jsx (v2.0alpha)

an erlang application for consuming, producing and manipulating json. inspired by yajl

jsx is built via rebar and continuous integration testing provided courtesy travis

current status:

jsx is released under the terms of the MIT license

copyright 2010-2013 alisdair sullivan

really important note

this is a preview of the 2.0 release. there are lots of changes. see CHANGES.md for the overview or read this document for the details. use master branch if you want the last 1.x version

index

• quickstart
• description
  • json <-> erlang mapping
  • incomplete input
• data types
  • json_term()
  • json_text()
  • event()
  • option()
• exports
  • encoder/3, decoder/3 & parser/3
  • decode/1,2
  • encode/1,2
  • format/1,2
  • minify/1
  • prettify/1
  • is_json/1,2
  • is_term/1,2
• callback exports
  • Module:init/1
  • Module:handle_event/2
• acknowledgements

quickstart

to build the library and run tests

$ rebar compile
$ rebar eunit

or, to build using hipe

$ rebar -C hipe.cfg compile
$ rebar -C hipe.cfg eunit

to convert a utf8 binary containing a json string into an erlang term

1> jsx:decode(<<"{\"library\": \"jsx\", \"awesome\": true}">>).
[{<<"library">>,<<"jsx">>},{<<"awesome">>,true}]
2> jsx:decode(<<"[\"a\",\"list\",\"of\",\"words\"]">>).
[<<"a">>, <<"list">>, <<"of">>, <<"words">>]

to convert an erlang term into a utf8 binary containing a json string

1> jsx:encode([{<<"library">>,<<"jsx">>},{<<"awesome">>,true}]).
<<"{\"library\": \"jsx\", \"awesome\": true}">>
2> jsx:encode([<<"a">>, <<"list">>, <<"of">>, <<"words">>]).
<<"[\"a\",\"list\",\"of\",\"words\"]">>

to check if a binary or a term is valid json

1> jsx:is_json(<<"[\"this is json\"]">>).
true
2> jsx:is_json("[\"this is not\"]").
false
3> jsx:is_term([<<"this is a term">>]).
true
4> jsx:is_term([this, is, not]).
false

to minify some json

1> jsx:minify(<<"{
  \"a list\": [
    1,
    2,
    3
  ]
}">>).
<<"{\"a list\":[1,2,3]}">>

to prettify some json

1> jsx:prettify(<<"{\"a list\":[1,2,3]}">>).
<<"{
  \"a list\": [
    1,
    2,
    3
  ]
}">>

description

jsx is an erlang application for consuming, producing and manipulating json

jsx follows the json spec as closely as possible with allowances for real world usage

jsx is pragmatic. the json spec allows extensions so jsx extends the spec in a number of ways. see the section on strict in options below though

json has no official comments but this parser allows c/c++ style comments. anywhere whitespace is allowed you can insert comments (both // ... and /* ... */)

all jsx decoder input should be utf8 encoded binaries. sometimes you get binaries that are almost but not quite valid utf8 whether due to improper escaping or poor encoding. jsx replaces invalid codepoints and poorly formed sequences with the unicode replacement character (u+FFFD)

json only allows keys and strings to be delimited by double quotes (u+0022) but javascript allows them to be delimited by single quotes (u+0027) as well. jsx follows javascript in this. strings that start with single quotes can contain double quotes but must end with single quotes and must escape any single quotes they contain

json and jsx only recognize escape sequences as outlined in the json spec. it just ignores bad escape sequences

json <-> erlang mapping

json	erlang
number	integer() and float()
string	binary() and atom()
true, false and null	true, false and null
array	[] and [JSON]
object	[{}] and [{binary() OR atom(), JSON}]

