document.rs

//! docx を作って読み戻すところ

use folio::document::{Block, Document, Writer};
use folio::paragraph::{Align, Paragraph, Run};
use folio::table::{Cell, Row, Table};
use std::path::PathBuf;


/// 試しに使うファイルの置き場。target の下なので消えても差し支えない
fn place(name: &str) -> PathBuf {
    let mut path = PathBuf::from("target");
    path.push(format!("document-{name}.docx"));
    path
}


/// 本文を読み切る
fn collect(document: &mut Document) -> Vec<Block> {
    let mut body = document.read().unwrap();
    let mut result = Vec::new();

    while let Some(block) = body.read().unwrap() {
        result.push(block);
    }

    result
}


/// 段落だけを取り出す
fn paragraph(block: &Block) -> &Paragraph {
    match block {
        Block::Paragraph(one) => one,
        Block::Table(_) => panic!("段落ではありません"),
    }
}


#[test]
fn paragraphs_come_back_as_they_were_written() {
    let path = place("paragraph");
    let mut writer = Writer::create(&path).unwrap();
    writer.paragraph(&Paragraph::styled("題", "Title")).unwrap();
    writer.paragraph(&Paragraph::styled("見出し", "Heading1")).unwrap();

    writer.paragraph(&Paragraph {
        style: String::new(),
        align: Align::Center,
        run: vec![
            Run::new("ふつう"),
            Run::bold("太字"),
            Run {text: "斜体".to_string(), italic: true, ..Run::default()},
            Run {text: "下線".to_string(), underline: true, ..Run::default()},
            Run {text: "取り消し".to_string(), strike: true, ..Run::default()},
            Run {text: "大きい".to_string(), size: 14.0, ..Run::default()},
            Run {text: "赤".to_string(), color: "FF0000".to_string(), ..Run::default()},
        ],
    }).unwrap();

    writer.finish().unwrap();

    let mut document = Document::open(&path).unwrap();
    let block = collect(&mut document);
    assert_eq!(block.len(), 3);

    assert_eq!(paragraph(&block[0]).style, "Title");
    assert_eq!(paragraph(&block[0]).text(), "題");
    assert_eq!(paragraph(&block[1]).style, "Heading1");

    let one = paragraph(&block[2]);
    assert_eq!(one.align, Align::Center);
    assert_eq!(one.run.len(), 7);
    assert!(one.run[1].bold);
    assert!(one.run[2].italic);
    assert!(one.run[3].underline);
    assert!(one.run[4].strike);
    assert_eq!(one.run[5].size, 14.0);
    assert_eq!(one.run[6].color, "FF0000");
    assert_eq!(one.text(), "ふつう太字斜体下線取り消し大きい赤");
}


#[test]
fn marks_and_breaks_survive_the_trip() {
    let path = place("mark");
    let mut writer = Writer::create(&path).unwrap();
    writer.paragraph(&Paragraph::new("記号 <&>\"' です")).unwrap();
    writer.paragraph(&Paragraph::new("改行\nと\tタブ")).unwrap();
    writer.finish().unwrap();

    let mut document = Document::open(&path).unwrap();
    let block = collect(&mut document);
    assert_eq!(paragraph(&block[0]).text(), "記号 <&>\"' です");
    assert_eq!(paragraph(&block[1]).text(), "改行\nと\tタブ");
}


#[test]
fn a_table_comes_back_with_its_cells() {
    let path = place("table");
    let mut writer = Writer::create(&path).unwrap();

    writer.table(&Table::new(vec![
        Row {
            cell: vec![
                Cell {span: 2, block: vec![Block::Paragraph(Paragraph::new("つながり"))]},
                Cell::new("三"),
            ],
        },
        Row::new(&["あ", "い", "う"]),
    ])).unwrap();

    writer.paragraph(&Paragraph::new("表のあと")).unwrap();
    writer.finish().unwrap();

    let mut document = Document::open(&path).unwrap();
    let block = collect(&mut document);

    let Block::Table(one) = &block[0] else {panic!("表ではありません")};
    assert!(one.border);
    assert_eq!(one.row.len(), 2);
    assert_eq!(one.row[0].cell[0].span, 2);
    assert_eq!(one.row[0].cell[0].text(), "つながり");
    assert_eq!(one.row[0].cell[1].text(), "三");
    assert_eq!(one.row[1].cell.len(), 3);
    assert_eq!(one.row[1].cell[2].text(), "う");
    assert_eq!(one.width(), 3);
}


#[test]
fn a_table_can_hold_another_table() {
    let path = place("nested");
    let inner = Table::new(vec![Row::new(&["おく"])]);

    let outer = Table::new(vec![Row {
        cell: vec![Cell {
            span: 1,
            block: vec![Block::Paragraph(Paragraph::new("そと")), Block::Table(inner)],
        }],
    }]);

    let mut writer = Writer::create(&path).unwrap();
    writer.table(&outer).unwrap();
    writer.finish().unwrap();

    let mut document = Document::open(&path).unwrap();
    let block = collect(&mut document);

    let Block::Table(one) = &block[0] else {panic!("表ではありません")};
    let inside = &one.row[0].cell[0].block;
    assert_eq!(inside.len(), 3, "段落、表、表のあとの空段落");

    let Block::Table(deep) = &inside[1] else {panic!("中に表がありません")};
    assert_eq!(deep.row[0].cell[0].text(), "おく");
}


