Haskell implementation of labeled one-dimensional arrays

Packages in this repository implement series, or labeled one-dimensional arrays, and associated functions.

Like Data.Map.Strict, series support efficient:

• random access by key ( $\mathcal{O}\left( \log n \right)$ ) ;
• slice by key ( $\mathcal{O}\left( \log n \right)$ ).

Like Data.Vector.Vector, series support efficient:

• random access by integer index ( $\mathcal{O}\left( 1 \right)$ );
• slice by integer index ( $\mathcal{O}\left( 1 \right)$ );
• numerical operations.

Tutorial and documentation

A tutorial and interface documentation for the most recent published version are available here.

Locally, you can generate documentation for all packages using haddock like so:

cabal haddock javelin
cabal haddock javelin-io

Get involved!

Do not hesitate to make feature requests or report bugs via the issue tracker.

Preliminary benchmarks

Looking up random integers:

Summing random integers:

Averaging elements using 'Control.Foldl.mean' from the 'foldl' package:

Concatenating with (<>):

Extracing a 10% chunk of the container:

You can run the benchmarks above by running:

# Run the benchmark
cabal bench comparison-containers --project=cabal.project.benchmarking
# Format the resulting CSV file
cabal run bench-report javelin/out.csv report.md