This RFC defines a simple schema for tabular data. The schema is designed to be expressible in JSON.

Last Updated17 November 2016
Created12 November 2012


The key words “MUST”, “MUST NOT”, “REQUIRED”, “SHALL”, “SHALL NOT”, “SHOULD”, “SHOULD NOT”, “RECOMMENDED”, “MAY”, and “OPTIONAL” in this document are to be interpreted as described in RFC 2119.


  • 1.0.0-pre15: add calendar units gyear and gyearmonth (#105), tweak pattern support for date/time types #260
  • 1.0.0-pre14: add support for missingValue (#97)
  • 1.0.0-pre13: remove null datatype (#262)
  • 1.0.0-pre12: add support for new number properties such as decimalChar(#246)
  • 1.0.0-pre11: add new field property: rdfType (#217)
  • 1.0.0-pre10: add new field types: duration (#210)
  • 1.0.0-pre9: make date formats stricter for default issue. Define value of fmt:PATTERN for dates issue
  • 1.0-pre8: Rename contraints.oneOf to contraints.enum issue
  • 1.0-pre7: Add contraints.oneOf issue
  • 1.0-pre6: clarify types and formats issue
  • 1.0-pre5: add type validation issue
  • 1.0-pre4: add foreign key support - see this issue
  • 1.0-pre3.2: add primary key support (see this issue)
  • 1.0-pre3.1: breaking changes.
    • label (breaking) changed to title - see Closer alignment with JSON Schema
    • id changed to name (with slight alteration in semantics - i.e. SHOULD be unique but no longer MUST be unique)

Table of Contents


A Table consists of a set of rows. Each row has a set of fields (columns). We usually expect that each Row has the same set of fields and thus we can talk about the fields for the table as a whole.

In cases of tables in spreadsheets or CSV files we often interpret the first row as a header row giving the names of the fields. By contrast, in other situations, e.g. tables in SQL databases the fields (columns) are explicitly designated.

To illustrate here’s a classic spreadsheet table:

field     field
  |         |
  |         |
  V         V

 A     |    B    |    C    |    D      <--- Row
 valA  |   valB  |  valC   |   valD    <--- Row

In JSON a table would be:

  { "A": value, "B": value, ... },
  { "A": value, "B": value, ... },


A JSON Table Schema consists of:

  • a required list of field descriptors
  • optionally, a primary key description
  • optionally, a foreign _key description

A schema is described using JSON. This might exist as a standalone document or may be embedded within another JSON structure, e.g. as part of a data package description.


A schema has the following structure:

  # fields is an ordered list of field descriptors
  # one for each field (column) in the table
  "fields": [
    # a field-descriptor
      "name": "name of field (e.g. column name)",
      "title": "A nicer human readable label or title for the field",
      "type": "A string specifying the type",
      "format": "A string specifying a format",
      "description": "A description for the field"
    ... more field descriptors
  # (optional) specification of the primary key
  "primaryKey": ...
  # (optional) specification of the foreign keys
  "foreignKeys": ...


That is, a JSON Table Schema is:

  • a Hash which MUST contain a key fields
  • fields MUST be an array where each entry in the array is a field descriptor. (Structure and usage described below)
  • the Hash MAY contain a property primaryKey (structure and usage specified below)
  • the Hash MAY contain a property foreignKeys (structure and usage specified below)
  • the Hash MAY contain any number of other properties (not defined in this spec)

Field Descriptors

A field descriptor is a simple JSON hash that describes a single field. The descriptor provides additional human-readable documentation for a field, as well as additional information that may be used to validate the field or create a user interface for data entry.

At a minimum a field descriptor will contain at least a name key, but MAY have additional keys as described below:

  "name": "name of field (e.g. column name)",
  "title": "A nicer human readable label or title for the field",
  "type": "A string specifying the type",
  "format": "A string specifying a format",
  "missingValue": "",
  "description": "A description for the field",
  "constraints": {
      # a constraints-descriptor
  • a field descriptor MUST be a Hash
  • the field descriptor Hash MUST contain a name property. This property SHOULD correspond to the name of field/column in the data file (if it has a name). As such it SHOULD be unique (though it is possible, but very bad practice, for the data file to have multiple columns with the same name). Additionally, name SHOULD be considered case sensitive. In practice, case sensitivity for names can be limiting in certain scenarios, so consumers MAY choose to ignore case sensitivity for name values.
  • the field descriptor Hash MAY contain any number of other properties
  • specific properties that MAY be included in the Hash and whose meaning is defined in this spec are:

