use rust_docx::{
build_doc_model, extract_document_blocks, extract_document_blocks_with_revisions,
DocumentBlock, OoxmlPackage, OoxmlPart, ReviewedDocument, RevisionKind, RevisionScope,
};
use std::collections::HashSet;
fn package(body: &str) -> OoxmlPackage {
let name = "word/document.xml".to_string();
let content = format!(
r#"<w:document xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"><w:body>{body}</w:body></w:document>"#
);
OoxmlPackage {
parts: [(name.clone(), OoxmlPart { name, content })].into(),
binary_assets: Default::default(),
warnings: vec![],
}
}
fn check_anchors(review: &ReviewedDocument) {
fn visit<'a>(blocks: &'a [DocumentBlock], ids: &mut HashSet<&'a str>) {
for b in blocks {
ids.insert(&b.id);
visit(&b.children, ids);
}
}
let mut ids = HashSet::new();
visit(&review.document.blocks, &mut ids);
let mut revisions = HashSet::new();
for r in &review.revisions {
assert!(revisions.insert(&r.id));
assert!(r
.source_refs
.iter()
.all(|s| ids.contains(s.node_id.as_str()) && s.offset_encoding == "utf-16"));
}
}
#[test]
fn replacements_and_deleted_paragraphs_use_current_text_coordinates() {
let model = build_doc_model(&package(
r#"<w:p><w:r><w:t>😀 </w:t></w:r>
<w:del w:id="9" w:author="A & B" w:date="2020-01-01T09:00:00Z"><w:r><w:delText>x&x</w:delText><w:tab/><w:delText>y</w:delText></w:r></w:del>
<w:ins w:id="10"><w:hyperlink><w:r><w:t>ne</w:t></w:r><w:r><w:t>w🚀</w:t></w:r></w:hyperlink></w:ins>
<w:r><w:rPr><w:strike/></w:rPr><w:t>!</w:t></w:r></w:p>
<w:p><w:del w:id="11"><w:r><w:delText>entire paragraph</w:delText></w:r></w:del></w:p>"#,
));
let plain = extract_document_blocks(&model);
assert_eq!(plain.blocks.len(), 1);
assert_eq!(plain.blocks[0].content, "😀 new🚀!");
assert!(!serde_json::to_string(&plain)
.unwrap()
.contains("entire paragraph"));
let review = extract_document_blocks_with_revisions(&model, Default::default());
assert_eq!(review.document.blocks.len(), 2);
assert_eq!(review.revisions.len(), 3);
let del = &review.revisions[0];
let ins = &review.revisions[1];
let gone = &review.revisions[2];
assert_eq!(
(
del.kind,
del.author.as_deref(),
del.text.as_deref(),
del.source_id.as_deref()
),
(
RevisionKind::Deletion,
Some("A & B"),
Some("x&x\ty"),
Some("9")
)
);
assert_eq!(del.date.as_deref(), Some("2020-01-01T09:00:00Z"));
let range = |r: &rust_docx::DocumentRevision| (r.source_refs[0].start, r.source_refs[0].end);
assert_eq!(range(del), (Some(3), Some(3)));
assert_eq!(
(ins.text.as_deref(), range(ins)),
(Some("new🚀"), (Some(3), Some(8)))
);
assert_eq!(
(gone.text.as_deref(), range(gone)),
(Some("entire paragraph"), (Some(0), Some(0)))
);
check_anchors(&review);
assert_eq!(
serde_json::from_value::<ReviewedDocument>(serde_json::to_value(&review).unwrap()).unwrap(),
review
);
}
#[test]
fn distinguishes_moves_mark_changes_and_direct_formatting() {
let model = build_doc_model(&package(
r#"<w:p><w:moveFromRangeStart w:id="1" w:name="m"/><w:moveFrom w:id="2"><w:r><w:delText>move</w:delText></w:r></w:moveFrom><w:moveFromRangeEnd w:id="1"/></w:p>
