not quite so
bleeding-edge?

Re: X-Face and depth

[Zlatko Calusic]

Karl Kleinpaste <karl@charcoal.com> writes:

> Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
>> I've now implemented a set of functions in gnus-fun.el that basically
>> splits up and puts back together again bit planes.
>
> Compiling under XEmacs 21.4.6:
>
> | Compiling /home/karl/Emacs/gnus/lisp/gnus-fun.el...
> | While compiling gnus-convert-gray-x-face-to-xpm in file /home/karl/Emacs/gnus/lisp/gnus-fun.el:
> |   ** shell-command-on-region called with 6 arguments, but accepts only 3-5
> | Wrote /home/karl/Emacs/gnus/lisp/gnus-fun.elc
>
> XEmacs' doc for this function observes its calling structure:
>
> (shell-command-on-region START END COMMAND &optional OUTPUT-BUFFER REPLACE)
>
> Now, I looked at gnus-fun.el to see how this function might be
> mis-called, and the code in use is...
>
>        (shell-command-on-region
> 	 (point-min) (point-max)
> 	 "uncompface -X | xbmtopbm | pnmtoplainpnm"
> 	 (current-buffer) t)
>
> ...which looks a lot like 5 arguments.  I'm confused.
>

Problematic invocation is around line 130:

      (shell-command-on-region
       (point-min) (point-max)
       "ppmtoxpm"
       (current-buffer) t (get-buffer-create " *junk"))

Can I safely remove the 6th argument and solve the problem with that
change?
-- 
Zlatko

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Zlatko Calusic <zlatko.calusic@iskon.hr> writes:

> Problematic invocation is around line 130:
>
>       (shell-command-on-region
>        (point-min) (point-max)
>        "ppmtoxpm"
>        (current-buffer) t (get-buffer-create " *junk"))
>
> Can I safely remove the 6th argument and solve the problem with that
> change?

Almost.  I've now fixed the problem.

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

Re: X-Face and depth

[Karl Kleinpaste]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
> Almost.  I've now fixed the problem.

I don't know what you fixed, but gnus-fun.el didn't get updated in my
latest "cvs update" 10 minutes ago, and the build is still complaining:

  ** shell-command-on-region called with 6 arguments, but accepts only 3-5

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Karl Kleinpaste <karl@charcoal.com> writes:

> I don't know what you fixed, but gnus-fun.el didn't get updated in my
> latest "cvs update" 10 minutes ago, and the build is still complaining:

I haven't checked in the fixes yet.  Sorry.

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

Re: X-Face and depth

[Karl Kleinpaste]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
> I haven't checked in the fixes yet.  Sorry.

No problem.

Now that I've got all the current fixes,
gnus-convert-gray-x-face-to-xpm is bombing out on every message of
yours because you invoke "uncompface -X", in turn because uncompface
doesn't take any args other than filenames -- the diagnostics that end
up in the temp buffer lead to a subsequent wrong-type-argument
complaint, against number-char-or-marker-p.

What had you expected "-X" to provide?

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Karl Kleinpaste <karl@charcoal.com> writes:

> Now that I've got all the current fixes,
> gnus-convert-gray-x-face-to-xpm is bombing out on every message of
> yours because you invoke "uncompface -X", in turn because uncompface
> doesn't take any args other than filenames -- the diagnostics that end
> up in the temp buffer lead to a subsequent wrong-type-argument
> complaint, against number-char-or-marker-p.
>
> What had you expected "-X" to provide?

It's another Debian compface extension.  They've extended it to
output/input xbm files in addition to that odd "ikon" format the
normal compface uses.

What would be a more "standard" sequence of commands to convert an
X-Face to ppm?

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

Re: X-Face and depth

[Karl Kleinpaste]

Karl Kleinpaste <karl@charcoal.com> writes:
>> What had you expected "-X" to provide?

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
> It's another Debian compface extension.  They've extended it to
> output/input xbm files in addition to that odd "ikon" format the
> normal compface uses.

> What would be a more "standard" sequence of commands to convert an
> X-Face to ppm?

uncompface | ikon2xbm

both of which are part of the faces package, as installed on my RH
system.  I would guess that Debian's -X extension to uncompface is
merely an internal shorthand for this anyway.

