setedit-users — General list about setedit

Re: latest build 5.5 build 1162 problems & questions

[Salvador E. T. <sal...@in...>]

Arnold Wiegert wrote:

>> Do you remmember if the make command recompiled all? Try doing "make 
>> clean" to force a recompilation.
>>

> I did use make clean first to ensure a recompile and it did - it took 
> a bit of time ;-)


:-)

>>> it also hangs, but when I kill it from another window, I do get the 
>>> gdb window and can do a backtrace, but I have not been able to cut & 
>>> paste again from that window - I know I did it last week - but this 
>>> week a thick "fog" has descended and I can't get it to work.
>>
>>
>>
>>
>> Strange, to what program are you trying to copy the text?
>
> I tried to cut&paste to kedit and any other editor I could find, 
> including setedit.
>
> I've tried various ways of marking and copying the text to the Windows 
> clipboard, but not luck at all - and yet I know I did it at least once.


The easiest way is to select with the mouse and paste with the middle 
button.

SET

-- 
Salvador Eduardo Tropea (SET). (Electronics Engineer)
Visit my home page: http://welcome.to/SetSoft or
http://www.geocities.com/SiliconValley/Vista/6552/
Alternative e-mail: se...@co... se...@ie... 
Address: Curapaligue 2124, Caseros, 3 de Febrero
Buenos Aires, (1678), ARGENTINA Phone: +(5411) 4759 0013

Re: latest build 5.5 build 1162 problems & questions

[Arnold W. <awi...@te...>]

Salvador Eduardo Tropea wrote:
> Arnold Wiegert wrote:
> 

>>
>> I have done in both tvision first and then setedit:
>>
>> 'make clean'
>> ./configure --debug
>> make
>>
>> then in setedit
>> make install
> 
> 
> 
> Do you remmember if the make command recompiled all? Try doing "make 
> clean" to force a recompilation.
> 
I did use make clean first to ensure a recompile and it did - it took a 
bit of time ;-)
>> it also hangs, but when I kill it from another window, I do get the 
>> gdb window and can do a backtrace, but I have not been able to cut & 
>> paste again from that window - I know I did it last week - but this 
>> week a thick "fog" has descended and I can't get it to work.
> 
> 
> 
> Strange, to what program are you trying to copy the text?
I tried to cut&paste to kedit and any other editor I could find, 
including setedit.

I've tried various ways of marking and copying the text to the Windows 
clipboard, but not luck at all - and yet I know I did it at least once.

Arnold

Re: latest build 5.5 build 1162 problems & questions

[Salvador E. T. <sal...@in...>]

Arnold Wiegert wrote:

>
>> Please double check you are running a copy with debug info. Also 
>> check that all the editor was recompiled when you selected the debug 
>> info. This traceback is showing "execDialog" for two addresses quite 
>> different.
>> The other strange thing is that this traceback is too small, doesn't 
>> look like a traceback of the editor loading the desktop file.
>
>
> I have done in both tvision first and then setedit:
>
> 'make clean'
> ./configure --debug
> make
>
> then in setedit
> make install


Do you remmember if the make command recompiled all? Try doing "make 
clean" to force a recompilation.

>
> I still get the same behavior. When I have no tcedit.dst file, it 
> hangs. If I copy an old one or rename an old dst file to tcedit.dst it 
> runs and tells me it is build 1162.
>
> When I run it under ddd, it also tells me there is no debug info - so 
> now I'm confused.
>
> If I run the program in makes i.e.
> cd makes
> editor.exe -d 3
>
> it also hangs, but when I kill it from another window, I do get the 
> gdb window and can do a backtrace, but I have not been able to cut & 
> paste again from that window - I know I did it last week - but this 
> week a thick "fog" has descended and I can't get it to work.


Strange, to what program are you trying to copy the text?

