[pdal] Building a PDAL pipeline in python

[Howard Butler]

> On Mar 2, 2017, at 3:34 PM, Peter Tittmann <ptittmann at gmail.com> wrote:
> 
> Howard,
> 
> I’m writing the pipeline in python as I’m going to use the multiprocessing module to parallelize pdal pipelines. I’m in the early stages of testing this. I’d like to be able to avoid using Fusion to derive gridded point cloud metrics and also be able to work directly on the points using PCL.

Can you get from numpy arrays to PCL very easily? I don't know if the path has been beaten down on that or how well worn the trail is.


> 
> Right now however, its not really working:
> 
>> In [5]: import pdal
>> 
>> In [6]: pString = unicode(makePlots(1112))
>> 
>> In [7]: pString
>> Out[7]:
>> u'{"pipeline": ["./pgc_input/chug_lidar/Kenai/Points/Ken_00403.las", {"distance": 46.330202145018255, "type": "filters.crop", "point": "POINT(671977.149785037 6621359.13979493)"}, "./pgc_input/plot_las/plot1112.las"]}'
>> 
>> In [8]: pipe = pdal.Pipeline(pString)
>> 
>> In [9]: pipe.validate()
>> Warning 1: Cannot find pcs.csv
>> Out[9]:
>> True
>> 
>> In [10]: pipe.loglevel = 9
>> 
>> In [11]: count =pipe.execute()
>> 
>> Process Python killed
>> 

If you privately send me the input, I'd be willing to try to work this for you. I have a suspicion about your pcs.csv complaint and Python/PDAL's logging...