Open jbfaden opened 3 weeks ago
Somehow the info thinks there will only be four parameters, but the data contains six parameters. Here are some useful URLS:
Used to create the info:
wget -O - "https://csa.esac.esa.int/csa-sl-tap/data?retrieval_type=HEADER&DATASET_ID=D1_CP_STA-DWP_PSD&FORCEPACK=false"
Info read while debugging:
wget -O - 'http://localhost:8080/HapiServer/hapi/info?dataset=D1_CP_STA-DWP_PSD'
Data read while debugging:
wget -O - 'http://localhost:8080/HapiServer/hapi/data?dataset=D1_CP_STA-DWP_PSD&start=2007-09-27T09:00Z&stop=2007-09-28T09:00Z'
Data read from TAP server:
wget -O D1_CP_ST.cef "https://csa.esac.esa.int/csa-sl-tap/data?RETRIEVAL_TYPE=product&RETRIEVAL_ACCESS=streamed&DATASET_ID=D1_CP_STA-DWP_PSD&START_DATE=2007-09-27T09:00:00Z&END_DATE=2007-09-28T09:00:00Z"
I found what I think is the bug, where the each of the six parameters is counted in the HapiRecord, though two of them are non-time-varying. I need to see why other datasets like this worked, if they did work.
Also, I'm unable to see how the DEPEND_1 connection is made. In the HEADER request, I don't see anything that declares the connection.
There's another case which is now broken, I think. This is D1_SP_HIA, which had code which would repeat "Half_interval" though it is non-time-varying.
Jonathan points out
results in a runtime error.