Regards, SET

-- 
Salvador Eduardo Tropea (SET). (Electronics Engineer)
Visit my home page: http://welcome.to/SetSoft or
http://www.geocities.com/SiliconValley/Vista/6552/
Alternative e-mail: se...@co... se...@ie... 
Address: Curapaligue 2124, Caseros, 3 de Febrero
Buenos Aires, (1678), ARGENTINA Phone: +(5411) 4759 0013

Re: latest build 5.5 build 1162 problems & questions

[Arnold W. <awi...@te...>]

Salvador Eduardo Tropea wrote:
> [snip]
> 
> That's a very strange calling sequence.
> It looks like the debug info isn't enough because you didn't get source 
> file and line number information. But at least you got the function names.
> The CLY_YieldProcessor() call (calling usleep and nanosleep) looks like 
> the "idle" loop when the library is waiting for input.
> The strange part is that it seems to be called from execDialog() ... I 
> tried to reproduce such an interruption and I got "TSetEditorApp::idle" 
> there. So I think you are mixing binaries. The TV calls seems to be OK, 
> but the functions inside the editor doesn't have much sense.
> Please double check you are running a copy with debug info. Also check 
> that all the editor was recompiled when you selected the debug info. 
> This traceback is showing "execDialog" for two addresses quite different.
> The other strange thing is that this traceback is too small, doesn't 
> look like a traceback of the editor loading the desktop file.

I have done in both tvision first and then setedit:

'make clean'
./configure --debug
make

then in setedit
make install

I still get the same behavior. When I have no tcedit.dst file, it hangs. 
If I copy an old one or rename an old dst file to tcedit.dst it runs and 
tells me it is build 1162.

When I run it under ddd, it also tells me there is no debug info - so 
now I'm confused.

If I run the program in makes i.e.
cd makes
editor.exe -d 3

it also hangs, but when I kill it from another window, I do get the gdb 
window and can do a backtrace, but I have not been able to cut & paste 
again from that window - I know I did it last week - but this week a 
thick "fog" has descended and I can't get it to work.

I rebooted and even as root I cannot cut-n-paste.

One thing, though, it does look different, but too much to accurately 
type by hand.

Arnold

Re: Can't use tags with SETEDIT's code

[Salvador E. T. <sal...@in...>]

Thiago F.G. Albuquerque wrote:

> I was trying to use ctags to browse SETEDIT's code, but it didn't 
> work. I generated the 'tags' file with the command line:
>
> ctags -R --fields=+i+l+m+z `find . -name "*.cc" -or -name "*.h"`
>
> Then I started the editor, put the cursor under a function name and 
> pressed ^F2.
>
> The editor "crunches" the HD for some seconds, and then exits with 
> status code 255. It doesn't crash, it doesn't core dump: it simply exits.


Are you sure it doesn't crash? I can't find any explicit exit there.

> The file 'tags' is 1.2M in size and I have 256Mb of memory.

Mine is 1.6M (automagically generated from the project, it includes perl 
and other files) and I always use it in combination with the TV one, 
which is 1.2M here. So I'm using 2.8M of tags. They seem to consume an 
ammount of memory similar to the file size.
Can you send me the problematic file? I think the problem could be in 
the parser, may be some difference with the ctags program.
If you do it using email I'll suggest you to compress it using bzip2 and 
using some silly extension like "piz", the administrators of our e-mail 
system are dummies and filters anything.

SET

-- 
Salvador Eduardo Tropea (SET). (Electronics Engineer)
Visit my home page: http://welcome.to/SetSoft or
http://www.geocities.com/SiliconValley/Vista/6552/
Alternative e-mail: se...@co... se...@ie... 
Address: Curapaligue 2124, Caseros, 3 de Febrero
Buenos Aires, (1678), ARGENTINA Phone: +(5411) 4759 0013

Re: latest build 5.5 build 1162 problems & questions

[Salvador E. T. <sal...@in...>]

Arnold Wiegert wrote:

> Wishing you all a very happy Christmas.


:-) Same for all.

