Add extractor key and owner email param support #39
Conversation
|
|
pyclowder/extractors.py
Outdated
| connector.connect() | ||
| connector.register_extractor(self.args.registration_endpoints) | ||
|
|
||
| # TODO: register extractor initially without _process_message? |
There was a problem hiding this comment.
how is this different from the register_extractor also I like to move away from this in the future.
There was a problem hiding this comment.
register_extractor registers with arg-specified endpoints, while this is a generic registration with the clowder instance in clowder_url (if not already registered there). this would allow extractor_info to not have registration_urls, but auto-register with Clowder. i was having trouble registering with just RMQ heartbeats in my test environment, but maybe this is unnecessary...
| @@ -519,8 +519,12 @@ def register_extractor(self, endpoints): | |||
|
|
|||
| for url in endpoints.split(','): | |||
| if "unique_key" in data: | |||
| if url.find("?") > -1: url += "&user=%s" % self.clowder_email | |||
| else: url += "?user=%s" % self.clowder_email # TODO: This will not work, need an auth key matching email | |||
| if url.find("?") > -1: | |||
There was a problem hiding this comment.
what if this is the first parameter?
There was a problem hiding this comment.
if the config has specified a registration URL without an API key, registering privately with just an email address won't be accepted. so it doesn't send the request.
Co-authored-by: Rob Kooper <kooper@illinois.edu>
Co-authored-by: Rob Kooper <kooper@illinois.edu>
…amework/pyclowder into extractor-key-support
Related to clowder-framework/clowder#300
add -e EXTRACTOR_KEY and -e CLOWDER_EMAIL to pyclowder extractors to register an initial user and unique key.