I propose an enhancement where Read and ProductSet-Reader are combined. Currently two graphs are required: One if there is only one input image, the other when two or more input images are used. Additional logic is required to select and execute the either the graph with Read or ProductSet-Reader respectively.
I’d like bump this request back up. I’m still maintaining duplicate graphs: one for AOIs requiring a single slice, and another for AOIs requiring two or more slices. @marpet seems this should be an easy change?
sorry, I can’t say. The ProductSet-Reader is not in my responsiblity. I hardly know it.
But maybe @lveci can asnwer it.
Without knowing the details I would agree, the ProductSet-Reader should also handle the case correctly when there is only one input.
This is a little more complicated since down stream operators would also need to support single instead of multiple inputs: Like slice assembly (and TOPSAR merge?).
I still promote this since a single graph would then support AOIs with one or more slices, (and one or more subswaths).
Any ideas how to call ProductSet-Reader in Python with snappy? Wanted to do S-1 slice assembly, but couldn’t figure out this one.
I’m not a python user. Just shell. Have you looked at gpt help:
gpt ProductSet-Reader -h
gpt ProductSet-Reader [options]
Description: Adds a list of sources Parameter Options: -PfileList=<string,string,string,...> Sets parameter 'fileList' to <string,string,string,...>. Graph XML Format: <graph id="someGraphId"> <version>1.0</version> <node id="someNodeId"> <operator>ProductSet-Reader</operator> <sources/> <parameters> <fileList>string,string,string,...</fileList> </parameters> </node> </graph>
Thanks, this helped.
Opened with readProduct and appended to a list . Then used list as input for createProduct.