> Salvador Eduardo Tropea wrote:
>
>
>>
>> Try compiling with debug info both (TV and the editor). Both 
>> configure scripts now have a "--debug" option.
>> Then run the editor under gdb or start the editor with the "-d 3" 
>> command line option.
>> Run the binary from the "setedit/makes" directory, the editor.exe file.
>> Then when the editor hangs kill it. This will show the point where 
>> you stopped it.
>> If you are using gdb the "backtrace" command will show the needed 
>> information. If you use "-d 3" the editor will try to start gdb and 
>> execute backtrace.
>>
> I've finally got a few minutes to send you the results of my tests.
>
> OK. I've recompiled with the debug info and have run it again, 
> although I can only get what I would expect from your comments above 
> when using "-d 3 " if I run makes/editor.exe not the one as installed 
> - perhaps I'm going about it all wrong, though?
>
> here is the traceback when it hangs:

[snip]

That's a very strange calling sequence.
It looks like the debug info isn't enough because you didn't get source 
file and line number information. But at least you got the function names.
The CLY_YieldProcessor() call (calling usleep and nanosleep) looks like 
the "idle" loop when the library is waiting for input.
The strange part is that it seems to be called from execDialog() ... I 
tried to reproduce such an interruption and I got "TSetEditorApp::idle" 
there. So I think you are mixing binaries. The TV calls seems to be OK, 
but the functions inside the editor doesn't have much sense.
Please double check you are running a copy with debug info. Also check 
that all the editor was recompiled when you selected the debug info. 
This traceback is showing "execDialog" for two addresses quite different.
The other strange thing is that this traceback is too small, doesn't 
look like a traceback of the editor loading the desktop file.

SET

-- 
Salvador Eduardo Tropea (SET). (Electronics Engineer)
Visit my home page: http://welcome.to/SetSoft or
http://www.geocities.com/SiliconValley/Vista/6552/
Alternative e-mail: se...@co... se...@ie... 
Address: Curapaligue 2124, Caseros, 3 de Febrero
Buenos Aires, (1678), ARGENTINA Phone: +(5411) 4759 0013

Re: latest build 5.5 build 1162 problems & questions

[Arnold W. <awi...@te...>]

Wishing you all a very happy Christmas.

Salvador Eduardo Tropea wrote:


> 
> Try compiling with debug info both (TV and the editor). Both configure 
> scripts now have a "--debug" option.
> Then run the editor under gdb or start the editor with the "-d 3" 
> command line option.
> Run the binary from the "setedit/makes" directory, the editor.exe file.
> Then when the editor hangs kill it. This will show the point where you 
> stopped it.
> If you are using gdb the "backtrace" command will show the needed 
> information. If you use "-d 3" the editor will try to start gdb and 
> execute backtrace.
> 
I've finally got a few minutes to send you the results of my tests.

OK. I've recompiled with the debug info and have run it again, although 
I can only get what I would expect from your comments above when using 
"-d 3 " if I run makes/editor.exe not the one as installed - perhaps I'm 
going about it all wrong, though?

here is the traceback when it hangs:
-------------------------
Program received signal SIGTERM, Terminated.
0x402cb50b in nanosleep () from /lib/tls/libc.so.6
(gdb) backtrace
#0  0x402cb50b in nanosleep () from /lib/tls/libc.so.6
#1  0x402f4c4b in usleep () from /lib/tls/libc.so.6
#2  0x40069033 in CLY_YieldProcessor () from /usr/lib/librhtv.so.2.0.4
#3  0x08088350 in execDialog ()
#4  0x400a8c3b in TProgram::getEvent () from /usr/lib/librhtv.so.2.0.4
#5  0x0808603e in execDialog ()
#6  0x4009b92c in TGroup::execute () from /usr/lib/librhtv.so.2.0.4
#7  0x400a93f1 in TProgram::run () from /usr/lib/librhtv.so.2.0.4
#8  0x0808a762 in execDialog ()
#9  0x402337f8 in __libc_start_main () from /lib/tls/libc.so.6
#10 0x40355edc in ?? () from /lib/tls/libc.so.6
(gdb)


