Editing the context causes save errors?

[John DeHart]

Anyone have an issues where by editing the context will result in the
following error:

****************************************************
I recovered from the following error:
invalid index
in the Save_action
****************************************************

When I create complex models this error starts showing up anytime I
edit the context. I then have to reopen the model edit the context and
save again to continue.

[FD]

This means that an error has been caught when doing a Save
operation. I assume it comes up when you try to save the model. Does
everything run smoothly when you close the context? Is the problem
reproducible, if so can you help up reproduce it? Thanks.
FD

[John DeHart]

The model will run once then on the save I recieve the error. I have
to reopen the model perform any changes then save the model before the
next run. This seems to only happen when I add variables to the
context menu. If I just make a model with out context edits I can save
and run any number of times.

[FD]

I was not able to reproduce the error on the latest version of
ScicosLab. If you have a systematic way to reproduce it on a model
that you can share with us, then send it to sci...@inria.fr and I will
check. If your model is confidential, then copy the file
macros/auto/scicos.sci
and edit as follows:
comment line 739 and uncomment line 740
comment line 840 and uncomment line 841

Then load the edited scicos.sci (by exec or getf) before opening
scicos and recreating the error. Now the scicos crashes but it gives
you more detailed information about where the "invalid index" error
occurs (line number and function name). Send us then the information.
Thanks.
FD

[John DeHart]

On Jan 13, 8:20 am, FD <Francois.Delebec...@inria.fr> wrote:

> On 11 jan, 14:27, JohnDeHart<j...@enginetix.com> wrote:
>
>
>
>
>
>
>
>
>
> > On Jan 9, 6:00 am, FD <Francois.Delebec...@inria.fr> wrote:
>

> you more detailed information about where the "invalidindex" error

> occurs (line number and function name). Send us then the information.
> Thanks.
> FD

I still have this issue but if I recompile the model in scicos I can
then save the document. FYI