6:10 a.m.
A couple of users have reported warnings on 21.5.31 when using
"gnuclient -nw" and XEmacs has a graphics image that was created using
(make-glyph ... :data ...). See issue778 for details (including the
values of the :data-based and :file-based glyphs that I talk about
below).
I'd like some guidance on how the code is supposed to work.
First question: as described in issue778, the test-case glyph from MH-E
is different depending on whether it was created using :data or :file.
With:data, the specifier has tty and stream clauses that it lacks when
created with :file. Is that to be expected?
Second question: suppose a buffer is created containing one of these
:data glyphs while the current frame is an X window, and then the buffer
is then viewed in a tty frame. With 21.5.31, the user gets warnings.
With 21.5.29+0d436a78c514, there are no warnings, even though the
glyph's specifier still has those tty and stream clauses. Is this
new warning expected?
thanks,
mike
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
7:56 a.m.
Mike Kupfer writes:
A couple of users have reported warnings on 21.5.31 when using
"gnuclient -nw" and XEmacs has a graphics image that was created using
(make-glyph ... :data ...). See issue778 for details (including the
values of the :data-based and :file-based glyphs that I talk about
below).
I'd need to see the exact code which is instantiating the images.
However remember that there are many source types for images. If you
only provide the :data element, the image instantiator code will have
to guess media type based on content. I suspect there's a bug in that
code that has a bad interaction with the generic specifier code that
tries to instantiate on all consoles. On the other hand the :file
element uses the file name extension to guess type, and I suspect is
more accurate.
First question: as described in issue778, the test-case glyph from
MH-E
is different depending on whether it was created using :data or :file.
With:data, the specifier has tty and stream clauses that it lacks when
created with :file. Is that to be expected?
No.
Second question: suppose a buffer is created containing one of these
:data glyphs while the current frame is an X window, and then the buffer
is then viewed in a tty frame. With 21.5.31, the user gets warnings.
With 21.5.29+0d436a78c514, there are no warnings, even though the
glyph's specifier still has those tty and stream clauses. Is this
new warning expected?
No.
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
8:36 a.m.
Mike Kupfer writes:
> First question: as described in issue778, the test-case glyph
from MH-E
> is different depending on whether it was created using :data or :file.
> With:data, the specifier has tty and stream clauses that it lacks when
> created with :file. Is that to be expected?
I replied:
No.
I'm sorry, I sent without reading carefully. I'm not entirely sure
without looking at the code more carefully, but I suspect the answer
is "Yes, at least sometimes." For example, if the :data were simply a
string, then it would be displayable in all those cases, and the
specifier code would try to instantiate. I suppose the media type
guessing code is invoked separately for each domain, maybe different
code is involved.
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
5:52 p.m.
Stephen J. Turnbull wrote:
Mike Kupfer writes:
>> First question: as described in issue778, the test-case glyph from MH-E
>> is different depending on whether it was created using :data or :file.
>> With:data, the specifier has tty and stream clauses that it lacks when
>> created with :file. Is that to be expected?
I replied:
> No.
I'm sorry, I sent without reading carefully. I'm not entirely sure
without looking at the code more carefully, but I suspect the answer
is "Yes, at least sometimes." For example, if the :data were simply a
string, then it would be displayable in all those cases, and the
specifier code would try to instantiate. I suppose the media type
guessing code is invoked separately for each domain, maybe different
code is involved.
Okay. Here's an example, based on the MH-E code.
-----8<-----8<-----
(let* ((load-path '("/home/kupfer/.xemacs/site-packages/etc/mh-e"))
(data (mh-icon-image 'mh-logo)))
(setq logo-data data))
logo-data
"/* XPM */
static char *mh-e[] = {
/* width height num_colors chars_per_pixel */
\" 18 13 2 1\",
/* colors */
\"# c #666699\",
\". c None s None\",
/* pixels */
\"........##........\",
\".......####.......\",
\"......######......\",
\"......######......\",
\"....#########.....\",
\"..##############..\",
\".##...######....#.\",
\"##...#.#.####...#.\",
\"....#..#.##.#...#.\",
\"...#..##.#.#.#....\",
\"...#..#..#..#.#...\",
\"...#..#.##..#.##..\",
\"...#..#.#..#....#.\"
};
"
;; Create glyph using :data:
(make-glyph `[xpm :data ,logo-data])
#<glyph (buffer) #<image-specifier global=(((x) . [xpm :color-symbols
(("foregroundToolBarColor" . #<color-specifier global=<unspecified>
fallback=#<color-specifier global=<unspecified> fallback=(((tty) . []) ((x) .
