search

clowder-framework / pyclowder

Library to assist in the development of extractors for clowder.
University of Illinois/NCSA Open Source License
10 stars 2 forks source link

fix #47 now the name = queue name #48

Closed robkooper closed 2 years ago

robkooper commented 2 years ago

pyclowder would connect to the queue specified as either the queue name specified on hte command line or rabbitmq_queue environment variable (if neither was set it would default to extractor_info[name]).

clowder on the other hand would use the extractor_info[name] always to send messages, resulting in a mismatch between what the extractor was listening on and where clowder send the messages

this will now set the extractor_info[name] to be the same as the queue, fixing this issue.

robkooper commented 2 years ago

for example:

docker run -ti --network clowder_clowder -e RABBITMQ_URI=amqp://guest:guest@rabbitmq/%2F clowder/wordcount
docker run -ti --network clowder_clowder -e RABBITMQ_QUEUE=ncsa.wordcount.rob -e RABBITMQ_URI=amqp://guest:guest@rabbitmq/%2F clowder/wordcount

results in:

image
ddey2 commented 2 years ago

Looks good to me. Tested on my local.