[Paraview-developers] Filter/pipeline design question - any helpful hints?

[Cornelis Bockemühl]

And regarding the "more complex geometrical setup" I was not sure whether
saving the data to some simple table (CSV) might be more difficult, because
there would be one part with associated data arrays and another one
without...

Now I simply created a test object with the characteristics manually - and
I realized that things are much more easy: Exporting "cell data" to CSV
simply creates two tables - one with the data I would need to export, and a
second one that can be deleted.

Regards,
Cornelis


2016-11-06 22:41 GMT+01:00 Cornelis Bockemühl <cornelis.bockemuehl at gmail.com
>:

> Thanks a lot for that quick and precise answer - and this even in the
> middle of the Sunday!
>
> Looks like indeed the "multiblock inspector" is doing what I need, so
> "option 2" comes obviously closest to what I want, so I will certainly go
> for that solution.
>
> Is there a way to include some "color hint" already in the data file, like
> "preferred color for part 2 would be red" etc.?
>
> My guess is that it is not - because it contradicts a bit the "paraview
> philosophy" that is to my understanding: define geometry and data in the
> data files, then define the details of the view interactively.
>
> Regards, Cornelis
>
>
> 2016-11-06 16:28 GMT+01:00 Utkarsh Ayachit <utkarsh.ayachit at kitware.com>:
>
>> Cornelis,
>>
>> Option 2 is probably the easiest. As for each of the disadvantages for
>> that option you mention, here's some more info:
>>
>> > However,
>> > now I have a problem with the independent coloring of the two parts: So
>> far
>> > I did not find a way to say that part 1 should be colored according to
>> > attribute X and part 2 should be colored in black, white or whatever
>> else I
>> > would define.
>>
>> Multiple options for this:
>> 1. For part 2, remove all attribute arrays. Now when you choose to
>> color by attribute X, since X is missing on part 2, ParaView will
>> simply color by the solid color set before choosing the attribute to
>> color with.
>> 2. Use the "Multiblock Inspector" [a] to manually set the solid color
>> to use for specific part/block. When such a color is specified, the
>> attribute chosen is ignored. This does need to be set after each
>> filter application, however.
>>
>> > Another disadvantage of the second option is the fact that it generates
>> a
>> > much more complex geometrical setup.
>>
>> Can you elaborate what you mean by that?
>>
>> Utkarsh
>>
>> [a]: https://blog.kitware.com/color-multiblock-datasets/
>>
>> >
>> > 3. As a 3-rd option I could imagine that there are possibilities that I
>> did
>> > not understand yet that would make my job much easier than I thought?
>> >
>> > 4. Derive a new cut filter that takes two input ports and works on both
>> > simultaneously...(??)
>> >
>> > It is to say that I am an experienced programmer (which is why I am so
>> > quickly "starting" with paraview filters), but I am still rather new to
>> > paraview, reading a lot in the docs, trying around...
>> >
>> > Thanks for any helpful hints, considerations or tipps!
>> > Cornelis
>> >
>> >
>> >
>> > --
>> > View this message in context: http://the-unofficial-paraview
>> -developers-forum.34153.x6.nabble.com/Filter-pipeline-
>> design-question-any-helpful-hints-tp4865.html
>> > Sent from the The Unofficial ParaView Developers Forum mailing list
>> archive at Nabble.com.
>> > _______________________________________________
>> > Powered by www.kitware.com
>> >
>> > Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html
>> >
>> > Search the list archives at: http://markmail.org/search/?q=
>> Paraview-developers
>> >
>> > Follow this link to subscribe/unsubscribe:
>> > http://public.kitware.com/mailman/listinfo/paraview-developers
>>
>
>
>
> --
> Cornelis Bockemühl
> Basel, Schweiz
>



-- 
Cornelis Bockemühl
Basel, Schweiz
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20161106/863f583a/attachment.html>