This code base implements AnyDB, parser and query executor for the raw data. AnyDB can be applied to any format by specifying the pattern of the extraction in regular expression faster than the exisiting parsers. Also, you can execute SQL-like query in AnyDB on the raw data for analytics tasks. Note that this repository is the beta-version of AnyDB.

make
./anydb

Then give the raw data and extraciton pattern like following one.

>>> .scan /path/to/large/file.json -t data -e "stars":(?P<stars>DOUBLE).*"categories":"(?P<categories>[a-zA-Z]*)"

You can execute the query on the specified pattern.

>>> select categories, avg(stars) from data where stars > 3.5 group by categories limit 30;