mirror of
https://github.com/storytold/craft-libs.git
synced 2026-11-13 22:12:12 +00:00
Dylan M. Taylor
7fe902292f
craft-rtf: WordCraft's RTF reader and writer on craft-wordmodel and craft-flow (feature doc) * 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. * Add craft-flow: the flow model the text formats map to, on craft-wordmodel WordCraft's crates/formats/src/model (07f97737) moved as it is: Flow, FBlock, Para, Inline, Fmt, tables and cells, pictures, to_doc and from_doc, picture sniffing, colours, base64 and the Dropped report. Only the paths changed (wordcraft_doc is craft_wordmodel); the report helpers are public so the format crates can call them; the image re-encoding helper stays with WordCraft's RTF writer. * wip: WordCraft's RTF interpreter, as it is * Take main's side of files outside craft-rtf * craft-rtf: WordCraft's RTF reader and writer as they are at wordcraft 07f97737 crates/formats/src/rtf.rs and crates/formats/src/rtf/, unchanged and not yet compiled (no module declares them), so the next commit shows only what the move changes. * craft-rtf: WordCraft's RTF reader and writer as craft_rtf::doc (feature doc) On craft-flow and craft-wordmodel, with this crate's code pages (craft-codepage), dates and escaping in place of WordCraft's copies, and craft-metafile's metafile sniffing. A picture in a format RTF cannot carry is re-encoded as PNG with a bounded decode (100 megapixels, 512 MiB), as WordCraft's picture decoder bounded it. The 76 tests moved with it.
1833 lines
93 KiB
Rust
1833 lines
93 KiB
Rust
#![allow(clippy::type_complexity)] // The tests came from an app whose lints allow this.
|
||
use super::*;
|
||
|
||
fn texts(f: &Flow) -> Vec<String> {
|
||
f.blocks
|
||
.iter()
|
||
.map(|b| match b {
|
||
FBlock::Para(p) => format!("{:?}:{}", p.kind, p.text()),
|
||
FBlock::Table(t) => {
|
||
format!("table{}x{}:{}", t.rows.len(), t.cols(), t.rows.iter().flatten().map(|c| c.text()).collect::<Vec<_>>().join("|"))
|
||
}
|
||
})
|
||
.collect()
|
||
}
|
||
|
||
#[test]
|
||
fn basic_reader() {
|
||
let rtf = br#"{\rtf1\ansi{\fonttbl{\f0 Arial;}{\f1 Courier New;}}{\colortbl;\red255\green0\blue0;}{\stylesheet{\s0 Normal;}{\s1 heading 1;}}
|
||
{\info{\title My Doc}{\author Ann}}
|
||
\pard\s1 Head\par
|
||
\pard Plain {\b bold} \i it\i0 {\cf1 red} caf\'e9 \u8364? and\line next\tab x\par
|
||
{\*\unknown skip me}{\field{\*\fldinst{HYPERLINK "http://a.b"}}{\fldrslt{link}}}\par
|
||
\trowd\cellx100\cellx200 \pard\intbl a\cell b\cell\row
|
||
\pard after\par}"#;
|
||
let f = parse(rtf).unwrap();
|
||
assert_eq!(f.meta.title, "My Doc");
|
||
assert_eq!(f.meta.author, "Ann");
|
||
let t = texts(&f);
|
||
assert_eq!(t[0], "Heading(1):Head");
|
||
assert_eq!(t[1], "Normal:Plain bold it red café € and\nnext\tx");
|
||
assert_eq!(t[2], "Normal:link");
|
||
assert_eq!(t[3], "table1x2:a|b");
|
||
assert_eq!(t[4], "Normal:after");
|
||
let FBlock::Para(p) = &f.blocks[1] else { panic!() };
|
||
assert!(p.inlines.iter().any(|i| matches!(i, Inline::Text(t, f) if t == "bold" && f.bold)));
|
||
assert!(p.inlines.iter().any(|i| matches!(i, Inline::Text(t, f) if t == "red" && f.color == Some(Rgb(255, 0, 0)))));
|
||
let FBlock::Para(l) = &f.blocks[2] else { panic!() };
|
||
assert!(matches!(&l.inlines[0], Inline::Text(_, f) if f.link.as_deref() == Some("http://a.b")));
|
||
}
|
||
|
||
#[test]
|
||
fn stylesheet_names_take_unicode_escapes() {
|
||
// \u116?=t \u105?=i \u116?=t \u108?=l \u101?=e — how Word writes a non-Latin style name.
|
||
let rtf = br"{\rtf1\ansi{\stylesheet{\s0 Normal;}{\s2 \u116?\u105?\u116?\u108?\u101?;}}\pard\s2 T\par}";
|
||
let f = parse(rtf).unwrap();
|
||
assert_eq!(texts(&f)[0], "Title:T");
|
||
}
|
||
|
||
#[test]
|
||
fn stylesheet_names_take_negative_escapes() {
|
||
// 't' as a signed 16-bit escape: 116 - 65536 = -65420.
|
||
let rtf = br"{\rtf1\ansi{\stylesheet{\s0 Normal;}{\s2 \u-65420?\u105?\u116?\u108?\u101?;}}\pard\s2 T\par}";
|
||
let f = parse(rtf).unwrap();
|
||
assert_eq!(texts(&f)[0], "Title:T");
|
||
}
|
||
|
||
#[test]
|
||
fn stylesheet_names_honour_the_uc_fallback_count() {
|
||
// \uc2: two fallback bytes follow each \uN? — here a second `?`.
|
||
let rtf = br"{\rtf1\ansi{\stylesheet{\s0 Normal;}{\s2 \uc2\u116??\u105??\u116??\u108??\u101??;}}\pard\s2 T\par}";
|
||
let f = parse(rtf).unwrap();
|
||
assert_eq!(texts(&f)[0], "Title:T");
|
||
}
|
||
|
||
#[test]
|
||
fn a_malformed_hex_escape_leaves_no_stray_text() {
|
||
// A `\'` that is not followed by two hex digits is malformed: the escape itself goes, any
|
||
// hex digits of a truncated pair go with it, and everything else — non-hex bytes, `}` —
|
||
// is read as what it is. `\'gZ` keeps "gZ" as text; `\'4}` keeps the closing brace.
|
||
let rtf = br"{\rtf1\ansi \'gZok\par}";
|
||
let f = parse(rtf).unwrap();
|
||
assert_eq!(texts(&f)[0], "Normal:gZok");
|
||
|
||
// The bytes after a truncated pair must not swallow the group's closing brace or the next
|
||
// control word: the group and its destination state end where they should. Paragraph text
|
||
// alone would not show that: `survivor` reads the same inside the bold group, so the run's
|
||
// formatting is checked too.
|
||
let rtf = br"{\rtf1\ansi{\b \'4}survivor\par}";
|
||
let f = parse(rtf).unwrap();
|
||
assert!(texts(&f).iter().any(|t| t.contains("survivor")), "{:?}", texts(&f));
|
||
let survivor_bold = f.blocks.iter().find_map(|b| match b {
|
||
FBlock::Para(p) => p.inlines.iter().find_map(|i| match i {
|
||
Inline::Text(t, fmt) if t.contains("survivor") => Some(fmt.bold),
|
||
_ => None,
|
||
}),
|
||
FBlock::Table(_) => None,
|
||
});
|
||
assert_eq!(survivor_bold, Some(false), "survivor must not be inside the bold group: {:?}", f.blocks);
|
||
// The truncated pair's hex digit goes with the escape: no stray "4" is left in the text.
|
||
assert!(!texts(&f).iter().any(|t| t.contains('4')), "{:?}", texts(&f));
|
||
|
||
// A leading `+` is not a hex digit: `\'+4` is no escape: it stays text, not the byte 0x04.
|
||
let rtf = br"{\rtf1\ansi \'+4z\par}";
|
||
let f = parse(rtf).unwrap();
|
||
assert_eq!(texts(&f)[0], "Normal:+4z");
|
||
|
||
let rtf = br"{\rtf1\ansi{\'}}after\par}";
|
||
let f = parse(rtf).unwrap();
|
||
assert!(texts(&f).iter().any(|t| t.contains("after")), "{:?}", texts(&f));
|
||
|
||
// A hex digit that starts a valid pair with the byte after it is part of the escape:
|
||
// \'4} reads '4' as the escape's byte and leaves `}` to close the group.
|
||
let rtf = br"{\rtf1\ansi\'4}x\par}";
|
||
let f = parse(rtf).unwrap();
|
||
assert!(texts(&f).iter().any(|t| t.contains('x') && !t.contains('}')), "{:?}", texts(&f));
|
||
}
|
||
|
||
#[test]
|
||
fn unicode_fallback_uses_the_declared_count() {
|
||
for (rtf, expected) in [
|
||
(br"{\rtf1\uc18\u8804less than or equal rest}".as_slice(), "≤ rest"),
|
||
(br"{\rtf1\uc10\u233\'30\'31\'32\'33\'34\'35\'36\'37\'38\'39 rest}", "é rest"),
|
||
(br"{\rtf1\uc2\u8804\:ab rest}", "≤b rest"),
|
||
(br"{\rtf1\uc18\u8804\:12345678901234567 rest}", "≤ rest"),
|
||
(br"{\rtf1\uc1\u8804\:ab rest}", "≤ab rest"),
|
||
(br"{\rtf1\uc0\u8804\:ab rest}", "≤ab rest"),
|
||
(br"{\rtf1\uc9\u233abcdefgh\par VISIBLE}", "éVISIBLE"),
|
||
(br"{\rtf1\uc1\u233\pard VISIBLE}", "éVISIBLE"),
|
||
(br"{\rtf1\uc1\u233\cell VISIBLE}", "éVISIBLE"),
|
||
(br"{\rtf1\uc1\u233\row VISIBLE}", "éVISIBLE"),
|
||
(br"{\rtf1\uc0\u233\par VISIBLE}", "é\nVISIBLE"),
|
||
(b"{\\rtf1\\uc9\\u233abcdefgh\\\nVISIBLE}", "éVISIBLE"),
|
||
(b"{\\rtf1\\uc9\\u233abcdefgh\\\rVISIBLE}", "éVISIBLE"),
|
||
(br"{\rtf1\uc9\u233abcdefgh\*VISIBLE}", "éVISIBLE"),
|
||
(b"{\\rtf1\\uc0\\u233\\\nVISIBLE}", "é\nVISIBLE"),
|
||
(b"{\\rtf1\\uc0\\u233\\\rVISIBLE}", "é\nVISIBLE"),
|
||
(br"{\rtf1\uc0\u233\*VISIBLE}", "éVISIBLE"),
|
||
(br"{\rtf1\uc0\u233X}", "éX"),
|
||
(br"{\rtf1\uc1\u233?X}", "éX"),
|
||
(br"{\rtf1{\uc2147483647\u233?}X}", "éX"),
|
||
(br"{\rtf1\uc-1\u233X}", "éX"),
|
||
] {
|
||
let doc = import(rtf).unwrap();
|
||
assert_eq!(doc.plain_text(craft_wordmodel::StoryRef::Body).trim(), expected, "{rtf:?}");
|
||
}
|
||
}
|
||
|
||
#[test]
|
||
fn unicode_fallback_info_words_preserve_following_metadata() {
|
||
for (rtf, title, body) in [
|
||
(br"{\rtf1{\info{\title\uc9\u233abcdefgh\info X}{\author Alice}} Text}".as_slice(), "éX", "Text"),
|
||
(br"{\rtf1{\info{\title\uc1\u233\info X}{\author Alice}} Text}", "éX", "Text"),
|
||
(br"{\rtf1{\info{\title\uc0\u233X}{\author Alice}} Text}", "éX", "Text"),
|
||
(br"{\rtf1\uc0\u233{\info{\title Plain}{\author Alice}}Text}", "Plain", "éText"),
|
||
] {
|
||
let doc = import(rtf).unwrap();
|
||
assert_eq!(doc.core.title, title, "{rtf:?}");
|
||
assert_eq!(doc.core.creator, "Alice", "{rtf:?}");
|
||
assert_eq!(doc.plain_text(craft_wordmodel::StoryRef::Body).trim(), body, "{rtf:?}");
|
||
}
|
||
}
|
||
|
||
#[test]
|
||
fn unicode_fallback_in_revision_authors_counts_control_words() {
|
||
for (rtf, expected) in [
|
||
(br"{\rtf1{\*\revtbl{Unknown;}{\uc9\u233abcdefgh\par Zed;}}{\revised\revauth1 text}}".as_slice(), "éZed"),
|
||
(br"{\rtf1{\*\revtbl{Unknown;}{\uc0\u233\par Zed;}}{\revised\revauth1 text}}", "éZed"),
|
||
] {
|
||
let doc = import(rtf).unwrap();
|
||
assert_eq!(doc.revisions.first().map(|r| r.author.as_str()), Some(expected), "{rtf:?}");
|
||
}
|
||
}
|
||
|
||
#[test]
|
||
fn paragraph_borders() {
|
||
// A heading with a grey rule under it, as hh.ru résumés write their section titles.
|
||
let rtf = br"{\rtf1\ansi{\colortbl;\red0\green0\blue0;\red216\green216\blue216;}
|
||
\pard\brdrb\brdrs\brdrw10\brdrcf2 Section\par
|
||
\pard Body\par
|
||
\pard\brdrt\brdrdb\brdrl\brdrs\brdrr\brdrs Boxed\par
|
||
\pard\brdrb\brdrs\par}";
|
||
let f = parse(rtf).unwrap();
|
||
let paras: Vec<&Para> = f.blocks.iter().filter_map(|b| if let FBlock::Para(p) = b { Some(p) } else { None }).collect();
|
||
let b = paras[0].borders.unwrap();
|
||
let bottom = b.bottom.unwrap();
|
||
assert_eq!((bottom.width, bottom.color), (0.5, Some(Rgb(216, 216, 216))));
|
||
assert!(b.top.is_none());
|
||
assert_eq!(paras[0].kind, Kind::Normal);
|
||
// \pard resets them.