• numbers

  javascript and thus json represent all numeric values with floats. there's no reason for erlang -- a language that supports arbitrarily large integers -- to restrict all numbers to the ieee754 range

  whenever possible, jsx will interpret json numbers that look like integers as integers. other numbers will be converted to erlang's floating point type, which is nearly but not quite iee754. negative zero is not representable in erlang (zero is unsigned in erlang and 0 is equivalent to -0) and will be interpreted as regular zero. numbers not representable are beyond the concern of this implementation, and will result in parsing errors

  when converting from erlang to json, floats are represented with their shortest representation that will round trip without loss of precision. this means that some floats may be superficially dissimilar (although functionally equivalent). for example, 1.0000000000000001 will be represented by 1.0

• strings

  json strings must be unicode encoded binaries or erlang atoms. in practice, because jsx only accepts utf8 binaries all binary strings must be utf8. in addition to being unicode json strings restrict a number of codepoints and define a number of escape sequences

  json string escapes of the form \uXXXX will be converted to their equivalent codepoints during parsing. this means control characters and other codepoints disallowed by the json spec may be encountered in resulting strings. the utf8 restriction means the surrogates are explicitly disallowed. if a string contains escaped surrogates (u+d800 to u+dfff) they are interpreted but only when they form valid surrogate pairs. surrogates encountered otherwise are replaced with the replacement codepoint (u+fffd)

  all erlang strings are represented by valid utf8 encoded binaries. the encoder will check strings for conformance. noncharacters (like u+ffff) are allowed in erlang utf8 encoded binaries, but will be replaced in strings passed to the encoder (although, again, see options)

  this implementation performs no normalization on strings beyond that detailed here. be careful when comparing strings as equivalent strings may have different utf8 encodings

• true, false and null

  the json primitives true, false and null are represented by the erlang atoms true, false and null. surprise

• arrays

  json arrays are represented with erlang lists of json values as described in this section

• objects

  json objects are represented by erlang proplists. the empty object has the special representation [{}] to differentiate it from the empty list. ambiguities like [true, false] prevent the use of the shorthand form of property lists using atoms as properties so all properties must be tuples. all keys must be encoded as in string or as atoms or integers (which will be escaped and converted to binaries for presentation to handlers). values should be valid json values

incomplete input

jsx can handle incomplete json texts. if the option stream is passed to the decoder or parser and if a partial json text is parsed, rather than returning a term from your callback handler, jsx returns {incomplete, F} where F is a function with an identical API to the anonymous fun returned from decoder/3, encoder/3 or parser/3. it retains the internal state of the parser at the point where input was exhausted. this allows you to parse as you stream json over a socket or file descriptor, or to parse large json texts without needing to keep them entirely in memory

however, it is important to recognize that jsx is conservative by default. jsx will not consider the parsing complete even when input is exhausted and the json text is unambiguously incomplete. to end parsing call the incomplete function with the argument end_stream like:

1> {incomplete, F} = jsx:decode(<<"[">>, [stream]).
{incomplete,#Fun<jsx_decoder.1.122947756>}
2> F(end_stream).
** exception error: bad argument
3> {incomplete, G} = F(<<"]">>).
{incomplete,#Fun<jsx_decoder.1.122947756>}
4> G(end_stream).
[]

data types

json_term()

json_term() = [json_term()]
    | [{binary() | atom(), json_term()}]
    | true
    | false
    | null
    | integer()
    | float()
    | binary()
    | atom()

the erlang representation of json. binaries should be utf8 encoded, or close at least

json_text()

json_text() = binary()

a utf8 encoded binary containing a json string

event()

event() = start_object
    | end_object
    | start_array
    | end_array
    | {key, binary()}
    | {string, binary()}
    | {integer, integer()}
    | {float, float()}
    | {literal, true}
    | {literal, false}
    | {literal, null}
    | end_json

the subset of token() emitted by the decoder and encoder to handlers

option()

option() = escaped_forward_slashes
    | escaped_strings
    | unescaped_jsonp
    | dirty_strings
    | strict
    | {strict, [strict_option()]}
    | stream
    | {incomplete_handler, fun()}
    | {error_handler, fun()}

strict_option() = comments
    | utf8
    | single_quotes
    | escapes

jsx functions all take a common set of options. not all flags have meaning in all contexts, but they are always valid options. functions may have additional options beyond these. see individual function documentation for details

