The following information may be provided at the verb or version level.

• description, description of the verb or a specific version
• note, information pertaining to the verb or a specific version

Each element of the list contains the following information:

• version number of the verb.
• arguments, a list of the names of the fixed and keyword arguments, if any, accepted by the verb in that version.
• example template of request to this repository, with fixed argument indicated in brackets
• returns, optional, contains information about response format.

Note that a service may implement more than one version of a verb.

The response to a Display verb request is an HTTP redirect, if the request is successful. A repository can display a document however it chooses. If the request is not successful, an error response is returned.

The response to Disseminate is the requested content, directly.

Each response includes one or more divReq elements.

• divReq Any div for which format information has been requested becomes a divReq (requested div) element in the response. A divReq element has the same attributes as a Structure response div (except that the order attribute is not used):
  • id :: [required] an ID value for this division, usable as is in Disseminate and other requests. This is not defined as an XML ID.
  • type :: type of division.
  • label :: a displayable label for this division. These are applied by the local repository and should make sense in a navigational context.
  The divReq element must contain at least one format or formatSeq element. It may contain more of either.
• format an empty element with attributes describing a single format available for its parent element, which may be divReq, div, or formatDiv. The format element has four attributes:
  • type :: [required] from a controlled vocabulary of format types.
  • mime :: [required] the mime-type of the format.
  • size :: in bytes.
  • label :: [required] a displayable label identifying this format for use in a user interface.
• formatSeq an element that contains a sequence of formatDiv elements. The formatSeq element has two attributes:
  • type :: [required] from a controlled vocabulary of format types.
  • label :: [required] a displayable label identifying this group of formats, for use in a user interface.
• formatDiv a division within formatSeq that contains one or more format elements. formatDiv is used for divisions that are not structural--not represented in Structure verb responses--but exist because of arbitrary segmenting of documents for delivery purposes. This element has three attibutes:
  • id :: [required] an ID value for this division, usable as is in Disseminate requests. This is not defined as an XML ID.
  • order :: [required] the order of this formatDiv among its siblings. Values are positive integers, beginning with 1.
  • label :: a displayable label for this division. These are applied by the local repository and should make sense in a navigational context.

The List Verbs response contains a single, repeatable element.

• verb an EMPTY element with two attributes:
  • name :: the verb name.
  • ver :: the version of the verb.

• version its attribute value is the version number, a positive integer, used when requesting a specific version in Structure, Formats, and Disseminate verbs. The element contains two subelements:
  • date the date (or date-time) the version was last modified, expressed in using ISO 8601.
  • comment a comment or description.

A default view is mandatory. One view, but only one, must be identified as the default view.

• view an empty element, one per view, with three attributes:
  • id :: [required] an id of the view (XML ID).
  • label :: [required] a displayable, short label describing this view. These are applied by the local repository and should help end users choose an appropriate document view.
  • default :: [required for default view] possible values are 1 or 0. If not included, default="0" is assumed.

Unless there is an error, all responses include a single resultsSummary and zero to many record elements.

A well-formed query that executes properly and yet has no matches is not considered an error. In this case, a resultsSummary would be returned, but no record elements.

• resultsSummary [required] an empty element with data about the results returned. All of the following attributes must be included, using defaults if not specified in request. These values describe the results actually returned; they may differ from values in the request.
  • repositoryIdentifier :: the repository identifier.
  • set :: if no set was indicated in the response, then return either 1) a list of all sets searched, if sets are supported (set="set1|set2|set3"); or 2) an empty value, if sets are not supported (set="0").
  • sort :: use default when not in request. Note that the requested sort order can be ignored by the indexer if the requested sort value is not supported by the indexer, or if the indexer's sortThreshold has been exceeded. An indexer's supported sort orders and sortThreshold value is publicized in the Describe Verb response.
  • totalResults :: the total number of matches on a query, regardless of how many or which results are returned.
  • startResult :: the order of the first result returned in this response. Value is 0 when no results are returned.
  • resultSize :: the number of results returned in this response. This may be less than the resultSize requested, if there are fewer results than requested. Note that the requested result size can be ignored and reset to any value by the indexer if the indexer's resultsThreshold has been exceeded. An indexer's resultsThreshold value is publicized in the Describe Verb response.
• record zero to many record elements allowed. Currently allowed record elements are the following:
  • identifier [required] the unique document identifier. Its format is determined by the repository.
  • title
  • author
  • pubdate The date of publication of the document.
  • rank Relevance data.
  • resultDivs in the response to a full-text search, this element may contain ID values of document subdivisions containing matches. Each ID value is contained within a separate <divID> element.

Each response will include one view element. There will always be one and only one default view of a document.

• view contains one high-level, or root, div element, within which nested div hierarchies are allowed. view element attributes are:
  • id :: [required] an id value for this view. This id is defined as an XML ID.
  • label :: [required] a displayable label identifying this view. These are applied by the local repository and should help end users choose an appropriate document view.
  • default :: [required for default view] equals 1 for the sole default view; equals 0 for all other views. If absent, default="0" is assumed.
• div each view element contains one immediate child div element. This single div may itself contain one or more div elements. div element attributes are:
  • id :: [required] an ID value for this division, usable as is in Formats and other requests. This is not defined as an XML ID.
  • type :: type of division.
  • order :: [required] the order of this div among its siblings. Values are positive integers, beginning with 1.
  • label :: a displayable label for this division. These are applied by the local repository and should make sense in a navigational context.
  • diss :: indicates whether this division can be requested successfully using the Disseminate verb. Values are "0" (cannot be disseminated), or "1" (can be requested with Disseminate).

• statement a single statement element is returned, the most current statement available for the document requested. A statement element can contain one or more p elements. p elements in turn contain text, a link to another document, or a combination of text and links.
• p a paragraph element, for encoding a text statement.
• link for creating a link. link can contain data or be empty. The behavior of the link is controlled with three attributes:
  • actuate :: defines whether a link occurs automatically or must be activated by the user.
    • onload :: the link is activated, and the resource is displayed, automatically.
    • onrequest :: the resource is displayed if the user activates the link.
  • show :: determines how a remote resource appears when the link is activated.
    • embed :: the target resource displays at the point of the link.
    • new :: the target resource appears in a new window.
    • replace :: the target resource displays in the same window as the resource that contained the link.
  • href :: the locator for a remote resource, expressed as a URL.