setedit-users — General list about setedit

Re: Request: new command: cmcSearchAgainBackward

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

Can you update the snapshots?

[],
Thiago

On Tue, 7 Feb 2006, Salvador Eduardo Tropea wrote:

> On 04/02/06 12:34, Thiago F.G. Albuquerque wrote:
>
>> Old thread...
>> 
>> On Thu, 2 Dec 2004, Salvador Eduardo Tropea wrote:
>> 
>>> Ok, about this "UltraEdit" feature:
>>> 
>>> I added what I understood from Sergio's description ;-)
>>> Please try the CVS code and tell me if that's more or less what you want.
>>> 
>>> The new commands are called:  cmcSearchSelForward and 
>>> cmcSearchSelForward.
>>> They have menu entries in the "Search" menu, but no shortcut yet.
>>> Not yet documented because they could change.
>>> 
>>> I coded them in a way that they shouldn't interfere with ^Q+F/^L.
>>> 
>>> A lot of details remains to be adjusted like:
>>> 1) I implemented it as a case sensitive search, but it could be case 
>>> sensitive only for case sensitive languages or just don't be case 
>>> sensitive. I'm not sure what's best.
>>> 2) I'm starting from the other end of the file when the search fails. This 
>>> is silently done and as the text comes from a selection the second search 
>>> *must* be successful so I never inform things like "not found".
>> 
>> 
>> Ok, but I'd like to see something along the lines of "search wrapped, 
>> continued from [ TOP | BOTTOM ]" in the status bar.
>
>
> Ok, added.
>
>> 2) 'Whole words only' should be turned on.
>
>
> Done, but only if we took a whole word.
>
>> 3) When there is no selection, select the current word. 
>
>
> Done.
>
> I updated CVS, please test the new code.
>
> Regards, SET
>
>

-- 
Thiago F. G. Albuquerque

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

Re: Request: new command: cmcSearchAgainBackward

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

On 04/02/06 12:34, Thiago F.G. Albuquerque wrote:

> Old thread...
>
> On Thu, 2 Dec 2004, Salvador Eduardo Tropea wrote:
>
>> Ok, about this "UltraEdit" feature:
>>
>> I added what I understood from Sergio's description ;-)
>> Please try the CVS code and tell me if that's more or less what you 
>> want.
>>
>> The new commands are called:  cmcSearchSelForward and  
>> cmcSearchSelForward.
>> They have menu entries in the "Search" menu, but no shortcut yet.
>> Not yet documented because they could change.
>>
>> I coded them in a way that they shouldn't interfere with ^Q+F/^L.
>>
>> A lot of details remains to be adjusted like:
>> 1) I implemented it as a case sensitive search, but it could be case 
>> sensitive only for case sensitive languages or just don't be case 
>> sensitive. I'm not sure what's best.
>> 2) I'm starting from the other end of the file when the search fails. 
>> This is silently done and as the text comes from a selection the 
>> second search *must* be successful so I never inform things like "not 
>> found".
>
>
> Ok, but I'd like to see something along the lines of "search wrapped, 
> continued from [ TOP | BOTTOM ]" in the status bar.


Ok, added.

> 2) 'Whole words only' should be turned on.


Done, but only if we took a whole word.

> 3) When there is no selection, select the current word. 


Done.

I updated CVS, please test the new code.

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: dual debugging

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

On 07/02/06 01:33, David Chmelik wrote:

>I need to dual debug with setedit.  Some debug options in docs that Salvador
>mentioned, like "force terminal;" do not show up.  
>

"Debug|Options|Program and mode ..." It brings a dialog with an option 
called "Forced terminal, leave blank for auto"

>Actually now my 2.6.13
>kernel has no main fb driver (nvidia,) just g450, and i am not sure what the
>heck to choose in config so it recompiles in (maybe it only worked in 2.4.31.)
>  
>

Linux framebuffer is full of bugs I don't recommend using it.
You need ttys in both monitors for dual head debugging.