|
||
assert!(paras[1].borders.is_none());
|
||
let b = paras[2].borders.unwrap();
|
||
assert_eq!(b.top.map(|x| x.style), Some(craft_wordmodel::props::BorderStyle::Double));
|
||
assert!(b.left.is_some() && b.right.is_some() && b.bottom.is_none());
|
||
// An empty paragraph with only a bottom border is still a rule.
|
||
assert_eq!(paras[3].kind, Kind::Rule);
|
||
// The document's paragraph carries them.
|
||
let d = import(rtf).unwrap();
|
||
let first = d.body.iter().find_map(|b| if let craft_wordmodel::Block::Para(p) = &**b { Some(p) } else { None }).unwrap();
|
||
assert!(first.props.borders.and_then(|b| b.bottom).is_some());
|
||
// Paragraph borders round-trip through export and import.
|
||
let exported = export(&d);
|
||
assert!(exported.contains("\\brdrb\\brdrs\\brdrw10\\brdrcf1"), "{exported}");
|
||
let back = import(exported.as_bytes()).unwrap();
|
||
let first = back.body.iter().find_map(|b| if let craft_wordmodel::Block::Para(p) = &**b { Some(p) } else { None }).unwrap();
|
||
let b = first.props.borders.unwrap().bottom.unwrap();
|
||
assert_eq!((b.style, b.width, b.color), (craft_wordmodel::props::BorderStyle::Single, 0.5, Some(Rgb(216, 216, 216))));
|
||
}
|
||
|
||
#[test]
|
||
fn headers_and_footers() {
|
||
let rtf = br"{\rtf1\ansi\titlepg
|
||
{\header \pard\qr Running head\par}
|
||
{\headerf \pard First head {\b bold}\par}
|
||
{\footer \pard Page \chpgn\par}
|
||
\pard Body text\par
|
||
\trowd\cellx2000\pard\intbl cell\cell\row
|
||
\pard after\par}";
|
||
let d = import(rtf).unwrap();
|
||
let s = &d.last_section;
|
||
assert!(s.title_page);
|
||
let text = |id: Option<u32>| {
|
||
let p = id.and_then(|id| d.parts.get(&id)).unwrap();
|
||
p.blocks
|
||
.iter()
|
||
.filter_map(|b| if let craft_wordmodel::Block::Para(p) = &**b { Some(p.text.clone()) } else { None })
|
||
.collect::<Vec<_>>()
|
||
.join("|")
|
||
};
|
||
assert_eq!(text(s.headers.default), "Running head");
|
||
assert_eq!(text(s.headers.first), "First head bold");
|
||
assert!(text(s.footers.default).starts_with("Page"));
|
||
assert!(s.footers.first.is_none() && s.headers.even.is_none());
|
||
// The body keeps only its own text, and its table survives the header groups.
|
||
let f = parse(rtf).unwrap();
|
||
let t = texts(&f);
|
||
assert_eq!(t, ["Normal:Body text", "table1x1:cell", "Normal:after"]);
|
||
// Without \titlepg there is no separate first page.
|
||
let d = import(br"{\rtf1\ansi{\headerf x\par} y\par}").unwrap();
|
||
assert!(!d.last_section.title_page && d.last_section.headers.first.is_some());
|
||
// An unterminated header still ends, and the body text after it isn't lost.
|
||
assert!(import(br"{\rtf1\ansi{\header x").is_ok());
|
||
}
|
||
|
||
#[test]
|
||
fn paragraph_spacing_and_indents() {
|
||
use craft_wordmodel::props::LineSpacing;
|
||
let rtf = br"{\rtf1\ansi
|
||
\pard\sb120\sa150\sl360\slmult1\li720\ri360\fi-360 a\par
|
||
\pard\sl-280 b\par
|
||
\pard\sl300\slmult0 c\par
|
||
\pard d\par}";
|
||
let f = parse(rtf).unwrap();
|
||
let sp: Vec<model::Spacing> = f.blocks.iter().filter_map(|b| if let FBlock::Para(p) = b { Some(p.spacing) } else { None }).collect();
|
||
assert_eq!(sp[0].before, Some(6.0));
|
||
assert_eq!(sp[0].after, Some(7.5));
|
||
assert_eq!(sp[0].line, Some(LineSpacing::Multiple(1.5)));
|
||
assert_eq!((sp[0].left, sp[0].right, sp[0].first), (Some(36.0), Some(18.0), Some(-18.0)));
|
||
assert_eq!(sp[1].line, Some(LineSpacing::Exactly(14.0)));
|
||
assert_eq!(sp[2].line, Some(LineSpacing::AtLeast(15.0)));
|
||
// \pard resets them.
|
||
assert_eq!(sp[3], model::Spacing::default());
|
||
// The document's defaults are RTF's: single, no space after; the paragraph's own win.
|
||
let d = import(rtf).unwrap();
|
||
assert_eq!(d.styles.default_para.space_after, Some(0.0));
|
||
assert_eq!(d.styles.default_para.line_spacing, Some(LineSpacing::Multiple(1.0)));
|
||
let first = d.body.iter().find_map(|b| if let craft_wordmodel::Block::Para(p) = &**b { Some(p) } else { None }).unwrap();
|
||
assert_eq!(first.props.space_after, Some(7.5));
|
||
assert_eq!(first.props.indent_first, Some(-18.0));
|
||
}
|
||
|
||
#[test]
|
||
fn default_font_is_deff_or_font_zero() {
|
||
// Text that never selects a font is in the default font: \deff, else font 0.
|
||
let d = import(br"{\rtf1\ansi{\fonttbl{\f0\fnil Times New Roman Cyr;}{\f1 Arial;}}\plain x\par}").unwrap();
|
||
assert_eq!(d.styles.default_chr.font.as_deref(), Some("Times New Roman Cyr"));
|
||
let d = import(br"{\rtf1\ansi\deff1{\fonttbl{\f0 Times New Roman;}{\f1 Arial;}} x\par}").unwrap();
|
||
assert_eq!(d.styles.default_chr.font.as_deref(), Some("Arial"));
|
||
// No font table: our default stays.
|
||
let d = import(br"{\rtf1\ansi x\par}").unwrap();
|
||
assert_eq!(d.styles.default_chr.font, craft_wordmodel::StyleSheet::builtin().default_chr.font);
|
||
}
|
||
|
||
#[test]
|
||
fn unicode_escapes_in_font_names() {
|
||
// Ordinary fallback, hex fallback and \uc0 all give "Café Sans"; the fallback is not kept.
|
||
for name in [r"Caf\u233? Sans", r"Caf\u233\'e9 Sans", r"\uc0 Caf\u233 Sans", r"\uc2 Caf\u233\'e9\'e9 Sans"] {
|
||
let rtf = format!(r"{{\rtf1\ansi\deff0{{\fonttbl{{\f0 {name};}}{{\f1 \u923?ambda;}}}}\plain x {{\f1 y}}\par}}");
|
||
let d = import(rtf.as_bytes()).unwrap();
|
||
assert_eq!(d.styles.default_chr.font.as_deref(), Some("Café Sans"), "{name}");
|
||
// A body run in a Unicode-named font keeps that name.
|
||
let f = parse(rtf.as_bytes()).unwrap();
|
||
let y_font = f.blocks.iter().find_map(|b| match b {
|
||
FBlock::Para(p) => p.inlines.iter().find_map(|i| match i {
|
||
Inline::Text(t, fmt) if t.trim() == "y" => fmt.font.clone(),
|
||
_ => None,
|
||
}),
|
||
FBlock::Table(_) => None,
|
||
});
|
||
assert_eq!(y_font.as_deref(), Some("Λambda"), "{name}");
|
||
// Our export writes font names as `\uN?`; reading it back gives the same name.
|
||
let back = import(export(&d).as_bytes()).unwrap();
|
||
assert_eq!(back.styles.default_chr.font.as_deref(), Some("Café Sans"), "{name}");
|
||
}
|
||
}
|
||
|
||
#[test]
|
||
fn page_size_and_margins() {
|
||
let d = import(br"{\rtf1\ansi\paperw11906\paperh16838\margl1134\margr850\margt1134\margb1134 x\par}").unwrap();
|
||
let s = &d.last_section;
|
||
assert!((s.page_w - 595.3).abs() < 0.1 && (s.page_h - 841.9).abs() < 0.1, "{} x {}", s.page_w, s.page_h);
|
||
assert!((s.margin_left - 56.7).abs() < 0.1 && (s.margin_right - 42.5).abs() < 0.1);
|
||
assert!(!s.landscape);
|
||
let d = import(br"{\rtf1\ansi\paperw11906\paperh16838\landscape x\par}").unwrap();
|
||
assert!(d.last_section.landscape && d.last_section.page_w > d.last_section.page_h);
|
||
let exported = export(&d);
|
||
assert!(exported.contains("\\landscape"), "exported RTF must include \\landscape: {exported}");
|
||
// Nonsense keeps Word's defaults.
|
||
let d = import(br"{\rtf1\ansi\paperw0\margl99999 x\par}").unwrap();
|
||
assert_eq!(d.last_section, craft_wordmodel::SectionProps::default());
|
||
}
|
||
|
||
#[test]
|
||
fn table_grid_from_cell_edges_and_borders() {
|
||
// Rows whose cells don't line up share one grid: edges 2000, 3000, 9000.
|
||
let rtf = br"{\rtf1\ansi
|
||
\trowd\cellx2000\cellx9000 \pard\intbl a\cell b\cell\row
|
||
\trowd\cellx3000\cellx9000 \pard\intbl c\cell d\cell\row
|
||
\pard after\par}";
|
||
let f = parse(rtf).unwrap();
|
||
let FBlock::Table(t) = &f.blocks[0] else { panic!() };
|
||
assert_eq!(t.widths, [100.0, 50.0, 300.0]);
|
||
assert_eq!(t.rows[0].iter().map(|c| c.colspan).collect::<Vec<_>>(), [1, 2]);
|
||
assert_eq!(t.rows[1].iter().map(|c| c.colspan).collect::<Vec<_>>(), [2, 1]);
|
||
assert_eq!(t.cols(), 3);
|
||
// RTF tables have no grid style: each cell draws the borders it names, nothing else.
|
||
assert!(t.borderless);
|
||
assert!(t.rows.iter().flatten().all(|c| c.borders.is_none()));
|
||
// A rule under a cell (a letterhead's line): the bottom side only, its width and colour.
|
||
let f = parse(br"{\rtf1\ansi{\colortbl;\red174\green174\blue174;}\trowd\clbrdrb\brdrs\brdrw15\brdrcf1\cellx2000\pard\intbl a\cell\row}").unwrap();
|
||
let FBlock::Table(t) = &f.blocks[0] else { panic!() };
|
||
let b = t.rows[0][0].borders.unwrap();
|
||
assert!(b.top.is_none() && b.left.is_none() && b.right.is_none());
|
||
let bottom = b.bottom.unwrap();
|
||
assert_eq!((bottom.style, bottom.width, bottom.color), (craft_wordmodel::props::BorderStyle::Single, 0.75, Some(Rgb(174, 174, 174))));
|
||
// Every side, in different styles, per cell.
|
||
let f = parse(
|
||
br"{\rtf1\ansi\trowd\clbrdrt\brdrdb\clbrdrl\brdrs\brdrw10\clbrdrb\brdrdot\clbrdrr\brdrth\cellx2000\cellx4000\pard\intbl a\cell b\cell\row}",
|
||
)
|
||
.unwrap();
|
||
let FBlock::Table(t) = &f.blocks[0] else { panic!() };
|
||
let b = t.rows[0][0].borders.unwrap();
|
||
use craft_wordmodel::props::BorderStyle as S;
|
||
assert_eq!(
|
||
[b.top.map(|x| x.style), b.left.map(|x| x.style), b.bottom.map(|x| x.style), b.right.map(|x| x.style)],
|
||
[Some(S::Double), Some(S::Single), Some(S::Dotted), Some(S::Thick)]
|
||
);
|
||
assert_eq!(b.left.map(|x| x.width), Some(0.5));
|
||
assert!(t.rows[0][1].borders.is_none(), "the second cell names no borders");
|
||
// A border side set to none isn't a border.
|
||
let f = parse(br"{\rtf1\ansi\trowd\clbrdrl\brdrnone\cellx2000\pard\intbl a\cell\row}").unwrap();
|
||
let FBlock::Table(t) = &f.blocks[0] else { panic!() };
|
||
assert!(t.rows[0][0].borders.is_none());
|
||
// The document's cells carry them.
|
||
let d = import(br"{\rtf1\ansi\trowd\clbrdrb\brdrs\cellx2000\pard\intbl a\cell\row}").unwrap();
|
||
let t = d.body.iter().find_map(|b| if let craft_wordmodel::Block::Table(t) = &**b { Some(t) } else { None }).unwrap();
|
||
assert!(t.props.style.is_none());
|
||
assert!(t.rows[0].cells[0].props.borders.and_then(|b| b.bottom).is_some());
|
||
// Borderless tables export without cell borders and round-trip borderless.
|
||
let d = import(br"{\rtf1\ansi\trowd\cellx2000\pard\intbl a\cell\row}").unwrap();
|
||
let exported = export(&d);
|
||
assert!(!exported.contains("\\clbrdr"), "{exported}");
|
||
let back = import(exported.as_bytes()).unwrap();
|
||
let tb = back.body.iter().find_map(|b| if let craft_wordmodel::Block::Table(t) = &**b { Some(t) } else { None }).unwrap();
|
||
assert!(tb.props.style.is_none());
|
||
assert!(tb.rows[0].cells[0].props.borders.is_none());
|
||
|
||
// Cell borders round-trip through export and import.
