v1.0.0

What's Changed

• tparquet: Parquet for time series, with the PCODEC compression codec (spec, Rust, Go, Python, Polars, DuckDB) by @almostintuitive in #1
• polars-tparquet: a Polars plugin package for tparquet files by @almostintuitive in #2
• README: lead with the result, measured numbers, and tparquet vs Vortex by @almostintuitive in #3
• Brand: tonal Squeeze logo in the README header by @almostintuitive in #6
• JVM, Spark, Trino, DataFusion and C/C++ readers for tparquet, with forks of their Parquet libraries by @almostintuitive in #4
• bench: tparquet on data providers' sample data, grouped by data type by @almostintuitive in #5
• README: lead with the real-data benchmark, synthetic numbers at the end by @almostintuitive in #11
• test: review every component, extend its test suite and fix the bugs found by @almostintuitive in #7
• fix(parquet): allocate only what a PCODEC page payload can produce by @almostintuitive in #8
• fix(tparquet): keep bloom filters and statistics settings in transcode, read intervals in verify by @almostintuitive in #9
• fix(go): check the linked C library's ABI version by @almostintuitive in #10
• ci: gate workflows on changed paths by @almostintuitive in #17
• docs: storage-tier guide and draft proposal to make PCODEC a standard Parquet codec by @almostintuitive in #15
• bench: tparquet against Parquet's ALP encoding by @almostintuitive in #14
• duckdb: DuckDB's own parquet extension with PCODEC, a small patch over DuckDB 1.5.6 and 2.0.0 by @almostintuitive in #16
• Install guide, and releases to crates.io and PyPI by @almostintuitive in #13
• Release 1.0.0 by @almostintuitive in #20

New Contributors

• @almostintuitive made their first contribution in #1

Full Changelog: https://github.com/aperiodic-io/tparquet/commits/v1.0.0