Re: X-Face and depth

[Karl Kleinpaste]

Karl Kleinpaste <karl@charcoal.com> writes:
> uncompface | ikon2xbm

But when I just tried this on one of Lars' messages, it bombed thus
anyhow:

Signaling: (void-function create-image)
  create-image("/* XPM */\nstatic char *noname[] = {\n/* width height ncolors chars_per_pixel */\n\"48 48 4 1\",\n/* colors */\n\"  c #000000\",\n\". c #555555\",\n\"X c #AAAAAA\",\n\"o c #FFFFFF\",\n/* pixels */\n\"                                                \",\n\"  .  .  .  .  .  .  .  .  .   .  .  .  .  .  .  \",\n\"                                                \",\n\"  .  .   .                                      \",\n\"       .   . . .  .  .  .  .        .  .   .    \",\n\" . . .                        . . .      .    . \",\n\"       . . .  . .  .  . .  .         .     .    \",\n\" . . .      .       .          .   .   . .   .  \",\n\"      . . .   . . .   .  . . .  . . . .    .    \",\n\"  . .       .          .        . . . . ..      \",\n\"     . . .    . . . .    . . .             . .  \",\n\" . .        .          .       . . . . . .      \",\n\"     . . . .  . . . .    .  .              .  . \",\n\"  . .       .          .  .   . . . . . . .     \",\n\"     . . .    . . . . .     .               .   \",\n\" . .       . .          . .   . . . . . . .  .  \",\n\"     . . .     . . . .      .                   \",\n\"  . .       . .         . .   . . . . . . . . . \",\n\"      . . .    . . . .     .                    \",\n\" . . .      . . ..     . .   . . . . . . . . .  \",\n\"      . . .   . X.. . .   .                     \",\n\"  . .     . . ......    .  . . . . . . . . . .  \",\n\"     . ...  . ..... . . .                      .\",\n\". . .  .  ...X.....X.. . . . . . . . . . . . .  \",\n\"      .. XXX XX....X..X. .                      \",\n\" . . .   XXX..X...X...XX.  . . . . . . .  .  .  \",\n\"       . . .  Xo.XX.....  .              .      \",\n\"  . . .    ..........   .    . . . . . .   . .  \",\n\"        . . .oX ...XX     . .            .      \",\n\" . . . .   .....X...... .      . . . . .     .  \",\n\"         ..  .. .....    . . .           . .    \",\n\" . . . .  . XXX. ..XX. .       . . . . .      . \",\n\"      .  ... X. ...X.... . . .           . . .  \",\n\"  . .  . .... .X  . .. .       . . . . .        \",\n\"     .   .   . ....  . . . . .          . . .   \",\n\" . .   . . .  .X..  .  .        . . . .       . \",\n\"     .   .   ..o.XX. .X  . . .          . .     \",\n\"  . .  . . .  .o.XX   . .     . . . . .      .  \",\n\" .    .  ..  .XX.XX . .. . . .          . .     \",\n\"    .  . X .  Xo.Xo  ...        . . . .      .  \",\n\" . .  .     . oo.Xo    . . . .         . . .    \",\n\"    .   . .   X. .X. .         . . . .        . \",\n\" .    .    . .X...X    . . . .        . . . .   \",\n\"  . .   . .         . .        . . . .          \",\n\"     .     . . . .      . . .          . . . .  \",\n\" . .  . . .        . . .      . . . . .   .     \",\n\"    .    .  . . . .      . . .    .    . .  . . \",\n\"  .   . .  . .      . . .       .  . .    .     \"\n};\n" xpm t)
  article-display-x-face()
[remainder deleted]

At least the ikon2xbm stage worked correctly, but now "create-image"
is a problem.

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Karl Kleinpaste <karl@charcoal.com> writes:

> But when I just tried this on one of Lars' messages, it bombed thus
> anyhow:
>
> Signaling: (void-function create-image)

Fixed now.

