Error: can't read debugstate: no such variable

classic Classic list List threaded Threaded
6 messages Options
Perry Hargrave Perry Hargrave
Reply | Threaded
Open this post in threaded view
|

Error: can't read debugstate: no such variable

i've x86 (core duo) with arch linux.

i've just built scilab-se-build-4_21422 (a build of 4.1.2 wouldn't work
at all) with these options:

./configure --prefix=/usr --with-gfortran --with-tk --with-pvm
--without-ocaml --without-java --with-gtk2

It compiles with no errors. and runs the command window and all demos
I've seen OK.

When I try to enter the scipad i get this error (when scipad is started
the first time) in the command window:

shared archive loaded
Link done
 !--error 999
TCL_EvalStr, tab stop "-34.38852765205531" is not at a positive distance
at line 1
at line      73 of function scipad called by :
scipad();(1)
while executing a callback



and the editor doesnt start at all. usually it will start the second
time I try.

Once it starts I keep getting and error about 'debugstate' not being
defined, and this renders SciPad useless (everytime I click in the text
field to edit the error occurs).

some googling turned up this:
http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/scilab/modules/scipad/tcl/db_states.tcl?view=markup&pathrev=18170

but I'm not sure how to proceed from here. Any suggestions?

thanks.


Francois Vogel-2 Francois Vogel-2
Reply | Threaded
Open this post in threaded view
|

Re: Error: can't read debugstate: no such variable

You have a problem with something Tcl I think related to setting tabs
in text widgets on your platform, not with Scipad.

You must fix the first error, not the problem with debugstate, which
is only a ocnsequence of the first TCL_EvalStr error.

Is it the same with the binary version? Perhaps what you see is due to
--with-gtk2

> It compiles with no errors. and runs the command window and all demos
> I've seen OK.

Did you try the Tcl demos? Try them.

Francois




Perry Hargrave said on 12/04/2008 01:03:

> i've x86 (core duo) with arch linux.
>
> i've just built scilab-se-build-4_21422 (a build of 4.1.2 wouldn't work
> at all) with these options:
>
> ./configure --prefix=/usr --with-gfortran --with-tk --with-pvm
> --without-ocaml --without-java --with-gtk2
>
> It compiles with no errors. and runs the command window and all demos
> I've seen OK.
>
> When I try to enter the scipad i get this error (when scipad is started
> the first time) in the command window:
>
> shared archive loaded
> Link done
> !--error 999
> TCL_EvalStr, tab stop "-34.38852765205531" is not at a positive distance
> at line 1
> at line      73 of function scipad called by : scipad();(1)
> while executing a callback
>
>
>
> and the editor doesnt start at all. usually it will start the second
> time I try.
>
> Once it starts I keep getting and error about 'debugstate' not being
> defined, and this renders SciPad useless (everytime I click in the text
> field to edit the error occurs).
>
> some googling turned up this:
> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/scilab/modules/scipad/tcl/db_states.tcl?view=markup&pathrev=18170 
>
>
> but I'm not sure how to proceed from here. Any suggestions?
>
> thanks.
>
>
>

Enrico Segre Enrico Segre
Reply | Threaded
Open this post in threaded view
|

Re: Error: can't read debugstate: no such variable

In reply to this post by Perry Hargrave
[resending - listserver down during the weekend?]

> i've x86 (core duo) with arch linux.
>
> i've just built scilab-se-build-4_21422 (a build of 4.1.2 wouldn't work
> at all) with these options:

I confirm that I can build BUILD4 on a multicore x64 system. I got
r24233, for the record. The only problem known to me specific of x64,
which is not in x32, is bug 2394/2448.

I can reproduce your error though, if I tamper with the file
~/.Scilab/scilab-4.1.2/.SciPadPreferences.tcl and write there an invalid
number for the entry tabsizeinchars (say -5). This causes a complaint in
the scilab shell analogous to yours (the decimal number in the error
message is slightly different, but I think this is a result of the
particular screen metrics), an apparent success in bringing up the main
editor window, but a tk_messageBox complaining about debugstate all the
time subsequently, and an unusable scipad. This story about debugstate
is understandable since the initialization of scipad couldn't complete
as it should have in first instance, and doesn't trouble me.

So, I suggest you to check what happened with the .SciPadPreferences.tcl
file - if you edited it just correct it (or remove it, it will be
recreated with defaults). If that is not the case please come back,
we'll see why it happens that way.

hth, Enrico




Perry Hargrave Perry Hargrave
Reply | Threaded
Open this post in threaded view
|

Re: Error: can't read debugstate: no such variable

it seems I don't have ~/.Scilab/scilab-<build>/.SciPadPreferences.tcl

'slocate SciPadPreferences' returns nothing at all.



