--- title: Overview of support for the Datapackage Specification v2 author: Jan van der Laan css: "style.css" --- Note that any property can always be obtained and set using `dpproperty()` and `dpproperty<-()` respectively. Therefore, when specific support for a property is missing from the table below, this poperty can still be obtained and set. ## Data Package

Property	Getting	Setting
`resources`	`dp_resource()`, `dp_resource_names()`	`dp_resource<-()`, `cp_resources<-()`
`$schema`
`name`	`dp_name()`	`dp_name<-()`
`id`	`dp_id()`	`dp_id<-()`
`licenses`
`title`	`dp_title()`	`dp_title<-()`
`description`	`dp_description()`	`dp_description<-()`
`homepage`
`image`
`version`
`created`	`dp_created()`	`dp_created<-()`
`keywords`	`dp_keywords()`	`dp_keywords<-()`
`contributors`	`dp_contributors()`	`dp_contributors<-()`, `dp_add_contributor()`, `dp_new_contributor()`
`sources`

## Data Resource

Property	Getting	Setting
`name`	`dp_name()`	`dp_name<-()`
`path`	`dp_path()`	`dp_path<-()`
`data`	`dp_get_data()` returns data either from 'data' property or by reading from the 'path'.	It is possible to write data to file using `dp_write_data()`, but not to the 'data' property inside the `datapackage.json`
`type`
`$schema`
`title`	`dp_title()`	`dp_title<-()`
`description`	`dp_description()`	`dp_description<-()`
`format`	`dp_format()`, `dp_get_data()` will use the 'format' as the primary determinant for the reader to use to read the data.	`dp_format<-()`
`mediatype`	`dp_mediatype()`, when 'format' is missing 'mediatype' will be used to determine which reader to use for reading the data by `dp_get_data()`	`dp_mediatype<-()`, `dp_generate_dataresource()`
`encoding`	`dp_encoding()`	`dp_encoding<-()`
`bytes`	`dp_bytes()`^a	`dp_bytes<-()`^a
`hash`	`dp_hash()`^a	`dp_hash<-()`^a
`sources`
`licences`

^a The number of bytes and the hash can be set and get. There is no functionality to check is the file indeed has the specified number of bytes or hash and/or to automatically calculate this from the given file(s). ## Tabular Data Resource

Property	Getting	Setting
`dialect`	See 'Table Dialect'. There is no function to specifically get the 'dialect' information. The data resource is passed to the reader functions that will access this information.	See 'Table Dialect'. The writer functions will use this information when writing. There is no specific function to change this information. By default the, safe, default values will be used.
`schema`	`dp_schema()`, also see 'Table Schema'	`dp_generate_dataresource()` will generate appropiate schema for a given data set.

## Table Dialect As mentioned above, the 'dialect' property cannot be directly set. The table below indicates what properties are recognised when reading and writing data. The items are marked as support or not or irrelevant based on the support by the CSV reader and writer.

Property	Reading	Writing
`$schema`
`header`	CSV	CSV
`headerRows`
`headerJoin`
`commentRows`
`commentChar`	CSV	CSV
`delimiter`	CSV	CSV
`lineTerminator`	CSV^a	CSV^a
`quoteChar`	CSV^b	CSV^b
`doubleQuote`	CSV^c	CSV^c
`escapeChar`
`nullSequence`	CSV	CSV
`skipInitialSpace`	CSV	CSV
`property`
`itemType`
`itemKeys`
`sheetNumber`
`sheetName`
`table`

^a Only \\n/\\r or \\r\\n is accepted. ^b Only '"' is accepted. ^c Only 'true' is accepted. ## Table Schema

Property	Getting	Setting
`$schema`
`fields`	`dp_field()`, `dp_field_names()`
`fieldsMatch`
`missingValues`
`primaryKeys`
`uniqueKeys`
`foreignKeys`

## Field Descriptor

Property	Getting	Setting
`name`	`dp_name()`; also used by `dp_get_data()`.	`dp_name<-()`
`type`	Used by `dp_get_data()`.
`format`	`dp_format()`	`dp_format<-()`
`title`	`dp_title()`	`dp_title<-()`
`description`	`dp_description()`	`dp_description<-()`
`example`
`constraints`	Used by `dp_check_field()` and `dp_check_dataresource()`; see 'Field Constraints'.
`categories`	Used by `dp_categorieslist()` and `dp_get_data()`.	Used by `dp_write_data()` and `dp_generate_dataresource()`.
`categoriesOrdered`
`missingValues`	Used by `dp_get_data()`.
`refType`

## Field Types As mentioned above, the field descriptors cannot be directly modified or read from. The table below indicates what properties are recognised when reading and writing data. The items are marked as support or not or irrelevant based on the support by the CSV reader and writer. When a type is not supported the data will be read as a character string.

Property	Reading	Writing
`string`	CSV	CSV
`format`
`number`	CSV	CSV
`NaN, INF, -INF`	CSV	CSV
`exponent`	CSV	CSV
`decimalChar`	CSV	CSV
`groupChar`	CSV	CSV
`bareNumber`	CSV	CSV
`integer`	CSV	CSV
`groupChar`	CSV	CSV
`bareNumber`	CSV	CSV
`boolean`	CSV	CSV
`trueValues`	CSV	CSV
`falseValues`	CSV	CSV
`object`
`array`
`list`
`delimiter`
`itemType`
`datetime`	CSV	CSV
`format ("default", "", "any")`	CSV	CSV
`date`	CSV	CSV
`format ("default", "", "any")`	CSV	CSV
`time`	CSV	CSV
`format ("default", "", "any")`	CSV	CSV
`year`	CSV	CSV
`yearmonth`	CSV	CSV
`duration`
`geopoint`
`format ("default", "array", "object")`
`geojson`
`format ("default", "topojson")`
`any`

## Field Constraints The functions `dp_check_dataresource()` and `dp_check_field()` checks if a given `data.frame` or vector is valid given the Data Resource or Field Descriptor. By default these will also check any constraints of fields. The default CSV and fixed width readers will not run these checks.

Property	Checking constraints	Getting	Setting
`required`	`dp_check_field()`, `dp_check_dataresource(..., constraints = TRUE)`
`unique`	`dp_check_field()`, `dp_check_dataresource(..., constraints = TRUE)`
`minLength`
`maxLength`
`minimum`	`dp_check_field()`, `dp_check_dataresource(..., constraints = TRUE)`
`maximum`	`dp_check_field()`, `dp_check_dataresource(..., constraints = TRUE)`
`exclusiveMinimum`	`dp_check_field()`, `dp_check_dataresource(..., constraints = TRUE)`
`exclusiveMaximum`	`dp_check_field()`, `dp_check_dataresource(..., constraints = TRUE)`
`jsonSchema`
`pattern`
`enum`	`dp_check_field()`, `dp_check_dataresource(..., constraints = TRUE)`