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

Re: X-Face and depth

[Karl Kleinpaste]

[-- Attachment #1: Type: text/plain, Size: 1058 bytes --]

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

I'd like to believe you, and I updated from CVS a few minutes ago, but
now one (just one) of your recent messages
(<m3d70svvis.fsf@quimbies.gnus.org>) still causes the old error:

Signaling: (wrong-type-argument number-char-or-marker-p nil)
  gnus-convert-gray-x-face-to-xpm(("%Hp1fbwa;SIy+'TltJ#4yll=PWK+N{,o/+yZ0DGDuGz-V!GBP60VYmywY0L,TJsDhSqd5n*\n (65ami.3pfVr9E85W=EjzWTVhk3L@we!]P=}I]jiuRG;@s[~=J1<p[[{F'A~LwLOuGpn4V8=mu1~OF\n (y;+QzfD6$;kVG/qs7|2<StKcx-./Im|&O$1GX9ZL_xc7yEpI:IKe[=,i\"`lkXbYC*v|$[?B.=WsCn\n &ep4n`PpcG=#}m~?+{=B.Tkc5b,|YX`CMD$}@k<Ae7U$<J$Xn}/I)GvD`\n" "\"f<CbgVRvS!!I\"s>x4YP?7~e:{]Rj}pRIj6Y@ScmCCC|?8!E3dV-yW,D9Sz%3ueas1`)Di%\n [J6h<:4~38-$iuv.C/lW_\\j'4/.5`-8|Ds7pT)DRFR1r0!L\"!tpfcZVhBj{s{GNvqQcPgG%4zGdGIH\n T,Z+GWh<U,t9;[&%3ExeZIzQHD1\"H!zz'U?B[5y<A?!GZ:>\n"))
  article-display-x-face()

And now that I can see them, I have trouble believing that these are
the images you intended to be displayed.

Notice that, in the 2nd, the image isn't placed properly, either.


[-- Attachment #2: 2002_01_03_135936_shot.jpg --]
[-- Type: image/jpeg, Size: 42766 bytes --]

[-- Attachment #3: 2002_01_03_140112_shot.jpg --]
[-- Type: image/jpeg, Size: 29006 bytes --]

[-- Attachment #4: 2002_01_03_140145_shot.jpg --]
[-- Type: image/jpeg, Size: 41698 bytes --]

Re: X-Face and depth

[Steve Youngs]

|--==> "KK" == Karl Kleinpaste <karl@charcoal.com> writes:

  KK> Notice that, in the 2nd, the image isn't placed properly, either.

And on those messages where the image is displayed to the left of
"From:" instead of to the right,  BBDB doesn't recognise that the From
header is the From header.  It picks up on the To header instead.


-- 
|---<Steve Youngs>---------------<GnuPG KeyID: 10D5C9C5>---|
|            XEmacs - It's not just an editor.             |
|                    It's a way of life.                   |
|------------------------------------<youngs@xemacs.org>---|

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Karl Kleinpaste <karl@charcoal.com> writes:

> I'd like to believe you, and I updated from CVS a few minutes ago, but
> now one (just one) of your recent messages
> (<m3d70svvis.fsf@quimbies.gnus.org>) still causes the old error:

Yup.  Fix in Oort Gnus v0.05 (i. e., CVS).

> And now that I can see them, I have trouble believing that these are
> the images you intended to be displayed.

Those were the images I sent out.

The first one demonstrates that not all article sleeves are instantly
recognizeable when reduced to 48x48x2, and the second demonstrates
that I didn't check error codes from compface.

> Notice that, in the 2nd, the image isn't placed properly, either.

Yup.  Fix in Oort Gnus v0.05 (i. e., CVS).

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

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Karl Kleinpaste <karl@charcoal.com> writes:

> uncompface | ikon2xbm
>
> both of which are part of the faces package, as installed on my RH
> system.  I would guess that Debian's -X extension to uncompface is
> merely an internal shorthand for this anyway.

My Debian machine doesn't seem to have ikon2xbm installed, and
"apt-cache search ikon2" returns nothing, so it's not part of Debian.

Is "icontopbm" something else?

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

Re: X-Face and depth

[Karl Kleinpaste]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
> My Debian machine doesn't seem to have ikon2xbm installed, and
> "apt-cache search ikon2" returns nothing, so it's not part of Debian.

This growing empty-intersection disparity among Linux systems will be
its deathknell, the way that BSD /vs/ SysV doomed UNIX generally
against the M$ onslaught.

/mutter/

> Is "icontopbm" something else?

