Crawler Architecture

Crawler works in the following way:

1. Extractor runs Crawler through the API using Crawler's ID.
   a. API sends a request to the Orchestra.
   b. Orchestra calls Crawler itself.
2. Crawler is checking job sites and looking for vacancies, in which search_word is found and writes them down in a raw_vacancy document.
3. Cron Job checks a raw_vacancy document and parsed_vacancy for new vacancies (status: "new").
   a. If there are new records in raw_vacancy document, Cron sends a request to Orchestra to run Parser and to process vacancy.
   b. If there are new records in parsed_vacancy document, Cron sends a request to Orchestra to run graph_macker and to process skills.