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

dual debugging

[David C. <dch...@nw...>]

I need to dual debug with setedit.  Some debug options in docs that Salvador
mentioned, like "force terminal;" do not show up.  Actually now my 2.6.13
kernel has no main fb driver (nvidia,) just g450, and i am not sure what the
heck to choose in config so it recompiles in (maybe it only worked in 2.4.31.)
David Melik

Re: [Linux] Bug in "Powered Grep"

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

On 06/02/06 14:08, Thiago F.G. Albuquerque wrote:

> Of course.


Thanks.

> [],
> Thiago
>
> Starting program: /home/tfga/packages/set/setedit/makes/editor.exe
>
> Program received signal SIGSEGV, Segmentation fault.
> 0x40298b77 in memcpy () from /lib/libc.so.6
> (gdb) bt
> #0  0x40298b77 in memcpy () from /lib/libc.so.6
> #1  0x080b8da8 in DynStrCat (Struct=0xbffff730, str=0xc81f7434 
> <Address 0xc81f7434 out of bounds>, len=1074527396)
>     at ../extra/dyncat.cc:49
> #2  0x080ab90c in TSOSListBoxMsg::selectItem (this=0x81c97d0, 
> item=223) at ../setedit/edmsg.cc:194


Killed ;-)
I found the code added FileInfo structures living the type field 
"blank". It looks like your system usually gets this field with 0x8000 
ored and the code interpreted it as a line with continuation ...
CVS is fixed.

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: [Linux] Bug in "Powered Grep"

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

Of course.

[],
Thiago

Starting program: /home/tfga/packages/set/setedit/makes/editor.exe

Program received signal SIGSEGV, Segmentation fault.
0x40298b77 in memcpy () from /lib/libc.so.6
(gdb) bt
#0  0x40298b77 in memcpy () from /lib/libc.so.6
#1  0x080b8da8 in DynStrCat (Struct=0xbffff730, str=0xc81f7434 <Address 0xc81f7434 out of bounds>, len=1074527396)
     at ../extra/dyncat.cc:49
#2  0x080ab90c in TSOSListBoxMsg::selectItem (this=0x81c97d0, item=223) at ../setedit/edmsg.cc:194
#3  0x080ac1b3 in TSOSListBoxMsg::selectNext (this=0x81c97d0, offset=1) at ../setedit/edmsg.cc:401
#4  0x080ad001 in EdMessageSelectNext () at ../setedit/edmsg.cc:799
#5  0x08099f8f in TSetEditorApp::handleEvent (this=0x8178ce8, event=@0xbffff7d0) at ../setedit/editmain.cc:1117
#6  0x4009ea38 in TGroup::execute () from /usr/lib/librhtv.so.2.1.0
#7  0x400ad321 in TProgram::run () from /usr/lib/librhtv.so.2.1.0
#8  0x0809e210 in main (argc=1, argv=0xbffff8e4) at ../setedit/editmain.cc:3317

> On 04/02/06 04:35, Thiago F.G. Albuquerque wrote:
>
> > Hi,
> >
> > I found a bug in "Powered Grep":
> >
> > 1. Go to the root directory of SETEDIT's source tree
> > 2. Start the editor
> > 3. Shift-F2 (Powered Grep)
> >      Pattern box: "cmcInsertKeyName"
> >      [X] Recurse in subdirs
> >      <enter>
> >
> > 4. After grep has finished, press the key bound to cmeNextMessage.
> 
> 
> I can't reproduce it on my system.
> Can you try to use gdb (standalone or from setedit) to find the point of 
> the crash?
> The traceback isn't really usefull (i.e. no line numbers)
> 
> Regards, SET

Re: [Linux] Bug in "Powered Grep"

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

On 04/02/06 04:35, Thiago F.G. Albuquerque wrote:

> Hi,
>
> I found a bug in "Powered Grep":
>
> 1. Go to the root directory of SETEDIT's source tree
> 2. Start the editor
> 3. Shift-F2 (Powered Grep)
>      Pattern box: "cmcInsertKeyName"
>      [X] Recurse in subdirs
>      <enter>
>
> 4. After grep has finished, press the key bound to cmeNextMessage.