After thinking some more, the problem seems to have been that the old 
file was owned by root. Changing it to have the user own it, got the 
editor running - BUT, deleting the old desk top file altogether made it 
hang again - in the same place:

Program received signal SIGTERM, Terminated.
0x402cb50b in nanosleep () from /lib/tls/libc.so.6
(gdb) backtrace
#0  0x402cb50b in nanosleep () from /lib/tls/libc.so.6
#1  0x402f4c4b in usleep () from /lib/tls/libc.so.6
#2  0x40069033 in CLY_YieldProcessor () from /usr/lib/librhtv.so.2.0.4
#3  0x08088350 in execDialog ()
#4  0x400a8c3b in TProgram::getEvent () from /usr/lib/librhtv.so.2.0.4
#5  0x0808603e in execDialog ()
#6  0x4009b92c in TGroup::execute () from /usr/lib/librhtv.so.2.0.4
#7  0x400a93f1 in TProgram::run () from /usr/lib/librhtv.so.2.0.4
#8  0x0808a762 in execDialog ()
#9  0x402337f8 in __libc_start_main () from /lib/tls/libc.so.6
#10 0x40355edc in ?? () from /lib/tls/libc.so.6
(gdb)

===============
Arnold

Can't use tags with SETEDIT's code

[Thiago F.G. A. <tf...@te...>]

I was trying to use ctags to browse SETEDIT's code, but it didn't work. I 
generated the 'tags' file with the command line:

ctags -R --fields=+i+l+m+z `find . -name "*.cc" -or -name "*.h"`

Then I started the editor, put the cursor under a function name and 
pressed ^F2.

The editor "crunches" the HD for some seconds, and then exits with status 
code 255. It doesn't crash, it doesn't core dump: it simply exits.

The file 'tags' is 1.2M in size and I have 256Mb of memory.

[],
Thiago

-- 
()  ascii ribbon campaign - against html mail 
/\                        - against microsoft attachments

Re: latest build 5.5 build 1162 problems & questions

[Salvador E. T. <sal...@in...>]

Arnold Wiegert wrote:

> Hello, Salvador,
>
> just got the latest snap shot from sourceforge and 'make'd it under 
> linux.
>
> One of the things I remember from an earlier version - somehow I don't 
> think my message got through last time because of problems at my end 
> and I forgot to verify it had gone out.
>
> After building the latest build,  it hangs with a blank window (except 
> for title bar with version info) when I run it unless there is an 
> existing, even if old tcedit.dst file in the current directory. I have 
> to kill it from another window to recover.
>
> Sometimes it tells me afterwards to contact you at se...@ie...
>
> Seems I tried that the last time and it bounced eventually, but I 
> forgot to follow up. Now it is back ;-)
>
>
> If I copy some old tcedit.dst file to the current directory, set 
> starts up ok, although it complains about a missing help file.


Try compiling with debug info both (TV and the editor). Both configure 
scripts now have a "--debug" option.
Then run the editor under gdb or start the editor with the "-d 3" 
command line option.
Run the binary from the "setedit/makes" directory, the editor.exe file.
Then when the editor hangs kill it. This will show the point where you 
stopped it.
If you are using gdb the "backtrace" command will show the needed 
information. If you use "-d 3" the editor will try to start gdb and 
execute backtrace.

> ========================================
> Questions:
>
> Various messages and the make script have mentioned program/utilities 
> that I don't seem to be able to find on my Debian/MEPIS system:
>
> symify

Where?

> makeinfo

I see you already found it.

> migdb

That's the debug support:
http://sourceforge.net/projects/libmigdb


>
> ===========================================
>
> If I can supply further information, please let me know.
>
> Arnold
>
> PS: still have not had a chance to work some more on my Windows 
> version - the UNC stuff still does not work as it should
>
Ok, I'll try to revisit this issue after the holidays.

