Serialize information about equivalences between miRBase identifiers and colloquial names using the two namespaces produced in #1

miRBase has an identifier system that looks like MIRXXXXX and a colloquial name system that's similar to the biological literature

This was a great first attempt:

Now, we should implement this function to be smarter and not re-download the file each time. Here's an example from Bio2BEL ChEBI:

import logging

# build the logger using this module's name at runtime
log = logging.getLogger(__name__)

def download_compounds(force_download=False):
    """Downloads the compounds information

    :param bool force_download: If true, overwrites a previously cached file
    :rtype: str
    """
    if os.path.exists(COMPOUNDS_DATA_PATH) and not force_download:
        log.info('using cached data at %s', COMPOUNDS_DATA_PATH)
    else:
        log.info('downloading %s to %s', COMPOUNDS_URL, COMPOUNDS_DATA_PATH)
        urlretrieve(COMPOUNDS_URL, COMPOUNDS_DATA_PATH)

    return COMPOUNDS_DATA_PATH

Basically, it adds a check if the data file already exists and doesn't download it again if it does (and gives some nice logging).

• make a tox script that has build entries for creating namespaces and equivalence files
• leverage Travis CI to deploy to artifactory using https://github.com/Parallels/artifactory. Perhaps in the future, travis will directly support deploying there

Make a file called /src/bio2bel_mirbase/cli.py inside which you should use the click package to make a simple command line interface.

• use absolute imports (i.e. from bio2bel_mirbase.parser import ...) so this script can be run from inside the src directory
• name your click function def main(...):
• add an if __name__ == '__main__': main() so it can be run from inside an IDE

Let's pair program this today.