tinyolap 0.8.12

TinyOlap: A multi-dimensional in-memory OLAP database in plain Python 3.

TinyOlap is a minimal in-process in-memory multi-dimensional database with numerical aggregations and calculations in mind. First a multi-dimensional data model needs to be defined, consisting of cubes, dimensions, members, hierarchies etc. Afterwards additional calculation logic can be added through arbitrary Python code. Data access is cell-based or range-based. A minimal support for SQL in also provided. All calculations will be executed on the fly. Optionally, persistence is provided through SQLite. TinyOlap is a byproduct of a research project, intended to mimic the behavior and capabilities of real-world MOLAP databases (e.g. IBM TM/1, SAP HANA or Jedox PALO) but with a super minimal footprint. TinyOlap is best suited for interactive planning, forecasting, simulation and general multidimensional numerical problems.

TinyOlap is also quite handy as a more comfortable alternative to Pandas DataFrames when your data is multidimensional, requires hierarchical aggregations or complex calculations.