In another mail you wrote:

    it looks like the problem is one of file permissions.

    If I change to root and make the file tcedit.dst writable by all, it
    all works as expected, except that it still complains about not
    finding the help file.

1) I think you didn't have makeinfo by this time and hence is quite 
logic that the editor fails to find the help file.
2) The editor shouldn't hang if the desktop file isn't 
readable/writable. It can say some nonsense like "wrong desktop file", 
but not hang. Please try to get a backtrace to find where is hanging.


    I had done all the work in building the editor as non-root - does
    that make a difference? i.e. must I be root to build the programs?

Nope. In fact is safier if you compile programs as a regular user and 
only use root to install them.
I think the problem was like this:

1) You compiled as a regular user.
2) To install you switched to superuser (root).
3) After installing you tried the new binary without switching back to 
the regular user. It created a desktop file owned by root.
4) You dropped root privileges and then started the editor again. This 
time the editor failed to open the desktop file.

If I do it in my system I get a (misleading) message: "wrong desktop file".

Regards, SET

-- 
Salvador Eduardo Tropea (SET). (Electronics Engineer)
Visit my home page: http://welcome.to/SetSoft or
http://www.geocities.com/SiliconValley/Vista/6552/
Alternative e-mail: se...@co... se...@ie... 
Address: Curapaligue 2124, Caseros, 3 de Febrero
Buenos Aires, (1678), ARGENTINA Phone: +(5411) 4759 0013

follow-up-2 - have found makeinfo

[Arnold W. <awi...@te...>]

after some googling, I've have found out where to find makeinfo and have 
installed it
Arnold

follow-up on build 1162 problems

[Arnold W. <awi...@te...>]

Hello Salvador,

it looks like the problem is one of file permissions.

If I change to root and make the file tcedit.dst writable by all, it all 
works as expected, except that it still complains about not finding the 
help file.

I had done all the work in building the editor as non-root - does that 
make a difference? i.e. must I be root to build the programs?

Arnold

latest build 5.5 build 1162 problems & questions

[Arnold W. <awi...@te...>]

Hello, Salvador,

just got the latest snap shot from sourceforge and 'make'd it under linux.

One of the things I remember from an earlier version - somehow I don't 
think my message got through last time because of problems at my end and 
I forgot to verify it had gone out.

After building the latest build,  it hangs with a blank window (except 
for title bar with version info) when I run it unless there is an 
existing, even if old tcedit.dst file in the current directory. I have 
to kill it from another window to recover.

Sometimes it tells me afterwards to contact you at se...@ie...

Seems I tried that the last time and it bounced eventually, but I forgot 
to follow up. Now it is back ;-)


If I copy some old tcedit.dst file to the current directory, set starts 
up ok, although it complains about a missing help file.


========================================
Questions:

Various messages and the make script have mentioned program/utilities 
that I don't seem to be able to find on my Debian/MEPIS system:

symify

makeinfo

migdb

===========================================

If I can supply further information, please let me know.

Arnold

PS: still have not had a chance to work some more on my Windows version 
- the UNC stuff still does not work as it should

Re: Currently Assigned to

[Thiago F.G. A. <tf...@te...>]

On Mon, 13 Dec 2004, Salvador Eduardo Tropea wrote:

> Thiago F.G. Albuquerque wrote:
>
>> In the key binding dialog, when you assign a key that was already assigned 
>> to something else, the editor warns you. But it doesn't tell you which 
>> stuff it was previously assigned to. That's something I miss. I'd like to 
>> have one extra field (widget?) in the "Sequence of keys" window with the 
>> caption "currently assigned to:"  that would display to what the currently 
>> selected item in the list box is assigned to.
>> 
>> I imagine this would involve doing a reverse lookup in the key binding 
>> table (if such thing exists).
>
> I implemented it, is in the CVS. Please try it and tell me if it works as you 
> describe.
> Currently what I'm doing is to simulate the "addition" process every time the 
> sequence is edited. The result of the simulated addition is displayed at the 
> top of the dialog.