    • title: A nicer human readable label or title for the field
    • description: A description for this field e.g. “The recipient of the funds”
    • type: The type of the field (string, number etc) - see below for more detail. If type is not provided a consumer should assume a type of “string”.
    • format: A description of the format e.g. “DD.MM.YYYY” for a date. See below for more detail.
    • missingValue: a field value which should be interpreted as missing data (null). See below for
    • constraints: A constraints descriptor that can be used by consumers to validate field values

Field Constraints

A set of constraints can be associated with a field. These constraints can be used to validate data against a JSON Table Schema. The constraints might be used by consumers to validate, for example, the contents of a data package, or as a means to validate data being collected or updated via a data entry interface.

A constraints descriptor is a JSON hash. It MAY contain any of the following keys.

  • required – A boolean value which indicates whether a field must have a value in every row of the table. An empty string is considered to be a missing value.
  • minLength – An integer that specifies the minimum length of a value. Supported field types are sequences, such as string and array, and collections containing items, such as object.
  • maxLength – An integer that specifies the maximum length of a value. Supported field types are sequences, such as string and array, and collections containing items, such as object.
  • unique – A boolean. If true, then all values for that field MUST be unique within the data file in which it is found. This defines a unique key for a row although a row could potentially have several such keys.
  • pattern – A regular expression that can be used to test field values. If the regular expression matches then the value is valid. Values will be treated as a string of characters. It is recommended that values of this field conform to the standard XML Schema regular expression syntax. See also this reference.
  • minimum – specifies a minimum value for a field. This is different to minLength which checks the number of items in the value. A minimum value constraint checks whether a field value is greater than or equal to the specified value. The range checking depends on the type of the field. E.g. an integer field may have a minimum value of 100; a date field might have a minimum date. If a minimum value constraint is specified then the field descriptor MUST contain a type key. Supported field types are integer, number, date, time and datetime.
  • maximum – as above, but specifies a maximum value for a field.
  • enum – An array of values, where each value MUST comply with the type and format of the field. The field value must exactly match a value in the enum array.

The constraints listed above may also define a list of supported field types. Implementations SHOULD report an error if an attempt is made to evaluate a value against an unsupported constraint.

A constraints descriptor may contain multiple constraints, in which case a consumer MUST apply all the constraints when determining if a field value is valid.

A data file, e.g. an entry in a data package, is considered to be valid if all of its fields are valid according to their declared type and constraints.

Field Types and Formats

A field’s type property is a string indicating the type of this field.

A field’s format property is a string, being a keyword indicating a format for the field type.

Both type and format are optional: in a field descriptor, the absence of a type property indicates that the field is of the type “string”, and the absence of a format property indicates that the field’s type format is “default”.

Types are based on the type set of json-schema with some additions and minor modifications (cf other type lists include those in Elasticsearch types).

The type list with associated formats and other related properties is as follows.


The field contains strings, that is, sequences of characters.


  • default: any valid string.
  • email: A valid email address.
  • uri: A valid URI.
  • binary: A base64 encoded string representing binary data.
  • uuid: A string that is a uuid.


The field contains numbers of any kind including decimals.

The lexical formatting follows that of decimal in XMLSchema: a non-empty finite-length sequence of decimal digits separated by a period as a decimal indicator. An optional leading sign is allowed. If the sign is omitted, “+” is assumed. Leading and trailing zeroes are optional. If the fractional part is zero, the period and following zero(es) can be omitted. For example: ‘-1.23’, ‘12678967.543233’, ‘+100000.00’, ‘210’.

The following special string values are permitted (case need not be respected):

  • NaN: not a number
  • INF: positive infinity
  • -INF: negative infinity

A number MAY also have a trailing:

  • exponent: this MUST consist of an E followed by an optional + or - sign followed by one or more decimal digits (0-9)
  • percentage: the percentage sign: “%. In conversion percentages should be divided by 100.

If both exponent and percentages are present the percentage MUST follow the exponent e.g. ‘53E10%’ (equals 5.3).

