1For the latest version of this document, please see [https://github.com/bos/aeson/blob/master/changelog.md](https://github.com/bos/aeson/blob/master/changelog.md).
5* Added `withEmbeddedJSON` to help parse JSON embedded inside a JSON string, thanks to Jesse Kempf.
6* Memory usage improvements to the default (pure) parser, thanks to Jonathan Paugh. Also thanks to Neil Mitchell & Oleg Grenrus for contributing a benchmark.
7* `omitNothingFields` now works for the `Option` newtype, thanks to Xia Li-yao.
8* Some documentation fixes, thanks to Jonathan Paug & Philippe Crama.
12* Add `FromJSON` and `ToJSON` instances for
13 * `DiffTime`, thanks to Víctor López Juan.
14 * `CTime`, thanks to Daniel Díaz.
15* Fix handling of fractions when parsing Natural, thanks to Yuriy Syrovetskiy.
16* Change text in error messages for Integral types to make them follow the common pattern, thanks to Yuriy Syrovetskiy.
17* Add missing `INCOHERENT` pragma for `RecordToPair`, thanks to Xia Li-yao.
18* Everything related to `Options` is now exported from `Data.Aeson`, thanks to Xia Li-yao.
19* Optimizations to not escape text in clear cases, thanks to Oleg Grenrus.
20* Some documentation fixes, thanks to Phil de Joux & Xia Li-yao.
24* Add `parserThrowError` and `parserCatchError` combinators, thanks to Oleg Grenrus.
26* Add `Generic` instance for `Value`, thanks to Xia Li-yao.
28* Fix a mistake in the 1.2.0.0 changelog, the `cffi` flag is disabled by default! Thanks to dbaynard.
32* `tagSingleConstructors`, an option to encode single-constructor types as tagged sums was added to `Options`. It is disabled by default for backward compatibility.
34* The `cffi` flag is now turned off (`False`) by default, this means C FFI code is no longer used by default. You can flip the flag to get C implementation.
36* The `Options` constructor is no longer exposed to prevent new options from being breaking changes, use `defaultOptions` instead.
38* The contents of `GToJSON` and `GToEncoding` are no longer exposed.
40* Some INLINE pragmas were removed to avoid GHC running out of simplifier ticks.
44* Fix an accidental change in the format of `deriveJSON`. Thanks to Xia Li-yao!
46* Documentation improvements regarding `ToJSON`, `FromJSON`, and `SumEncoding`. Thanks to Xia Li-yao and Lennart Spitzner!
50* Added a pure implementation of the C FFI code, the C FFI code. If you wish to use the pure haskell version set the `cffi` flag to `False`. This should make aeson compile when C isn't available, such as for GHCJS. Thanks to James Parker & Marcin Tolysz!
52* Using the `fast` flag can no longer cause a test case to fail. As far as we know this didn't affect any users of the library itself. Thanks to Xia Li-yao!
56* Added instances for `UUID`.
58* The operators for parsing fields now have named aliases:
59 - `.:` => `parseField`
60 - `.:?` => `parseFieldMaybe`
61 - `.:!` => `parseFieldMaybe'`
62 - These functions now also have variants with explicit parser functions: `explicitParseField`, `explicitParseFieldMaybe`, "explicitParseFieldMaybe'`
63Thanks to Oleg Grenrus.
65* `ToJSONKey (Identity a)` and `FromJSONKey (Identity a)` no longer require the unnecessary `FromJSON a` constraint. Thanks to Oleg Grenrus.
67* Added `Data.Aeson.Encoding.pair'` which is a more general version of `Data.Aeson.Encoding.pair`. Thanks to Andrew Martin.
69* `Day`s BCE are properly encoded and `+` is now a valid prefix for `Day`s CE. Thanks to Matt Parsons.
71* Some commonly used ToJSON instances are now specialized in order to improve compile time. Thanks to Bartosz Nitka.
74[JSONTestSuite](https://github.com/nst/JSONTestSuite) cleanups, all
75motivated by tighter RFC 7159 compliance:
77* The parser now rejects numbers for which
78 [the integer portion contains a leading zero](https://github.com/bos/aeson/commit/3fb7c155f2255482b1b9566ec5c1eaf9895d630e).