Sorry for the delay.

It is working, but I have some remarks.

1) The "assigned to" field doesn't show the command assigned to the key 
currently selected it the list box: it shows the command assigned to the 
last key pressed. So, if you change the selection in the edit box (in 
case this is a multi-key assignment), it still shows the command assigned 
to the other key.

But, now that I am thinking about it, in the case of multi-key 
assignments, this wouldn't make sense anyway. Because what the user would 
want to know in a case like this is: "to what command this *key sequence* 
is assigned?" and not "to what each individual key of the sequence is 
assigned?"

2) Cosmetic stuff: maybe the "assigned to" label should be below the boxes 
(edit box and radio group) or just below the edit box.

[],
Thiago

-- 
()  ascii ribbon campaign - against html mail 
/\                        - against microsoft attachments

Re: TVision: build errors

[Thiago F.G. A. <tf...@te...>]

Everything is ok. Thank you.

Thiago

On Wed, 15 Dec 2004, Salvador Eduardo Tropea wrote:

> Thiago F.G. Albuquerque wrote:
>
>> This is the last snapshot.
>> 
>> ../classes/tcluster.cc: In method `const char *TCluster::getItemText
>> (int)':
>> ../classes/tcluster.cc:97: no matching function for call to
>> `TStringCollection::atI (ccIndex &)'
>> make[1]: *** [obj/tcluster.o] Error 1
>> make[1]: Leaving directory `/home/thiago/packages/set/tvision/makes'
>> make: *** [static-lib] Error 2
>
> Now is fixed and the snapshot contains updated.
>
> SET

-- 
()  ascii ribbon campaign - against html mail 
/\                        - against microsoft attachments

Re: Adding support for Python list of functions

[Grzegorz A. H. <gr...@ti...>]

On 2004-11-30, Salvador Eduardo Tropea <sal...@in...> wrote:
> >Cool. I guess you are still preparing some release and this could
> >get in after that?
> >
> I already tagged the CVS and changed version to 0.5.5 ;-)

Thanks. In the future I might give you another patch to improve
the detection support for this other C++ like mode. In the meantime
you might want to apply this nitpick cleanup patch.

Index: ppython.cc
===================================================================
RCS file: /cvsroot/setedit/setedit/mainsrc/ppython.cc,v
retrieving revision 1.1
diff -u -r1.1 ppython.cc
--- ppython.cc  29 Nov 2004 18:10:59 -0000      1.1
+++ ppython.cc  16 Dec 2004 09:15:34 -0000
@@ -6,7 +6,7 @@
 
   Description:
   Parses a .py file looking for function and class definitions.
-  It can be compiled as an standalone program by defining STANDALONE.
+  It can be compiled as an standalone program by defining STANDALONE_TEST.
 
 ***************************************************************************/
 
@@ -19,8 +19,6 @@
 #include <bufun.h>
 #include <assert.h>
 
-#define STANDALONE
-
 #define STATE_NOTHING               0
 #define STATE_PARAGRAPH_QUOTE       1
 #define STATE_CLASS                 2

Re: TVision: build errors

[Salvador E. T. <sal...@in...>]

Thiago F.G. Albuquerque wrote:

> This is the last snapshot.
>
> ../classes/tcluster.cc: In method `const char *TCluster::getItemText
> (int)':
> ../classes/tcluster.cc:97: no matching function for call to
> `TStringCollection::atI (ccIndex &)'
> make[1]: *** [obj/tcluster.o] Error 1
> make[1]: Leaving directory `/home/thiago/packages/set/tvision/makes'
> make: *** [static-lib] Error 2

Now is fixed and the snapshot contains updated.

SET

-- 
Salvador Eduardo Tropea (SET). (Electronics Engineer)
Visit my home page: http://welcome.to/SetSoft or
http://www.geocities.com/SiliconValley/Vista/6552/
Alternative e-mail: se...@co... se...@ie... 
Address: Curapaligue 2124, Caseros, 3 de Febrero
Buenos Aires, (1678), ARGENTINA Phone: +(5411) 4759 0013

TVision: build errors

[Thiago F.G. A. <tf...@te...>]

This is the last snapshot.