This lexical formatting may be modified using these additional properties:

  • decimalChar: A string whose value is used to represent a decimal point within the number. The default value is “.”.
  • groupChar: A string whose value is used to group digits within the number. The default value is null. A common value is “,” e.g. “100,000”.
  • currency: A number that may include additional currency symbols.

format: no options (other than the default).


The field contains integers - that is whole numbers.

Integer values are indicated in the standard way for any valid integer.

format: no options (other than the default).


The field contains boolean (true/false) data.

Boolean values can be indicated with the following strings (case-insensitive so, for example, “Y” and “y” are both acceptable):

  • true: ‘yes’, ‘y’, ‘true’, ‘t’, ‘1’
  • false: ‘no’, ‘n’, ‘false’, ‘f’, ‘0’

format: no options (other than the default).


The field contains data which is valid JSON.

format: no options (other than the default).


The field contains data that is a valid JSON format arrays.

format: no options (other than the default).


datetime; date; time

Tthe field contains temporal values such as dates, times and date-times.

format: (datetime, date and time share the these same options)

  • default: An ISO8601 format string.
    • date: This MUST be in ISO8601 format YYYY-MM-DD
    • datetime: a date-time. This MUST be in ISO 8601 format of YYYY-MM-DDThh:mm:ssZ in UTC time
    • time: a time without a date
  • any: Any parsable representation of the type. The implementing library can attempt to parse the datetime via a range of strategies. An example is dateutil.parser.parse from the python-dateutils library.
  • {PATTERN}: date/time values in this field can be parsed according to {PATTERN}. {PATTERN} MUST follow the syntax of standard Python / C strptime. (That is, values in the this field should be parseable by Python / C standard strptime using PATTERN). Example: %d %b %y would correspond to dates like: 30 Nov 14


A calendar year as per XMLSchema gYear.

Usual lexical reprentation is YYYY. There are no format options.


A specific month in a specific year as per XMLSchema gYearMonth.

Usual lexical representation is: YYYY-MM. There are no format options.


A duration of time.

The lexical representation for duration is the ISO 8601 extended format PnYnMnDTnHnMnS, where nY represents the number of years, nM the number of months, nD the number of days, ‘T’ is the date/time separator, nH the number of hours, nM the number of minutes and nS the number of seconds. The number of seconds can include decimal digits to arbitrary precision. Date and time elements including their designator may be omitted if their value is zero, and lower order elements may also be omitted for reduced precision. Here we follow the definition of XML Schema duration datatype directly and that definition is implicitly inlined here.

format: no options (other than the default).


The field contains data describing a geographic point.


  • default: A string of the pattern “lon, lat”, where lon is the longitude and lat is the latitude.
  • array: An array of exactly two items, where each item is either a number, or a string parsable as a number, and the first item is lon and the second item is lat.
  • object: A JSON object with exactly two keys, lat and lon


The field contains a JSON object according to GeoJSON or TopoJSON spec.



Any type or format is accepted.

Missing Values

By “missing” we simply mean null or “not present for whatever reason”. Many datasets arrive with missing data values, either because a value was not collected or it never existed.

Missing values may be indicated simply by the value being empty in other cases a special value may have been used e.g. -, NaN, 0, -9999 etc.

The missingValue property provides a way to indicate that these values should be interpreted as equivalent to null.

Strings that indicate missing (null) data values for a field MAY be provided for a field using the the missingValue property.

If present, the missingValue MUST be a single string or an array of strings, for example:

"missingValue": ""
"missingValue": "-"
"missingValue": ["Nan", "-"]

Note: missingValue are strings rather than being the data type of the particular field. This allows for comparison prior to casting and for fields to have missing value which are not of their type, for example a number field to have missing values indicated by -.

The default value of missingValue for a non-string type field is the empty string "". For string type fields there is no default for missingValue (for string fields the empty string "" is a valid value and need not indicate null).

Processing: if a missing value is encountered it SHOULD be converted into the NULL or equivalent value.

Rich Field Types

A richer, “semantic”, description of the “type” of data in a given column MAY be provided using a rdfType property on a field descriptor.

