[Scilab-users] xload mess ...

classic Classic list List threaded Threaded
4 messages Options
amonmayr amonmayr
Reply | Threaded
Open this post in threaded view
|

[Scilab-users] xload mess ...

Hi all,

I'm fighting with xload to reload saved figures, modify them (aesthetics only) and export.
There are several issues with xload that does not seem to match the documentation:

- If no graphic windows are opened, xload('figfile.scg') is creating a new one and restoring the figure size ( gcf().figure_size ), contrary to what is stated in the documentation "xload does not restore the window number, the window size nor the window dimensions."
- If there are graphic windows opened, xload('figfile.scg') adds the saved figure to the current window, usually resulting in a terrible mess, figure size being lost during the battle. The fact the the current window is modified by calling xload() is not mentioned in the documentation.
-- if h=scf() and using xload('figfile.scg',h) the figure is restored in h and the original figure size is lost (h.figure_size is used).

So here is my question: is there a way to restore a figure from a file, including its figure size, without resorting to closing all the other figures beforehand?

Cheers,

Antoine

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users
Samuel GOUGEON Samuel GOUGEON
Reply | Threaded
Open this post in threaded view
|

Re: xload mess ...

Hello Antoine,

As discussed with you there, xload() has been restored and upgraded for Scilab 6.0.2.
However, its help page was updated only after the 6.0.2 release. Its update is pending
here
, and visible in PDF there.

As it is not possible to reproduce the issue on win7 and on Fedora, it is clear that it depends
on the OS, is likely specific to Ubuntu, and very likely linked to HDF5 issues reported with Ubuntu.

However, after your recent report about the same issue on Windows 10, the confirmation
of the issue by any user working with 6.0.2 on Windows 10 would be welcome.
The test is :

So here is my question: is there a way to restore a figure from a file, including its figure size, without resorting to closing all the other figures beforehand?
With Scilab 6.0.2 on my win7 64 bits, load("c.scg") does it.

Regards
Samuel


_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users
Antoine Monmayrant Antoine Monmayrant
Reply | Threaded
Open this post in threaded view
|

Re: xload mess ...


Le 13/09/2019 à 13:38, Samuel Gougeon a écrit :
Hello Antoine,
Hello Samuel,

As discussed with you there, xload() has been restored and upgraded for Scilab 6.0.2.
However, its help page was updated only after the 6.0.2 release. Its update is pending
here
, and visible in PDF there.

As it is not possible to reproduce the issue on win7 and on Fedora, it is clear that it depends
on the OS, is likely specific to Ubuntu, and very likely linked to HDF5 issues reported with Ubuntu.

However, after your recent report about the same issue on Windows 10, the confirmation
of the issue by any user working with 6.0.2 on Windows 10 would be welcome.
The test is :
No, that's not the test!
Just xload("c.scg") does not crash scilab.
As stated in the bug report ( http://bugzilla.scilab.org/16189 ) The test is the following:
// XLOAD Segfault bug
xdel(winsid())
xload('c.scg'); // from 5.5.2
h=gcf()
xsave('c2.scg',h);
xload('c2.scg'); //bye

That is you  have to xload, save a copy and xload again the copy to crash.
I confirm this bug under another Win10 box with a colleague.
Can you try again?


Antoine



So here is my question: is there a way to restore a figure from a file, including its figure size, without resorting to closing all the other figures beforehand?
With Scilab 6.0.2 on my win7 64 bits, load("c.scg") does it.
It's not the case under Ubuntu.

Regards
Samuel


_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users
Samuel GOUGEON Samuel GOUGEON
Reply | Threaded
Open this post in threaded view
|

Re: xload mess ...

Le 16/09/2019 à 11:31, Antoine Monmayrant a écrit :


Le 13/09/2019 à 13:38, Samuel Gougeon a écrit :
Hello Antoine,
Hello Samuel,

As discussed with you there, xload() has been restored and upgraded for Scilab 6.0.2.
However, its help page was updated only after the 6.0.2 release. Its update is pending
here
, and visible in PDF there.

As it is not possible to reproduce the issue on win7 and on Fedora, it is clear that it depends
on the OS, is likely specific to Ubuntu, and very likely linked to HDF5 issues reported with Ubuntu.

However, after your recent report about the same issue on Windows 10, the confirmation
of the issue by any user working with 6.0.2 on Windows 10 would be welcome.
The test is :
No, that's not the test!
Just xload("c.scg") does not crash scilab.
As stated in the bug report ( http://bugzilla.scilab.org/16189 ) The test is the following:
// XLOAD Segfault bug
xdel(winsid())
xload('c.scg'); // from 5.5.2
h=gcf()
xsave('c2.scg',h);
xload('c2.scg'); //bye

That is you  have to xload, save a copy and xload again the copy to crash.
I confirm this bug under another Win10 box with a colleague.
Can you try again?


Done. As reported on the Bugzilla thread, the instability more likely comes from xsave()
-- actually save() --, that changes Scilab and makes it instable.



_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users