Enrico Segre wrote:

> [resending - listserver down during the weekend?]
>
>> i've x86 (core duo) with arch linux.
>>
>> i've just built scilab-se-build-4_21422 (a build of 4.1.2 wouldn't work
>> at all) with these options:
>
> I confirm that I can build BUILD4 on a multicore x64 system. I got
> r24233, for the record. The only problem known to me specific of x64,
> which is not in x32, is bug 2394/2448.
>
> I can reproduce your error though, if I tamper with the file
> ~/.Scilab/scilab-4.1.2/.SciPadPreferences.tcl and write there an invalid
> number for the entry tabsizeinchars (say -5). This causes a complaint in
> the scilab shell analogous to yours (the decimal number in the error
> message is slightly different, but I think this is a result of the
> particular screen metrics), an apparent success in bringing up the main
> editor window, but a tk_messageBox complaining about debugstate all the
> time subsequently, and an unusable scipad. This story about debugstate
> is understandable since the initialization of scipad couldn't complete
> as it should have in first instance, and doesn't trouble me.
>
> So, I suggest you to check what happened with the .SciPadPreferences.tcl
> file - if you edited it just correct it (or remove it, it will be
> recreated with defaults). If that is not the case please come back,
> we'll see why it happens that way.
>
> hth, Enrico
>
>
>
>

Perry Hargrave Perry Hargrave
Reply | Threaded
Open this post in threaded view
|

Re: Error: can't read debugstate: no such variable

In reply to this post by Francois Vogel-2
I did try all the TCL/TK demos and all succeeded with no errors. I can even see the same function

I haven't gotten SciLab to compile at all with any combination of options that lacked the --with-gtk2 flag.

Fran├žois Vogel wrote:

> You have a problem with something Tcl I think related to setting tabs in
> text widgets on your platform, not with Scipad.
>
> You must fix the first error, not the problem with debugstate, which is
> only a ocnsequence of the first TCL_EvalStr error.
>
> Is it the same with the binary version? Perhaps what you see is due to
> --with-gtk2
>
>> It compiles with no errors. and runs the command window and all demos
>> I've seen OK.
>
> Did you try the Tcl demos? Try them.
>
> Francois
>
>
>
>
> Perry Hargrave said on 12/04/2008 01:03:
>> i've x86 (core duo) with arch linux.
>>
>> i've just built scilab-se-build-4_21422 (a build of 4.1.2 wouldn't
>> work at all) with these options:
>>
>> ./configure --prefix=/usr --with-gfortran --with-tk --with-pvm
>> --without-ocaml --without-java --with-gtk2
>>
>> It compiles with no errors. and runs the command window and all demos
>> I've seen OK.
>>
>> When I try to enter the scipad i get this error (when scipad is
>> started the first time) in the command window:
>>
>> shared archive loaded
>> Link done
>> !--error 999
>> TCL_EvalStr, tab stop "-34.38852765205531" is not at a positive
>> distance at line 1
>> at line      73 of function scipad called by : scipad();(1)
>> while executing a callback
>>
>>
>>
>> and the editor doesnt start at all. usually it will start the second
>> time I try.
>>
>> Once it starts I keep getting and error about 'debugstate' not being
>> defined, and this renders SciPad useless (everytime I click in the
>> text field to edit the error occurs).
>>
>> some googling turned up this:
>> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/scilab/modules/scipad/tcl/db_states.tcl?view=markup&pathrev=18170 
>>
>>
>> but I'm not sure how to proceed from here. Any suggestions?
>>
>> thanks.
>>
>>
>>
>

Perry Hargrave Perry Hargrave
Reply | Threaded
Open this post in threaded view
|

Re: Error: can't read debugstate: no such variable

In reply to this post by Francois Vogel-2
 > Is it the same with the binary version?

I just tried the binary 4.1.2 and it does work.

perry

Fran├žois Vogel wrote:

> You have a problem with something Tcl I think related to setting tabs in
> text widgets on your platform, not with Scipad.
>
> You must fix the first error, not the problem with debugstate, which is
> only a ocnsequence of the first TCL_EvalStr error.
>
> Is it the same with the binary version? Perhaps what you see is due to
> --with-gtk2
>
>> It compiles with no errors. and runs the command window and all demos
>> I've seen OK.
>
> Did you try the Tcl demos? Try them.
>
> Francois
>
>
>
>
> Perry Hargrave said on 12/04/2008 01:03:
>> i've x86 (core duo) with arch linux.
>>
>> i've just built scilab-se-build-4_21422 (a build of 4.1.2 wouldn't
>> work at all) with these options:
>>
>> ./configure --prefix=/usr --with-gfortran --with-tk --with-pvm
>> --without-ocaml --without-java --with-gtk2
>>
>> It compiles with no errors. and runs the command window and all demos
>> I've seen OK.
>>
>> When I try to enter the scipad i get this error (when scipad is
>> started the first time) in the command window:
>>
>> shared archive loaded
>> Link done
>> !--error 999
>> TCL_EvalStr, tab stop "-34.38852765205531" is not at a positive
>> distance at line 1
>> at line      73 of function scipad called by : scipad();(1)
>> while executing a callback
>>
>>
>>
>> and the editor doesnt start at all. usually it will start the second
>> time I try.
>>
>> Once it starts I keep getting and error about 'debugstate' not being
>> defined, and this renders SciPad useless (everytime I click in the
>> text field to edit the error occurs).
>>
>> some googling turned up this:
>> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/scilab/modules/scipad/tcl/db_states.tcl?view=markup&pathrev=18170 
>>
>>
>> but I'm not sure how to proceed from here. Any suggestions?
>>
>> thanks.
>>
>>
>>
>