#[test]
fn the_whole_text_can_be_taken_at_once() {
    let path = place("text");
    let mut writer = Writer::create(&path).unwrap();
    writer.paragraph(&Paragraph::new("一行目")).unwrap();
    writer.table(&Table::new(vec![Row::new(&["あ", "い"])])).unwrap();
    writer.paragraph(&Paragraph::new("三行目")).unwrap();
    writer.finish().unwrap();

    let mut document = Document::open(&path).unwrap();
    let text = document.text().unwrap();
    assert!(text.contains("一行目"));
    assert!(text.contains("あ\tい"));
    assert!(text.contains("三行目"));
}


#[test]
fn many_paragraphs_go_in_and_come_back() {
    let path = place("many");
    let count = 30000;

    let mut writer = Writer::create(&path).unwrap();

    for index in 0..count {
        writer.paragraph(&Paragraph::new(&format!("{index} 番目の段落です。"))).unwrap();
    }

    writer.finish().unwrap();

    let mut document = Document::open(&path).unwrap();
    let mut body = document.read().unwrap();
    let mut seen = 0;

    while let Some(block) = body.read().unwrap() {
        assert_eq!(block.text(), format!("{seen} 番目の段落です。"));
        seen += 1;
    }

    assert_eq!(seen, count);
}


/// 用紙の向きと、ヘッダのパートを持った docx を、手で組んで置く
///
/// folio は用紙もヘッダも作れないので、直しても残ることを確かめるにはこれが要る。
fn build_with_extras(path: &std::path::Path) {
    let head = "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"yes\"?>";
    let main = "http://schemas.openxmlformats.org/wordprocessingml/2006/main";
    let relation = "http://schemas.openxmlformats.org/officeDocument/2006/relationships";
    let base = "application/vnd.openxmlformats-officedocument.wordprocessingml";

    let body = format!(
        "{head}<w:document xmlns:w=\"{main}\" xmlns:r=\"{relation}\"><w:body>\
         <w:p><w:r><w:t>まえ</w:t></w:r></w:p>\
         <w:p><w:pPr><w:sectPr><w:pgSz w:w=\"1\" w:h=\"1\"/></w:sectPr></w:pPr></w:p>\
         <w:sectPr><w:headerReference r:id=\"rId2\" w:type=\"default\"/>\
         <w:pgSz w:w=\"16838\" w:h=\"11906\" w:orient=\"landscape\"/>\
         <w:pgMar w:top=\"567\" w:right=\"567\" w:bottom=\"567\" w:left=\"567\"/></w:sectPr>\
         </w:body></w:document>"
    );

    let header = format!(
        "{head}<w:hdr xmlns:w=\"{main}\"><w:p><w:r><w:t>ヘッダ</w:t></w:r></w:p></w:hdr>"
    );

    let types = format!(
        "{head}<Types xmlns=\"http://schemas.openxmlformats.org/package/2006/content-types\">\
         <Default Extension=\"rels\" ContentType=\"application/vnd.openxmlformats-package.relationships+xml\"/>\
         <Default Extension=\"xml\" ContentType=\"application/xml\"/>\
         <Override PartName=\"/word/document.xml\" ContentType=\"{base}.document.main+xml\"/>\
         <Override PartName=\"/word/header1.xml\" ContentType=\"{base}.header+xml\"/>\
         </Types>"
    );

    let root = format!(
        "{head}<Relationships xmlns=\"http://schemas.openxmlformats.org/package/2006/relationships\">\
         <Relationship Id=\"rId1\" Type=\"{relation}/officeDocument\" Target=\"word/document.xml\"/>\
         </Relationships>"
    );

    let link = format!(
        "{head}<Relationships xmlns=\"http://schemas.openxmlformats.org/package/2006/relationships\">\
         <Relationship Id=\"rId2\" Type=\"{relation}/header\" Target=\"header1.xml\"/>\
         </Relationships>"
    );

    let mut writer = folio::zip::Writer::create(path).unwrap();
    writer.add("[Content_Types].xml", types.as_bytes()).unwrap();
    writer.add("_rels/.rels", root.as_bytes()).unwrap();
    writer.add("word/_rels/document.xml.rels", link.as_bytes()).unwrap();
    writer.add("word/document.xml", body.as_bytes()).unwrap();
    writer.add("word/header1.xml", header.as_bytes()).unwrap();
    writer.finish().unwrap();
}


#[test]
fn editing_keeps_the_page_setup_and_the_other_parts() {
    let source_path = place("extras-source");
    let target_path = place("extras-target");
    build_with_extras(&source_path);

    let source = Document::open(&source_path).unwrap();
    let mut writer = Writer::edit(source, &target_path).unwrap();
    writer.paragraph(&Paragraph::new("あと")).unwrap();
    writer.finish().unwrap();

    let mut document = Document::open(&target_path).unwrap();
    let block = collect(&mut document);
    assert_eq!(block.len(), 1);
    assert_eq!(paragraph(&block[0]).text(), "あと");

    // 本文の後ろにある用紙の指定と、ヘッダのパートが残っている
    let body = String::from_utf8(document.package().unwrap().read("word/document.xml").unwrap()).unwrap();
    assert!(body.contains("w:orient=\"landscape\""), "{body}");
    assert!(body.contains("<w:headerReference r:id=\"rId2\""), "{body}");
    assert!(!body.contains("まえ"), "{body}");

    // 段落の中にあった途中の区切りは、本文と一緒に消える
    assert!(!body.contains("w:w=\"1\""), "{body}");

    let header = document.package().unwrap().read("word/header1.xml").unwrap();
    assert!(String::from_utf8(header).unwrap().contains("ヘッダ"));
}


#[test]
fn a_file_that_is_not_a_document_is_an_error() {
    let path = place("broken");
    std::fs::write(&path, "これは docx ではありません").unwrap();
    assert!(Document::open(&path).is_err());
}