Skip to content

Latest commit

 

History

History
150 lines (118 loc) · 4.45 KB

CHANGELOG.md

File metadata and controls

150 lines (118 loc) · 4.45 KB
Raw

v0.16.4 - 2024-01-04

  • Remove remove ahash 0.7.6 from dependencies (cargo audit)

v0.16.3 - 2023-08-16

  • Add serde feature (Serialize and Deserialize for Lang and Script).

v0.16.2 - 2022-10-23

v0.16.1 - 2022-08-30

  • Fix bug in Czech alphabet (improved quality of Czech language detection)

v0.16.0 - 2022-05-07

  • [breaking] Add Armenian script (Script::Armenian) and language (Lang::Hye)

v0.15.0 - 2022-05-01

  • Update enum-map dependency to version 2
  • Optimize alphabet method for Cyrillic: almost 2x improved performance for Cyrillic languages and 7% for the average detect() benchmark.

v0.14.0 - 2022-04-15

  • Improve performance of detect() almost twice (see PR)

v0.13.0 - 2022-01-02

  • [breaking] - Support of Tagalog (Tgl)
  • Rename whatlang::Error -> whatlang::ParseError

v0.12.0 - 2021-04-18

  • [breaking] - Drop languages:
    • Tigrinya (Tir)
    • Hausa (Hau)
    • Chewa (Nya)
    • Bhojpuri (Bho)
    • Igbo (Ibo)
    • Maithili (Mai)
    • Oromo (Orm)
    • Rundi (Run)
    • Saraiki (Srk)
    • Kurdish (Kur)
    • Cebuano (Ceb)
    • Malagasy (Mlg)
    • Kinyarwanda (Kin)
    • Somali (Som)
    • Ilocano (Ilo)
    • Uyghur (Uig)
    • Tagalog ('Tgl')
    • Haitian Creole (Hat)
    • Nynorsk (Nno)
    • Yoruba (Yor)
  • [breaking] - Rename Yiddish: Ydd -> Yid
  • [breaking] - Rename Azerbaijani: Azj -> Aze
  • [breaking] Rename List -> FilterList
  • [breaking] rename whitelist and blacklist to allowlist and denylist respectively
  • Drop support of Cyrillic Azerbaijani and Turkmen
  • Add Script::all() to iterate over all scripts.
  • Add Lang::all() to iterate over all languages.
  • Add integration with enum-map
  • Implement FromStr for Script and Lang
  • Implement Script::langs(&self) -> &[Lang]
  • About 7% slower than v0.11.1 due to new detection method introduced. It's still much faster than v0.11.0

v0.11.1 - 2020-11-28

  • Use Trigram tuple instead of heap allocated String. (~68% faster)

v0.11.0 - 2020-11-03

  • [breaking] - rename code for Arabic: Arb -> Ara

v0.10.0 - 2020-09-04

  • Support Catalan

v0.9.0 - 2020-06-26

  • Support Slovak

v0.8.0 - 2020-05-08

  • Support Latin

v0.7.4 - 2020-04-26 (yanked version)

  • Support Latin

v0.7.2 - 2019-10-19

  • (fix) respect japanese whitelisting when mandarin characters are given (#44)

v0.7.1 - 2019-05-06

  • Update dependency hashbrown 0.1.8 -> 0.3.0 (10% faster)

v0.7.0 - 2019-03-03

  • Support Afrikaans language (afr)
  • Get rid of build dependencies: installation is much faster now

v0.6.0 - 2018-11-09

  • Use hashbrown instead of fnv (detect() is 30% faster)
  • Use array on stack instead of vector for detect_script (1-2% faster)
  • Use build.rs to generate lang.rs file
  • Add property based testing

v0.5.0 - 2017-08-06

  • (breaking) Rename Lang::to_code(&self) to Lang::code(&self)
  • (fix) Fix bug with zero division in confidence calculation
  • (fix) Confidence can not exceed 1.0
  • Implement Lang::eng_name(&self) -> &str function
  • Implement Lang::name(&self) -> &str function
  • Implement Script::name(&self) -> &str function
  • Implement trait Dislpay for Script
  • Implement Display trait for Lang

v0.4.1 - 2017-07-31

  • Calculate confidence in the range from 0 to 1 for Info

v0.4.0 - 2017-07-30

  • Calculate is_reliable bool for Info struct.
  • Breaking changes for Info. Make fields private. Now one should use methods.
  • Remove support of Latin version of Serbo-Croatian, because it conflicts a lot with modern Croatian.

v0.3.3 - 2017-07-26

  • Replace HashMap with FnvHashMap (~ 33% faster)

v0.3.2 - 2017-06-04

  • Small performance improvement: preallocate memory for counter_hash in trigrams.rs (~ 2-3% faster)

v0.3.1 - 2017-02-10

  • Fix build
  • Add link to doc at crates.io

v0.3.0 - 2017-02-10

  • Support New 14 languages
  • (breaking) New API

v0.2.1 - 2017-02-07

  • Support 10 new languages
  • Optimize trigram algorithms

v0.2.0 - 2017-02-06

  • Optimize script detection
  • Accept text, blacklist and whitelist as references
  • 10 new languages
  • Fix: always guarantee same result on same input data (fix sorting issue)

v0.1.4 - 2017-02-04

  • Support whitelist and blacklist

v0.1.3 - 2017-02-03

  • Support more than 50 languages

v0.1.2 - 2017-01-29

  • Support about 20 languages

v0.1.1 - 2016-12-25

  • Tiny improvements

v0.1.0 - 2016-12-25

  • First public release