[Paraview] questions

[Berk Geveci]

> Thanks for the pointers.  The reader appears to treat the utf-8
> sequence as regular characters.  Would you happen to know where the
> field names are converted read from the field names and converted to
> Qt strings?  If I can tell Qt the encoding of the string it is getting
> from the vtk objects, I think it may display the characters correctly.

In many places. There are a few ways of getting field (array) names.
One is through a class called vtkPVInformation. Another one through
vtkSMArraySelectionDomain. Another one, probably, through
vtkSMArrayListDomain. Happy digging.

-berk

> On Wed, Jun 4, 2008 at 3:02 PM, Berk Geveci <berk.geveci at kitware.com> wrote:
>> Start with vtkDataSetReader. That class points to bunch of other
>> classes. Just  a word of caution: it is not a straightforward task. I
>> would estimate that it would take someone a few months of effort to
>> fully, properly integrate unicode support to VTK and ParaView.
>>
>> -berk
>>
>> On Wed, Jun 4, 2008 at 3:59 PM, John Doe <ufnoise at gmail.com> wrote:
>>> I'd be interested in what is referred to as the legacy file format and
>>> has the header:
>>> # vtk DataFile Version 3.1
>>>
>>> Since QT has native unicode support, I'd be interested in seeing how
>>> hard it would be to allow utf-8 scalar names.  I would consider either
>>> writing vtk files in utf-8 or in escape sequences (e.g. \u03c8).
>>>
>>> I'd be happy to provide a patch or a plugin enabling such a capability.
>>>
>>> Regards,
>>>
>>> Juan
>>>
>>> On Wed, Jun 4, 2008 at 2:47 PM, Berk Geveci <berk.geveci at kitware.com> wrote:
>>>> Which file format are we talking about?
>>>>
>>>> On Wed, Jun 4, 2008 at 3:39 PM, John Doe <ufnoise at gmail.com> wrote:
>>>>> Could you please let me know which source file reads the ascii format?
>>>>>
>>>>> Regards,
>>>>>
>>>>> Juan
>>>>>
>>>>>
>>>>> On Wed, Jun 4, 2008 at 2:26 PM, Berk Geveci <berk.geveci at kitware.com> wrote:
>>>>>> On Wed, Jun 4, 2008 at 11:18 AM, John Doe <ufnoise at gmail.com> wrote:
>>>>>>> Comments inline:
>>>>>>>
>>>>>>> On Wed, Jun 4, 2008 at 7:25 AM, Berk Geveci <berk.geveci at kitware.com> wrote:
>>>>>>>>> I am starting to evaluate paraview for a project and I am curious
>>>>>>>>> about a few things:
>>>>>>>>>
>>>>>>>>> File formats:
>>>>>>>>>
>>>>>>>>> Does the vtk ascii format support unicode data set names?  If it
>>>>>>>>> doesn't, is there a format that does?
>>>>>>>>
>>>>>>>> I am afraid that you will find VTK (the underlying computing layer of
>>>>>>>> ParaView) and ParaView lacking quite a bit in supporting anything but
>>>>>>>> ASCII. Can you elaborate a bit more what you need? Do you need the
>>>>>>>> whole user interface to support unicode or just the dataset names? And
>>>>>>>> what do you mean by "dataset name"? Names of the field variables?
>>>>>>>
>>>>>>> My program uses utf-8.  I want to allow users to use symbols for their
>>>>>>> data, such as psi (u03c8) for potential, and script E (u2130) for
>>>>>>> electric field.
>>>>>>
>>>>>> I am afraid this is not supported yet.
>>>>>>
>>>>>> -berk
>>>>>>
>>>>>
>>>>
>>>
>>
>