"black")) 0x50> 0x68>) ("backgroundToolBarColor" .
#<color-specifier global=<unspecified> fallback=#<color-specifier
global=<unspecified> fallback=(((tty) . []) ((x) . "Gray80")) 0x51>
0x69>) ("background" . #<color-specifier global="#ffffe6"
fallback=(((tty) . []) ((x) . "gray80")) 0x45>) ("foreground" .
#<color-specifier global=<unspecified> fallback=(((tty) . []) ((x) .
"black")) 0x44>)) :data "/* XPM */
static char *mh-e[] = {
/* width height num_colors chars_per_pixel */
\" 18 13 2 1\",
/* colors */
\"# c #666699\",
\". c None s None\",
/* pixels */
\"........##........\",
\".......####.......\",
\"......######......\",
\"......######......\",
\"....#########.....\",
\"..##############..\",
\".##...######....#.\",
\"##...#.#.####...#.\",
\"....#..#.##.#...#.\",
\"...#..##.#.#.#....\",
\"...#..#..#..#.#...\",
\"...#..#.##..#.##..\",
\"...#..#.#..#....#.\"
};
"]) ((tty) . [xpm :color-symbols (("foregroundToolBarColor" .
#<color-specifier global=<unspecified> fallback=#<color-specifier
global=<unspecified> fallback=(((tty) . []) ((x) . "black")) 0x50>
0x68>) ("backgroundToolBarColor" . #<color-specifier
global=<unspecified> fallback=#<color-specifier global=<unspecified>
fallback=(((tty) . []) ((x) . "Gray80")) 0x51> 0x69>)
("background" . #<color-specifier global="#ffffe6" fallback=(((tty)
. []) ((x) . "gray80")) 0x45>) ("foreground" . #<color-specifier
global=<unspecified> fallback=(((tty) . []) ((x) . "black")) 0x44>))
:data "/* XPM */
static char *mh-e[] = {
/* width height num_colors chars_per_pixel */
\" 18 13 2 1\",
/* colors */
\"# c #666699\",
\". c None s None\",
/* pixels */
\"........##........\",
\".......####.......\",
\"......######......\",
\"......######......\",
\"....#########.....\",
\"..##############..\",
\".##...######....#.\",
\"##...#.#.####...#.\",
\"....#..#.##.#...#.\",
\"...#..##.#.#.#....\",
\"...#..#..#..#.#...\",
\"...#..#.##..#.##..\",
\"...#..#.#..#....#.\"
};
"]) ((stream) . [xpm :color-symbols (("foregroundToolBarColor" .
#<color-specifier global=<unspecified> fallback=#<color-specifier
global=<unspecified> fallback=(((tty) . []) ((x) . "black")) 0x50>
0x68>) ("backgroundToolBarColor" . #<color-specifier
global=<unspecified> fallback=#<color-specifier global=<unspecified>
fallback=(((tty) . []) ((x) . "Gray80")) 0x51> 0x69>)
("background" . #<color-specifier global="#ffffe6" fallback=(((tty)
. []) ((x) . "gray80")) 0x45>) ("foreground" . #<color-specifier
global=<unspecified> fallback=(((tty) . []) ((x) . "black")) 0x44>))
:data "/* XPM */
static char *mh-e[] = {
/* width height num_colors chars_per_pixel */
\" 18 13 2 1\",
/* colors */
\"# c #666699\",
\". c None s None\",
/* pixels */
\"........##........\",
\".......####.......\",
\"......######......\",
\"......######......\",
\"....#########.....\",
\"..##############..\",
\".##...######....#.\",
\"##...#.#.####...#.\",
\"....#..#.##.#...#.\",
\"...#..##.#.#.#....\",
\"...#..#..#..#.#...\",
\"...#..#.##..#.##..\",
\"...#..#.#..#....#.\"
};
"])) fallback=((nil . [nothing])) 0xd0f>0x17b>
;; Create glyph using :file:
(make-glyph `[xpm :file
"/home/kupfer/.xemacs/site-packages/etc/mh-e/mh-logo.xpm"])
#<glyph (buffer) #<image-specifier global=((x) . [xpm :color-symbols
(("foregroundToolBarColor" . #<color-specifier global=<unspecified>
fallback=#<color-specifier global=<unspecified> fallback=(((tty) . []) ((x) .
