Subscribe to
Posts
Comments

Updated ICU bindings for Haskell: 0.4.0.0

September 10th, 2010 by Bryan O'Sullivan

I've released version 0.4.0.0 of the text-icu library, my Haskell bindings for the International Components for Unicode (ICU) library.

The text-icu library is the practical cousin to the text library: it implements a myriad of Unicode-related functionality that is not yet otherwise available in Haskell.

Features:

  • Locale sensitive case mapping

  • Text normalization

  • [NEW] Locale sensitive string collation

  • Conversion to and from a huge number of native encodings

  • [NEW] Efficient comparison of ByteString and Text values

The library is thoroughly documented, and most interfaces are both pure and easy to use.

Look for further releases over the coming weeks as I complete the ICU integration.

5 Comments »

5 Responses to “Updated ICU bindings for Haskell: 0.4.0.0”

  1. on 10 Sep 2010 at 16:121 donatas

    i got following error trying to build it:

    Data/Text/ICU/Collate/Pure.hs:30:1: lexical error at character ‘i’

  2. on 10 Sep 2010 at 16:182 Bryan O'Sullivan

    Thanks, Donatas. Fixed now.

  3. on 10 Sep 2010 at 23:513 AndyStewart

    I think you should add text >= 0.8.0 in build-depends.
    Otherwise, you will got below error :

    …

    [11 of 14] Compiling Data.Text.ICU.Break ( dist/build/Data/Text/ICU/Break.hs, dist/build/Data/Text/ICU/Break.o )

    Data/Text/ICU/Break.hsc:46:26:
    Module `Data.Text.Foreign’ does not export `dropWord16′

    Data/Text/ICU/Break.hsc:46:38:
    Module `Data.Text.Foreign’ does not export `takeWord16′
    cabal: Error: some packages failed to install:
    text-icu-0.4.0.1 failed during the building phase. The exception was:
    ExitFailure 1

    BTW, can you add module to binding “Charset Detection” (http://userguide.icu-project.org/conversion/detection) in icu?
    I need “automatic detect encoding” module to get the encode information from unknown string, then i can use it in my file-manager.

    Thanks!

    — Andy

  4. on 13 Sep 2010 at 08:344 Ersin Er

    It handles the “Turkish İ Problem”* correctly. So it should be quite stable.. :-)

    * http://www.i18nguy.com/unicode/turkish-i18n.html

  5. on 27 Sep 2010 at 00:505 Case Mapping for Turkish in Haskell « Ersin Er ~ Computer Scientist & Software Engineer

    [...] for various Unicode services through text and text-icu packages. Some recent updates to text and text-icu are explained in the author’s [...]

Leave a Reply