I can't reproduce it on my system.
Can you try to use gdb (standalone or from setedit) to find the point of 
the crash?
The traceback isn't really usefull (i.e. no line numbers)

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: Request: new command: cmcSearchAgainBackward

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

Old thread...

On Thu, 2 Dec 2004, Salvador Eduardo Tropea wrote:

> Ok, about this "UltraEdit" feature:
>
> I added what I understood from Sergio's description ;-)
> Please try the CVS code and tell me if that's more or less what you want.
>
> The new commands are called:  cmcSearchSelForward and  cmcSearchSelForward.
> They have menu entries in the "Search" menu, but no shortcut yet.
> Not yet documented because they could change.
>
> I coded them in a way that they shouldn't interfere with ^Q+F/^L.
>
> A lot of details remains to be adjusted like:
> 1) I implemented it as a case sensitive search, but it could be case 
> sensitive only for case sensitive languages or just don't be case sensitive. 
> I'm not sure what's best.
> 2) I'm starting from the other end of the file when the search fails. This is 
> silently done and as the text comes from a selection the second search *must* 
> be successful so I never inform things like "not found".

Ok, but I'd like to see something along the lines of "search wrapped, 
continued from [ TOP | BOTTOM ]" in the status bar.

2) 'Whole words only' should be turned on.

3) When there is no selection, select the current word. I implemented this
with macros, but maybe this could be part of the behaviour of the command. 
In case anyone is interested:

(defmacro 'SearchSelForward'
  (progn
   (if (not (SelectionExists))   (SendCommands cmcMarkWord))

   (SendCommands cmcSearchSelForward)
  )
)

(defmacro 'SearchSelBackward'
  (progn
   (if (not (SelectionExists))   (SendCommands cmcMarkWord))

   (SendCommands cmcSearchSelBackward)
  )
)

Maybe this whole stuff can be implemented entirely in sLisp, now that we 
have the commands FindString and FindAgain.

-- 
Thiago F. G. Albuquerque

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

[Linux] Bug in "Powered Grep"

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

Hi,

I found a bug in "Powered Grep":

1. Go to the root directory of SETEDIT's source tree
2. Start the editor
3. Shift-F2 (Powered Grep)
      Pattern box: "cmcInsertKeyName"
      [X] Recurse in subdirs
      <enter>

4. After grep has finished, press the key bound to cmeNextMessage.

  Panic! got a signal, quiting
  That's from SET's editor v0.5.5
  Signal number: 11

Here's the backtrace:

#0  0x080a6ca8 in virtual function thunk (delta:-4) for TListBox::read ()
#0  0x080a6ca8 in virtual function thunk (delta:-4) for TListBox::read ()
#1  0x0809bc6d in virtual function thunk (delta:-4) for TStaticText::write ()
#2  0x0809c336 in virtual function thunk (delta:-4) for TStaticText::write ()
#3  0x0809d009 in virtual function thunk (delta:-4) for TStaticText::write ()
#4  0x0808bee7 in execDialog ()
#5  0x4008d237 in TGroup::execute () from /usr/lib/librhtv.so.2.1.0
#6  0x40097ba7 in TProgram::run () from /usr/lib/librhtv.so.2.1.0
#7  0x0809004b in execDialog ()
#8  0x42017499 in __libc_start_main () from /lib/i686/libc.so.6
#9  0x08054cc1 in ?? ()

It has happened several times since I discovered it -- in other source 
trees as well, and with other search strings. It is a highly reproducible 
bug. The crash might not occur in the first call to cmeNextMessage; it 
might be the second or the last.