|
||
let d =
|
||
import(br"{\rtf1\ansi{\colortbl;\red174\green174\blue174;}\trowd\clbrdrb\brdrdb\brdrw15\brdrcf1\cellx2000\pard\intbl a\cell\row}").unwrap();
|
||
let exported = export(&d);
|
||
assert!(exported.contains("\\clbrdrb\\brdrdb\\brdrw15\\brdrcf1"), "{exported}");
|
||
let back = import(exported.as_bytes()).unwrap();
|
||
let tb = back.body.iter().find_map(|b| if let craft_wordmodel::Block::Table(t) = &**b { Some(t) } else { None }).unwrap();
|
||
let b = tb.rows[0].cells[0].props.borders.unwrap().bottom.unwrap();
|
||
assert_eq!((b.style, b.width, b.color), (craft_wordmodel::props::BorderStyle::Double, 0.75, Some(Rgb(174, 174, 174))));
|
||
|
||
// The table sizes to the document's page, not Letter.
|
||
let d = import(br"{\rtf1\ansi\paperw11906\margl1134\margr1134\trowd\cellx4818\cellx9638\pard\intbl a\cell b\cell\row}").unwrap();
|
||
let t = d.body.iter().find_map(|b| if let craft_wordmodel::Block::Table(t) = &**b { Some(t) } else { None }).unwrap();
|
||
assert_eq!(t.grid, [240.9, 241.0]);
|
||
}
|
||
|
||
#[test]
|
||
fn code_pages_from_ansicpg_and_font_charsets() {
|
||
// The document's code page: Russian RTF as accounting systems write it.
|
||
let f =
|
||
parse(br"{\rtf1\ansi\ansicpg1251\deff0{\fonttbl{\f0 Times New Roman;}}\f0 \'cf\'f0\'e8\'e2\'e5\'f2, \'b8\'eb\'ea\'e0 \'b9 5\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:Привет, ёлка № 5"]);
|
||
// Font charsets switch inside a Windows-1252 document: Western, Cyrillic, Central
|
||
// European, then back to the first font.
|
||
let f = parse(
|
||
br"{\rtf1\ansi\ansicpg1252\deff0{\fonttbl{\f0 Arial;}{\f1\fcharset204 Arial CYR;}{\f2\fcharset238 Arial CE;}}\f0 caf\'e9 {\f1 \'c4\'e0} \f2 \'a3\'f3d\'9f \f0 na\'efve\par}",
|
||
)
|
||
.unwrap();
|
||
assert_eq!(texts(&f), ["Normal:café Да Łódź naïve"]);
|
||
// The default font's charset applies before any \f; an explicit \cpg wins over \fcharset.
|
||
let f = parse(br"{\rtf1\ansi\deff1{\fonttbl{\f1\fcharset204 Times;}{\f2\fcharset0\cpg1253 Greek;}}\'cc\'e8\'f0 \f2 \'c1\'e2\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:Мир Αβ"]);
|
||
// Raw 8-bit text (not escaped) follows the same code page.
|
||
let f = parse(b"{\\rtf1\\ansi\\ansicpg1251 \xc4\xee\xec\\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:Дом"]);
|
||
// #1653: a Macintosh document (LibreOffice's fdo85889-mac.rtf) is Mac OS Roman, and so is
|
||
// a font with the Mac charset in a Windows document.
|
||
let f = parse(br"{\rtf1\mac \deff0{\fonttbl{\f0\fmodern Helvetica;}}\pard\f0\fs20 \'f1\'f2\'f3\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:ÒÚÛ"]);
|
||
let f = parse(br"{\rtf1\ansi\deff0{\fonttbl{\f0 Arial;}{\f1\fcharset77 Geneva;}}caf\'e9 {\f1 caf\'8e \'d2ok\'d3}\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:café café “ok”"]);
|
||
// \fcharset255 is the standard OEM character set (CP437).
|
||
let f = parse(br"{\rtf1\ansi\deff0{\fonttbl{\f0 Arial;}{\f1\fcharset255 Terminal;}}caf\'e9 {\f1 caf\'82}\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:café café"]);
|
||
// DOS documents: `\pc` is code page 437, `\pca` 850; a font can name one with `\cpg`.
|
||
let f = parse(br"{\rtf1\pc\deff0{\fonttbl{\f0 Courier;}}caf\'82 \'a0 la carte\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:café á la carte"]);
|
||
let f = parse(br"{\rtf1\pca\deff0{\fonttbl{\f0 Courier;}{\f1\cpg437 Courier;}}\'9d {\f1 \'9d}\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:Ø ¥"]);
|
||
// A code page number out of range is ignored, not truncated to another one.
|
||
let f = parse(br"{\rtf1\ansi\ansicpg65973 caf\'e9\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:café"]);
|
||
// The OEM charset (255) is DOS code page 437, like 254.
|
||
let f = parse(br"{\rtf1\ansi\deff0{\fonttbl{\f0\fcharset255 Courier New;}}\f0 caf\'82\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:café"]);
|
||
}
|
||
|
||
/// #1653: the East Asian code pages (932 Shift-JIS, 936 GBK, 949 UHC, 950 Big5) decode their
|
||
/// bytes as a run, so a lead byte and its trail byte decode together, escaped or raw.
|
||
#[test]
|
||
fn east_asian_multibyte_code_pages() {
|
||
// Japanese `\ansicpg932` (CP932: 日 = 93 fa, 本 = 96 7b, 語 = 8c ea).
|
||
let f = parse(br"{\rtf1\ansi\ansicpg932 \'93\'fa\'96\'7b\'8c\'ea\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:日本語"]);
|
||
// Raw 8-bit text (not `\'hh`), and raw bytes mixed with one escaped byte: RTF can't carry a
|
||
// raw `{`, `}` or `\`, so 本's trail byte 7b is written `\'7b` here.
|
||
let f = parse(b"{\\rtf1\\ansi\\ansicpg932 \x8a\xbf\x8e\x9a\\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:漢字"]);
|
||
let f = parse(b"{\\rtf1\\ansi\\ansicpg932 \x93\xfa\x96\\'7b\x8c\xea\\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:日本語"]);
|
||
// A font names the code page with `\fcharset128` or with `\cpg932`, and `\f0` selects it.
|
||
let f = parse(br"{\rtf1\ansi\ansicpg1252\deff0{\fonttbl{\f0\fcharset128 MS Mincho;}}\f0 \'93\'fa\'96\'7b\'8c\'ea\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:日本語"]);
|
||
let f = parse(br"{\rtf1\ansi\ansicpg1252\deff0{\fonttbl{\f0\cpg932 MS Mincho;}}\f0 \'93\'fa\'96\'7b\'8c\'ea\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:日本語"]);
|
||
// Raw 8-bit bytes in the font table (the font's name) decode in the document's code page.
|
||
let d = import(b"{\\rtf1\\ansi\\ansicpg932\\deff0{\\fonttbl{\\f0\\fcharset128 \x96\xbe\x92\xa9;}}\\f0 x\\par}").unwrap();
|
||
assert_eq!(d.styles.default_chr.font.as_deref(), Some("明朝"));
|
||
// Korean `\ansicpg949` (UHC: 한국어 = c7 d1 b1 b9 be ee).
|
||
let f = parse(br"{\rtf1\ansi\ansicpg949 \'c7\'d1\'b1\'b9\'be\'ee\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:한국어"]);
|
||
// Simplified Chinese `\ansicpg936` (GBK: 中文 = d6 d0 ce c4).
|
||
let f = parse(br"{\rtf1\ansi\ansicpg936 \'d6\'d0\'ce\'c4\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:中文"]);
|
||
// Traditional Chinese `\ansicpg950` (Big5: 中文 = a4 a4 a4 e5).
|
||
let f = parse(br"{\rtf1\ansi\ansicpg950 \'a4\'a4\'a4\'e5\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:中文"]);
|
||
// `\uN?` escapes still decode alongside multibyte text (`\u26412?` is 本, its `?` fallback
|
||
// skipped by `\uc1`).
|
||
let f = parse(br"{\rtf1\ansi\ansicpg932 \'93\'fa\u26412?\'8c\'ea x\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:日本語 x"]);
|
||
// An invalid sequence decodes as U+FFFD, as Word writes it, and doesn't lose the rest.
|
||
let f = parse(b"{\\rtf1\\ansi\\ansicpg932 \x93\\par after\\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:\u{FFFD}", "Normal:after"]);
|
||
let f = parse(b"{\\rtf1\\ansi\\ansicpg932 \x93\xfa\x93 x\x96\x8c\\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:日\u{FFFD} x膜"]);
|
||
}
|
||
|
||
/// The decoder never sniffs a BOM: the code page is named by `\ansicpg`, `\cpg` or `\fcharset`,
|
||
/// so a run that starts `EF BB BF` is a character pair and a lead byte, not a UTF-8 BOM (which
|
||
/// would switch the rest of the document to UTF-8 and garble it).
|
||
#[test]
|
||
fn a_bom_like_byte_triple_is_text_in_a_multibyte_code_page() {
|
||
// GBK EF BB = 锘, then BF B5 = 康; UHC EF BB = 癤, then BF C0 = 오; Big5 EF BB = 嚜, then
|
||
// BF 40 = 濃.
|
||
let f = parse(br"{\rtf1\ansi\ansicpg936 \'ef\'bb\'bf\'b5\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:锘康"]);
|
||
let f = parse(br"{\rtf1\ansi\ansicpg949 \'ef\'bb\'bf\'c0\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:癤오"]);
|
||
let f = parse(br"{\rtf1\ansi\ansicpg950 \'ef\'bb\'bf\'40\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:嚜濃"]);
|
||
}
|
||
|
||
/// A multibyte run longer than any fixed buffer keeps every character: the streaming decoder
|
||
/// holds a lead byte and its trail byte together across decode steps instead of cutting the
|
||
/// character at a buffer size (the review's finding on the byte-capped flush).
|
||
#[test]
|
||
fn a_long_multibyte_run_never_splits_a_character() {
|
||
let mut bytes = b"{\\rtf1\\ansi\\ansicpg932 ".to_vec();
|
||
bytes.push(b'A');
|
||
// 2048 × あ (CP932 82 a0) is 4096 bytes after the one ASCII byte: a fixed 4096-byte buffer
|
||
// would flush with the last あ's lead byte in one step and its trail byte in the next.
|
||
for _ in 0..2048 {
|
||
bytes.extend_from_slice(&[0x82, 0xA0]);
|
||
}
|
||
bytes.extend_from_slice(b"\\par}");
|
||
let f = parse(&bytes).unwrap();
|
||
let expected: String = std::iter::once('A').chain((0..2048).map(|_| 'あ')).collect();
|
||
assert_eq!(texts(&f), [format!("Normal:{expected}")]);
|
||
}
|
||
|
||
/// A lead byte with no trail byte at the end of the document becomes U+FFFD, as Word writes a
|
||
/// malformed sequence — the streaming decoder is finalised, not left holding it.
|
||
#[test]
|
||
fn an_incomplete_lead_byte_at_the_end_is_replacement() {
|
||
let f = parse(b"{\\rtf1\\ansi\\ansicpg932 \x93}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:\u{FFFD}"]);
|
||
}
|
||
|
||
/// #1653: a code page switch (`\f`, `\fcharset`, `\cpg`, `\ansicpg`), a group end and the end
|
||
/// of the document decode the multibyte bytes read so far, before the new code page applies.
|
||
#[test]
|
||
fn multibyte_code_page_switches_flush_pending_bytes() {
|
||
// Japanese (932) → ASCII → Cyrillic (`\fcharset204` with `\cpg1251`): each run decodes in
|
||
// its own code page, the last bytes before the switch included.
|
||
let f = parse(
|
||
br"{\rtf1\ansi\ansicpg932\deff0{\fonttbl{\f0 MS Mincho;}{\f1\fcharset204\cpg1251 Arial CYR;}}\f0 \'93\'fa\'96\'7b\'8c\'ea mixed \f1 \'cf\'f0\'e8\'e2\'e5\'f2\par}",
|
||
)
|
||
.unwrap();
|
||
assert_eq!(texts(&f), ["Normal:日本語 mixed Привет"]);
|
||
// A group end flushes: the group's 932 bytes decode before the outer 1252 text.
|
||
let f = parse(br"{\rtf1\ansi\ansicpg1252\deff0{\fonttbl{\f0 Arial;}{\f1\fcharset128 MS Mincho;}}\f0 {\f1 \'93\'fa\'96\'7b\'8c\'ea}caf\'e9\par}")
|
||
.unwrap();
|
||
assert_eq!(texts(&f), ["Normal:日本語café"]);
|
||
// An `\ansicpg` switch mid-document flushes too.
|
||
let f = parse(br"{\rtf1\ansi\ansicpg932 \'93\'fa\ansicpg1251 \'cf\'f0\'e8\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:日При"]);
|
||
// A run left open at the end of the document isn't lost.
|
||
let f = parse(b"{\\rtf1\\ansi\\ansicpg932 \x93\xfa").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:日"]);
|
||
}
|
||
|
||
/// A Windows-1252 document (the default) reads exactly as before: high bytes one at a time.
|
||
#[test]
|
||
fn windows_1252_text_is_unchanged() {
|
||
let f = parse(b"{\\rtf1\\ansi\\ansicpg1252 caf\xe9 na\xefve \x93quoted\x94 \\'80 5\\'b0\\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:café naïve “quoted” € 5°"]);
|
||
// No `\ansicpg`: the default is 1252; the byte values of an East Asian code page keep
|
||
// their 1252 meaning.
|
||
let f = parse(b"{\\rtf1\\ansi \xe9\\'92\\'93 \\'fa\\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:é’“ ú"]);
|
||
}
|
||
|
||
#[test]
|
||
fn shppict_imports_one_image_and_skips_legacy_duplicate() -> Result<(), String> {
|
||
let pixels = image::RgbaImage::from_pixel(1, 1, image::Rgba([40, 80, 160, 255]));
|
||
let mut png = Vec::new();
|
||
pixels.write_to(&mut std::io::Cursor::new(&mut png), image::ImageFormat::Png).map_err(|e| e.to_string())?;
|
||
let hex: String = png.iter().map(|b| format!("{b:02x}")).collect();
|
||
let pict = format!("{{\\pict\\pngblip\\picwgoal1440\\pichgoal720 {hex}}}");
|
||
// Also use a supported PNG as the legacy duplicate: it must still be skipped.