I have the following icon* programs:

Program:          Part of:
icon2gif          libungif-progs-4.1.0-7
icon2ikon         faces-1.6.1-10
icontopbm         netpbm-progs-9.9-5
iconv             glibc-common-2.2.4-19.3

Re: X-Face and depth

[Karl Kleinpaste]

Perhaps what we need is a (sick) conditional use of "uncompface -X"
/vs/ "uncompface | ikon2xbm", based on whether the latter program is
available.

Yuck.  That means Gnus will have to reach out into the filesystem to
hunt down ikon2xbm.

Weirder still: Use a variable to hold the pipeline, trying "uncompface
-X" by default.  But wrap its use in an error catch which replaces
"uncompface -X" with "uncompface | ikon2xbm" and retries it.

I can't believe I'm even contemplating this.

Re: X-Face and depth

[Simon Josefsson]

Karl Kleinpaste <karl@charcoal.com> writes:

> Yuck.  That means Gnus will have to reach out into the filesystem to
> hunt down ikon2xbm.

At least nnimap already does this for OpenSSL vs SSLeay, different
incompatible versions of "imtest", rsh vs ssh, STARTTLS etc.

> Weirder still: Use a variable to hold the pipeline, trying "uncompface
> -X" by default.  But wrap its use in an error catch which replaces
> "uncompface -X" with "uncompface | ikon2xbm" and retries it.
>
> I can't believe I'm even contemplating this.

Maybe

(uncompface -X) || (uncompface | ikon2xbm)

works.

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Karl Kleinpaste <karl@charcoal.com> writes:

> Perhaps what we need is a (sick) conditional use of "uncompface -X"
> /vs/ "uncompface | ikon2xbm", based on whether the latter program is
> available.

I just had a look at gnus-ems.el.  `gnus-article-display-xface' uses a
`gnus-article-compface-xbm' variable, and it checks stuff and stuff.

I think we should roll an uncompface.el library that does these
checks, and just provides a `uncompface' function that can be used.
We can use `gnus-article-display-xface' as a model.

I'll do that, and alter Gnus to use this library.

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

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Karl Kleinpaste <karl@charcoal.com> writes:

> Perhaps what we need is a (sick) conditional use of "uncompface -X"
> /vs/ "uncompface | ikon2xbm", based on whether the latter program is
> available.

Incidentally, XEmacs has built-in X-Face support, so all this isn't
really pertinent for XEmacs.  I think.  :-)

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

Re: X-Face and depth

[Karl Kleinpaste]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
> Incidentally, XEmacs has built-in X-Face support, so all this isn't
> really pertinent for XEmacs.  I think.  :-)

Well, I'm running XEmacs 21.4.6.  The built-in support is fine for
regular B&W X-Face; it's just the new multi-depth X-Face support that
is causing difficulty.

Re: X-Face and depth

[David S. Goldberg]

>>>>> On Thu, 03 Jan 2002 19:02:51 +0100, Lars Magne Ingebrigtsen <larsi@gnus.org> said:

> Incidentally, XEmacs has built-in X-Face support, so all this isn't
> really pertinent for XEmacs.  I think.  :-)

Unfortunately it is.  I'm getting the same traceback as Michael Cook
using XEmacs 21.1.14 with xface support compiled in.
-- 
Dave Goldberg
david.goldberg6@verizon.net

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

david.goldberg6@verizon.net (David S. Goldberg) writes:

>> Incidentally, XEmacs has built-in X-Face support, so all this isn't
>> really pertinent for XEmacs.  I think.  :-)
>
> Unfortunately it is.  I'm getting the same traceback as Michael Cook
> using XEmacs 21.1.14 with xface support compiled in.

That's another thing.  :-)

But I've fixed it now.

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

Re: X-Face and depth

[Vincent Bernat]

OoO Pendant le repas du jeudi 03 janvier 2002, vers 19:32, Lars Magne
Ingebrigtsen <larsi@gnus.org> disait:

> That's another thing.  :-)

> But I've fixed it now.

When all this will working, will the default config work even when we
don't have the requested utils on the system (all these imaging stuff)
?
-- 
I AM NOT DELIGHTFULLY SAUCY
I AM NOT DELIGHTFULLY SAUCY
I AM NOT DELIGHTFULLY SAUCY
-+- Bart Simpson on chalkboard in episode 1F14

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Vincent Bernat <bernat@scientist.com> writes:

> When all this will working, will the default config work even when
> we don't have the requested utils on the system (all these imaging
> stuff) ?

