Changelog for aeson-

If this renders badly as markdown, see the plain text version

* The performance of encoding to and decoding of bytestrings
  have both improved by up to 2x, while also using less

* New dependency: the scientific package lets us parse
  floating point numbers more quickly and accurately.

* eitherDecode, decodeStrictWith: fixed bugs.

* Added FromJSON and ToJSON instances for Tree and Scientific.

* Fixed the ToJSON instances for UTCTime and ZonedTime.

0.6 series

* Much improved documentation.

* Angle brackets are now escaped in JSON strings, to help avoid XSS

* Fixed up handling of nullary constructors when using generic

* Added ToJSON/FromJSON instances for:

  * The Fixed class
  * ISO-8601 dates: UTCTime, ZonedTime, and TimeZone

* Added accessor functions for inspecting Values.

* Added eitherDecode function that returns an error message if
  decoding fails.

0.5 to 0.6

* This release introduces a slightly obscure, but
  backwards-incompatible, change.

  In the generic APIs of versions 0.4 and 0.5, fields whose
  names began with a "_" character would have this character
  removed.  This no longer occurs, as it was both buggy and
  surprising (

* Fixed a bug in generic decoding of nullary constructors

0.4 to 0.5

* When used with the UTF-8 encoding performance improvements
  introduced in version of the text package, this
  release improves aeson's JSON encoding performance by 33%
  relative to aeson 0.4.

  As part of achieving this improvement, an API change was
  necessary.  The fromValue function in the
  Data.Aeson.Encode module now uses the text package's
  Builder type instead of the blaze-builder package's
  Builder type.

0.3 to 0.4

* The new decode function complements the longstanding
  encode function, and makes the API simpler.

* New examples make it easier to learn to use the package

* Generics support

  aeson's support for data-type generic programming makes it
  possible to use JSON encodings of most data types without
  writing any boilerplate instances.

  Thanks to Bas Van Dijk, aeson now supports the two major
  schemes for doing datatype-generic programming:

  * the modern mechanism, built into GHC itself

  * the older mechanism, based on SYB (aka "scrap your

  The modern GHC-based generic mechanism is fast and terse: in
  fact, its performance is generally comparable in performance
  to hand-written and TH-derived ToJSON and FromJSON
  instances.  To see how to use GHC generics, refer to

  The SYB-based generics support lives in Data.Aeson.Generic
  and is provided mainly for users of GHC older than 7.2.  SYB
  is far slower (by about 10x) than the more modern generic
  mechanism.  To see how to use SYB generics, refer to

* We switched the intermediate representation of JSON objects
  from Data.Map to Data.HashMap which has improved type
  conversion performance.

* Instances of ToJSON and FromJSON for tuples are between 45%
  and 70% faster than in 0.3.

* Evaluation control

  This version of aeson makes explicit the decoupling between
  *identifying* an element of a JSON document and *converting*
  it to Haskell.  See the Data.Aeson.Parser documentation for

  The normal aeson decode function performs identification
  strictly, but defers conversion until needed.  This can
  result in improved performance (e.g. if the results of some
  conversions are never needed), but at a cost in increased
  memory consumption.

  The new decode' function performs identification and
  conversion immediately.  This incurs an up-front cost in CPU
  cycles, but reduces reduce memory consumption.