|
||
for legacy in [format!("{{\\pict\\wmetafile8 {hex}}}"), pict.clone()] {
|
||
for picture in [format!("{{\\*\\shppict{pict}}}"), pict.clone()] {
|
||
let f = parse(format!("{{\\rtf1\\ansi Before{picture}{{\\nonshppict{legacy}}}After\\par}}").as_bytes())?;
|
||
let images: Vec<_> = f
|
||
.blocks
|
||
.iter()
|
||
.filter_map(|b| match b {
|
||
FBlock::Para(p) => Some(&p.inlines),
|
||
FBlock::Table(_) => None,
|
||
})
|
||
.flatten()
|
||
.filter_map(|i| match i {
|
||
Inline::Image(img) => Some(img),
|
||
_ => None,
|
||
})
|
||
.collect();
|
||
assert_eq!(images.len(), 1, "{picture}");
|
||
let img = images.first().ok_or("missing picture")?;
|
||
assert_eq!(img.data.as_ref(), &png);
|
||
assert_eq!(img.ext, "png");
|
||
assert_eq!((img.w, img.h), (72.0, 36.0));
|
||
assert_eq!(texts(&f), vec!["Normal:BeforeAfter"]);
|
||
}
|
||
}
|
||
// Recognizing shppict must not resurrect pictures inside skipped destinations.
|
||
for dest in ["nonshppict", "unknown"] {
|
||
let f = parse(format!("{{\\rtf1{{\\*\\{dest}{{\\*\\shppict{pict}}}}}Before\\par}}").as_bytes())?;
|
||
assert_eq!(texts(&f), vec!["Normal:Before"]);
|
||
assert!(f.blocks.iter().all(|b| match b {
|
||
FBlock::Para(p) => p.inlines.iter().all(|i| !matches!(i, Inline::Image(_))),
|
||
FBlock::Table(_) => false,
|
||
}));
|
||
}
|
||
Ok(())
|
||
}
|
||
|
||
/// #4259: `\wmetafile8` and `\emfblip` pictures are kept as WMF/EMF media (they were dropped),
|
||
/// sized from `\picwgoal`/`\pichgoal` and `\picscalex`/`\picscaley`, else from `\picw`/`\pich`
|
||
/// (0.01 mm for a metafile). As Word writes a floating picture: the `\shprslt` copy is read.
|
||
#[test]
|
||
fn metafile_pictures_are_imported() -> Result<(), String> {
|
||
// A plain WMF: the 9-word header (type 1, version 0x300, 12 words, max record 3), then EOF.
|
||
let wmf: Vec<u8> = [1u16, 9, 0x300, 12, 0, 0, 3, 0, 0, 3, 0, 0].iter().flat_map(|w| w.to_le_bytes()).collect();
|
||
let hex: String = wmf.iter().map(|b| format!("{b:02x}")).collect();
|
||
let images = |rtf: String| -> Result<Vec<model::Img>, String> {
|
||
let f = parse(rtf.as_bytes())?;
|
||
Ok(f.blocks
|
||
.iter()
|
||
.filter_map(|b| match b {
|
||
FBlock::Para(p) => Some(p.inlines.clone()),
|
||
FBlock::Table(_) => None,
|
||
})
|
||
.flatten()
|
||
.filter_map(|i| match i {
|
||
Inline::Image(img) => Some(img),
|
||
_ => None,
|
||
})
|
||
.collect())
|
||
};
|
||
let pict = format!("{{\\pict\\wmetafile8\\picw38761\\pich12832\\picwgoal21972\\pichgoal7272\\picscalex45\\picscaley45 {hex}}}");
|
||
let shape = format!("{{\\shp{{\\*\\shpinst{{\\sp{{\\sn pib}}{{\\sv {pict}}}}}}}{{\\shprslt{pict}}}}}");
|
||
let got = images(format!("{{\\rtf1\\ansi Before{shape}After\\par}}"))?;
|
||
let [img] = got.as_slice() else { return Err(format!("{} pictures", got.len())) };
|
||
assert_eq!((img.ext.as_str(), img.data.as_ref()), ("wmf", &wmf));
|
||
// 21972 twips at 45% is 494.4 pt, shrunk to the 468 pt column like every imported picture.
|
||
assert_eq!(img.w, 468.0);
|
||
assert!((img.h - 7272.0 / 20.0 * 0.45 * 468.0 / 494.37).abs() < 0.1, "{}", img.h);
|
||
// No goal size: the natural size, 2540 × 1270 hundredths of a millimetre = 72 × 36 pt; under
|
||
// MM_TEXT (`\wmetafile1`) the same numbers are pixels.
|
||
let hex_of = |b: &[u8]| -> String { b.iter().map(|b| format!("{b:02x}")).collect() };
|
||
let mut emf = vec![0u8; 88];
|
||
emf[0] = 1;
|
||
emf[4] = 88;
|
||
emf[40..44].copy_from_slice(b" EMF");
|
||
let emf_hex = hex_of(&emf);
|
||
let got = images(format!("{{\\rtf1\\ansi{{\\pict\\emfblip\\picw2540\\pich1270 {emf_hex}}}\\par}}"))?;
|
||
let [img] = got.as_slice() else { return Err(format!("{} pictures", got.len())) };
|
||
assert_eq!((img.ext.as_str(), img.w, img.h, img.data.as_ref()), ("emf", 72.0, 36.0, &emf));
|
||
let got = images(format!("{{\\rtf1\\ansi{{\\pict\\wmetafile1\\picw96\\pich48 {hex}}}\\par}}"))?;
|
||
let [img] = got.as_slice() else { return Err(format!("{} pictures", got.len())) };
|
||
assert_eq!((img.ext.as_str(), img.w, img.h), ("wmf", 72.0, 36.0));
|
||
// `\picscalex`/`\picscaley` scale the natural size when there is no goal size.
|
||
let got = images(format!("{{\\rtf1\\ansi{{\\pict\\wmetafile8\\picw2540\\pich1270\\picscalex50\\picscaley200 {hex}}}\\par}}"))?;
|
||
let [img] = got.as_slice() else { return Err(format!("{} pictures", got.len())) };
|
||
assert_eq!((img.w, img.h), (36.0, 72.0));
|
||
// The other mapping modes use their own units: 0.1 mm, 0.01 in, 0.001 in and twips.
|
||
for (mode, w, h) in [(2, 254, 127), (4, 100, 50), (5, 1000, 500), (6, 1440, 720)] {
|
||
let got = images(format!("{{\\rtf1\\ansi{{\\pict\\wmetafile{mode}\\picw{w}\\pich{h} {hex}}}\\par}}"))?;
|
||
let [img] = got.as_slice() else { return Err(format!("{} pictures", got.len())) };
|
||
assert!((img.w - 72.0).abs() < 0.01 && (img.h - 36.0).abs() < 0.01, "mode {mode}: {} x {}", img.w, img.h);
|
||
}
|
||
// Hostile: a zero or negative natural size falls back to the default 72 pt square, and bytes
|
||
// that are no picture at all are dropped, whatever the control word says.
|
||
let got = images(format!("{{\\rtf1\\ansi{{\\pict\\wmetafile8\\picw0\\pich-5 {hex}}}\\par}}"))?;
|
||
let [img] = got.as_slice() else { return Err(format!("{} pictures", got.len())) };
|
||
assert_eq!((img.w, img.h), (72.0, 72.0));
|
||
for blip in ["wmetafile8", "emfblip"] {
|
||
assert!(images(format!("{{\\rtf1\\ansi{{\\pict\\{blip}\\picw2540\\pich1270 0123456789abcdef}}\\par}}"))?.is_empty(), "{blip}");
|
||
}
|
||
// Saved as RTF and read back: the same picture, the EMF byte for byte, the WMF without the
|
||
// placeable header RTF has no room for.
|
||
let placeable: Vec<u8> = [0xD7u8, 0xCD, 0xC6, 0x9A].into_iter().chain([0u8; 18]).chain(wmf.iter().copied()).collect();
|
||
for (blip, data, want, ext) in [("emfblip", &emf, &emf, "emf"), ("wmetafile8", &wmf, &wmf, "wmf"), ("wmetafile8", &placeable, &wmf, "wmf")] {
|
||
let rtf = format!("{{\\rtf1\\ansi{{\\pict\\{blip}\\picw2540\\pich1270 {}}}\\par}}", hex_of(data));
|
||
let got = images(export(&import(rtf.as_bytes())?))?;
|
||
let [img] = got.as_slice() else { return Err(format!("{blip}: {} pictures", got.len())) };
|
||
assert_eq!((img.ext.as_str(), img.data.as_ref(), img.w, img.h), (ext, want, 72.0, 36.0), "{blip}");
|
||
}
|
||
// An EMF stored under `.wmf` (a DOCX can) is written as the EMF its bytes are.
|
||
let mut doc = import(format!("{{\\rtf1\\ansi{{\\pict\\wmetafile8\\picw2540\\pich1270 {hex}}}\\par}}").as_bytes())?;
|
||
let (key, bytes) = doc.media.iter_mut().next().ok_or("no media")?;
|
||
assert!(key.ends_with(".wmf"), "{key}");
|
||
*bytes = std::sync::Arc::new(emf.clone());
|
||
let rtf = export(&doc);
|
||
assert!(rtf.contains("\\emfblip") && !rtf.contains("\\wmetafile"), "{rtf}");
|
||
let got = images(rtf)?;
|
||
let [img] = got.as_slice() else { return Err(format!("{} pictures", got.len())) };
|
||
assert_eq!((img.ext.as_str(), img.data.as_ref()), ("emf", &emf));
|
||
// A placeable header with no records after it has nothing to write: no empty picture group.
|
||
*doc.media.values_mut().next().ok_or("no media")? = std::sync::Arc::new(placeable[..22].to_vec());
|
||
assert!(!export(&doc).contains("\\pict"), "an empty metafile is not written");
|
||
Ok(())
|
||
}
|
||
|
||
#[test]
|
||
fn unicode_in_info_bookmarks_and_field_instructions() {
|
||
// #435: `\uN` (with its `\ucN` fallback skipped) decodes in every text destination.
|
||
let rtf = br#"{\rtf1\ansi\uc1{\info{\title Caf\u233?}{\author Caf\u233\'3f}}{\*\bkmkstart caf\u233?}{\*\bkmkend caf\u233?}{\field{\*\fldinst{HYPERLINK "https://example.test/caf\u233?"}}{\fldrslt{link}}}\par Caf\u233?\par}"#;
|
||
let f = parse(rtf).unwrap();
|
||
assert_eq!(f.meta.title, "Café");
|
||
assert_eq!(f.meta.author, "Café");
|
||
let inl: Vec<&Inline> = f.blocks.iter().flat_map(|b| if let FBlock::Para(p) = b { p.inlines.iter().collect() } else { vec![] }).collect();
|
||
assert!(inl.iter().any(|i| matches!(i, Inline::Anchor(n) if n == "café")), "{inl:?}");
|
||
assert!(inl.iter().any(|i| matches!(i, Inline::Text(t, f) if t == "link" && f.link.as_deref() == Some("https://example.test/café"))), "{inl:?}");
|
||
assert_eq!(texts(&f).last().map(String::as_str), Some("Normal:Café"));
|
||
}
|
||
|
||
/// A hand-written RTF with tracked changes: Alice inserts "added ", Bob deletes "gone", and
|
||
/// "both" is inserted by Alice then deleted by Bob.
|
||
pub(crate) const TRACKED: &[u8] = br"{\rtf1\ansi\deff0{\fonttbl{\f0 Arial;}}{\*\revtbl {Unknown;}{Alice;}{Bob;}}
|
||
\pard Keep {\revised\revauth1\revdttm-1480361314 added }text{\deleted\revauthdel2\revdttmdel-2016341499 gone} end{\revised\revauth1\revdttm-1480361314\deleted\revauthdel2\revdttmdel-2016341499 both}.\par}";
|
||
|
||
/// A revision's (author, date).
|
||
type Stamp = (String, String);
|
||
|
||
/// (text, insertion (author, date), deletion (author, date)) of each run of the first paragraph.
|
||
fn tracked_runs(doc: &Document) -> Vec<(String, Option<Stamp>, Option<Stamp>)> {
|
||
let rev = |i: Option<u32>, kind: craft_wordmodel::RevisionKind| {
|
||
i.map(|i| {
|
||
let r = doc.revisions.get(i as usize).unwrap();
|
||
assert_eq!(r.kind, kind);
|
||
(r.author.clone(), r.date.clone())
|
||
})
|
||
};
|
||
let Some(craft_wordmodel::Block::Para(p)) = doc.body.first().map(|b| &**b) else { panic!("no paragraph") };
|
||
p.run_ranges()
|
||
.map(|(r, cp)| {
|
||
(p.text[r].to_string(), rev(cp.ins, craft_wordmodel::RevisionKind::Insert), rev(cp.del, craft_wordmodel::RevisionKind::Delete))
|
||
})
|
||
.collect()
|
||
}
|
||
|
||
fn expected_tracked() -> Vec<(String, Option<(String, String)>, Option<(String, String)>)> {
|
||
let a = Some(("Alice".to_string(), "2024-03-15T10:30:00Z".to_string()));
|
||
let b = Some(("Bob".to_string(), "2025-01-02T08:05:00Z".to_string()));
|
||
vec![
|
||
("Keep ".into(), None, None),
|
||
("added ".into(), a.clone(), None),
|
||
("text".into(), None, None),
|
||
(" gone".into(), None, b.clone()),
|
||
(" end".into(), None, None),
|
||
(" both".into(), a, b),
|
||
(".".into(), None, None),
|
||
]
|
||
}
|
||
|
||
#[test]
|
||
fn tracked_changes_import_with_authors_and_dates() {
|
||
let doc = import(TRACKED).unwrap();
|
||
assert_eq!(tracked_runs(&doc), expected_tracked());
|
||
// One revision per (kind, author, date).
