1.1 Terminology

The key words MUST, MUST NOT, REQUIRED, SHALL, SHALL NOT, SHOULD, SHOULD NOT, RECOMMENDED, MAY, and OPTIONAL, in this specification, are to be interpreted as described in [IETF RFC 2119].

The keywords expanded name, NCName and QName are to be interpreted as described in the XML Names specification [XML Names].

1.2 Namespaces and namespace prefixes

Many values in the Open Information Model are XML expanded names, that is, the combination of a namespace URI and a localname. This specification uses QName notation in the form prefix:localname to refer to these, where the prefix is a short string representing a full namespace URI. The prefixes used by this specification are defined below. These prefixes are reserved prefixes.

Prefix	Namespace URI
iso4217	http://www.xbrl.org/2003/iso4217
utr	http://www.xbrl.org/2009/utr
xbrl	https://xbrl.org/2021
xbrli	http://www.xbrl.org/2003/instance
xs	http://www.w3.org/2001/XMLSchema

This specification also uses the following prefix for error codes. This prefix is not a reserved prefix.

Prefix	Namespace URI
oimce	https://xbrl.org/2021/oim-common/error

3.1 SQName

The Open Information Model makes extensive use of expanded names ([XML Names]) to provide globally unique identifiers. These are a namespace URI/localname pair. In order to reduce size and improve readability, it is convenient to use a QName-like mechanism for representing these (and some other) values, where the globally unique namespace-URI is replaced with a compact, document-scoped prefix. The mechanism used is referred to as an SQName ("Simplified QName").

An SQName is a compact, colon-separated represention of a URI/value pair, where a short, document-scoped, prefix is used to represent the URI.

An SQName MUST have the following format: (oimce:invalidSQName)

prefix:localname

Where prefix is an NCName and localname is any string. The prefix is resolved to a URI using the prefix map.

Unlike QNames, the localname can consist of any non-whitespace characters, including colon (":"); the prefix and localname in an SQName are separated by the first colon character.

3.2 Prefix map

A prefix map is a URI map that is used to resolve the prefix component of a QName or SQName to a URI.

The prefix for a QName or SQName  MUST correspond to a key in the prefix map (oimce:unboundPrefix).

A reserved prefix is a reserved URI alias that is specified as applying to a prefix map.

3.3 URI map

A URI map is a set of key/value pairs that is used to resolve URI aliases to full URIs. A URI alias is an NCName used to provide a limited scope short name for a URI. The location and format of a URI map is defined by the specification for the representation in which it is used, but it MUST conform to the following constraints:

This specification uses the term URI to refer to a value that conforms to the xs:anyURI datatype [XML Schema Datatypes]. Such a value can be transformed into a value that conforms with the [RFC2396]/[RFC2732] definition of a URI reference by applying the URI reference escaping procedure described in [XML Schema Datatypes]. This specification uses the term encoded URI to refer to the result of this applying the URI reference escaping procedure.

1. URI aliases MUST NOT be empty, and MUST be valid NCNames (oimce:invalidURIAlias).
2. A URI MUST NOT be bound to more than one URI alias within the map (oimce:multipleAliasesForURI).
3. A URI alias MUST NOT be bound to more than one URI within the map (oimce:multipleURIsForAlias).
4. A reserved URI alias MUST only be bound to the URI prescribed for that URI alias (oimce:invalidURIForReservedAlias).
5. URIs MUST be valid according to the xs:anyURI datatype [XML Schema Datatypes], and the encoded URI MUST be an absolute URI Reference [RFC2396]/[RFC2732] (oimce:invalidURI).
6. A URI for which a reserved URI alias is prescribed MUST NOT be bound to any other URI alias (oimce:invalidAliasForReservedURI).

A reserved URI alias is any URI alias identified as being reserved by either this specification (see Section 1.2) or the specification defining the inclusion of the URI map.

URIs in this specification are used purely as unique identifiers, and should be compared using simple, case-sensitive string comparison. Percent-encoded characters are not considered equivalent to literal, unencoded characters.

It should be noted that some of the constraints listed above, such as the requirement for URI aliases to be unique within the map, may be enforced by syntactic or structural constraints of the format in which the map is represented. In this case, implementations should raise the appropriate error for that format, rather than the code listed above.

3.4 Unit string representation

The unit core dimension defines units in terms of the combination of QNames by defining an unordered collection of numerators and an unordered collection of denominators. Reflecting this in concrete representations can be somewhat cumbersome, particularly as analysis the structure of units is expected to be a relatively uncommon operation compared to simply comparing units for equivalence, or rendering them to a user.

This section defines a standardised unit string representation, allowing units to be captured using a unique string value. This has the property that, provided that a consistent set of namespace prefixes are used, a given unit will have a single representation meaning that equivalence of units can be determined by string comparison.

The unit string representation is obtained as follows:

1. If {denominators} is not present, the QNames representing the {numerators} are combined in lexicographic order, separated by the "*" character.
2. Otherwise, lists of {numerators} and {denominators} are created by combining QNames in lexicographic order and separating with the "*" character, as above. If a list contains more than one item, the list is enclosed in parentheses, "(" and ")". The final string is then obtained by concatenating the list of numerators, the "/" character, and the list of denominators.

Whitespace is not permitted anywhere in the string representation, and the use of QNames ensures that a 1:1 mapping between prefixes and namespaces is available thus ensuring that equivalent units appearing within the same document are represented with the same string.

Lexicographic order means ordered based on comparison of Unicode code point values of each character.

Processors MUST raise oimce:invalidUnitStringRepresentation if they encounter a unit string representation that does not conform to the syntax described above. If any QNames in the unit string representation use a prefix that is not defined in the prefix map then oimce:unboundPrefix MUST be raised.

3.5 Period string representation

The period core dimension takes a value that is a time interval, which may be zero-length (i.e. an instant). This section defines a standard period string representation which can be used to express such values.

The period string representation is obtained as follows:

1. If the value is a non-zero length interval, then the representation is a pair of ISO 8601 date-time values, separated by the "/" character;
2. Otherwise, the value is a single ISO 8601 date-time value.

The date-time values used in this representation MUST be valid according to the xs:dateTime datatype and MUST be in the canonical lexical representation. The error code oimce:invalidPeriodRepresentation MUST be raised for values that are required to be a period string representation and which do not conform to the above format.

It should be noted that unlike the xbrli:dateUnion type used in xBRL-XML, the xs:dateTime datatype used here does not allow the time component to be omitted.

3.6 Space-separated list

A space-separated list is a string containing items that are separated by a single space character (unicode 0x20). The list MUST NOT contain sequences of more than one space character, or space characters at the beginning or end of the string.