• escaped_forward_slashes

  json strings are escaped according to the json spec. this means forward slashes (solidus) are only escaped when this flag is present. otherwise they are left unescaped. you may want to use this if you are embedding json directly into a html or xml document

• escaped_strings

  by default both the encoder and decoder return strings as utf8 binaries appropriate for use in erlang. escape sequences that were present in decoded terms are converted into the appropriate codepoint while encoded terms are unaltered. this flag escapes strings as if for output in json, removing control codes and problematic codepoints and replacing them with the appropriate escapes

• unescaped_jsonp

  javascript interpreters treat the codepoints u+2028 and u+2029 as significant whitespace. json strings that contain either of these codepoints will be parsed incorrectly by some javascript interpreters. by default, these codepoints are escaped (to \u2028 and \u2029, respectively) to retain compatibility. this option simply removes that escaping

• dirty_strings

  json escaping is lossy; it mutates the json string and repeated application can result in unwanted behaviour. if your strings are already escaped (or you'd like to force invalid strings into "json" you monster) use this flag to bypass escaping. this can also be used to read in really invalid json strings. everything between unescaped quotes are passed as is to the resulting string term. note that this takes precedence over any other options

• strict

  as mentioned earlier, jsx is pragmatic. if you're more of a json purist or you're really into bdsm stricter adherence to the spec is possible. the following restrictions are available

  • comments

    comments are disabled and result in a badarg error

  • utf8

    invalid codepoints and malformed unicode result in badarg errors

  • single_quotes

    only keys and strings delimited by double quotes (u+0022) are allowed. the single quote (u+0027) results in a badarg error

  • escapes

    escape sequences not adhering to the json spec result in a badarg error

  any combination of these can be passed to jsx by using {strict, [strict_option()]}. strict is equivalent to {strict, [comments, bad_utf8, single_quotes, escapes]}

• stream

  see incomplete input

exports

encoder/3, decoder/3 & parser/3

decoder(Module, Args, Opts) -> Fun((JSONText) -> any())
encoder(Module, Args, Opts) -> Fun((JSONTerm) -> any())
parser(Module, Args, Opts) -> Fun((Tokens) -> any())

  Module = atom()
  Args = any()
  Opts = [option()]
  JSONText = json_text()
  JSONTerm = json_term()
  Tokens = event() | [event()]

jsx is a json compiler with interleaved tokenizing, syntactic analysis and semantic analysis stages. included are two tokenizers; one that handles json texts (decoder/3) and one that handles erlang terms (encoder/3). there is also an entry point to the syntactic analysis stage for use with user-defined tokenizers (parser/3)

all three functions return an anonymous function that takes the appropriate type of input and returns the result of performing semantic analysis, the tuple {incomplete, F} where F is a new anonymous function (see incomplete input) or a badarg error exception if syntactic analysis fails

Module is the name of the callback module

Args is any term that will be passed to Module:init/1 prior to syntactic analysis to produce an initial state

Opts are detailed here

check out callback module documentation for details of the callback module interface

decode/1,2

decode(JSON) -> Term
decode(JSON, Opts) -> Term

  JSON = json_text()
  Term = json_term()
  Opts = [option() | labels | {labels, Label}]
    Label = binary | atom | existing_atom | attempt_atom
    F = fun((any()) -> any())

decode parses a json text (a utf8 encoded binary) and produces an erlang term

the option labels controls how keys are converted from json to erlang terms. binary (the default behavior) does no conversion beyond normal escaping. atom converts keys to erlang atoms and results in a badarg error if the keys fall outside the range of erlang atoms. existing_atom is identical to atom except it will not add new atoms to the atom table and will result in a badarg error if the atom does not exist. attempt_atom will convert keys to atoms when they exist, and leave them as binary otherwise