79* The parser now rejects numbers for which
80 [a decimal point is not followed by at least one digit](https://github.com/bos/aeson/commit/ecfca35a45286dbe2bbaf5f62354be393bc59b66),
81* The parser now rejects documents that contain [whitespace outside the
82 set {space, newline, carriage return, tab}](https://github.com/bos/aeson/commit/8ef622c2ad8d4a109884e17c2792238a2a320e44).
84Over 90% of JSONTestSuite tests currently pass. The remainder can be
85categorised as follows:
87* The string parser is strict with Unicode compliance where the RFC
88 leaves room for implementation-defined behaviour (tests prefixed
89 with "`i_string_`". (This is necessary because the `text` library
90 cannot accommodate invalid Unicode.)
92* The parser does not (and will not) support UTF-16, UTF-32, or byte
94* The parser accepts unescaped control characters, even though the RFC
95 states that control characters must be escaped. (This may change at
96 some point, but doesn't seem important.)
100* Fixes a regression where a bunch of valid characters caused an
101 "Invalid UTF8-Stream" error when decoding. Thanks to Vladimir
102 Shabanov who investigated and fixed this.
106* Fixes a regression where it was no longer possible to derive
107 instances for types such as `data T a = T { f1 :: a, f2 :: Maybe a
110Thanks to Sean Leather for fixing this, and to Ryan Scott for helping out.
114* Decoding performance has been significantly improved (see
115 https://github.com/bos/aeson/pull/452). Thanks to @winterland1989.
117* Add `ToJSON`/`FromJSON` instances for newtypes from
118 `Data.Semigroup`: `Min`, `Max`, `First`, `Last`, `WrappedMonoid`,
119 `Option`. Thanks to Lennart Spitzner.
121* Make the documentation for `.:!` more accurate. Thanks to Ian Jeffries.
127* Introduced new `FromJSONKey` and `ToJSONKey` type classes that are
128 used to encode maps without going through HashMap. This also allows arbitrary
129 serialization of keys where a string-like key will encode into an object and
130 other keys will encode into an array of key-value tuples.
132* Added higher rank classes: `ToJSON1`, `ToJSON2`, `FromJSON1`, and
135* Added `Data.Aeson.Encoding` with functions to safely write `ToJSON`
136 instances using `toEncoding`.
140* A Cabal `fast` flag was added to disable building with optimizations. This drastically speeds up compiling both aeson ***and*** libraries using aeson so it is recommended to enable it during development. With cabal-install you can `cabal install aeson -ffast` and with stack you can add a flag section to your stack.yaml:
147* Added list specific members to `ToJSON` and `FromJSON` classes. In
148 the same way `Read` and `Show` handle lists specifically. This
149 removes need for overlapping instances to handle `String`.
151* Added a new `sumEncoding` option `UntaggedValue` which prevents
152 objects from being tagged with the constructor name.
154* JSONPaths are now tracked in instances derived with template-haskell
157* Get rid of redundancy of JSONPath error messages in nested records.
159 `eitherDecode "{\"x\":{\"a\": [1,2,true]}}" :: Either String Y`
161 `Error in $.x.a[2]: failed to parse field" x: failed to parse field a: expected Int, encountered Boolean`
162 and now yields `Error in $.x.a[2]: expected Int, encountered Boolean"`.
164 Some users might prefer to insert `modifyFailure` themselves to
165 customize error messages, which previously prevented the use of
168* Backwards compatibility with `bytestring-0.9` using the
169 `bytestring-builder` compatibility package.
171* Export `decodeWith`, `decodeStrictWith`, `eitherDecodeWith`, and
172 `eitherDecodeStrictWith` from `Data.Aeson.Parser`. This allows
173 decoding using explicit parsers instead of using `FromJSON`
176* Un-orphan internal instances to render them in haddocks.
180* Integral `FromJSON` instances now only accept integral
181 values. E.g. parsing `3.14` to `Int` fails instead of succeeding
184* Over/underflows are now caught for bounded numeric types.
