DiscourseElement

A part of a document’s structure: a section, paragraph, sentence, table, footnote, or a rhetorical unit like an introduction or conclusion. Discourse elements can be nested and ordered. Created via GraphSet.add_de(resp_agent). Short name: de. OCDM class: deo:DiscourseElement.

Properties

title

Functional. RDF predicate: dcterms:title.

has_title(string: str) / get_title() -> str | None / remove_title()

contained discourse elements

Non-functional. RDF predicate: frbr:part. Accepts: DiscourseElement.

contains_discourse_element(de_res) / get_contained_discourse_elements() -> list[DiscourseElement] / remove_contained_discourse_element(de_res=None)

Nested children: a section contains paragraphs, a paragraph contains sentences.

next discourse element

Functional. RDF predicate: oco:hasNext. Accepts: DiscourseElement.

has_next_de(de_res) / get_next_de() -> DiscourseElement | None / remove_next_de()

Ordering among sibling elements at the same nesting level.

context of reference pointers

Non-functional. RDF predicate: c4o:isContextOf. Accepts: ReferencePointer.

is_context_of_rp(rp_res) / get_is_context_of_rp() -> list[ReferencePointer] / remove_is_context_of_rp(rp_res=None)

The in-text reference pointers that appear within this discourse element.

context of pointer lists

Non-functional. RDF predicate: c4o:isContextOf. Accepts: PointerList.

is_context_of_pl(pl_res) / get_is_context_of_pl() -> list[PointerList] / remove_is_context_of_pl(pl_res=None)

content

Functional. RDF predicate: c4o:hasContent.

has_content(string: str) / get_content() -> str | None / remove_content()

The literal text of this discourse element.

number

Functional. RDF predicate: fabio:hasSequenceIdentifier.

has_number(string: str) / get_number() -> str | None / remove_number()

Type classification

Discourse elements carry two kinds of type: structural and rhetorical. They behave differently.

Structural types (replace existing secondary type)

These work like type methods on other entities: each call replaces the previous secondary rdf:type.

Method	RDF type
`create_discourse_element(de_class=None)`	`deo:DiscourseElement` (or custom IRI via `de_class`)
`create_section()`	`doco:Section`
`create_section_title()`	`doco:SectionTitle`
`create_paragraph()`	`doco:Paragraph`
`create_sentence()`	`doco:Sentence`
`create_text_chunk()`	`doco:TextChunk`
`create_table()`	`doco:Table`
`create_footnote()`	`doco:Footnote`
`create_caption()`	`deo:Caption`

Rhetorical types (additive)

These are not mutually exclusive. Calling create_introduction() does not remove an existing structural type; it adds a new rdf:type triple. A single discourse element can combine a structural type (e.g. doco:Section) with one or more rhetorical types (e.g. deo:Introduction).

Method	RDF type
`create_introduction()`	`deo:Introduction`
`create_methods()`	`deo:Methods`
`create_materials()`	`deo:Materials`
`create_related_work()`	`deo:RelatedWork`
`create_results()`	`deo:Results`
`create_discussion()`	`deo:Discussion`
`create_conclusion()`	`deo:Conclusion`

Example

g_set = GraphSet("https://w3id.org/oc/meta/")
resp_agent = "https://w3id.org/oc/meta/prov/pa/1"

intro = g_set.add_de(resp_agent)
intro.create_section()
intro.create_introduction()
intro.has_title("Introduction")
intro.has_number("1")

methods = g_set.add_de(resp_agent)
methods.create_section()
methods.create_methods()
methods.has_title("Methods")
methods.has_number("2")
intro.has_next_de(methods)

para = g_set.add_de(resp_agent)
para.create_paragraph()
intro.contains_discourse_element(para)

article = g_set.add_br(resp_agent)
article.contains_discourse_element(intro)
article.contains_discourse_element(methods)