raises a badarg error exception if input is not valid json

encode/1,2

encode(Term) -> JSON
encode(Term, Opts) -> JSON

  Term = json_term()
  JSON = json_text()
  Opts = [option() | space | {space, N} | indent | {indent, N}]
    F = fun((any()) -> any())
    N = pos_integer()

encode converts an erlang term into json text (a utf8 encoded binary)

the option {space, N} inserts N spaces after every comma and colon in your json output. space is an alias for {space, 1}. the default is {space, 0}

the option {indent, N} inserts a newline and N spaces for each level of indentation in your json output. note that this overrides spaces inserted after a comma. indent is an alias for {indent, 1}. the default is {indent, 0}

raises a badarg error exception if input is not a valid erlang representation of json

format/1,2

format(JSON) -> JSON
format(JSON, Opts) -> JSON

  JSON = json_text()
  Opts = [option() | space | {space, N} | indent | {indent, N}]
    N = pos_integer()

format parses a json text (a utf8 encoded binary) and produces a new json text according to the format rules specified by Opts

the option {space, N} inserts N spaces after every comma and colon in your json output. space is an alias for {space, 1}. the default is {space, 0}

the option {indent, N} inserts a newline and N spaces for each level of indentation in your json output. note that this overrides spaces inserted after a comma. indent is an alias for {indent, 1}. the default is {indent, 0}

raises a badarg error exception if input is not valid json

minify/1

minify(JSON) -> JSON

  JSON = json_text()

minify parses a json text (a utf8 encoded binary) and produces a new json text stripped of whitespace

raises a badarg error exception if input is not valid json

prettify/1

prettify(JSON) -> JSON

  JSON = json_text()

prettify parses a json text (a utf8 encoded binary) and produces a new json text equivalent to format(JSON, [{space, 1}, {indent, 2}])

raises a badarg error exception if input is not valid json

is_json/1,2

is_json(MaybeJSON) -> true | false
is_json(MaybeJSON, Opts) -> true | false

  MaybeJSON = any()
  Opts = options()

returns true if input is a valid json text, false if not

what exactly constitutes valid json may be altered

is_term/1,2

is_term(MaybeJSON) -> true | false
is_term(MaybeJSON, Opts) -> true | false

  MaybeJSON = any()
  Opts = options()

returns true if input is a valid erlang representation of json, false if not

what exactly constitutes valid json may be altered via options

callback exports

the following functions should be exported from a jsx callback module

Module:init/1

Module:init(Args) -> InitialState

  Args = any()
  InitialState = any()

whenever any of encoder/3, decoder/3 or parser/3 are called, this function is called with the Args argument provided in the calling function to obtain InitialState

Module:handle_event/2

Module:handle_event(Event, State) -> NewState

  Event = [event()]
  State = any()
  NewState = any()

semantic analysis is performed by repeatedly calling handle_event/2 with a stream of events emitted by the tokenizer and the current state. the new state returned is used as the input to the next call to handle_event/2. the following events must be handled:

• start_object

  the start of a json object

• end_object

  the end of a json object

• start_array

  the start of a json array

• end_array

  the end of a json array

• {string, binary()}

  a json string. it will usually be a utf8 encoded binary. see the options for possible exceptions. note that keys are also json strings

• {integer, integer()}

  an erlang integer (bignum)

• {float, float()}

  an erlang float

• {literal, true}

  the atom true

• {literal, false}

  the atom false

• {literal, null}

  the atom null

• end_json

  this event is emitted when syntactic analysis is completed. you should do any cleanup and return the result of your semantic analysis

acknowledgements

jsx wouldn't be what it is without the contributions of paul davis, lloyd hilaiel, john engelhart, bob ippolito, fernando benavides, alex kropivny, steve strong, michael truog, devin torres, dmitry kolesnikov, emptytea, john daily, ola bäckström, joseph crowe, patrick gombert, eskuat and max lapshin