186* Remove the `contents` field encoding with `allNullaryToStringTag = False`,
187 giving us `{ "tag" : "c1" }` instead of `{ "tag" : "c1", contents : [] }`.
188 The contents field is optional when parsing so this is only a breaking
189 change for ToJSON instances.
191* Fix a bug where `genericToEncoding` with `unwrapUnaryRecords = True`
192 would produce an invalid encoding: `"unwrap\":""`.
194* `ToJSON` instances using `genericToEncoding` and `omitNothingFields`
195 no longer produce invalid JSON.
197* Added instances for `DList`, `Compose`, `Product`, `Sum`.
201* Enable `PolyKinds` to generalize `Proxy`, `Tagged`, and `Const` instances.
202* Add `unsafeToEncoding` in `Data.Aeson.Types`, use with care!
206* Fix build with `base >= 4.8` and `unordered-containers < 0.2.6`.
210* Fix build on TH-less GHCs
214* Fix build with `base < 4.8` and `unordered-containers < 0.2.6`.
215* Add missing field in docs for `defaultOptions`.
219* Fixes a bug where the hashes of equal values could differ.
223The only changes are added instances.
226* `ToJSON a => ToJSON (NonEmpty a)`
227* `FromJSON a => FromJSON (NonEmpty a)`
229* `FromJSON (Proxy a)`
230* `ToJSON b => ToJSON (Tagged a b)`
231* `FromJSON b => FromJSON (Tagged a b)`
232* `ToJSON a => ToJSON (Const a b)`
233* `FromJSON a => FromJSON (Const a b)`
235These are now available for older GHCs:
241This release should be close to backwards compatible with aeson 0.9.
243If you are upgrading from aeson 0.10 it might be easier to go back in
244history to the point you were still using 0.9.
248* Revert `.:?` to behave like it did in 0.9. If you want the 0.10
249 behavior use `.:!` instead.
251* Revert JSON format of `Either` to 0.9, `Left` and `Right` are now
252 serialized with an initial uppercase letter. If you want the names
253 in lowercase you can add a newtype with an instance.
255* All `ToJSON` and `FromJSON` instances except for `[a]` are no longer
256 `OVERLAPPABLE`. Mark your instance as `OVERLAPPING` if it overlaps
257 any of the other aeson instances.
259* All `ToJSON` and `FromJSON` instances except for `[Char]` are no
260 longer incoherent, this means you may need to replace your
261 incoherent instances with a newtyped instance.
265* Introduce `.:!` that behaves like `.:?` did in 0.10.
267* Allow `HH:MM` format for `ZonedTime` and `UTCTime`.
268 This is one of the formats allowed by
269 [ISO 8601](https://en.wikipedia.org/wiki/ISO_8601#Times).
271* Added `ToJSON` and `FromJSON` instances for the
272 `Version`, `Ordering`, and `Natural` types.
276* JSONPath identifiers are now escaped if they contain invalid characters.
278* Fixed JSONPath messages for Seq to include indices.
280* Fixed JSONPath messages for Either to include `left`/`right`.
282* Fix missing quotes surrounding time encodings.
284* Fix #293: Type error in TH when using `omitNothingFields = True`.
288* Various updates to support GHC 8.
293## Performance improvements
295* Direct encoding via the new `toEncoding` method is over 2x faster
296 than `toJSON`. (You must write or code-gen a `toEncoding`
297 implementation to unlock this speedup. See below for details.)
299* Improved string decoding gives a 12% speed win in parsing
300 string-heavy JSON payloads (very common).
302* Encoding and decoding of time-related types are 10x faster (!!) as a
303 result of bypassing `Data.Time.Format` and the arbitrary-precision
306* When using `toEncoding`, `[Char]` can be encoded without a conversion to
307 `Text`. This is fast and efficient.
309* Parsing into an `Object` is now 5% faster and more
310 allocation-efficient.