|
||
assert_eq!(doc.revisions.len(), 2);
|
||
}
|
||
|
||
#[test]
|
||
fn tracked_changes_survive_export_and_import() {
|
||
let doc = import(TRACKED).unwrap();
|
||
let rtf = export(&doc);
|
||
assert!(rtf.contains("{\\*\\revtbl {Unknown;}{Alice;}{Bob;}}"), "{rtf}");
|
||
assert!(rtf.contains("\\deleted\\revauthdel2\\revdttmdel-2016341499"), "{rtf}");
|
||
let back = import(rtf.as_bytes()).unwrap();
|
||
assert_eq!(tracked_runs(&back), expected_tracked());
|
||
}
|
||
|
||
#[test]
|
||
fn plain_text_export_still_drops_deleted_text() {
|
||
// Only RTF keeps the deletions: the shared flow for other formats leaves them out.
|
||
let doc = import(TRACKED).unwrap();
|
||
let flow = model::from_doc(&doc);
|
||
assert_eq!(texts(&flow), vec!["Normal:Keep added text end."]);
|
||
}
|
||
|
||
#[test]
|
||
fn dttm_converts_to_and_from_iso() {
|
||
assert_eq!(dttm_to_iso(-1480361314i32 as u32).as_deref(), Some("2024-03-15T10:30:00Z"));
|
||
assert_eq!(dttm_to_iso(-2016341499i32 as u32).as_deref(), Some("2025-01-02T08:05:00Z"));
|
||
assert_eq!(iso_to_dttm("2024-03-15T10:30:00Z"), Some(-1480361314i32 as u32));
|
||
assert_eq!(iso_to_dttm("2025-01-02T08:05:59.123+02:00"), Some(-2016341499i32 as u32));
|
||
assert_eq!(iso_to_dttm("2000-02-29T00:00:00Z").and_then(dttm_to_iso).as_deref(), Some("2000-02-29T00:00:00Z"));
|
||
// Zero, out-of-range fields and impossible days are no date.
|
||
assert_eq!(dttm_to_iso(0), None);
|
||
assert_eq!(dttm_to_iso(13 << 16 | 1 << 11 | 100 << 20), None);
|
||
assert_eq!(dttm_to_iso(2 << 16 | 30 << 11 | 101 << 20), None);
|
||
assert_eq!(dttm_to_iso(1 << 16 | 1 << 11 | 100 << 20 | 60), None);
|
||
assert_eq!(iso_to_dttm(""), None);
|
||
assert_eq!(iso_to_dttm("1899-12-31T00:00:00Z"), None);
|
||
assert_eq!(iso_to_dttm("2024-02-30T00:00:00Z"), None);
|
||
assert_eq!(iso_to_dttm("2024-1é-01T00:00:00Z"), None);
|
||
assert_eq!(iso_to_dttm("2024-03-15"), iso_to_dttm("2024-03-15T00:00:00Z"));
|
||
}
|
||
|
||
#[test]
|
||
fn hostile_revision_tables_and_indices() {
|
||
// An author index past the table, a negative one, and a malformed table.
|
||
let rtf = br"{\rtf1{\*\revtbl {Unknown;}{Ann}{;}{\b;}}{\revised\revauth99999999 a}{\deleted\revauthdel-5\revdttmdel4294967295 b}{\revised\revauth1 c}\par}";
|
||
let doc = import(rtf).unwrap();
|
||
let runs = tracked_runs(&doc);
|
||
assert_eq!(runs.first().and_then(|r| r.1.clone()), Some(("Unknown".to_string(), String::new())));
|
||
assert_eq!(runs.get(1).and_then(|r| r.2.clone()), Some(("Unknown".to_string(), String::new())));
|
||
assert_eq!(runs.get(2).and_then(|r| r.1.clone()), Some(("Ann".to_string(), String::new())));
|
||
// A huge table is capped, and indices past the cap are "Unknown".
|
||
let mut rtf = String::from("{\\rtf1{\\*\\revtbl ");
|
||
for i in 0..20_000 {
|
||
rtf.push_str(&format!("{{A{i};}}"));
|
||
}
|
||
rtf.push_str("}{\\revised\\revauth15000 x}{\\revised\\revauth9999 y}\\par}");
|
||
let doc = import(rtf.as_bytes()).unwrap();
|
||
let runs = tracked_runs(&doc);
|
||
assert_eq!(runs.first().and_then(|r| r.1.clone()).map(|a| a.0).as_deref(), Some("Unknown"));
|
||
assert_eq!(runs.get(1).and_then(|r| r.1.clone()).map(|a| a.0).as_deref(), Some("A9999"));
|
||
// Unterminated table and revision words outside the body never panic.
|
||
for bad in [&br"{\rtf1{\*\revtbl {Al"[..], br"{\rtf1{\*\revtbl \revised\deleted{x}}y\par}", br"{\rtf1\revauth\revdttm\deleted"] {
|
||
let _ = import(bad);
|
||
}
|
||
}
|
||
|
||
#[test]
|
||
fn unicode_author_names_survive_import_and_round_trip() {
|
||
// `\uN` in the author table: "José" and "Дмитрий" (not "Jos?" / "???????").
|
||
let rtf = br"{\rtf1\uc1{\*\revtbl {Unknown;}{Jos\u233?;}{\u1044?\u1084?\u1080?\u1090?\u1088?\u1080?\u1081?;}}{\revised\revauth1 a}{\deleted\revauthdel2 b}\par}";
|
||
let authors = |d: &Document| tracked_runs(d).into_iter().map(|r| (r.1.map(|a| a.0), r.2.map(|a| a.0))).collect::<Vec<_>>();
|
||
let want = vec![(Some("José".to_string()), None), (None, Some("Дмитрий".to_string()))];
|
||
let doc = import(rtf).unwrap();
|
||
assert_eq!(authors(&doc), want);
|
||
let back = import(export(&doc).as_bytes()).unwrap();
|
||
assert_eq!(authors(&back), want);
|
||
}
|
||
|
||
#[test]
|
||
fn pard_keeps_revision_marks() {
|
||
// Revision marks are character formatting: `\pard` resets paragraph formatting only.
|
||
let rtf = br"{\rtf1{\*\revtbl {Unknown;}{Bob;}}\deleted\revauthdel1 gone\par\pard more\par}";
|
||
let doc = import(rtf).unwrap();
|
||
let Some(craft_wordmodel::Block::Para(p)) = doc.body.get(1).map(|b| &**b) else { panic!("no second paragraph") };
|
||
assert_eq!(p.text, "more");
|
||
let del = p.props_of_char(0).del.and_then(|i| doc.revisions.get(i as usize));
|
||
assert_eq!(del.map(|r| (r.kind, r.author.as_str())), Some((craft_wordmodel::RevisionKind::Delete, "Bob")));
|
||
}
|
||
|
||
/// The body's paragraphs.
|
||
fn doc_paras(doc: &Document) -> Vec<&craft_wordmodel::Paragraph> {
|
||
doc.body
|
||
.iter()
|
||
.filter_map(|b| match &**b {
|
||
craft_wordmodel::Block::Para(p) => Some(p),
|
||
_ => None,
|
||
})
|
||
.collect()
|
||
}
|
||
|
||
/// A document of `paras` with `revisions`.
|
||
fn doc_with(paras: Vec<craft_wordmodel::Paragraph>, revisions: Vec<craft_wordmodel::Revision>) -> Document {
|
||
let mut doc = Document::new();
|
||
doc.body = paras.into_iter().map(craft_wordmodel::para_block).collect();
|
||
doc.revisions = revisions;
|
||
doc
|
||
}
|
||
|
||
/// Kind, author and date of revision `i`.
|
||
fn rev_of(doc: &Document, i: Option<u32>) -> Option<(craft_wordmodel::RevisionKind, String, String)> {
|
||
i.and_then(|i| doc.revisions.get(i as usize)).map(|r| (r.kind, r.author.clone(), r.date.clone()))
|
||
}
|
||
|
||
#[test]
|
||
fn character_formatting_change_round_trips() {
|
||
use craft_wordmodel::props::{CharProps, PropChange};
|
||
use craft_wordmodel::{Revision, RevisionKind};
|
||
// "Bold" was made bold by Alice; before, it was plain italic.
|
||
let old = CharProps { italic: Some(true), ..Default::default() };
|
||
let cp = CharProps { bold: Some(true), fmt_change: PropChange::boxed(0, old), ..Default::default() };
|
||
let doc =
|
||
doc_with(vec![craft_wordmodel::Paragraph::with_text("Bold", cp)], vec![Revision::new(RevisionKind::Format, "Alice", "2024-03-15T10:30:00Z")]);
|
||
let rtf = export(&doc);
|
||
assert!(rtf.contains("\\crauth1\\crdate-1480361314"), "{rtf}");
|
||
assert!(rtf.contains("{\\*\\oldcprops"), "{rtf}");
|
||
let back = import(rtf.as_bytes()).unwrap();
|
||
let paras = doc_paras(&back);
|
||
let p = paras.first().unwrap();
|
||
assert_eq!(p.text, "Bold");
|
||
let cp = p.props_of_char(0);
|
||
assert_eq!(cp.bold, Some(true));
|
||
let change = cp.fmt_change.as_deref().expect("formatting change read back");
|
||
assert_eq!(rev_of(&back, Some(change.rev)), Some((RevisionKind::Format, "Alice".into(), "2024-03-15T10:30:00Z".into())));
|
||
assert_eq!(change.old.italic, Some(true));
|
||
assert_ne!(change.old.bold, Some(true));
|
||
}
|
||
|
||
#[test]
|
||
fn paragraph_formatting_change_round_trips() {
|
||
use craft_wordmodel::props::{CharProps, ParaProps, PropChange};
|
||
use craft_wordmodel::{Align, Revision, RevisionKind};
|
||
// Bob centred a right-aligned paragraph.
|
||
let mut p = craft_wordmodel::Paragraph::with_text("Centred", CharProps::default());
|
||
p.props.align = Some(Align::Center);
|
||
p.props.fmt_change = PropChange::boxed(0, ParaProps { align: Some(Align::Right), ..Default::default() });
|
||
let doc = doc_with(vec![p], vec![Revision::new(RevisionKind::Format, "Bob", "2025-01-02T08:05:00Z")]);
|
||
let rtf = export(&doc);
|
||
assert!(rtf.contains("{\\*\\oldpprops"), "{rtf}");
|
||
assert!(rtf.contains("\\prauth1\\prdate-2016341499"), "{rtf}");
|
||
let back = import(rtf.as_bytes()).unwrap();
|
||
let paras = doc_paras(&back);
|
||
let p = paras.first().unwrap();
|
||
assert_eq!(p.text, "Centred");
|
||
assert_eq!(p.props.align, Some(Align::Center));
|
||
let change = p.props.fmt_change.as_deref().expect("paragraph formatting change read back");
|
||
assert_eq!(rev_of(&back, Some(change.rev)), Some((RevisionKind::Format, "Bob".into(), "2025-01-02T08:05:00Z".into())));
|
||
assert_eq!(change.old.align, Some(Align::Right));
|
||
}
|
||
|
||
#[test]
|
||
fn paragraph_mark_revisions_round_trip() {
|
||
use craft_wordmodel::props::CharProps;
|
||
use craft_wordmodel::{Revision, RevisionKind};
|
||
// Bob deleted the mark after "One" (joining it to "Two"); Alice inserted the one after "Two".
|
||
let mut one = craft_wordmodel::Paragraph::with_text("One", CharProps::default());
|
||
one.mark.del = Some(0);
|
||
let mut two = craft_wordmodel::Paragraph::with_text("Two", CharProps::default());
|
||
two.mark.ins = Some(1);
|
||
let three = craft_wordmodel::Paragraph::with_text("Three", CharProps::default());
|
||
let doc = doc_with(
|
||
vec![one, two, three],
|
||
vec![
|
||
Revision::new(RevisionKind::Delete, "Bob", "2025-01-02T08:05:00Z"),
|
||
Revision::new(RevisionKind::Insert, "Alice", "2024-03-15T10:30:00Z"),
|
||
],
|
||
);
|
||
let rtf = export(&doc);
|
||
let back = import(rtf.as_bytes()).unwrap();
|
||
let paras = doc_paras(&back);
|
||
let texts: Vec<&str> = paras.iter().map(|p| p.text.as_str()).collect();
|
||
assert_eq!(texts.get(..3), Some(&["One", "Two", "Three"][..]), "{rtf}");
|
||
let marks: Vec<_> = paras.iter().take(3).map(|p| (rev_of(&back, p.mark.ins), rev_of(&back, p.mark.del))).collect();
|
||
assert_eq!(
|
||
marks,
|
||
vec![
|
||
(None, Some((RevisionKind::Delete, "Bob".into(), "2025-01-02T08:05:00Z".into()))),
|
||
(Some((RevisionKind::Insert, "Alice".into(), "2024-03-15T10:30:00Z".into())), None),
|
||
(None, None),
|
||
],
|
||
"{rtf}"
|
||
);
|
||
}
|
||
|
||
#[test]
|
||
fn formatting_revisions_read_as_the_spec_writes_them() {
|
||
use craft_wordmodel::{Align, RevisionKind};
|
||
// RTF 1.9.1 "Revision Marks": `\crauth`/`\crdate` with `{\*\oldcprops …}` on a run,
|
||
// `\prauth`/`\prdate` with `{\*\oldpprops …}` on a paragraph, and a `\par` inside a
|
||
// `\revised` group.