../classes/tcluster.cc: In method `const char *TCluster::getItemText
(int)':
../classes/tcluster.cc:97: no matching function for call to
`TStringCollection::atI (ccIndex &)'
make[1]: *** [obj/tcluster.o] Error 1
make[1]: Leaving directory `/home/thiago/packages/set/tvision/makes'
make: *** [static-lib] Error 2

Thiago

-- 
()  ascii ribbon campaign - against html mail 
/\                        - against microsoft attachments

Re: Tvision's configure can't find install

[Salvador E. T. <sal...@in...>]

Thiago F.G. Albuquerque wrote:

> On Mon, 13 Dec 2004, Salvador Eduardo Tropea wrote:
>
>> I don't think that's the reason.
>> You have to send me errormsg.txt it contains the errors from the tests=
=2E
>> Last time I tried with djgpp the test worked. Are you sure that GNU=20
>> install is the first install command in your path?
>> The test runs "install --version > file" and checks if the output=20
>> contains "Free Software Foundation"
>
>
> Ok, here it goes. I am sending you the output of configure,=20
> errormsg.txt and the output of install --version.
>
>install (GNU fileutils) 4.0
> =20
>
That's strange, the Linux version that I have says:

$ install --version
install (fileutils) 4.1
Escrito por David MacKenzie.

Copyright (C) 2001 Free Software Foundation, Inc.
Esto es software libre; vea el c=F3digo fuente para las condiciones de co=
pia.
No hay NINGUNA garant=EDa; ni siquiera de COMERCIABILIDAD o IDONEIDAD PAR=
A UN
FIN DETERMINADO.

Anyways, I modified the test to accept any "install.exe" in the path,=20
should be in the CVS.

SET

--=20
Salvador Eduardo Tropea (SET). (Electronics Engineer)
Visit my home page: http://welcome.to/SetSoft or
http://www.geocities.com/SiliconValley/Vista/6552/
Alternative e-mail: se...@co... se...@ie...=20
Address: Curapaligue 2124, Caseros, 3 de Febrero
Buenos Aires, (1678), ARGENTINA Phone: +(5411) 4759 0013

Re: Currently Assigned to

[Salvador E. T. <sal...@in...>]

Thiago F.G. Albuquerque wrote:

> On Mon, 13 Dec 2004, Salvador Eduardo Tropea wrote:
>
>> I implemented it, is in the CVS. Please try it and tell me if it 
>> works as you describe.
>
> [...]
>
> I can't use CVS (the conection times out every time). Can you make me 
> a snapshot?

Done, I updated the snapshots.

SET

-- 
Salvador Eduardo Tropea (SET). (Electronics Engineer)
Visit my home page: http://welcome.to/SetSoft or
http://www.geocities.com/SiliconValley/Vista/6552/
Alternative e-mail: se...@co... se...@ie... 
Address: Curapaligue 2124, Caseros, 3 de Febrero
Buenos Aires, (1678), ARGENTINA Phone: +(5411) 4759 0013

Re: Tvision's configure can't find install

[Thiago F.G. A. <tf...@te...>]

On Mon, 13 Dec 2004, Salvador Eduardo Tropea wrote:

> I don't think that's the reason.
> You have to send me errormsg.txt it contains the errors from the tests.
> Last time I tried with djgpp the test worked. Are you sure that GNU install 
> is the first install command in your path?
> The test runs "install --version > file" and checks if the output contains 
> "Free Software Foundation"

Ok, here it goes. I am sending you the output of configure, errormsg.txt 
and the output of install --version.

[],
Thiago

Re: Currently Assigned to

[Thiago F.G. A. <tf...@te...>]

On Mon, 13 Dec 2004, Salvador Eduardo Tropea wrote:

> I implemented it, is in the CVS. Please try it and tell me if it works as you 
> describe.
[...]

