ggi terminal for gnuplot

ggi terminal for gnuplot

[Peter Rosin]

[-- Attachment #1: Type: text/plain, Size: 659 bytes --]

Hi!

I noticed the following in the cygport file of gnuplot

	# This gives parse errors
	#  --with-ggi

and thought I'd have a look.

The cause is a type cleanup in libggi, and here's a patch to
make gnuplot follow.

Unfortunately the build is killed later by some tex config thing
that I don't have the energy to follow up on, so I have not
tested if this does anything useful other than fixing the first
ggi related build error. Since I don't know if there are more
problems later in the build I'm sending this anyway.

Another variant would be to build with GG_NEED_OLD_INTTYPES
defined, but I think the attached patch is a better approach.

Cheers,
Peter

[-- Attachment #2: ggi.patch --]
[-- Type: text/plain, Size: 1265 bytes --]

--- gnuplot-4.2.3/term/ggi.trm.old	2008-06-25 00:02:24.951334700 +0200
+++ gnuplot-4.2.3/term/ggi.trm	2008-06-25 00:02:42.826509600 +0200
@@ -116,9 +116,9 @@
 } GGI_vertex_t;
 
 TERM_PUBLIC long int GGI_SetTime(const struct timeval* current);
-TERM_PUBLIC int GGI_from_keysym __PROTO((uint32 keysym));
-TERM_PUBLIC int GGI_from_button __PROTO((uint32 button));
-TERM_PUBLIC int GGI_y __PROTO((sint32 y));
+TERM_PUBLIC int GGI_from_keysym __PROTO((uint32_t keysym));
+TERM_PUBLIC int GGI_from_button __PROTO((uint32_t button));
+TERM_PUBLIC int GGI_y __PROTO((int32_t y));
 TERM_PUBLIC int GGI_dispatch_event __PROTO((const ggi_event* event));
 TERM_PUBLIC int GGI_eventually_update_modifiers __PROTO((const ggi_event* event, const int add));
 TERM_PUBLIC int GGI_waitforinput __PROTO((void));
@@ -645,7 +645,7 @@
 
 /* translate ggi keysym to gnuplot keysym */
 TERM_PUBLIC int
-GGI_from_keysym(uint32 keysym)
+GGI_from_keysym(uint32_t keysym)
 {
     switch (keysym) {
 	case GIIUC_BackSpace:
@@ -811,7 +811,7 @@
 }
 
 TERM_PUBLIC int
-GGI_from_button(uint32 button)
+GGI_from_button(uint32_t button)
 {
     switch (button) {
     case GII_PBUTTON_LEFT:
@@ -827,7 +827,7 @@
 }
 
 TERM_PUBLIC int
-GGI_y(sint32 y)
+GGI_y(int32_t y)
 {
     return GGIymax - y;
 }


[-- Attachment #3: Type: text/plain, Size: 218 bytes --]

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: ggi terminal for gnuplot

[Peter Rosin]

Peter Rosin skrev:
> Hi!
> 
> I noticed the following in the cygport file of gnuplot
> 
>     # This gives parse errors
>     #  --with-ggi
> 
> and thought I'd have a look.
> 
> The cause is a type cleanup in libggi, and here's a patch to
> make gnuplot follow.

Just writing back to inform that I posted a variant of this patch
(adapted for gnuplot-cvs) in the upstream patch tracker[1].

Cheers,
Peter

[1] 
https://sourceforge.net/tracker/index.php?func=detail&aid=2003010&group_id=2055&atid=302055

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: ggi terminal for gnuplot

[Peter Rosin]

Peter Rosin skrev:
> Peter Rosin skrev:
>> Hi!
>>
>> I noticed the following in the cygport file of gnuplot
>>
>>     # This gives parse errors
>>     #  --with-ggi
>>
>> and thought I'd have a look.
>>
>> The cause is a type cleanup in libggi, and here's a patch to
>> make gnuplot follow.
> 
> Just writing back to inform that I posted a variant of this patch
> (adapted for gnuplot-cvs) in the upstream patch tracker[1].

Another day, another update. The patch was accepted upstream and
I also noticed that the build failed *after* generating the main
executable. So I was in fact able to check if the patch does
anything useful. And it does! :-)

With "set terminal ggi" in my ~/.gnuplot file, I'm now able to see
gnuplot output w/o an X-server.

Cheers,
Peter

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: ggi terminal for gnuplot

[Peter Rosin]

Peter Rosin skrev:
> Peter Rosin skrev:
>> Peter Rosin skrev:
>>> Hi!
>>>
>>> I noticed the following in the cygport file of gnuplot
>>>
>>>     # This gives parse errors
>>>     #  --with-ggi
>>>
>>> and thought I'd have a look.
>>>
>>> The cause is a type cleanup in libggi, and here's a patch to
>>> make gnuplot follow.
>>
>> Just writing back to inform that I posted a variant of this patch
>> (adapted for gnuplot-cvs) in the upstream patch tracker[1].
> 
> Another day, another update. The patch was accepted upstream and
> I also noticed that the build failed *after* generating the main
> executable. So I was in fact able to check if the patch does
> anything useful. And it does! :-)
> 
> With "set terminal ggi" in my ~/.gnuplot file, I'm now able to see
> gnuplot output w/o an X-server.

Ping.

It's been a while and vacations (or whatever) are over...

Cheers,
Peter

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: ggi terminal for gnuplot

[Dr. Volker Zell]

>>>>> Peter Rosin writes:

    > Ping.

    > It's been a while and vacations (or whatever) are over...

Sorry I'm busy right now with other things, but I noticed your patch and
will release an new gnuplot as soon as I find some time.

    > Cheers,
    > Peter

Thanks
  Volker

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/