Files
craft-libs/crates/docx/tests/it/bidi.rs
Dylan M. Taylor e3e31693c7 Add craft-docx: WordCraft's DOCX reader and writer on craft-wordmodel, craft-opc and craft-xml (#1015)
* Add craft-wordmodel: WordCraft's document model without its editing code

WordCraft's crates/doc (wordcraft-doc, at 07f97737) moved as it is, so the DOCX,
RTF, ODT and other text-format crates can read into and write from one model.
Module paths, type names and serde attributes are unchanged: an app that
re-exports the crate keeps its native file format.

Left in the app: range editing (edit.rs), IME text and equation editing. Not
model, and already in craft-libs: code pages (craft-rtf), the ZIP entry-count
check (craft-opc), the compound-file guard (craft-cfb). The paragraph walks
moved to walk.rs and the unit helpers of wordcraft-geom to geom.rs.

* craft-wordmodel: math_tree, the arguments of an equation node and a walk over its runs

The DOCX reader and writer walk an equation's runs. The walk and the child
accessors it uses were in WordCraft's equation editor (math_edit.rs), which
stays in the app; they are structure, not editing.

* craft-wordmodel: keep the document operations (edit.rs) with the model

The DOCX and RTF readers use them: alternative-format chunks are merged into
a document with adopt_blocks, and a reader formats ranges with format_range
and inserts objects. They operate on the document, not on a view, so they
belong beside it. The tests that drove them come back with them; the
paragraph walks stay in walk.rs and edit re-exports them.

base64 0.23, as WordCraft and craft-offcrypto use.

* craft-docx: WordCraft's crates/docx as it is at wordcraft 07f97737

The source, unchanged, so the next commit shows only what the move changes.
This commit does not build on its own (it still names wordcraft-doc and
wordcraft-geom, and is not a workspace member yet).

* craft-docx: on craft-wordmodel, craft-opc, craft-xml and craft-offcrypto

- The package (zip reading, part names, relationships, content types,
  zip writing) is craft-opc's; the module keeps WordCraft's bookkeeping
  (passthrough keys, the embedded-part manifest) and its own Rel/Rels shape.
- Parts are parsed with craft-xml's canonical names (the Strict drawing
  names and mc:Choice/@Requires through its hooks), then given the tree
  shape the reader was written against.
- Encryption and Restrict Editing hashes are craft-offcrypto's; this crate
  maps its errors and the w:documentProtection attributes.
- wordcraft-doc is craft-wordmodel; wordcraft-geom's helpers are its geom.

Tests that covered what now lives in craft-opc and craft-offcrypto are left to
those crates; the DIFAT regression is craft-cfb's. Test-only zip use follows
zip 9's fallible names. The attribute-cap test times the read only.
2026-11-11 20:31:40 -05:00