If you don't have all the image stuff installed, you should see no
images, and get no breakage.

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

Re: X-Face and depth

[Russ Allbery]

Karl Kleinpaste <karl@charcoal.com> writes:

> uncompface | ikon2xbm

> both of which are part of the faces package, as installed on my RH
> system.  I would guess that Debian's -X extension to uncompface is
> merely an internal shorthand for this anyway.

I wonder what faces package this is and where the upstream source is.

The canonical compface package from ftp.cs.indiana.edu contains only
compface and uncompface and doesn't support a -X option.

-- 
Russ Allbery (rra@stanford.edu)             <http://www.eyrie.org/~eagle/>

Re: X-Face and depth

[Karl Kleinpaste]

Karl Kleinpaste <karl@charcoal.com> writes:
>> uncompface | ikon2xbm

Russ Allbery <rra@stanford.edu> writes:
> I wonder what faces package this is and where the upstream source is.

Hm?

[1489] [02:10pm] cinnamon:~> rpm -ql faces
/usr/bin/addxface
/usr/bin/compface
/usr/bin/face_update
/usr/bin/faces
/usr/bin/faces.sendmail
/usr/bin/facesaddr
/usr/bin/facesall
/usr/bin/from.faces
/usr/bin/fs2ikon
/usr/bin/fs2xbm
/usr/bin/icon2ikon
/usr/bin/ikon2icon
/usr/bin/lpqall.faces
/usr/bin/mailq.faces
/usr/bin/mkfacesindex
/usr/bin/newscheck.faces
/usr/bin/newsfrom.faces
/usr/bin/rotary.faces
/usr/bin/rs2icon
/usr/bin/xbm2ikon
/usr/bin/xbmcut48
/usr/bin/xbmsize48
/usr/man/man1/compface.1
/usr/man/man1/face_update.1
/usr/man/man1/faces.1
/usr/man/man3/compface.3
[1490] [02:10pm] cinnamon:~> rpm -ql faces-xface
/usr/bin/ikon2xbm
/usr/bin/uncompface
/usr/man/man1/uncompface.1
/usr/man/man3/uncompface.3

Oh, waitaminit...

[ rummage dig shovel throw ]

The faces RPMs I have installed date from April 1998 and, as best I
can tell, were part of RH5.x.  I'm running RH7.1 now, and it seems
I've been carting these around on my own since 6.x came out and they
weren't included by RH themselves any more.

Re: X-Face and depth

[Andrew J Cosgriff]

Russ Allbery wrote :

> Karl Kleinpaste <karl@charcoal.com> writes:
>
>> uncompface | ikon2xbm
>
>> both of which are part of the faces package, as installed on my RH
>> system.  I would guess that Debian's -X extension to uncompface is
>> merely an internal shorthand for this anyway.
>
> I wonder what faces package this is and where the upstream source is.
>
> The canonical compface package from ftp.cs.indiana.edu contains only
> compface and uncompface and doesn't support a -X option.

It's not a Debian-specific extension, per-se : the patch to add -X was
floating around for a couple of years back in the early to mid 90's I
vaguely remember.

A quick google search shows that it's distributed with the EXMH source
(eg. http://www.mit.edu/afs/sipb/project/tcl/src/exmh-1.6.1/misc/uncompface.patch)
...which'd be where I remember it from - EXMH being my primary source
of joy before I changed to Gnus back in...95 or 96...


-- 
Andrew J Cosgriff <ajc@polydistortion.net> calling from a country phone

Re: X-Face and depth

[Zlatko Calusic]

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

> Zlatko Calusic <zlatko.calusic@iskon.hr> writes:
>
>> Problematic invocation is around line 130:
>>
>>       (shell-command-on-region
>>        (point-min) (point-max)
>>        "ppmtoxpm"
>>        (current-buffer) t (get-buffer-create " *junk"))
>>
>> Can I safely remove the 6th argument and solve the problem with that
>> change?
>
> Almost.  I've now fixed the problem.

