Skip to content

Latest commit

 

History

History
91 lines (70 loc) · 3.6 KB

README.md

File metadata and controls

91 lines (70 loc) · 3.6 KB
Raw

Médonay

This is a set of microservices to extract metadata from various sources (your emails, your online accounts) and taking actions on it. Using a microservice approach has the great advantage of being able to write any parser in the best fitted language, as is perfectly well described on this blog post.

The core idea is to have a set of interchangeable microservices:

  • A set of crawlers which are crawling through various places (emails, accounts, …) to fetch your data and handing it over to parsers.
  • A set of parsers to extract metadata from your emails.
  • An indexer which is basically exposing an API endpoint to store metadata extracted from your emails.
  • A viewer to browse the fetched metadata.

Each microservice could be easily replaced by an alternative microservice. For instance, the current implementation of the indexer uses an SQLite backend, but you might want to use MongoDB or CouchDB, which is just a matter of changing the indexer component.

Supported crawlers so far are:

  • An email crawler to fetch data from your emails. It can either fetch automatically over IMAP or read emails you pipe to it.
  • A Web Outside of Browsers based crawler to fetch data from your online accounts (by scraping them). This one is a bit particular as Web Outside of Browsers already returns structured data, so it does not call any parser under the hood.

Supported extractors (parsers) so far are:

Each microservice accepts environment variables to customize its behavior. These are HOST (for the host to bind to), PORT (for the port to bind to) and DEBUG (for enabling or not debug mode). Other environment variables might be available, have a look at scripts/run.sh file.

This repository started as a fork of iamthefij's email-assistant, but is now differing quite a lot.

Detailed documentation and design guides, including guides to extend the available parsers, are available in the doc folder.

Testing it out

Here are some basic steps to get started:

git clone https:/Phyks/medonay
cd medonay
# Create a virtual env and install all required dependencies. Same for Ruby
# scripts.
./scripts/setup.sh
# Next, start the indexer, parsers and viewer. You can edit the variables at
# the beginning of the file.
./scripts/run.sh

Then, for instance, you can easily start parsing emails by cating them to the crawler:

cd crawlers/imap-crawler && cat /path/to/email.eml | INDEXER_URL=http://127.0.0.1:4100 PARSER_1=http://127.0.0.1:4001 PARSER_2=http://127.0.0.1:4002 python -m imapCrawler

Pass it INDEXER_TOKEN=<token> as well if you used a token to protect your indexer.

Note about Docker files: The initial setup was using Docker files and Docker compose. I'm not using Docker, hence these files are left for documentation purpose, but not actively maintained. Any contribution to maintain these Docker files would be greatly appreciated! :)

License

This repository is licensed under MIT license.