Regards,
-- 
Thiago F. G. Albuquerque

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

Transparent support for GPG encrypted files

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

Hi All!

I'm adding it and I'm looking for betatesters and ideas.
The main goal is to support ASCII encrypted files (--armor option).
I'm not designing anything "ultra-secure", just a wrapper for gpg to 
avoid en/decrypting the files all the time.
The idea is that you just encrypt a text file, then you load it from the 
editor, it asks for the passphrase and loads the file decrypted. Then 
you can modify it and each time you save it the editor will use gpg to 
encrypt the file. In this way the file remains encrypted all the time 
(well, most of the time). Without it you have to manually decrypt the 
file and at the end of the edition remmember to encrypt it again.

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: [DOS] Can't assign to cmeSyntaxHelp

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

Hi, SET,

Syntax Help stopped working because I had commented the binding in 
menubind.smn and had forgotten about it (:-/ sorry about that). Still, in 
Linux 'cmeSyntaxHelp' shows in the command list and in DOS it doesn't. It 
doesn't bother me anymore, tough.

On Fri, 4 Nov 2005, Salvador Eduardo Tropea wrote:
> On 04/11/05 04:52, Thiago F.G. Albuquerque wrote:
>
>> Hi,
>> 
>> Syntax help stopped working after my last update of SETEDIT. I went to the 
>> 'Key Assignment' window and found out that ^F1 was unassigned. Then I tried 
>> to reassign it to cmeSyntaxHelp, but it doesn't show in the command list.
>> 
>> This only happens in the DOS port.
>
> I can't debug the DOS version because I don't have access to a DOS/Windows 
> machine.
> I don't know when I'll have access to such a setup.
> Double check you are using the same source code.

No, it's not the same source code. In DOS, I'm using the snapshot from

    2005/06/17

and in Linux, it's the snapshot from

    Mon Sep 12 11:35:17 PDT 2005

> Which compiler are you using for DOS?

DJGPP

[],
Thiago

PS: Sorry about the huge delay in this reply.

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

ebuild file for gentoo

[Szasz P. <sp...@sp...>]

Hi!

I'm switching over Gentoo, and I observed there is no setedit ebuild 
file on the gentoo servers (however there is one for rhide which uses 
setedit). Anyway, does anyone has an ebuild file? (I want to keep the 
system clean, that's why I would prefer building using ebuild/emerge 
then the normal make process)

Thanks in advance!

-- 
Best regards,
Szasz Pal
----------------
Space Software Studio
http://www.spacesoftwarestudio.com

Re: new DuplicateLineOrSelection

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

On 01/11/05 12:54, Thiago F.G. Albuquerque wrote:

> I wrote a new version of the macro 'DuplicateLineOrSelection'. 
> DuplicateLineOrSelection used the clipboard when you duplicated a 
> line. So, whatever you had in the clipboard before invoking the macro 
> was lost and replaced with the contents of the current line. The new 
> version uses a temporary variable instead.
>

Applied, thanks!

-- 
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: [DOS] Can't assign to cmeSyntaxHelp

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

On 04/11/05 04:52, Thiago F.G. Albuquerque wrote:

> Hi,
>
> Syntax help stopped working after my last update of SETEDIT. I went to 
> the 'Key Assignment' window and found out that ^F1 was unassigned. 
> Then I tried to reassign it to cmeSyntaxHelp, but it doesn't show in 
> the command list.
>
> This only happens in the DOS port.

I can't debug the DOS version because I don't have access to a 
DOS/Windows machine.
I don't know when I'll have access to such a setup.
Double check you are using the same source code.
Which compiler are you using for DOS?

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

[DOS] Can't assign to cmeSyntaxHelp

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

Hi,

Syntax help stopped working after my last update of SETEDIT. I went to the 
'Key Assignment' window and found out that ^F1 was unassigned. Then I 
tried to reassign it to cmeSyntaxHelp, but it doesn't show in the command 
list.

