[Paraview] Status of BoxLib AMR reader

Tue Aug 30 07:51:06 EDT 2016

Thanks, Tim. I'll take a look in next few days.

Utkarsh

On Mon, Aug 29, 2016 at 4:12 PM, Gallagher, Timothy P <
tim.gallagher at gatech.edu> wrote:

> I have to throw in the towel on this one for now and get back to some
> other deadlines... but, I put what I have so far online and created a merge
> request with information about the error and an example dataset attached.
>
>
> Hopefully somebody better with VTK than I am can take a look and figure
> out what I'm missing. I tried comparing the files for Chombo against the
> original Visit source files to see if any special modifications were
> required and none are (the only changes are HDF5 related). It looks like
> the Chombo reader is from Visit 2.7.0 -- the Boxlib files I added are from
> 2.10.3, but I tried it with the Boxlib reader from Visit 2.7.0 also and
> it's the same issue.
>
>
> If anybody has a chance to take a look, I'd appreciate it. Hopefully it's
> something simple. If not, I'll try to pick it back up in a few days. I
> apologize for the hacked CMakeLists to link against the library, I will fix
> it to do it properly once everything else works!
>
> Tim
>
>
> ------------------------------
> *From:* ParaView <paraview-bounces at paraview.org> on behalf of Gallagher,
> Timothy P <tim.gallagher at gatech.edu>
> *Sent:* Monday, August 29, 2016 12:41 PM
>
> *To:* Utkarsh Ayachit
> *Cc:* paraview at paraview.org
> *Subject:* Re: [Paraview] Status of BoxLib AMR reader
>
>
> Thanks for the quick response, the case didn't match between the two files
> and that was the issue. I'm able to use the reader to select the file and
> it sees the names of the variables in it properly, but when I hit Apply it
> throws:
>
>
> ERROR: In /data4/PVBoxLibTest/paraview/VTK/Common/DataModel/vtkAMRInformation.cxx,
> line 573
> vtkAMRInformation (0x44210f0): Invalid grid description for a
> vtkUniformGrid
>
>
> I have a feeling this might get beyond my paraview knowledge to fix.
>
>
> One potential source of the problem -- what version of Visit are the other
> readers pulled from? I have the Boxlib files from the latest version, but
> looking at some of the other readers, they are from older versions.
>
> The latest date I see in the copyright for existing readers is 2013 and
> looking through the Visit Release notes, the last release in 2013 says it
> was upgraded to use VTK-6.0.
>
>
> Is it possible the bridge is based on Visit before that major VTK upgrade
> and I need to pull the reader code from a version earlier than that?
>
>
> Thanks for your help. If I can't find something I know how to fix soon, I
> can create a merge request and see if anybody else can get it the rest of
> the way.
>
>
> Tim
>
> ------------------------------
> *From:* Utkarsh Ayachit <utkarsh.ayachit at kitware.com>
> *Sent:* Monday, August 29, 2016 12:17 PM
> *To:* Gallagher, Timothy P
> *Cc:* paraview at paraview.org
> *Subject:* Re: [Paraview] Status of BoxLib AMR reader
>
> Tim,
>
> It doesn't know what the class is. Multiple things could have gone wrong:
> 1. is there call to ADD_VISIT_READER() for the new reader?
> 2. does the case match between the XML's class="" and the name specified
> in 1?
>
> If it makes it easier, create a MR on the https://gitlab.kitware.
> com/paraview/visitbridge repo with your changes, and I can do the
> necessary fixes.
> <https://gitlab.kitware.com/paraview/visitbridge>
> ParaView / VisItBridge <https://gitlab.kitware.com/paraview/visitbridge>
> gitlab.kitware.com
> Bridge between VisIt and ParaView to use VisIt readers in ParaView.
>
> Utkarsh
>
> On Mon, Aug 29, 2016 at 12:06 PM, Gallagher, Timothy P <
> tim.gallagher at gatech.edu> wrote:
>
>> Utkarash,
>>
>>
>> Thanks, I had added it to the xml file, but I didn't reconfigure (just
>> rebuilt) so it didn't get added to the list.
>>
>>
>> I'm making some progress it seems, it is picking the right reader now.
>> But it is crashing with a rather unhelpful message:
>>
>>
>> ERROR: In /data4/PVBoxLibTest/paraview/ParaViewCore/ServerImplementation/Core/vtkSIProxy.cxx,
>> line 310
>> vtkSISourceProxy (0x3e87420): Failed to create vtkVisItBoxlib2DReader.
>> Aborting for debugging purposes.
>>
>> I thought compiling with a Debug build type would give more verbose
>> messages, but it's the same.
>>
>>
>> Is there documentation or suggestions on debugging paraview? Is there a
>> way to get more verbose output on why it isn't able to create the reader?
>>
>>
>> Thanks again,
>>
>>
>> Tim
>>
>> ------------------------------
>> *From:* Utkarsh Ayachit <utkarsh.ayachit at kitware.com>
>> *Sent:* Monday, August 29, 2016 8:46 AM
>>
>> *To:* Gallagher, Timothy P
>> *Cc:* paraview at paraview.org