312## SUBTLE API CHANGES, READ CAREFULLY
314With the exception of long-deprecated code, the API changes below
315**should be upwards compatible** from older versions of `aeson`. If you run
316into upgrade problems, please file an issue with details.
318* The `ToJSON` class has a new method, `toEncoding`, that allows
319 direct encoding from a Haskell value to a lazy bytestring without
320 construction of an intermediate `Value`.
322 The performance benefits of direct encoding are significant: more
323 than 2x faster than before, with less than 1/3 the memory usage.
325 To preserve API compatibility across upgrades from older versions of
326 this library, the default implementation of `toEncoding` uses
327 `toJSON`. You will *not* see any performance improvement unless you
328 write an implementation of `toEncoding`, which can be very simple:
331 instance ToJSON Coord where
332 toEncoding = genericToEncoding defaultOptions
335 (Behind the scenes, the `encode` function uses `toEncoding` now, so
336 if you implement `toEncoding` for your types, you should see a
337 speedup immediately.)
339 If you use Template Haskell or GHC Generics to auto-generate your
340 `ToJSON` instances, you'll benefit from fast toEncoding
341 implementations for free!
343* When converting from a `Value` to a target Haskell type, `FromJSON`
344 instances now provide much better error messages, including a
345 complete JSON path from the root of the object to the offending
346 element. This greatly eases debugging.
348* It is now possible to use Template Haskell to generate `FromJSON`
349 and `ToJSON` instances for types in data families.
351* If you use Template Haskell or generics, and used to use the
352 `camelTo` function to rename fields, the new `camelTo2` function is
353 smarter. For example, `camelTo` will rename `CamelAPICase` to
354 `camelapi_case` (ugh!), while `camelTo2` will map it to
355 `camel_api_case` (yay!).
357* New `ToJSON` and `FromJSON` instances for the following time-related
358 types: `Day`, `LocalTime`.
360* `FromJSON` `UTCTime` parser accepts the same values as for `ZonedTime`,
361 but converts any time zone offset into a UTC time.
363* The `Result` type is now an instance of `Foldable` and `Traversable`.
365* The `Data.Aeson.Generic` module has been removed. It was deprecated in
368* GHC 7.2 and older are no longer supported.
370* The instance of `Monad` for the `Result` type lacked an implementation
371 of `fail` (oops). This has been corrected.
373* Semantics of `(.:?)` operator are changed. It's doesn't anymore accept
374 present `Null` value.
376* Added `(Foldable t, ToJSON a) => ToJSON (t a)` overlappable instance.
377 You might see `No instance for (Foldable YourPolymorphicType) arising from a
378 use of ‘.=’` -errors due this change.
382* A stray export of `encodeToBuilder` got away!
386* The `json` and `json'` parsers are now synonyms for `value` and
387 `value'`, in conformance with the looser semantics of RFC 7159.
389* Renamed `encodeToByteStringBuilder` to the more compact
394* The dependency on the `unordered-containers` package was too lax,
395 and has been corrected.
399* Encoding a `Scientific` value with a huge exponent is now handled
400 efficiently. (This would previously allocate a huge
401 arbitrary-precision integer, potentially leading to a denial of
404* Handling of strings that contain backslash escape sequences is
405 greatly improved. For a pathological string containing almost a
406 megabyte of consecutive backslashes, the new implementation is 27x
407 faster and uses 42x less memory.
409* The `ToJSON` instance for `UTCTime` is rendered with higher
410 (picosecond) resolution.
412* The `value` parser now correctly handles leading whitespace.
414* New instances of `ToJSON` and `FromJSON` for `Data.Sequence` and
415 `Data.Functor.Identity`. The `Value` type now has a `Read` instance.
417* `ZonedTime` parser ordering now favours the standard `JSON` format,
418 increasing efficiency in the common case.
420* Encoding to a `Text.Builder` now escapes `'<'` and `'>'` characters,
425* Fix `ToJSON` instance for 15-tuples (see #223).