Now, when I access (only) your article with multiple X-Faces: I get:

Symbol's function definition is void: create-image

May be it isn't directly connected to the problem above.
Yup, XEmacs here!
-- 
Zlatko

Re: X-Face and depth

[Lars Magne Ingebrigtsen]

Karl Kleinpaste <karl@charcoal.com> writes:

> Also, for the record, I'm running Linux RH7.1 with most things
> installed, but I haven't got any program "pnmtoplainpnm".  What does
> it do?  Is this an excessively recent addition to the PBM package, and
> could the same result be achieved with something else not quite so
> bleeding-edge?

It seems to be a Debian alias for pnmnoraw.  I've now altered gnus-fun
to use that instead.

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

Re: X-Face and depth

[Simon Josefsson]

On Wed, 2 Jan 2002, Lars Magne Ingebrigtsen wrote:

> However, I'm not sure this is a good idea.  Do we want to encourage
> people to add even more data to the headers?  On the other hand, this
> is 2002, and bandwidth and disk space is so much cheaper than it was
> when X-Faces were dreamt up.  On the third hand, why not define a new
> format altogether?  X-Gif-Face, or something.  Or even referring to an
> outside source via HTTP.

Maybe it is time to write the Face: specification?  Suggestion: Make it 
contain a list of URLs inside <>'s. Then you could have things such as:

Face: <data:image/gif;base64,ABHCMKLKAKJNCKJNE=>
Face: <cid:MyFace>     (together with a MIME part with Content-ID MyFace)
Face: <http://www.gnus.org/myface.gif>   (possibly together with cache)
Face: <data:video/mpeg;base64,AUASHUINVJKNDK>

Etc.

Re: X-Face and depth

[Steve Youngs]

|--==> "LMI" == Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

  LMI> However, I'm not sure this is a good idea.  Do we want to
  LMI> encourage people to add even more data to the headers?

Probably not, but the lusers who'd abuse it are more likely to be
using Internet Exploder or Netscum rather than Gnus anyway.  Most of
them probably wouldn't even know what a header was.

  LMI> is 2002, and bandwidth and disk space is so much cheaper than
  LMI> it was when X-Faces were dreamt up.

True, but even today it's still out of reach of some people.  I'd love
to have a 24/7 cable connection, but I can't afford it so I stick with
my 56k modem dialup.  My 30 gig disk is rapidly filling up, but I bet
I'm deleting stuff before I can afford to add more disk space.

Just reminding you that there are still people in the world on a tight
budget. :-)

  LMI> On the third hand, why not define a new format altogether?
  LMI> X-Gif-Face, or something.  

X-Png-Face would be better (keeps away from the Gif patent issue). 

  LMI> Or even referring to an outside source via HTTP.

So when do we get X-Streaming-Video? :-)

-- 
|---<Steve Youngs>---------------<GnuPG KeyID: 10D5C9C5>---|
|            XEmacs - It's not just an editor.             |
|                    It's a way of life.                   |
|------------------------------------<youngs@xemacs.org>---|

Re: X-Face and depth

[Steinar Bang]

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

> However, I'm not sure this is a good idea.  Do we want to encourage
> people to add even more data to the headers?  On the other hand,
> this is 2002, and bandwidth and disk space is so much cheaper than
> it was when X-Faces were dreamt up.  On the third hand, why not
> define a new format altogether?  X-Gif-Face, or something.
                                     ^^^eeeek!
X-png-face, please!

It's 2002, after all. :-)

Re: X-Face and depth

[Andrew J Cosgriff]

Steinar Bang wrote :

>>>>>> Lars Magne Ingebrigtsen <larsi@gnus.org>:
>
>> However, I'm not sure this is a good idea.  Do we want to encourage
>> people to add even more data to the headers?  On the other hand,
>> this is 2002, and bandwidth and disk space is so much cheaper than
>> it was when X-Faces were dreamt up.  On the third hand, why not
>> define a new format altogether?  X-Gif-Face, or something.
>                                      ^^^eeeek!
> X-png-face, please!
>
> It's 2002, after all. :-)


I think I'd prefer an URL and perhaps some kind of caching mechanism
for people who don't have an HTTP cache their side of a slow link.

-- 
Andrew J Cosgriff <ajc@polydistortion.net> feeling like a robot by the river