throughout the code you will read references to something called representation (occasionally shortened to rep), this is supposed to mean some kind of simplification of a media item. For instance an image may be represented by a string hash or a 512 dimensional vector. So, the word representation is a generic way to encapsulate this idea. This allows us to write generic interfaces like store(rep) to store a representation of a media item in a database (mongo, sql, elasticsearch) as opposed to creating functions like store(hash), store(vector) etc.
within every module, inject the following code
import logging
log = logging.getLogger(__name__)
Continue to use log.info(), log.warning() and log.error() as a way to log results. During development these logs will be available for viewing to the developer in their terminal whereas in production, these will be collected by a log agent and sent to a centralized dashboard like kibana or xyz
For exception handling, please use the log.exception() method instead of log.error(). this ensures that the stack trace is printed along with the exception and is helpful in debugging.
log.info("hello")
log.warning("missing field : age")
log.error("somethign terrible happened")
log.exception("Error Initializing App")
to pretty print dict, use this log.info(json.dumps(my_dict, indent=4))
for any changes to how API endpoints are handled, look for corresponding functions in the handlers function todo : collocate handler functions and the mapping between API endpoints and the handler functions
Ensure that the operators you need are enabled in the config.yml and that only those operators are reffered to within your code base.
elements of a feature : controllers, routes, model
a lot of the config and payloads are dicts. pprint is a handy way to see them in a formatted way. import pprint pp = pprint.PrettyPrinter(indent=4) pp.pprint(my_dict)
in the request payload, either look for the media item in the files OR look for a media_url or text field in the post_data
pip list pip install <package_name> -U // upgrades the package to its latest version.
RESUME Open code in vscode open 3 terminals 1. docker-compose up 2. docker exec -it <container_id> /bin/sh -> python -m debugpy --listen 0.0.0.0:5678 --wait-for-client app.py 3. docker exec -it <container_id> /bin/sh -> nose2 feature.index.test_index.TestIndex.testRepresetnVideo