|
||
let rtf = br"{\rtf1\ansi{\fonttbl{\f0 Arial;}}{\*\revtbl {Unknown;}{Alice;}{Bob;}}
|
||
\pard\qc\prauth2\prdate-2016341499{\*\oldpprops \pard\ql}Plain {\b\crauth1\crdate-1480361314{\*\oldcprops \plain\i}bold}\par
|
||
\pard Next{\revised\revauth1\revdttm-1480361314 \par}After\par}";
|
||
let doc = import(rtf).unwrap();
|
||
let paras = doc_paras(&doc);
|
||
let texts: Vec<&str> = paras.iter().map(|p| p.text.as_str()).collect();
|
||
assert_eq!(texts.get(..3), Some(&["Plain bold", "Next", "After"][..]));
|
||
let p = paras.first().unwrap();
|
||
assert_eq!(p.props.align, Some(Align::Center));
|
||
let pc = p.props.fmt_change.as_deref().expect("paragraph change");
|
||
assert_eq!(rev_of(&doc, Some(pc.rev)), Some((RevisionKind::Format, "Bob".into(), "2025-01-02T08:05:00Z".into())));
|
||
assert_eq!(pc.old.align, Some(Align::Left));
|
||
assert!(p.props_of_char(0).fmt_change.is_none());
|
||
let cp = p.props_of_char(6);
|
||
assert_eq!(cp.bold, Some(true));
|
||
let cc = cp.fmt_change.as_deref().expect("character change");
|
||
assert_eq!(rev_of(&doc, Some(cc.rev)), Some((RevisionKind::Format, "Alice".into(), "2024-03-15T10:30:00Z".into())));
|
||
assert_eq!(cc.old.italic, Some(true));
|
||
assert_ne!(cc.old.bold, Some(true));
|
||
let next = paras.get(1).unwrap();
|
||
assert_eq!(rev_of(&doc, next.mark.ins), Some((RevisionKind::Insert, "Alice".into(), "2024-03-15T10:30:00Z".into())));
|
||
assert!(paras.first().unwrap().mark.ins.is_none());
|
||
assert!(paras.get(2).unwrap().mark.ins.is_none());
|
||
}
|
||
|
||
#[test]
|
||
fn list_item_with_a_formatting_change_keeps_its_list_on_reject() {
|
||
use model::{FmtChange, ListInfo};
|
||
// A numbered item Alice centred (it was left-aligned), and a second plain item.
|
||
let li = Some(ListInfo { ordered: true, level: 0, start: 1, ..Default::default() });
|
||
let mark = RevMark { author: "Alice".into(), date: "2024-03-15T10:30:00Z".into(), id: None };
|
||
let old = Para { list: li, align: Some(Align::Left), ..Default::default() };
|
||
let mut item = Para { list: li, align: Some(Align::Center), fmt_change: Some(Box::new(FmtChange { mark, old })), ..Default::default() };
|
||
item.push_text("Item", &Fmt::default());
|
||
let mut next = Para { list: li, ..Default::default() };
|
||
next.push_text("Next", &Fmt::default());
|
||
let flow = Flow { blocks: vec![FBlock::Para(item), FBlock::Para(next)], meta: Meta::default() };
|
||
let check = |doc: &Document, what: &str| {
|
||
let paras = doc_paras(doc);
|
||
let p = paras.first().unwrap();
|
||
assert!(p.props.numbering.is_some(), "{what}: the item is in a list");
|
||
let change = p.props.fmt_change.as_deref().unwrap_or_else(|| panic!("{what}: change kept"));
|
||
assert_eq!(change.old.align, Some(Align::Left), "{what}");
|
||
// Reject puts the old properties back: the item stays in its list, with its style.
|
||
let rejected = change.old.formatting();
|
||
assert_eq!(rejected.numbering, p.props.numbering, "{what}");
|
||
assert_eq!(rejected.style, p.props.style, "{what}");
|
||
};
|
||
let doc = model::to_doc(&flow);
|
||
check(&doc, "flow to document");
|
||
let rtf = export(&doc);
|
||
check(&import(rtf.as_bytes()).unwrap(), &format!("RTF round trip\n{rtf}"));
|
||
}
|
||
|
||
#[test]
|
||
fn paragraph_that_was_a_list_item_gets_its_old_list_back_on_reject() {
|
||
use model::{FmtChange, ListInfo};
|
||
// Alice took the paragraph out of a numbered list: it is no list item now, but was one.
|
||
let mark = RevMark { author: "Alice".into(), date: "2024-03-15T10:30:00Z".into(), id: None };
|
||
let old = Para { list: Some(ListInfo { ordered: true, level: 0, start: 1, ..Default::default() }), ..Default::default() };
|
||
let mut item = Para { fmt_change: Some(Box::new(FmtChange { mark, old })), ..Default::default() };
|
||
item.push_text("Item", &Fmt::default());
|
||
let flow = Flow { blocks: vec![FBlock::Para(item)], meta: Meta::default() };
|
||
let doc = model::to_doc(&flow);
|
||
let paras = doc_paras(&doc);
|
||
let p = paras.first().unwrap();
|
||
assert!(p.props.numbering.is_none());
|
||
let change = p.props.fmt_change.as_deref().expect("change kept");
|
||
assert!(change.old.formatting().numbering.is_some(), "Reject restores the list");
|
||
}
|
||
|
||
#[test]
|
||
fn numbered_item_made_bulleted_gets_its_numbered_list_back_on_reject() {
|
||
use model::{FmtChange, ListInfo};
|
||
let mark = RevMark { author: "Alice".into(), date: "2024-03-15T10:30:00Z".into(), id: None };
|
||
let old = Para { list: Some(ListInfo { ordered: true, level: 0, start: 1, ..Default::default() }), ..Default::default() };
|
||
let mut item = Para {
|
||
list: Some(ListInfo { ordered: false, level: 0, start: 1, ..Default::default() }),
|
||
fmt_change: Some(Box::new(FmtChange { mark, old })),
|
||
..Default::default()
|
||
};
|
||
item.push_text("Item", &Fmt::default());
|
||
let flow = Flow { blocks: vec![FBlock::Para(item)], meta: Meta::default() };
|
||
let doc = model::to_doc(&flow);
|
||
let paras = doc_paras(&doc);
|
||
let p = paras.first().unwrap();
|
||
let change = p.props.fmt_change.as_deref().expect("change kept");
|
||
let (now, was) = (p.props.numbering.unwrap(), change.old.formatting().numbering.unwrap());
|
||
assert_ne!(now.num, was.num, "the old list is a list of its own");
|
||
}
|
||
|
||
#[test]
|
||
fn heading_formatting_change_has_no_style_formatting_in_the_old_formatting() {
|
||
// A heading's runs repeat its style's bold and size; the old formatting does too.
|
||
let rtf = br"{\rtf1{\stylesheet{\s0 Normal;}{\s1 heading 1;}}{\*\revtbl {Unknown;}{Alice;}}
|
||
\pard\s1\b\fs32 {\i\crauth1{\*\oldcprops \plain\b\fs32}Head}\par}";
|
||
let doc = import(rtf).unwrap();
|
||
let paras = doc_paras(&doc);
|
||
let cp = paras.first().unwrap().props_of_char(0);
|
||
let old = cp.fmt_change.as_deref().map(|c| c.old.clone()).expect("change kept");
|
||
assert_eq!((old.bold, old.size), (cp.bold, cp.size));
|
||
}
|
||
|
||
#[test]
|
||
fn word_oldcprops_hold_only_the_changed_properties() {
|
||
use craft_wordmodel::RevisionKind;
|
||
// Word for Mac 16.113 saving tracked formatting changes as RTF: "beta" made bold,
|
||
// "gamma" (bold) grown from 12 to 16 pt. `\oldcprops` holds only what changed, with
|
||
// `\crauth`/`\crdate` inside the group; the rest of the old formatting is the run's.
|
||
let rtf = br"{\rtf1\ansi\deff0{\fonttbl{\f0 Times New Roman;}{\f1 Arial;}}{\*\revtbl {Unknown;}{Alice;}}
|
||
\pard\plain \f1\fs24 alpha {\f1\fs24\b\insrsid12079894\charrsid12079894 {\*\oldcprops \b0\crauth1\crdate-2014675576\insrsid12079894\charrsid12079894}beta} {\b\f1\fs32 {\*\oldcprops \fs24\crauth1\crdate-2014675576\insrsid12079894\charrsid12079894}gamma}\par}";
|
||
let doc = import(rtf).unwrap();
|
||
let paras = doc_paras(&doc);
|
||
let p = paras.first().unwrap();
|
||
assert_eq!(p.text, "alpha beta gamma");
|
||
let word_rev = Some((RevisionKind::Format, "Alice".to_string(), "2026-10-15T22:08:00Z".to_string()));
|
||
// beta: Reject takes the bold off; font and size stay.
|
||
let beta = p.props_of_char(6);
|
||
assert_eq!(beta.bold, Some(true));
|
||
let old = beta.fmt_change.as_deref().expect("beta's change");
|
||
assert_eq!(rev_of(&doc, Some(old.rev)), word_rev);
|
||
assert_ne!(old.old.bold, Some(true), "{:?}", old.old);
|
||
assert_eq!((old.old.font.as_deref(), old.old.size), (beta.font.as_deref(), Some(12.0)), "{:?}", old.old);
|
||
// gamma: Reject puts 12 pt back; bold and font stay.
|
||
let gamma = p.props_of_char(11);
|
||
assert_eq!(gamma.size, Some(16.0));
|
||
let old = gamma.fmt_change.as_deref().expect("gamma's change");
|
||
assert_eq!(rev_of(&doc, Some(old.rev)), word_rev);
|
||
assert_eq!((old.old.bold, old.old.font.as_deref(), old.old.size), (Some(true), gamma.font.as_deref(), Some(12.0)), "{:?}", old.old);
|
||
assert_eq!(gamma.font.as_deref(), Some("Arial"));
|
||
}
|
||
|
||
#[test]
|
||
fn word_oldpprops_hold_the_full_old_paragraph_formatting() {
|
||
use craft_wordmodel::{Align, RevisionKind};
|
||
// Word for Mac 16.113: a paragraph centred with tracking on. `\oldpprops` is the full
|
||
// old set from `\pard`; `\prauth`/`\prdate` are on the paragraph.
|
||
let rtf = br"{\rtf1\ansi\deff0{\fonttbl{\f0 Times New Roman;}}{\*\revtbl {Unknown;}{Alice;}}
|
||
\pard\plain \ltrpar\qc \li0\ri0\sa160\sl278\slmult1\widctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0\prauth1\prdate-2014675576{\*\oldpprops \pard \ltrpar\ql \li0\ri0\sa160\sl278\slmult1\widctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0\pararsid12079894}\fs24 Centred\par}";
|
||
let doc = import(rtf).unwrap();
|
||
let paras = doc_paras(&doc);
|
||
let p = paras.first().unwrap();
|
||
assert_eq!(p.text, "Centred");
|
||
assert_eq!(p.props.align, Some(Align::Center));
|
||
let c = p.props.fmt_change.as_deref().expect("paragraph change");
|
||
assert_eq!(rev_of(&doc, Some(c.rev)), Some((RevisionKind::Format, "Alice".into(), "2026-10-15T22:08:00Z".into())));
|
||
assert_eq!(c.old.align, Some(Align::Left));
|
||
assert_eq!(c.old.space_after, p.props.space_after);
|
||
assert_eq!(c.old.line_spacing, p.props.line_spacing);
|
||
assert!(p.props.space_after.is_some());
|
||
}
|
||
|
||
#[test]
|
||
fn oldcprops_are_written_as_the_differences() {
|
||
use craft_wordmodel::props::{CharProps, PropChange};
|
||
use craft_wordmodel::{Revision, RevisionKind};
|
||
let rev = || vec![Revision::new(RevisionKind::Format, "Alice", "2024-03-15T10:30:00Z")];
|
||
// Made bold: the group holds `\b0` and who changed it, nothing else.
|
||
let old = CharProps { font: Some("Arial".into()), size: Some(14.0), ..Default::default() };
|
||
let cp =
|
||
CharProps { bold: Some(true), font: Some("Arial".into()), size: Some(14.0), fmt_change: PropChange::boxed(0, old), ..Default::default() };
|
||
let rtf = export(&doc_with(vec![craft_wordmodel::Paragraph::with_text("Bold", cp)], rev()));
|
||
assert!(rtf.contains("{\\*\\oldcprops \\b0\\crauth1\\crdate-1480361314}"), "{rtf}");
|
||
// Several changes, on and off, round-trip.
|
||
let red = craft_wordmodel::props::TextColor::Rgb(craft_wordmodel::props::Rgb(255, 0, 0));
|
||
let old = CharProps {
|
||
italic: Some(true),
|
||
underline: Some(craft_wordmodel::props::Underline::Single),
|
||
size: Some(14.0),
|
||
color: Some(red),
|
||
font: Some("Arial".into()),
|
||
..Default::default()
|
||
};
|
||
let cp = CharProps {
|
||
bold: Some(true),
|
||
size: Some(16.0),
|
||
font: Some("Courier New".into()),
|
||
fmt_change: PropChange::boxed(0, old.clone()),
|
||
..Default::default()
|
||
};
|
||
let rtf = export(&doc_with(vec![craft_wordmodel::Paragraph::with_text("Mixed", cp)], rev()));
|
||
let back = import(rtf.as_bytes()).unwrap();
|
||
let paras = doc_paras(&back);
|
||
let cp = paras.first().unwrap().props_of_char(0);
|
||
assert_eq!((cp.bold, cp.size), (Some(true), Some(16.0)), "{rtf}");
|
||
let got = &cp.fmt_change.as_deref().expect("change read back").old;
|
||
assert_ne!(got.bold, Some(true), "{rtf}");
|
||
assert_eq!(
|
||
(got.italic, got.underline, got.size, got.color, got.font.as_deref()),
|
||
(old.italic, old.underline, old.size, old.color, old.font.as_deref()),
|
||
"{rtf}"
|
||
);
|
||
}
|
||
|
||
#[test]
|
||
fn old_formatting_in_styled_paragraphs_is_the_styles() {
|
||
use model::{FmtChange, Kind};
|
||
let mark = || RevMark { author: "Alice".into(), date: "2024-03-15T10:30:00Z".into(), id: None };
|
||
// A heading 2 run grown to 20 pt and a code run grown to 14 pt, both from no direct
|
||
// formatting: Reject gives back the style's size, not RTF's default 12 pt. (A code run's
|
||
// own font doesn't reach the flow from a document, so the code font is checked below.)