>> *Subject:* Re: [Paraview] Status of BoxLib AMR reader
>>
>> Tim,
>>
>> Did you also add it to visit_readers.xml [1]? The proxy definition should
>> also have a  <hints> section like so [2].
>>
>> Utkarsh
>>
>> [1] https://gitlab.kitware.com/paraview/visitbridge/blob/mas
>> ter/databases/visit_readers.xml
>> [2] https://gitlab.kitware.com/paraview/visitbridge/blob/mas
>> ter/databases/visit_readers.xml#L2562-2565
>>
>>
>>
>> On Fri, Aug 26, 2016 at 6:26 PM, Gallagher, Timothy P <
>> tim.gallagher at gatech.edu> wrote:
>>
>>> I took a stab at adding the Boxlib reader from the visit source to the
>>> VisitBridge internal readers. I managed to get everything to compile and
>>> link, Paraview built fine, but I can't seem to figure out how to make
>>> Paraview aware the reader exists.
>>>
>>>
>>> There isn't an option for Boxlib in the dropdown for the file type when
>>> I try to open a file, and it doesn't appear as an available reader when I
>>> try to select the file -- although all the others from the VisitBridge
>>> appear. I missed a configuration file somewhere to make the file loading
>>> aware of the reader.
>>>
>>>
>>> Any suggestions on what I missed? I added a section in the
>>> Utilities/VisItBridge/databases/visit_readers.xml file, but that wasn't
>>> enough I guess.
>>>
>>>
>>> Thanks,
>>>
>>>
>>> Tim
>>>
>>>
>>> ------------------------------
>>> *From:* ParaView <paraview-bounces at paraview.org> on behalf of
>>> Gallagher, Timothy P <tim.gallagher at gatech.edu>
>>> *Sent:* Friday, August 26, 2016 11:53 AM
>>> *To:* Utkarsh Ayachit
>>>
>>> *Cc:* paraview at paraview.org
>>> *Subject:* Re: [Paraview] Status of BoxLib AMR reader
>>>
>>>
>>> Hi Utkarsh,
>>>
>>>
>>> I just checked with the latest head of the git repo and when I compile
>>> with the VisitBridge, there are no readers for BoxLib built.
>>>
>>>
>>> This page lists it as supported: http://www.paraview.org/ParaVi
>>> ew/index.php/VisIt_avt_Integration
>>>
>>>
>>> But then this later one doesn't include it: http://www.paraview.org/Wi
>>> ki/VisIt_Database_Bridge
>>>
>>>
>>> I have a dataset for you, I will send it off-list.
>>>
>>>
>>> Thanks,
>>>
>>>
>>> Tim
>>>
>>>
>>> ------------------------------
>>> *From:* Utkarsh Ayachit <utkarsh.ayachit at kitware.com>
>>> *Sent:* Friday, August 26, 2016 8:35 AM
>>> *To:* Gallagher, Timothy P
>>> *Cc:* paraview at paraview.org
>>> *Subject:* Re: [Paraview] Status of BoxLib AMR reader
>>>
>>> Tim,
>>>
>>> I am not exactly sure what's the status of the reader, but I can
>>> certainly look it up. Do you have a sample dataset also to test that
>>> it's working (or not)?
>>>
>>> Utkarsh
>>>
>>> On Thu, Aug 25, 2016 at 2:22 PM, Gallagher, Timothy P
>>> <tim.gallagher at gatech.edu> wrote:
>>> > Hi,
>>> >
>>> >
>>> > Our code is moving away from traditional structured grids towards an
>>> AMR
>>> > capability using BoxLib. That means, unfortunately, that all of the
>>> > development we have done to make our code Paraview and CoProcessing
>>> friendly
>>> > is not usable with the BoxLib file format.
>>> >
>>> >
>>> > I have seen some information on various websites about using the
>>> VisitBridge
>>> > to add the readers from Visit into Paraview, but when I build with it,
>>> the
>>> > BoxLib reader was not included in the list of formats I could use for
>>> files
>>> > -- even though it is listed online.
>>> >
>>> >
>>> > Does anybody know what the status is of the reader in PV and if it is
>>> > usable? Or is there a different format we can try to write out from
>>> our code
>>> > that would be usable? I really don't want to lose the work we've done
>>> with
>>> > PV and I would rather spend time writing code to get things working
>>> again
>>> > for PV than learn to use new software!
>>> >
>>> >
>>> > Thanks,
>>> >
>>> >
>>> > Tim
>>> >
>>> >
>>> > _______________________________________________
>>> > Powered by www.kitware.com
>>> <http://www.kitware.com/>
>>> Kitware Inc. - leading edge, high-quality software
>>> <http://www.kitware.com/>
>>> www.kitware.com
>>> Kitware's mission is to create state-of-the-art software products and
>>> services in visualization and data processing using advanced quality
>>> software methods and ...
>>>
>>> >
>>> > Visit other Kitware open-source projects at
>>> > http://www.kitware.com/opensource/opensource.html
>>> >
>>> > Please keep messages on-topic and check the ParaView Wiki at:
>>> > http://paraview.org/Wiki/ParaView
>>> >
>>> > Search the list archives at: http://markmail.org/search/?q=ParaView
>>> >
>>> > Follow this link to subscribe/unsubscribe:
>>> > http://public.kitware.com/mailman/listinfo/paraview
>>> >
>>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview/attachments/20160830/8f58bcc2/attachment.html>