178 lines
8.8 KiB
Rust

//! Right-to-left and complex-script OOXML: `w:bidi` paragraphs, `w:rtl` / `w:cs` runs, the
//! complex-script font, size, bold and italic, and the bidi language. Fixtures are hand-written
//! from ECMA-376 Part 1 §17.3.1.6 (bidi), §17.3.2 (run properties).
use std::io::{Read, Write};
use craft_wordmodel::props::{Align, CharProps, ParaProps};
use craft_wordmodel::{Document, Paragraph, para_block};
fn zip(entries: &[(&str, &[u8])]) -> Vec<u8> {
let mut zw = zip::ZipWriter::new(std::io::Cursor::new(Vec::new()));
for (name, bytes) in entries {
zw.start_file(*name, zip::write::SimpleFileOptions::default()).unwrap();
zw.write_all(bytes).unwrap();
}
zw.finish().unwrap().into_inner()
}
/// A package whose body is `body`.
fn docx(body: &str) -> Vec<u8> {
let root = r#"<?xml version="1.0" encoding="UTF-8"?><Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument" Target="word/document.xml"/></Relationships>"#;
let doc = format!(
r#"<?xml version="1.0" encoding="UTF-8" standalone="yes"?><w:document xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"><w:body>{body}</w:body></w:document>"#
);
zip(&[("_rels/.rels", root.as_bytes()), ("word/document.xml", doc.as_bytes())])
}
fn paras(d: &Document) -> Vec<&Paragraph> {
d.body.iter().filter_map(|b| b.as_para()).collect()
}
/// `word/document.xml` of a written package.
fn document_xml(bytes: &[u8]) -> String {
let mut z = zip::ZipArchive::new(std::io::Cursor::new(bytes)).unwrap();
let mut s = String::new();
z.by_name("word/document.xml").unwrap().read_to_string(&mut s).unwrap();
s
}
/// Persian text with a ZWNJ (نیم‌فاصله), Persian digits, Latin and punctuation.
const FA: &str = "می‌خواهم نسخهٔ ۲٫۵ از WordCraft را (امروز) نصب کنم.";
fn persian_run() -> CharProps {
CharProps {
rtl: Some(true),
font_cs: Some("B Nazanin".into()),
size_cs: Some(14.0),
bold_cs: Some(true),
italic_cs: Some(false),
lang_bidi: Some("fa-IR".into()),
..Default::default()
}
}
#[test]
fn persian_paragraph_round_trips() {
let mut rtl = Paragraph::with_text(FA, persian_run());
rtl.props = ParaProps { bidi: Some(true), align: Some(Align::Justify), ..Default::default() };
let ltr = Paragraph::with_text("English with فارسی inside", CharProps { cs: Some(true), ..Default::default() });
let mut d = Document::new();
d.body = vec![para_block(rtl.clone()), para_block(ltr.clone())];
let bytes = craft_docx::write(&d).unwrap();
let r = craft_docx::read(&bytes).unwrap();
let ps = paras(&r);
assert_eq!(ps[0].text, FA, "Persian text, ZWNJ and digits survive byte for byte");
assert_eq!(ps[0].props.bidi, Some(true));
assert_eq!(ps[0].props.align, Some(Align::Justify));
assert_eq!(ps[0].runs[0].props, persian_run());
assert_eq!(ps[1].text, ltr.text);
assert_eq!(ps[1].props.bidi, None);
assert_eq!(ps[1].runs[0].props.cs, Some(true));
}
#[test]
fn bidi_markup_is_written_in_schema_order() {
let mut p =
Paragraph::with_text(FA, CharProps { bold: Some(true), italic: Some(true), size: Some(12.0), font: Some("Arial".into()), ..persian_run() });
p.props.bidi = Some(true);
let mut d = Document::new();
d.body = vec![para_block(p)];
let xml = document_xml(&craft_docx::write(&d).unwrap());
assert!(xml.contains("<w:bidi/>"), "paragraph direction: {xml}");
assert!(xml.contains(r#"w:cs="B Nazanin""#), "complex-script font");
assert!(xml.contains(r#"<w:szCs w:val="28"/>"#), "complex-script size in half points");
assert!(xml.contains(r#"w:bidi="fa-IR""#), "bidi language");
// CT_RPr is a sequence: b, bCs, i, iCs … sz, szCs … rtl, cs, lang.
let pos = |t: &str| xml.find(t).unwrap_or_else(|| panic!("{t} missing: {xml}"));
assert!(pos("<w:rFonts") < pos("<w:b/>"));
assert!(pos("<w:b/>") < pos("<w:bCs/>") && pos("<w:bCs/>") < pos("<w:i/>"));
assert!(pos("<w:i/>") < pos(r#"<w:iCs w:val="0"/>"#));
assert!(pos(r#"<w:sz w:val="24"/>"#) < pos("<w:szCs"));
assert!(pos("<w:rtl/>") < pos("<w:lang"));
}
#[test]
fn reads_word_style_bidi_markup() {
// What a right-to-left paragraph looks like in a Word document: w:bidi, logical
// alignment and indents (left = start), runs with w:rtl and the complex-script values.
let body = r#"
<w:p>
