[WIP] Convert to python package with cli

[ahmed-shariff]

Following the discussion in #1 , this PR aims for the following:

• Convert the project into a package
  I am using poetry to manage the packaging aspect. The package is called paperscraper.
• Wrap package functions to cli. The cli command is also using the same name as the package.
  • process : command group for the different processes to process data. All subcommand can take one optional flag -f/--force. When this flag is not a given processes will run only when the corresponding output of the process doesn't exist.
    • run_all: run all the processes.
    • db: clean the xml file
    • venues: extract unique venues
    • data-extraction: extract the data from dblp snapshot
    • collect-data: scrape additional information
    • postprocess: clean and extract unique data
  • search: takes pattern string and returns any entries that has a match in the title or abstract. By default uses fuzzy matching. Has the following options:
    • --venue: filter by venue. Can have multiple --venue. Each can be a partial match to either full name or short name.
    • --author: filter by author. Can have multiple --author. Each can be a partial match.
    • --re: a flag, when set, the pattern will be treated as regex.
    • --fuzzy-max-difference: the maximum number of differences allowed from the pattern to get a match.
  • list: summery of the data (lists venues)

[arpitnarechania]

Thanks @ahmed-shariff, especially for the kind of API you have in mind - looks very promising. I will set it up locally this weekend and get back to you with a more detailed response.