"black")) 0x50> 0x68>) ("backgroundToolBarColor" .
#<color-specifier global=<unspecified> fallback=#<color-specifier
global=<unspecified> fallback=(((tty) . []) ((x) . "Gray80")) 0x51>
0x69>) ("background" . #<color-specifier global="#ffffe6"
fallback=(((tty) . []) ((x) . "gray80")) 0x45>) ("foreground" .
#<color-specifier global=<unspecified> fallback=(((tty) . []) ((x) .
"black")) 0x44>)) :file
"/home/kupfer/.xemacs/site-packages/etc/mh-e/mh-logo.xpm" :data "/* XPM
*/static char *pixmap[] = {\"18 13 2 1\",\"# c #666699\",\". s
None c
None\",\"........##........\",\".......####.......\",\"......######......\",\"......######......\",\"....#########.....\",\"..##############..\",\".##...######....#.\",\"##...#.#.####...#.\",\"....#..#.##.#...#.\",\"..!
.#..##.#.#.#....\",\"...#..#..#..#.#...\",\"...#..#.##..#.##..\",\"...#..#.#..#....#.\"};"])
fallback=((nil . [nothing])) 0xd12>0x17c>
----->8----->8-----
In both cases, the specifier includes the format (xpm). So there
shouldn't be any need for XEmacs to guess the format, should there?
mike
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
3:02 p.m.
Mike Kupfer writes:
Stephen J. Turnbull wrote:
> Mike Kupfer writes:
> >> With:data, the specifier has tty and stream clauses
that it lacks when
> >> created with :file. Is that to be expected?
> "Yes, at least sometimes." For example, if the :data
were simply a
> string, then it would be displayable in all those cases, and the
> specifier code would try to instantiate. I suppose the media type
> guessing code is invoked separately for each domain, maybe different
> code is involved.
logo-data
"/* XPM */
[...]
};
"
;; Create glyph using :data:
(make-glyph `[xpm :data ,logo-data])
#<glyph (buffer) #<image-specifier global=
(((x) . [xpm :color-symbols ...
:data "/* XPM */ ..."])
((tty) . [xpm :color-symbols ... :data "/* XPM */ ..."])
((stream) . [xpm :color-symbols ... :data "/* XPM */ ..."]))
fallback=((nil . [nothing])) 0xd0f>0x17b>
In both cases, the specifier includes the format (xpm). So there
shouldn't be any need for XEmacs to guess the format, should there?
I guess not. I don't know why this happens. You can dig into the
specifier code if you like. It will be a couple of weeks before I
can, though.
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
10:33 p.m.
Stephen J. Turnbull wrote:
Mike Kupfer writes:
> In both cases, the specifier includes the format (xpm). So
there
> shouldn't be any need for XEmacs to guess the format, should there?
I guess not. I don't know why this happens. You can dig into the
specifier code if you like. It will be a couple of weeks before I
can, though.
Okay. It's not urgent for me because there's a workaround. I've put it
on my list, and we'll see who finds a round tuit first. ;-)
mike
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
4674
days inactive
4676
days old
5 comments
2 participants
participants (2)
-
Mike Kupfer -
Stephen J. Turnbull