<w:pPr><w:bidi/><w:ind w:left="720" w:right="0"/><w:jc w:val="left"/></w:pPr>
<w:r><w:rPr><w:rFonts w:ascii="Times New Roman" w:hAnsi="Times New Roman" w:cs="B Lotus"/><w:b/><w:bCs/><w:sz w:val="24"/><w:szCs w:val="32"/><w:rtl/><w:lang w:val="en-US" w:bidi="fa-IR"/></w:rPr><w:t>سلام دنیا</w:t></w:r>
<w:r><w:rPr><w:rtl/></w:rPr><w:t xml:space="preserve"> </w:t></w:r>
<w:r><w:t>Word 2025</w:t></w:r>
</w:p>
<w:p><w:pPr><w:bidi w:val="0"/></w:pPr><w:r><w:rPr><w:cs/><w:iCs/></w:rPr><w:t>x</w:t></w:r></w:p>"#;
let d = craft_docx::read(&docx(body)).unwrap();
let ps = paras(&d);
assert_eq!(ps[0].text, "سلام دنیا Word 2025");
assert_eq!(ps[0].props.bidi, Some(true));
assert_eq!(ps[0].props.indent_left, Some(36.0), "w:left is the start indent");
assert_eq!(ps[0].props.align, Some(Align::Left), "jc=left is the start edge");
let fa = &ps[0].runs[0].props;
assert_eq!(fa.font.as_deref(), Some("Times New Roman"));
assert_eq!(fa.font_cs.as_deref(), Some("B Lotus"));
assert_eq!((fa.size, fa.size_cs), (Some(12.0), Some(16.0)));
// Both explicit values survive, even when they happen to be equal.
assert_eq!((fa.bold, fa.bold_cs, fa.rtl), (Some(true), Some(true), Some(true)));
assert_eq!((fa.lang.as_deref(), fa.lang_bidi.as_deref()), (Some("en-US"), Some("fa-IR")));
assert_eq!(ps[1].props.bidi, Some(false));
let x = &ps[1].runs[0].props;
assert_eq!((x.cs, x.italic_cs, x.italic), (Some(true), Some(true), None));
// The resolved complex-script formatting is what Persian characters are drawn with.
let rc = d.styles.resolve_char(None, fa);
let cs = rc.complex();
assert_eq!((cs.font.as_str(), cs.size, cs.bold), ("B Lotus", 16.0, true));
assert!(rc.uses_complex('س') && rc.uses_complex('W'), "every character of an rtl run is complex script");
}
#[test]
fn hostile_bidi_markup_is_bounded() {
let long = "x".repeat(10_000);
let body = format!(
r#"<w:p><w:pPr><w:bidi w:val="banana"/></w:pPr><w:r><w:rPr><w:rFonts w:cs="{long}"/><w:szCs w:val="-5"/><w:szCs w:val="99999999"/><w:lang w:bidi="{long}"/><w:rtl w:val="maybe"/></w:rPr><w:t>ok</w:t></w:r></w:p>"#
);
let d = craft_docx::read(&docx(&body)).unwrap();
let c = &paras(&d)[0].runs[0].props;
assert_eq!(c.font_cs, None, "absurd font names are dropped");
assert_eq!(c.lang_bidi, None);
assert!(c.size_cs.is_none_or(|s| (1.0..=1638.0).contains(&s)));
}
/// Right-to-left tables (`w:tblPr/w:bidiVisual`, ECMA-376 Part 1 §17.4.1) and sections
/// (`w:sectPr/w:bidi`, §17.6.1) read, write in schema order and round-trip (#362).
#[test]
fn rtl_tables_and_sections_round_trip() {
let body = r#"
<w:tbl>
<w:tblPr><w:bidiVisual/><w:tblW w:w="0" w:type="auto"/><w:tblInd w:w="360" w:type="dxa"/></w:tblPr>
<w:tblGrid><w:gridCol w:w="2000"/><w:gridCol w:w="2000"/></w:tblGrid>
<w:tr><w:tc><w:p><w:r><w:t>اول</w:t></w:r></w:p></w:tc><w:tc><w:p><w:r><w:t>دوم</w:t></w:r></w:p></w:tc></w:tr>
</w:tbl>
<w:tbl>
<w:tblPr><w:bidiVisual w:val="0"/></w:tblPr>
<w:tblGrid><w:gridCol w:w="2000"/></w:tblGrid>
<w:tr><w:tc><w:p/></w:tc></w:tr>
</w:tbl>
<w:p/>
<w:sectPr><w:cols w:num="2" w:space="720"/><w:bidi/></w:sectPr>"#;
let d = craft_docx::read(&docx(body)).unwrap();
let tables: Vec<&craft_wordmodel::Table> = d.body.iter().filter_map(|b| b.as_table()).collect();
assert_eq!(tables.len(), 2);
assert!(tables[0].props.bidi_visual, "w:bidiVisual reads as a right-to-left table");
assert_eq!(tables[0].props.indent, Some(18.0), "the indent is kept (measured from the right)");
assert!(!tables[1].props.bidi_visual, "w:val=0 turns it off");
assert!(d.last_section.rtl, "w:bidi in sectPr is a right-to-left section");
let bytes = craft_docx::write(&d).unwrap();
let xml = document_xml(&bytes);
assert_eq!(xml.matches("<w:bidiVisual/>").count(), 1, "{xml}");
// CT_TblPr is a sequence: tblStyle, tblpPr, tblOverlap, bidiVisual, … tblW, jc, … tblInd.
let pos = |t: &str| xml.find(t).unwrap_or_else(|| panic!("{t} missing: {xml}"));
assert!(pos("<w:bidiVisual/>") < pos("<w:tblW"));
let r = craft_docx::read(&bytes).unwrap();
let again: Vec<bool> = r.body.iter().filter_map(|b| b.as_table()).map(|t| t.props.bidi_visual).collect();
assert_eq!(again, vec![true, false]);
assert!(r.last_section.rtl && r.last_section.columns.count == 2);
}