prepare KFlate 1.1.0 - #41

Closed

rafambn wants to merge 55 commits into

Commits on May 5, 2026

Infra: migrate to kotlinx-benchmark and update project identity
rafambn
committed
e90be34
Refactor benchmarks to simplify raw compression testing
rafambn
committed
2cbb8f2
Docs: update documentation snippets and benchmarking instructions
rafambn
committed
6cbee92
Build: add automated benchmark comparison reporting
rafambn
committed
9897535
Build: improve benchmark metadata handling and increase measurement precision
rafambn
committed
ba67a1f

Commits on Sep 4, 2026

Separate compression APIs without type collisions
rafambn
committed
bddfc68

Commits on Sep 5, 2026

Reject reserved DEFLATE literal codes
rafambn
committed
432848f
Clarify fflate attribution and package the license
rafambn
committed
f27e457
Remove unsupported GZIP dictionaries (#5)
rafambn
authored
c9577ef
Handle partial ZLIB streaming headers (#6)
rafambn
authored
84f22cd
Reject empty raw compressed input (#7)
rafambn
authored
025226d
Validate GZIP metadata before encoding (#8)
rafambn
authored
87a65ed
Bound decompression output before allocation
rafambn
committed
2cf8a2a
Simplify decompression output limits (#17)
rafambn
authored
bfd79b5
Reject sizes that overflow internal buffers (#12)
rafambn
authored
a91961b
Decode DEFLATE streams across chunk boundaries (#13)
rafambn
authored
ee03df8
Keep the web demo responsive on worker failures (#14)
rafambn
authored
c54cd40
Cap web demo decompression at 128 MiB (#16)
rafambn
authored
ed1a1da
Keep the error factory internal (#19)
rafambn
authored
aee9b08
Give internal result types their own files (#20)
rafambn
authored
9f7e2ac
Stop repeating large fixtures across tests (#22)
rafambn
authored
ba56d57

Commits on Sep 6, 2026

Build every target on pull requests (#23)
rafambn
authored
331b183
Keep JVM coverage at 100%
rafambn
authored
9abb15c
Throw FlateError directly
rafambn
authored
766c95b
Prepare the 1.1.0 release docs and artifacts
rafambn
committed
4af757e
Improve higher DEFLATE compression levels
rafambn
committed
c993daf
Let compression levels manage hash-table memory
rafambn
committed
91ad6c4
Cut level 9 output with cost-aware parsing
rafambn
committed
71d4753
Make benchmark results representative and reviewable
rafambn
committed
d0fec3b
Reject invalid measurements in benchmark reports
rafambn
committed
e5992df

Commits on Sep 7, 2026

Commits on Sep 8, 2026

Make the compression demo easier to use and inspect
rafambn
committed
b995b5d
Merge pull request #27 from rafambn/release/prepare-1.1.0
rafambn
authored
047692d

Commits on Sep 9, 2026

cache benchmark baselines and publish level history
rafambn
committed
36a54b1
remove trailing blank lines from benchmark report
rafambn
committed
f980d13
Simplify benchmarks with library selection and retain collected data
rafambn
committed
db56184
Keep archived benchmark commits usable in shallow checkouts
rafambn
committed
38e1cd8
Merge pull request #31 from rafambn/feat/cached-benchmark-baselines
rafambn
authored
fe2ff51
Choose stored blocks using their actual padding cost
rafambn
committed
5e34cfc
Reuse fixed token prices in level 9 parsing
rafambn
committed
374b326
Expand inflate matches with bulk copies
rafambn
committed
4c78b02
Cut scratch allocation for short compression inputs
rafambn
committed
4cc3f89
Merge pull request #40 from rafambn/agent/deflate-scratch-sizing
rafambn
authored
ba01d65
Shorten benchmark runs with one JVM fork and ten measurements
rafambn
committed
7592ba5
Merge pull request #39 from rafambn/agent/inflate-copy-speed
rafambn
authored
060f957
Merge branch 'dev-1.1.0' into cost-parser-speed
rafambn
authored
8978d9f
Merge pull request #35 from rafambn/cost-parser-speed
rafambn
authored
c2ffe0d

Commits on Sep 11, 2026

Merge pull request #33 from rafambn/agent/huffman-ratio
rafambn
authored
611c594