This only happens in the DOS port.

[],
-- 
Thiago F. G. Albuquerque

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

new DuplicateLineOrSelection

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

Hi,

I wrote a new version of the macro 'DuplicateLineOrSelection'. 
DuplicateLineOrSelection used the clipboard when you duplicated a line. 
So, whatever you had in the clipboard before invoking the macro was lost 
and replaced with the contents of the current line. The new version uses a 
temporary variable instead.

[],
Thiago

-------------------------------------------

(defmacro 'DuplicateLineOrSelection'
   (if (SelectionExists)  (InsertText (GetSelection))

                          (progn  ; else
                                  (SendCommands cmcPushCursorPos cmcLineStart cmcSelectOn cmcLineEnd
                                                cmcSelectOff)
                                  (= "temp" (GetSelection))
                                  (SendCommands cmcLineEnd)
                                  (InsertText (+ "\n" temp))
                                  (SendCommands cmcPopCursorPos cmcLineDown)
                          )
   )
)


-- 
Thiago F. G. Albuquerque

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

Re: Cursor problem under console in Fedora core 3/Suse 9.2

[Jan C. <Jan...@ho...>]

(on the problem of incorrect screen refresh when moving the cursor).

There is a simple workaround for this problem. Simply write macros for all 
cursor movements, so that the redraw function is called. For instance, the 
left arrow should execute the functions cmcCharLeft cmcRedraw.

Glad I can use my favorite editor again.

Greetings,

Jan Cnops

Re: 'GDB suddenly died'

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

On 15/10/05 05:58, Szasz Pal wrote:

> Hi!
>
> I'm debugging a C++ program with setedit, and I keep getting these 
> images. I'm not using optimizations (ie. no -O switch in my makefile), 
> and I added the -g switch. Sometimes debugging works, sometimes not.
>
> Setedit version: 0.5.5 revision 1185
> GDB version: GNU gdb 6.0-2mdk (Mandrake Linux)
>
> Maybe the GDB is too new or too old? WHich version do you use?
>
I'm using 6.3 and during development I used 6.1.1.

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

'GDB suddenly died'

[Szasz P. <sp...@sp...>]

Hi!

I'm debugging a C++ program with setedit, and I keep getting these 
images. I'm not using optimizations (ie. no -O switch in my makefile), 
and I added the -g switch. Sometimes debugging works, sometimes not.

Setedit version: 0.5.5 revision 1185
GDB version: GNU gdb 6.0-2mdk (Mandrake Linux)

Maybe the GDB is too new or too old? WHich version do you use?

Thanks in advance!

-- 
Best regards,
Szasz Pal
----------------
Space Software Studio
http://www.spacesoftwarestudio.com

Re: debug java application

[Szasz P. <sp...@sp...>]

Salvador Eduardo Tropea wrote:
> Does it answer your question? Ask me if you have doubts.

Yes, thank you very much. I just hope I will have time to try to do 
something with it. I work with java and I want to avoid those 
extra-over-complicated IDEs, like Sun one studio, eclipse.

Setedit uses ~1MB of memory and starts within 1 second. Eclipse needs 
40MB and starts in ~10seconds...

-- 
Best regards,
Szasz Pal
----------------
Space Software Studio
http://www.spacesoftwarestudio.com

Re: debug java application

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

On 28/09/05 15:06, Szasz Pal wrote:

> Hi!
>
> How complicated would be to add debuggin support for java programs?
> Java also has a console mode debugger, jdb, but I don't know how similar
> it is to gdb.
>
> And in general how complicated would be to make a general debugging 
> framework? Ie. SETEDIT would connect to the mid-layer and communicate 
> to it with it's own protocoll. Then the mid-layer can communicate with 
> the real debugger. The advantage would be that new debuggers could be 
> added withouth needing to recompile setedit.