<w:p><w:pPr><w:rPr><w:del w:id="2"/></w:rPr><w:jc w:val="right"/><w:pPrChange w:id="3"><w:pPr><w:jc w:val="left"/></w:pPr></w:pPrChange></w:pPr>
<w:moveToRangeStart w:id="4" w:name="m"/><w:moveTo w:id="5"><w:r><w:rPr><w:b/><w:rPrChange w:id="6"><w:rPr><w:i/></w:rPr></w:rPrChange></w:rPr><w:t>move</w:t></w:r></w:moveTo><w:moveToRangeEnd w:id="4"/></w:p>"#,
));
let review = extract_document_blocks_with_revisions(&model, Default::default());
check_anchors(&review);
assert_eq!(
review.revisions.len(),
5,
"{:?}",
review
.revisions
.iter()
.map(|r| (&r.source_id, r.kind, r.scope))
.collect::<Vec<_>>()
);
let moves: Vec<_> = review
.revisions
.iter()
.filter(|r| matches!(r.kind, RevisionKind::MoveFrom | RevisionKind::MoveTo))
.collect();
assert_eq!(moves.len(), 2);
assert!(moves
.iter()
.all(|r| r.move_name.as_deref() == Some("m") && r.text.as_deref() == Some("move")));
let mark = review
.revisions
.iter()
.find(|r| r.scope == RevisionScope::ParagraphMark)
.unwrap();
assert_eq!(mark.source_refs[0].start, Some(4));
let formatting = review
.revisions
.iter()
.find(|r| r.scope == RevisionScope::Text && r.kind == RevisionKind::Formatting)
.unwrap();
assert_eq!(
(
formatting.source_refs[0].start,
formatting.source_refs[0].end
),
(Some(0), Some(4))
);
let props = formatting.properties.as_ref().unwrap();
assert_eq!(props.before[0].name, "w:i");
assert_eq!(props.after[0].name, "w:b");
}
#[test]
fn row_and_cell_revisions_link_to_cells_and_nested_text_is_reported_once() {
let model = build_doc_model(&package(
r#"<w:tbl><w:tblPr><w:tblPrChange w:id="33"><w:tblPr/></w:tblPrChange></w:tblPr>
<w:tr><w:trPr><w:ins w:id="34"/></w:trPr><w:tc><w:tcPr><w:cellIns w:id="35"/></w:tcPr><w:p><w:ins w:id="36"><w:r><w:t>cell</w:t></w:r></w:ins></w:p></w:tc><w:tc><w:p/></w:tc></w:tr></w:tbl>"#,
));
let review = extract_document_blocks_with_revisions(&model, Default::default());
check_anchors(&review);
assert_eq!(review.revisions.len(), 4);
let row = review
.revisions
.iter()
.find(|r| r.scope == RevisionScope::TableRow)
.unwrap();
assert_eq!(row.source_refs.len(), 2);
assert!(row
.source_refs
.iter()
.all(|s| s.start.is_none() && s.node_id.contains("/row/0/cell/")));
}
#[test]
fn section_changes_attach_to_the_closing_block_once() {
let section = r#"<w:sectPr><w:pgSz w:w="8000"/><w:sectPrChange w:id="12"><w:sectPr><w:pgSz w:w="7000"/></w:sectPr></w:sectPrChange></w:sectPr>"#;
for body in [
format!("<w:p><w:r><w:t>p</w:t></w:r></w:p>{section}"),
format!("<w:p><w:pPr>{section}</w:pPr><w:r><w:t>p</w:t></w:r></w:p>"),
] {
let model = build_doc_model(&package(&body));
let review = extract_document_blocks_with_revisions(&model, Default::default());
check_anchors(&review);
assert_eq!(review.revisions.len(), 1);
assert_eq!(review.revisions[0].scope, RevisionScope::Section);
let old = &review.revisions[0].properties.as_ref().unwrap().before[0];
assert_eq!(old.attributes["w:w"], "7000");
}
}