I can't use CVS (the conection times out every time). Can you make me a 
snapshot?

[],
Thiago

Re: Currently Assigned to

[Salvador E. T. <sal...@in...>]

Thiago F.G. Albuquerque wrote:

> In the key binding dialog, when you assign a key that was already 
> assigned to something else, the editor warns you. But it doesn't tell 
> you which stuff it was previously assigned to. That's something I 
> miss. I'd like to have one extra field (widget?) in the "Sequence of 
> keys" window with the caption "currently assigned to:"  that would 
> display to what the currently selected item in the list box is 
> assigned to.
>
> I imagine this would involve doing a reverse lookup in the key binding 
> table (if such thing exists).

I implemented it, is in the CVS. Please try it and tell me if it works 
as you describe.
Currently what I'm doing is to simulate the "addition" process every 
time the sequence is edited. The result of the simulated addition is 
displayed at the top of the dialog.

SET

-- 
Salvador Eduardo Tropea (SET). (Electronics Engineer)
Visit my home page: http://welcome.to/SetSoft or
http://www.geocities.com/SiliconValley/Vista/6552/
Alternative e-mail: se...@co... se...@ie... 
Address: Curapaligue 2124, Caseros, 3 de Febrero
Buenos Aires, (1678), ARGENTINA Phone: +(5411) 4759 0013

Re: Tvision's configure can't find install

[Salvador E. T. <sal...@in...>]

Thiago F.G. Albuquerque wrote:

> Hi,
>
> I was trying to build TV 2.0.3 with DJGPP, but the configure script 
> can't find 'install'. I am sending the output attached.
>
> [],
> Thiago
>
>
> PS: I am using the UNIX .tar.gz instead of the DJGPP .zip. Does it 
> have anything to do with it?


I don't think that's the reason.
You have to send me errormsg.txt it contains the errors from the tests.
Last time I tried with djgpp the test worked. Are you sure that GNU 
install is the first install command in your path?
The test runs "install --version > file" and checks if the output 
contains "Free Software Foundation"

SET

-- 
Salvador Eduardo Tropea (SET). (Electronics Engineer)
Visit my home page: http://welcome.to/SetSoft or
http://www.geocities.com/SiliconValley/Vista/6552/
Alternative e-mail: se...@co... se...@ie... 
Address: Curapaligue 2124, Caseros, 3 de Febrero
Buenos Aires, (1678), ARGENTINA Phone: +(5411) 4759 0013

Key bindings in menubind.smn

[Thiago F.G. A. <tf...@te...>]

Currently, the key bindings are configured in two places in SETEDIT: in the key 
assignment dialog and in menubind.smn. menubind.smn has precedence over the key 
assignment dialog. So, if you change a keybinding using the key assignment 
dialog and it was assigned by menubind.smn, it simply doesn't work. The user 
gets puzzled, until he figures out that he has to edit menubind.smn. I see this 
as a problem: this is confusing, this is not user-friendly.

So, how the other editors do it? I think they do it like this: the key names in 
the menu items are dinamically generated, using some sort of reverse-lookup 
function like the one I mentioned in the other email [Currently Assigned to].

[],
Thiago

-- 
()  ascii ribbon campaign - against html mail
/\                        - against microsoft attachments

Currently Assigned to

[Thiago F.G. A. <tf...@te...>]

Hi,

In the key binding dialog, when you assign a key that was already assigned to 
something else, the editor warns you. But it doesn't tell you which stuff it 
was previously assigned to. That's something I miss. I'd like to have one extra 
field (widget?) in the "Sequence of keys" window with the caption "currently 
assigned to:"  that would display to what the currently selected item in the 
list box is assigned to.

I imagine this would involve doing a reverse lookup in the key binding table 
(if such thing exists).

[],
Thiago

-- 
()  ascii ribbon campaign - against html mail
/\                        - against microsoft attachments