Answer: not so easy and not so complicated ;-)
Currently SETEdit uses an abstraction layer called "libmigdb", that's a 
separated library (currently only static version available). So you 
could just write a library with the same API as "libmigdb" to 
communicate with the debugger.
But this isn't the best approach. The best (and a little bit more 
complicated) is to write a binary that translates "GDB/MI" commands into 
the commands used by your debugger.
The GDB/MI protocol is documented in GDB docs (some missing and 
missleading stuff, be careful!).
This protocol uses plain text (to avoid the problems associated with 
binary data) and is designed for "Machine Interaction". It means the 
format of the responses isn't designed to be human readable, but they 
are quite easy to be parsed.
This approach can be done without altering SETEdit (nor libmidbg) 
sources (or with very little changes). But the most important thing you 
get: you can use any other gdb frontend that implements GDB/MI protocol. 
I know Eclipse is one.
The "ultimate" approach is to just cooperate with the debugger team and 
help to just add GDB/MI protocol to the debugger itself.

Does it answer your question? Ask me if you have doubts.

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

debug java application

[Szasz P. <sp...@sp...>]

Hi!

How complicated would be to add debuggin support for java programs?
Java also has a console mode debugger, jdb, but I don't know how similar
it is to gdb.

And in general how complicated would be to make a general debugging 
framework? Ie. SETEDIT would connect to the mid-layer and communicate to 
it with it's own protocoll. Then the mid-layer can communicate with the 
real debugger. The advantage would be that new debuggers could be added 
withouth needing to recompile setedit.

-- 
Best regards,
Szasz Pal
----------------
Space Software Studio
http://www.spacesoftwarestudio.com

Re: Cursor problem under console in Fedora core 3/Suse 9.2

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

On 27/09/05 09:27, Jan Cnops wrote:

>Right on spot. I tried it out, and when I use a framebuffer mode 
>(vga=0x317) the problem is there; if I switch to a text mode (vga=0F05, or 
>something like that) everything is OK. Any idea whom I should report the 
>problem too?
>  
>

Not sure, try to find which framebuffer module are you using (plain VGA? 
VESA? something specific for your video board?) then try to find who is 
the maintainer (the kernel sources have a list of maintainers) and 
contact this person.
Lamentably Linux framebuffer is full of bugs, the cursor seems to be one 
of the most buggy parts in most modules :-(

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: Cursor problem under console in Fedora core 3/Suse 9.2

[Jan C. <Jan...@ho...>]

Hello,

On Mon, 26 Sep 2005, Salvador Eduardo Tropea wrote:

> On 26/09/05 13:31, Jan Cnops wrote:
> 
> >Hello,
> >
> >I have the following problem when I use setedit v0.5.4 under FC3 or Suse 
> >9.2. If I use it in a terminal window under X everything is OK, but there 
> >something wrong if I use it with the console.
> >
> >The cursor leaves a trace (the cursor here has the form of a blinking 
> >underscore): every letter (and blank, ...)  where the cursor has been is 
> >now underlined. These underlines remain as long as the text isn't 
> >scrolled, so I assume it is a problem with refreshing the screen.
> >
> >Any solution for this problem?
> >  
> >
> 
> Looks like a bug in the kernel's frambuffer code. Are you using 
> framebuffer or a real text mode?

Right on spot. I tried it out, and when I use a framebuffer mode 
(vga=0x317) the problem is there; if I switch to a text mode (vga=0F05, or 
something like that) everything is OK. Any idea whom I should report the 
problem too?

-- 
Jan Cnops

P.S. Thanks for the help. I like using setedit, so it makes a lot of 
difference.

Re: Multithreaded and forking version of former email-fmt.py

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

On 2005-09-16, Salvador Eduardo Tropea <sal...@in...> wrote:
> I tried using the "Test RunProgram Filter" macro. It took around
> a second to execute the formater and replace the whole text.

Yes, I was using the version with "-soft", which does more
processing. That highlights more the slowdown of running something
inside setedit.