The value of of the rdfType property MUST be the URI of a RDF Class, that is an instance or subclass of RDF Schema Class object

Here is an example using the RDF Class

Country Year Date Value
US 2010
# JSON Table Schema
  fields: [
      "name": "Country",
      "type": "string",
      "rdfType": ""

Primary Key

A primary key is a field or set of fields that uniquely identifies each row in the table.

The primaryKey entry in the schema Hash is optional. If present it specifies the primary key for this table.

The primaryKey, if present, MUST be:

  • Either: an array of strings with each string corresponding to one of the field name values in the fields array (denoting that the primary key is made up of those fields). It is acceptable to have an array with a single value (indicating just one field in the primary key). Strictly, order of values in the array does not matter. However, it is RECOMMENDED that one follow the order the fields in the fields has as client applications may utitlize the order of the primary key list (e.g. in concatenating values together).
  • Or: a single string corresponding to one of the field name values in the fields array (indicating that this field is the primary key). Note that this version corresponds to the array form with a single value (and can be seen as simply a more convenient way of specifying a single field primary key).

Here’s an example:

  "fields": [
      "name": "a"
  "primaryKey": "a"

Here’s an example with an array primary key:

"schema": {
  "fields": [
      "name": "a"
      "name": "b"
      "name": "c"
  "primaryKey": ["a", "c"]

Foreign Keys

Foreign Keys by necessity must be able to reference other data objects. These data objects require a specific structure for the spec to work. Therefore, this spec makes one of two assumptions:
  • Your Foreign Key points to another field/fields within the current JSON Table Schema. In this case you use a special self keyword is employed.
  • Your Foreign Key points to a field/fields in a JSON Table Schema "elsewhere". In this case, the JSON Table Schema must be inside of a resource on Data Package. There are two situations here: EITHER this JSON Table Schema is already situated within a (Tabular) Data Package and the reference is to a resource within this Data Package; OR we are pointing out to a (Tabular) Data Package stored elsewhere e.g. online.

A foreign key is a reference where entries in a given field (or fields) on this table (‘resource’ in data package terminology) is a reference to an entry in a field (or fields) on a separate resource.

The foreignKeys property, if present, MUST be an Array. Each entry in the array must be a foreignKey. A foreignKey MUST be a Hash and:

  • MUST have a property fields. fields is a string or array specifying the field or fields on this resource that form the source part of the foreign key. The structure of the string or array is as per primaryKey above.
  • MUST have a property reference which MUST be a Hash. The Hash
    • MAY have a property datapackage. This property is a string being a url pointing to a Data Package or is the name of a datapackage. If absent the implication is that this is a reference to a resource within the current data package or this is self-referencing foreign key.
    • MUST have a property resource which is the name of the resource within the referenced data package. For self-referencing foreign keys, the value of resource MUST be self.
    • MUST have a property fields which is a string if the outer fields is a string, else an array of the same length as the outer fields, describing the field (or fields) references on the destination resource. The structure of the string or array is as per primaryKey above.

Here’s an example:

  "fields": [
      "name": "state"
  "foreignKeys": [
      "fields": "state"
      "reference": {
        "datapackage": "",
        "resource": "the-resource",
        "fields": "state_id"

An example of a self-referencing foreign key:

  "fields": [
      "name": "parent"
      "name": "id"
  "foreignKeys": [
      "fields": "parent"
      "reference": {
        "datapackage": "",
        "resource": "self",
        "fields": "id"

Appendix: Related Work

See Web-Oriented Data Formats for more details and links for each format.


See Allowed values:

  • string
  • float
  • integer
  • boolean
  • date
  • concept

Google BigQuery

Example schema:

"schema": {
        "mode": "nullable",
        "name": "placeName",
        "type": "string"
        "mode": "nullable",
        "name": "kind",
        "type": "string"
     },  ...


  • string - UTF-8 encoded string up to 64K of data (as opposed to 64K characters).
  • integer - IEEE 64-bit signed integers: [-263-1, 263-1]
  • float - IEEE 754-2008 formatted floating point values
  • boolean - “true” or “false”, case-insensitive
  • record (JSON only) - a JSON object; also known as a nested record

XML Schema


  • string
  • boolean
  • decimal
  • float
  • double
  • duration
  • dateTime
  • time
  • date
  • gYearMonth
  • gYear
  • gMonthDay
  • gDay
  • gMonth
  • hexBinary
  • base64Binary
  • anyURI

Type Lists