|
||
let mut head = Para { kind: Kind::Heading(2), ..Default::default() };
|
||
head.push_text(
|
||
"Head",
|
||
&Fmt { size: Some(20.0), fmt_change: Some(Box::new(FmtChange { mark: mark(), old: Fmt::default() })), ..Default::default() },
|
||
);
|
||
let mut code = Para { kind: Kind::Code, ..Default::default() };
|
||
code.push_text(
|
||
"let x",
|
||
&Fmt { size: Some(14.0), fmt_change: Some(Box::new(FmtChange { mark: mark(), old: Fmt::default() })), ..Default::default() },
|
||
);
|
||
let flow = Flow { blocks: vec![FBlock::Para(head), FBlock::Para(code)], meta: Meta::default() };
|
||
let rtf = export(&model::to_doc(&flow));
|
||
let back = parse(rtf.as_bytes()).unwrap();
|
||
let runs: Vec<&Fmt> = back
|
||
.blocks
|
||
.iter()
|
||
.filter_map(|b| match b {
|
||
FBlock::Para(p) => p.inlines.first().and_then(|i| if let Inline::Text(_, f) = i { Some(f) } else { None }),
|
||
_ => None,
|
||
})
|
||
.collect();
|
||
let head = runs.first().expect("heading run");
|
||
// The import takes a heading's direct size as the style's (as on main); the old size
|
||
// written is the style's, so Reject gives it whichever way the import reads it.
|
||
assert!(rtf.contains("{\\*\\oldcprops \\fs32\\crauth1"), "{rtf}");
|
||
let old = &head.fmt_change.as_deref().expect("heading change").old;
|
||
assert!(old.size.is_none_or(|s| s == 16.0), "{old:?}\n{rtf}");
|
||
let code = runs.get(1).expect("code run");
|
||
assert!(rtf.contains("{\\*\\oldcprops \\fs20\\crauth1"), "{rtf}");
|
||
let old = &code.fmt_change.as_deref().expect("code change").old;
|
||
assert!(old.size.is_none_or(|s| s == 10.0), "{old:?}\n{rtf}");
|
||
assert!(old.font.is_none() || old.code, "{old:?}\n{rtf}");
|
||
}
|
||
|
||
#[test]
|
||
fn old_formatting_in_a_code_paragraph_has_the_code_font() {
|
||
use model::{FmtChange, Kind};
|
||
// A code run set in Arial from no direct font: the old font written is the code font.
|
||
let mark = RevMark { author: "Alice".into(), date: String::new(), id: None };
|
||
let mut p = Para { kind: Kind::Code, ..Default::default() };
|
||
p.push_text("x", &Fmt { font: Some("Arial".into()), fmt_change: Some(Box::new(FmtChange { mark, old: Fmt::default() })), ..Default::default() });
|
||
let mut w = Writer {
|
||
marks: Vec::new(),
|
||
t: Tables { fonts: vec!["Aptos".into()], colors: Vec::new() },
|
||
lists: Vec::new(),
|
||
authors: None,
|
||
base: Fmt::default(),
|
||
sections: Vec::new(),
|
||
sect_i: 0,
|
||
};
|
||
let mut out = String::new();
|
||
w.para(&p, false, None, "\\par", &mut out);
|
||
let mono = w.t.font(model::MONO_FONT);
|
||
assert!(out.contains(&format!("{{\\*\\oldcprops \\f{mono}\\crauth1}}")), "{out}");
|
||
}
|
||
|
||
/// The text runs of `rtf`'s paragraphs, in order, with their formatting.
|
||
fn text_runs(rtf: &[u8]) -> Vec<(String, Fmt)> {
|
||
let f = parse(rtf).unwrap();
|
||
f.blocks
|
||
.iter()
|
||
.filter_map(|b| if let FBlock::Para(p) = b { Some(p) } else { None })
|
||
.flat_map(|p| p.inlines.iter())
|
||
.filter_map(|i| if let Inline::Text(t, f) = i { Some((t.clone(), f.clone())) } else { None })
|
||
.collect()
|
||
}
|
||
|
||
/// The formatting of the run whose text is `t`.
|
||
fn run_fmt<'a>(runs: &'a [(String, Fmt)], t: &str) -> &'a Fmt {
|
||
runs.iter().find(|r| r.0 == t).map(|r| &r.1).unwrap_or_else(|| panic!("no run {t:?} in {runs:?}"))
|
||
}
|
||
|
||
#[test]
|
||
fn heading_and_code_runs_take_their_style_formatting_on_import() {
|
||
// A stylesheet entry without formatting words (WordCraft's older files, minimal
|
||
// stylesheets): WordCraft's own heading and code formatting stand in for the style's.
|
||
let rtf = br"{\rtf1\ansi{\fonttbl{\f0 Times New Roman;}{\f1 Arial;}{\f2 Courier New;}}{\stylesheet{\s0 Normal;}{\s2 heading 2;}{\s9 Code;}}
|
||
\pard\s2\b\fs32 Head\par
|
||
\pard\s2\b\fs26 Small\par
|
||
\pard\s9\f2\b\fs21 code\par
|
||
\pard\s9\f1\fs21 arial\par}";
|
||
let runs = text_runs(rtf);
|
||
let head = run_fmt(&runs, "Head");
|
||
assert_eq!((head.size, head.bold), (None, false));
|
||
// 13 pt isn't heading 2's 16 pt: a size set by hand.
|
||
let small = run_fmt(&runs, "Small");
|
||
assert_eq!((small.size, small.bold), (Some(13.0), false));
|
||
let code = run_fmt(&runs, "code");
|
||
assert_eq!((code.size, code.bold, code.font.as_deref(), code.code), (None, false, None, false));
|
||
// Arial isn't the code font: set by hand.
|
||
let arial = run_fmt(&runs, "arial");
|
||
assert_eq!((arial.size, arial.font.as_deref(), arial.code), (None, Some("Arial"), false));
|
||
}
|
||
|
||
#[test]
|
||
fn heading_runs_are_compared_with_the_files_heading_style() {
|
||
// Word writes resolved values: a 13 pt bold Heading 2 has `\b\fs26` in its stylesheet
|
||
// entry and on its runs. Only what differs from the entry is direct formatting.
|
||
let rtf = br"{\rtf1\ansi{\fonttbl{\f0 Times New Roman;}}{\stylesheet{\ql \f0\fs24 Normal;}
|
||
{\s2\ql \keepn\outlinelevel1 \b\fs26 \sbasedon0 \snext0 heading 2;}{\s3\ql \keepn\b0\fs24 \sbasedon0 \snext0 heading 3;}}
|
||
\pard\s2\b\fs26 Head\par
|
||
\pard\s2\b\fs40 Big\par
|
||
\pard\s3\b\fs24 Bolded\par}";
|
||
let runs = text_runs(rtf);
|
||
let head = run_fmt(&runs, "Head");
|
||
assert_eq!((head.size, head.bold), (None, false));
|
||
let big = run_fmt(&runs, "Big");
|
||
assert_eq!((big.size, big.bold), (Some(20.0), false));
|
||
// Heading 3 isn't bold in this file: the bold is the user's.
|
||
let bolded = run_fmt(&runs, "Bolded");
|
||
assert_eq!((bolded.size, bolded.bold), (None, true));
|
||
}
|
||
|
||
#[test]
|
||
fn style_formatting_is_resolved_through_sbasedon() {
|
||
// LibreOffice's stylesheet: Heading 2 is based on Heading, which sets the boldness and
|
||
// font; Code is based on a style that sets the font.
|
||
let rtf = br"{\rtf1\ansi{\fonttbl{\f0 Liberation Serif;}{\f1 Liberation Sans;}{\f2 Iosevka;}{\f3 Courier New;}}
|
||
{\stylesheet{\s0\f0\fs24 Normal;}{\s1\sbasedon0\snext2\f1\b0\fs28 Heading;}{\s2\sbasedon1\snext3\ql\fs32 Heading 2;}
|
||
{\s10\sbasedon0\f2\fs20 Source Base;}{\s9\sbasedon10 Code;}}
|
||
\pard\s2\f1\fs32 Plain\par
|
||
\pard\s2\f1\b\fs32 Bolded\par
|
||
\pard\s9\f2\fs20 base\par
|
||
\pard\s9\f3\fs20 courier\par}";
|
||
let runs = text_runs(rtf);
|
||
let plain = run_fmt(&runs, "Plain");
|
||
assert_eq!((plain.size, plain.bold), (None, false));
|
||
// Heading isn't bold, so neither is Heading 2.
|
||
let bolded = run_fmt(&runs, "Bolded");
|
||
assert_eq!((bolded.size, bolded.bold), (None, true));
|
||
let base = run_fmt(&runs, "base");
|
||
assert_eq!((base.size, base.font.as_deref(), base.code), (None, None, false));
|
||
// Code's font is Iosevka here, so Courier New was set by hand.
|
||
let courier = run_fmt(&runs, "courier");
|
||
assert_eq!((courier.size, courier.font.as_deref(), courier.code), (None, None, true));
|
||
}
|
||
|
||
#[test]
|
||
fn code_runs_in_the_files_code_font_have_no_direct_font() {
|
||
let rtf = br"{\rtf1\ansi{\fonttbl{\f0 Times New Roman;}{\f1 Courier New;}{\f2 Arial;}}{\stylesheet{\s0 Normal;}{\s9\sbasedon0\sa0\f1\fs21 Code;}}
|
||
\pard\s9\f1\fs21 mono\par
|
||
\pard\s9\f2\fs21 arial\par}";
|
||
let runs = text_runs(rtf);
|
||
let mono = run_fmt(&runs, "mono");
|
||
assert_eq!((mono.size, mono.font.as_deref(), mono.code), (None, None, false));
|
||
let arial = run_fmt(&runs, "arial");
|
||
assert_eq!((arial.size, arial.font.as_deref(), arial.code), (None, Some("Arial"), false));
|
||
}
|
||
|
||
#[test]
|
||
fn hostile_sbasedon_chains_are_harmless() {
|
||
let mut chain = String::new();
|
||
// s100 (heading 5) based on s101 … s139, the last one not bold: past the bound, so
|
||
// heading 5 takes WordCraft's bold.
|
||
for i in 100..140 {
|
||
let name = if i == 100 { "heading 5".to_string() } else { format!("x{i}") };
|
||
let b0 = if i == 139 { "\\b0" } else { "" };
|
||
chain.push_str(&format!("{{\\s{i}\\sbasedon{}{b0} {name};}}", i + 1));
|
||
}
|
||
let rtf = format!(
|
||
r"{{\rtf1{{\stylesheet{{\s0 Normal;}}{{\s1\sbasedon1 heading 1;}}{{\s3\sbasedon4\b0 heading 3;}}{{\s4\sbasedon3 heading 4;}}
|
||
{{\s6\sbasedon-5 heading 6;}}{{\s2\sbasedon2147483647\fs-40 heading 2;}}{{\s-2147483648\sbasedon-2147483648 Title;}}{chain}}}
|
||
\pard\s1\b\fs40 one\par
|
||
\pard\s3\b\fs24 three\par
|
||
\pard\s4\b\fs24 four\par
|
||
\pard\s6\b\fs22 six\par
|
||
\pard\s2\b\fs32 two\par
|
||
\pard\s100\b\fs24 five\par
|
||
\pard\s-2147483648\fs56 title\par}}"
|
||
);
|
||
let runs = text_runs(rtf.as_bytes());
|
||
// Self-based heading 1: WordCraft's values.
|
||
assert_eq!((run_fmt(&runs, "one").size, run_fmt(&runs, "one").bold), (None, false));
|
||
// A loop between heading 3 and 4: both take heading 3's `\b0`.
|
||
assert!(run_fmt(&runs, "three").bold);
|
||
assert!(run_fmt(&runs, "four").bold);
|
||
assert_eq!((run_fmt(&runs, "six").size, run_fmt(&runs, "six").bold), (None, false));
|
||
assert_eq!((run_fmt(&runs, "two").size, run_fmt(&runs, "two").bold), (None, false));
|
||
assert!(!run_fmt(&runs, "five").bold);
|
||
let _ = run_fmt(&runs, "title");
|
||
}
|
||
|
||
#[test]
|
||
fn a_word_365_heading_is_not_bold_so_a_bold_run_there_is_the_users() {
|
||
// Word 365's Heading 1 isn't bold: its entry sets the font and size but no `\b`, so a
|
||
// bold run, or a tracked change to or from bold, is the user's.
