[QGIS-Developer] How to handle "RuntimeError: wrapped C/C++ object of type AlgorithmDialog has been deleted"

[Tobias Schmetzer]

Dear developers,

 

I've rewritten three python scripts as processing scripts. Two of them don't terminate correctly showing only a not very helpful stack trace dialog stating:


An error has occurred while executing Python code:

RuntimeError: wrapped C/C++ object of type AlgorithmDialog has been deleted
Traceback (most recent call last):
  File "C:/OSGEO4~1/apps/qgis/./python/plugins\processing\gui\AlgorithmDialog.py", line 246, in runAlgorithm
on_complete(ok, results)
  File "C:/OSGEO4~1/apps/qgis/./python/plugins\processing\gui\AlgorithmDialog.py", line 201, in on_complete
self.tr('Execution completed in {0:0.2f} seconds').format(time.time() - start_time))
RuntimeError: wrapped C/C++ object of type AlgorithmDialog has been deleted


 

I couldn't tell if my script is error free (it most likely isn't). Yet the user would need to have some more information besides just a runtime error.

 

So the question is: Given that QGIS doesn't provide any more information is it considered a bug or a feature request to provide information about the runtime error? I have already created a bug ticket with my script attached (https://github.com/qgis/QGIS/issues/42069) assuming QGIS should provide information about the error but I am unsure if I shouldn't rather put the issue as a question on gis.stackexchange.com assuming my script is just wrong. Maybe the topic can also be solved by adding the most common mistakes to the docs (which I'm happy to do once the reasons are clear). Any opinions to what extent QGIS error messages should be meaningful and how to treat the topic?

 

Thanks in advance for your thoughts!

 

Regards, Tobias

 

 
___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [QGIS-Developer] WITH_PDAL and WITH_EPT doubt

[Alexandre Neto]

Sorry for the basic questions, I was about to say. I am lagging a bit on
all this point cloud business, but I wanted to make it available on
conda-forge.

Thanks,

Alexandre Neto


A sábado, 6/03/2021, 09:53, Alexandre Neto  escreveu:

> Hi Sylvain,
>
> Thank you for you answer. Yes, the build in conda is building untwine for
> sure. I know because the windows build is failing there. So that's a result
> of with_ept=true?
>
>  entwine or untwine serve the same purpose? Is there a prefered one?
>
> If I build using WITH_EPT = true, does it make sense to install entwine
> (from packaging system) anyway? Or, in that case, I should use WITH_EPT =
> FALSE?
>
> Sorry for
>
> A sábado, 6/03/2021, 09:28, Sylvain POULAIN 
> escreveu:
>
>> Hi Alexandre,
>>
>> When you build qgis with cmake, it will build untwine (not entwine) in
>> qgis structure.
>> I guess it could be the same with conda but not tested.
>>
>> Regards,
>>
>> *Sylvain*
>>
>>
>> Le mer. 3 mars 2021 à 12:46, Alexandre Neto  a
>> écrit :
>>
>>> Hello again,
>>>
>>> Does this mean that, if entwine isn't present in the system, the
>>> compilation won't fail or complaint, but some functionality will be missing?
>>>
>>> Should I include entwine as a dependency for 3.18.0 on conda-forge?
>>>
>>> Thanks,
>>> Alexandre
>>>
>>> A quarta, 3/03/2021, 03:40, Alexandre Neto 
>>> escreveu:
>>>
 Great, thanks!

 A quarta, 3/03/2021, 02:03, Nyall Dawson 
 escreveu:

> On Wed, 3 Mar 2021 at 11:59, Alexandre Neto 
> wrote:
> >
> > Hi,
> >
> > I am trying to add point cloud support to QGIS on conda-forge [0],
> but I am unsure if I need to use
> > WITH_PDAL=TRUE
>
> yes, you'll need to add this even if you have pdal library available.
> it defaults to false to avoid breaking the compilation if pdal isn't
> available (and we want explicit breaks here rather than silently
> compiling without pdal provider if the library isn't found).
>
> > and WITH_EPT=TRUE
>
> That defaults to true anyway, so you can omit it if you want.
>
> Nyall
>
 ___
>>> QGIS-Developer mailing list
>>> QGIS-Developer@lists.osgeo.org
>>> List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>>> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>>>
>>
___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer