rdocx-oxml

rdocx-oxml provides typed WordprocessingML elements with prefix-tolerant parsing, schema-ordered serialization, and verbatim preservation of unmodelled XML. Use it when a tool already owns the XML model. Most applications should use the package-preserving rdocx facade instead.

use rdocx_oxml::{BodyContent, CT_Document, CT_P};

let mut document = CT_Document::new();
let mut paragraph = CT_P::new();
paragraph.add_run("Low-level WordprocessingML");
document.body.content.push(BodyContent::Paragraph(paragraph));

assert_eq!(document.body.paragraphs().count(), 1);

[dependencies]
rdocx-oxml = "0.5"

Migrating from 0.4

Version 0.5 adds explicit preservation state to the public numbering structs. Code that constructs CT_Lvl, CT_AbstractNum, CT_Num, or CT_Numbering with a full struct literal must supply the new raw XML fields. Prefer CT_Lvl::new, CT_AbstractNum::new, and CT_Numbering::new. When a literal is required, CT_Lvl adds empty extra_xml and extra_attributes vectors plus ppr_raw: None and rpr_raw: None. CT_AbstractNum and CT_Num add empty extra_xml and extra_attributes vectors. CT_Numbering adds empty root_attributes and extra_xml vectors. These fields retain producer XML during typed numbering edits and are intentionally part of the 0.5 model.

CT_TabStop full literals also add source_occurrence: None. Parsed numbering tabs use this field to keep producer XML attached to the same occurrence after typed edits, insertions, or removals. CT_TabStop::new initializes it for a new tab. Semantic equality ignores the field and continues to compare alignment, position, and leader values. CT_Tabs::from_xml_with_prefixes accepts the in-scope WordprocessingML prefixes when callers parse aliased tab collections directly. Numbering, style, and document-family readers route complete paragraph property subtrees through one internal namespace projection rather than exposing a partially contextual public CT_PPr parser.