433* Add `ToJSON` and `FromJSON` instances for tuples of up to 15
438* Major compiler and library compatibility changes: we have dropped
439 support for GHC older than 7.4, `text` older than 1.1, and
440 `bytestring` older than 0.10.4.0. Supporting the older versions had
441 become increasingly difficult, to the point where it was no longer
446* The performance of encoding to and decoding of bytestrings have both
447 improved by up to 2x, while also using less memory.
449* New dependency: the `scientific` package lets us parse floating point
450 numbers more quickly and accurately.
452* `eitherDecode`, `decodeStrictWith`: fixed bugs.
454* Added `FromJSON` and `ToJSON` instances for `Tree` and `Scientific`.
456* Fixed the `ToJSON` instances for `UTCTime` and `ZonedTime`.
460* Much improved documentation.
462* Angle brackets are now escaped in JSON strings, to help avoid XSS
465* Fixed up handling of nullary constructors when using generic
468* Added `ToJSON`/`FromJSON` instances for:
471 * ISO-8601 dates: `UTCTime`, `ZonedTime`, and `TimeZone`
473* Added accessor functions for inspecting `Value`s.
475* Added `eitherDecode` function that returns an error message if
480* This release introduces a slightly obscure, but
481 backwards-incompatible, change.
483 In the generic APIs of versions 0.4 and 0.5, fields whose names
484 began with a `"_"` character would have this character removed. This
485 no longer occurs, as it was both buggy and surprising
486 (https://github.com/bos/aeson/issues/53).
488* Fixed a bug in generic decoding of nullary constructors
489 (https://github.com/bos/aeson/issues/62).
493* When used with the UTF-8 encoding performance improvements
494 introduced in version 0.11.1.12 of the `text` package, this release
495 improves `aeson`'s JSON encoding performance by 33% relative to
498 As part of achieving this improvement, an API change was necessary.
499 The `fromValue` function in the `Data.Aeson.Encode` module now uses
500 the `text` package's `Builder` type instead of the `blaze-builder`
501 package's `Builder` type.
505* The new `decode` function complements the longstanding `encode`
506 function, and makes the API simpler.
508* New examples make it easier to learn to use the package
509 (https://github.com/bos/aeson/tree/master/examples).
513 `aeson`'s support for data-type generic programming makes it
514 possible to use JSON encodings of most data types without writing
515 any boilerplate instances.
517 Thanks to Bas Van Dijk, `aeson` now supports the two major schemes
518 for doing datatype-generic programming:
520 * the modern mechanism, built into GHC itself
521 (http://www.haskell.org/ghc/docs/latest/html/users_guide/generic-programming.html)
523 * the older mechanism, based on SYB (aka "scrap your
526 The modern GHC-based generic mechanism is fast and terse: in fact,
527 its performance is generally comparable in performance to
528 hand-written and TH-derived `ToJSON` and `FromJSON` instances. To
529 see how to use GHC generics, refer to `examples/Generic.hs`.
531 The SYB-based generics support lives in `Data.Aeson.Generic` and is
532 provided mainly for users of GHC older than 7.2. SYB is far slower
533 (by about 10x) than the more modern generic mechanism. To see how
534 to use SYB generics, refer to `examples/GenericSYB.hs`.
536* We switched the intermediate representation of JSON objects from
537 `Data.Map` to `Data.HashMap` which has improved type conversion
540* Instances of `ToJSON` and `FromJSON` for tuples are between 45% and 70%
545 This version of aeson makes explicit the decoupling between
546 *identifying* an element of a JSON document and *converting* it to
547 Haskell. See the `Data.Aeson.Parser` documentation for details.
549 The normal `aeson` `decode` function performs identification
550 strictly, but defers conversion until needed. This can result in
551 improved performance (e.g. if the results of some conversions are
552 never needed), but at a cost in increased memory consumption.
554 The new `decode'` function performs identification and conversion
555 immediately. This incurs an up-front cost in CPU cycles, but
556 reduces reduce memory consumption.
keysesc to close
- j k
- move down, up
- enter
- open
- esc
- back, or up a level
- 1 2 3 4
- readme, files, commits, refs
- /
- filter
- y
- copy the clone command
- c
- compare the chosen commit with another (or shift-click both)
- ?
- show or hide this list