|
||
let rtf = br"{\rtf1\ansi{\fonttbl{\f1 Aptos;}{\f2 Aptos Display;}}{\colortbl;\red15\green71\blue97;}
|
||
{\stylesheet{\ql \f1\fs24 Normal;}{\s1\ql \keepn\outlinelevel0 \f2\fs40\cf1 \sbasedon0 \snext0 heading 1;}}
|
||
{\*\revtbl {Unknown;}{Alice;}}
|
||
\pard\s1\f2\fs40 Plain\par
|
||
\pard\s1\f2\b\fs40 Bold\par
|
||
\pard\s1\f2\fs40 {\b{\*\oldcprops \b0\crauth1}Bolded}{\b0{\*\oldcprops \b\crauth1}Unbolded}\par}";
|
||
let runs = text_runs(rtf);
|
||
let plain = run_fmt(&runs, "Plain");
|
||
assert_eq!((plain.size, plain.bold), (None, false));
|
||
let bold = run_fmt(&runs, "Bold");
|
||
assert_eq!((bold.size, bold.bold), (None, true));
|
||
let bolded = run_fmt(&runs, "Bolded");
|
||
let old = &bolded.fmt_change.as_deref().expect("change kept").old;
|
||
assert_eq!((bolded.bold, old.bold, old.size), (true, false, None));
|
||
let unbolded = run_fmt(&runs, "Unbolded");
|
||
let old = &unbolded.fmt_change.as_deref().expect("change kept").old;
|
||
assert_eq!((unbolded.bold, old.bold), (false, true));
|
||
}
|
||
|
||
#[test]
|
||
fn a_style_chain_without_a_size_has_rtfs_default_size() {
|
||
// The entries carry formatting words but no `\fs`: the style's size is RTF's 12 pt.
|
||
let rtf = br"{\rtf1\ansi{\fonttbl{\f0 Times New Roman;}}{\stylesheet{\s0\f0 Normal;}{\s2\sbasedon0\b heading 2;}}
|
||
\pard\s2\b\fs24 Twelve\par
|
||
\pard\s2\b\fs40 Twenty\par}";
|
||
let runs = text_runs(rtf);
|
||
let twelve = run_fmt(&runs, "Twelve");
|
||
assert_eq!((twelve.size, twelve.bold), (None, false));
|
||
let twenty = run_fmt(&runs, "Twenty");
|
||
assert_eq!((twenty.size, twenty.bold), (Some(20.0), false));
|
||
}
|
||
|
||
#[test]
|
||
fn a_code_paragraph_round_trips_without_a_direct_code_font() {
|
||
// WordCraft's Code entry has no `\f` (the code font is on the paragraph): the code font
|
||
// stands in for the style's.
|
||
let mut code = Para { kind: Kind::Code, ..Default::default() };
|
||
code.push_text("let x", &Fmt::default());
|
||
let flow = Flow { blocks: vec![FBlock::Para(code)], meta: Meta::default() };
|
||
let rtf = export(&model::to_doc(&flow));
|
||
let runs = text_runs(rtf.as_bytes());
|
||
let f = run_fmt(&runs, "let x");
|
||
assert_eq!((f.size, f.bold, f.code, f.font.as_deref()), (None, false, false, None), "{rtf}");
|
||
}
|
||
|
||
/// Escaped semicolons in a font name stop at the limit, as other characters do (review of #3862).
|
||
#[test]
|
||
fn escaped_semicolons_in_a_font_name_stop_at_the_limit() {
|
||
let mut r = reader::Reader::default();
|
||
r.st.dest = state::Dest::FontTbl;
|
||
for _ in 0..100_000 {
|
||
r.ch(';');
|
||
}
|
||
assert_eq!(r.font_name.len(), model::MAX_FONT_NAME + 1);
|
||
let rtf = format!(r"{{\rtf1{{\fonttbl{{\f0 {};}}}}\f0 x\par}}", r"\'3b".repeat(model::MAX_FONT_NAME + 1));
|
||
let runs = text_runs(rtf.as_bytes());
|
||
assert_eq!(run_fmt(&runs, "x").font, None);
|
||
}
|
||
|
||
#[test]
|
||
fn a_semicolon_in_a_font_name_round_trips() {
|
||
// `;` ends a `\fonttbl` entry, so it is written `\'3b` and read back as part of the name (#3740).
|
||
let fonts = [("one", "Semi;Colon"), ("two", "Semi;Other"), ("three", "Tail;"), ("four", "Line\nBreak")];
|
||
let mut p = Para::default();
|
||
for (t, font) in fonts {
|
||
p.push_text(t, &Fmt { font: Some(font.into()), ..Default::default() });
|
||
}
|
||
let flow = Flow { blocks: vec![FBlock::Para(p)], meta: Meta::default() };
|
||
let rtf = export(&model::to_doc(&flow));
|
||
assert!(rtf.contains(r"Semi\'3bColon;") && !rtf.contains(r"\line"), "{rtf}");
|
||
let runs = text_runs(rtf.as_bytes());
|
||
for (t, font) in fonts {
|
||
assert_eq!(run_fmt(&runs, t).font.as_deref(), Some(font.replace('\n', "").as_str()), "{rtf}");
|
||
}
|
||
// A Unicode escape is part of the name too, in a multibyte code page as well; a literal `;` ends it.
|
||
let runs = text_runs(br"{\rtf1\ansi\ansicpg932{\fonttbl{\f0 A\u59?B;}{\f1 C\'3bD;}}\f0 x {\f1 y}\par}");
|
||
assert_eq!((run_fmt(&runs, "x ").font.as_deref(), run_fmt(&runs, "y").font.as_deref()), (Some("A;B"), Some("C;D")));
|
||
}
|
||
|
||
#[test]
|
||
fn a_stylesheet_entry_without_a_number_is_normal() {
|
||
// An entry without `\sN` is style 0; it doesn't replace the entry before it.
|
||
let rtf = br"{\rtf1\ansi{\stylesheet{\s2\b\fs26 heading 2;}{\b0\fs30 Normal;}}
|
||
\pard\s2\b\fs26 Head\par
|
||
\pard\s2\b\fs30 Other\par}";
|
||
let runs = text_runs(rtf);
|
||
let head = run_fmt(&runs, "Head");
|
||
assert_eq!((head.size, head.bold), (None, false));
|
||
let other = run_fmt(&runs, "Other");
|
||
assert_eq!((other.size, other.bold), (Some(15.0), false));
|
||
}
|
||
|
||
#[test]
|
||
fn a_heading_size_set_by_hand_survives_in_the_old_formatting() {
|
||
// Reject puts back a size set by hand: it isn't the file's heading size.
|
||
let rtf = br"{\rtf1\ansi{\fonttbl{\f0 Times New Roman;}}{\stylesheet{\ql \fs24 Normal;}{\s2\ql \b\fs26 \sbasedon0 \snext0 heading 2;}}
|
||
{\*\revtbl {Unknown;}{Alice;}}
|
||
\pard\s2 {\b\fs26{\*\oldcprops \fs40\crauth1}Shrunk}{\b\fs40{\*\oldcprops \fs26\crauth1}Grown}\par}";
|
||
let runs = text_runs(rtf);
|
||
let shrunk = run_fmt(&runs, "Shrunk");
|
||
assert_eq!(shrunk.size, None);
|
||
let old = &shrunk.fmt_change.as_deref().expect("change kept").old;
|
||
assert_eq!((old.size, old.bold), (Some(20.0), false));
|
||
let grown = run_fmt(&runs, "Grown");
|
||
assert_eq!(grown.size, Some(20.0));
|
||
let old = &grown.fmt_change.as_deref().expect("change kept").old;
|
||
assert_eq!((old.size, old.bold), (None, false));
|
||
}
|
||
|
||
#[test]
|
||
fn pard_inside_oldcprops_keeps_the_words_before_it() {
|
||
let doc = import(br"{\rtf1{\*\revtbl {Unknown;}{Alice;}}{\b{\*\oldcprops \i\crauth1\pard\b0}x}\par}").unwrap();
|
||
let paras = doc_paras(&doc);
|
||
let cp = paras.first().unwrap().props_of_char(0);
|
||
let c = cp.fmt_change.as_deref().expect("change kept");
|
||
assert_eq!(c.old.italic, Some(true));
|
||
assert_ne!(c.old.bold, Some(true));
|
||
assert_eq!(rev_of(&doc, Some(c.rev)).map(|r| r.1).as_deref(), Some("Alice"));
|
||
}
|
||
|
||
#[test]
|
||
fn a_section_break_under_a_deletion_has_a_deleted_mark() {
|
||
let rtf = br"{\rtf1{\*\revtbl {Unknown;}{Bob;}}\deleted\revauthdel1 one\sect\pard\plain two\par}";
|
||
let doc = import(rtf).unwrap();
|
||
let paras = doc_paras(&doc);
|
||
assert_eq!(paras.first().map(|p| p.text.as_str()), Some("one"));
|
||
assert_eq!(rev_of(&doc, paras.first().and_then(|p| p.mark.del)).map(|r| r.1).as_deref(), Some("Bob"));
|
||
}
|
||
|
||
#[test]
|
||
fn hostile_formatting_revisions_never_panic() {
|
||
let text = |rtf: &[u8]| import(rtf).map(|d| doc_paras(&d).iter().map(|p| p.text.clone()).collect::<Vec<_>>());
|
||
// Old properties without an author: the text is kept and nothing else appears.
|
||
assert_eq!(text(br"{\rtf1{\*\oldcprops \b}x\par}").unwrap().first().map(String::as_str), Some("x"));
|
||
// Huge and negative indices and dates.
|
||
let doc =
|
||
import(br"{\rtf1{\b\crauth2147483647\crdate4294967295{\*\oldcprops \i}x}\pard\prauth-5\prdate4294967295{\*\oldpprops \qc}y\par}").unwrap();
|
||
let paras = doc_paras(&doc);
|
||
let author = |i: Option<u32>| rev_of(&doc, i).map(|r| r.1);
|
||
assert_eq!(paras.first().map(|p| p.text.as_str()), Some("xy"));
|
||
assert_eq!(author(paras.first().and_then(|p| p.props_of_char(0).fmt_change.as_deref().map(|c| c.rev))).as_deref(), Some("Unknown"));
|
||
// Structure inside old properties is ignored: no paragraphs, cells or text come from it.
|
||
assert_eq!(text(br"{\rtf1\pard{\*\oldpprops \pard\qc\par\cell\row\trowd\cellx100 junk}z\par}").unwrap().first().map(String::as_str), Some("z"));
|
||
assert_eq!(text(br"{\rtf1{\b\crauth1{\*\oldcprops {\*\oldcprops \i}\par x}y}\par}").unwrap().first().map(String::as_str), Some("y"));
|
||
// Border words in old properties don't edit the cell border being defined.
|
||
let f = parse(br"{\rtf1\ansi\trowd\clbrdrt\brdrs\brdrw10{\*\oldpprops \brdrdb\brdrw120\brdrcf1}\cellx2000\pard\intbl a\cell\row}").unwrap();
|
||
let FBlock::Table(t) = &f.blocks[0] else { panic!() };
|
||
let top = t.rows[0][0].borders.and_then(|b| b.top).unwrap();
|
||
assert_eq!((top.style, top.width, top.color), (craft_wordmodel::props::BorderStyle::Single, 0.5, None));
|
||
// Unterminated groups and deep nesting.
|
||
for bad in [
|
||
&br"{\rtf1\pard\prauth1\prdate5{\*\oldpprops \pard\qc"[..],
|
||
br"{\rtf1{\crauth1{\*\oldcprops \b",
|
||
br"{\rtf1\crauth\crdate\prauth\prdate{\*\oldcprops}{\*\oldpprops}",
|
||
] {
|
||
let _ = import(bad);
|
||
}
|
||
let mut deep = String::from("{\\rtf1\\crauth1");
|
||
for _ in 0..20_000 {
|
||
deep.push_str("{\\*\\oldcprops \\b");
|
||
}
|
||
deep.push_str(" x\\par}");
|
||
let _ = import(deep.as_bytes());
|
||
}
|
||
|
||
#[test]
|
||
fn sections_keep_their_page_columns_numbering_and_headers() {
|
||
// A4 portrait → landscape restarting at 1 with its own header → continuous two columns
|
||
// → portrait. A `\sectd` inside a footer group is the footer's, not the body's.
|
||
let rtf = br"{\rtf1\ansi\paperw11906\paperh16838\margl1134\margr1134
|
||
\sectd\headery708\footery708
|
||
{\header \pard One head\par}
|
||
{\footer \pard Page {\field{\*\fldinst PAGE}{\fldrslt 1}}\sectd\par}
|
||
\pard First\par\sect
|
||
\sectd\lndscpsxn\pgwsxn16838\pghsxn11906\headery708\pgnrestart\pgnstarts1\pgnlcrm\titlepg
|
||
{\header \pard Two head\par}
|
||
{\headerf \pard Two first\par}
|
||
{\footer \pard Two foot\sectd\lndscpsxn\par}
|
||
\pard Second\par
|
||
\trowd\cellx2000\pard\intbl cell\cell\row
|
||
\pard{\sect}
|
||
\sectd\lndscpsxn\pgwsxn16838\pghsxn11906\headery708\sbknone\pgnrestart\cols2\colsx567\linebetcol
|
||
\pard Third\sect
|
||
\sectd\headery708\sbkodd
|
||
\pard Fourth\par}";
|
||
let d = import(rtf).unwrap();
|
||
let secs = d.sections();
|
||
assert_eq!(secs.len(), 4, "{secs:?}");
|
||
let s: Vec<&craft_wordmodel::SectionProps> = secs.iter().map(|(_, s)| *s).collect();
|
||
assert_section_geometry(&s);
|
||
assert_section_breaks_and_layout(&s);
|
||
assert_section_headers_and_footers(&d, &s);
|
||
assert_section_body_boundaries(&d, &secs);
|
||
assert_section_page_defaults();
|
||
}
|
||
|
||
fn assert_section_geometry(s: &[&craft_wordmodel::SectionProps]) {
|
||
// Page size and orientation; the document's margins apply to every section.
|
||
for (i, land) in [(0, false), (1, true), (2, true), (3, false)] {
|
||
assert_eq!(s[i].landscape, land, "section {i}");
|
||
let (w, h) = if land { (841.9, 595.3) } else { (595.3, 841.9) };
|
||
assert!((s[i].page_w - w).abs() < 0.1 && (s[i].page_h - h).abs() < 0.1, "section {i}: {} x {}", s[i].page_w, s[i].page_h);
|
||
assert!((s[i].margin_left - 56.7).abs() < 0.1, "section {i}");
|
||
assert!((s[i].header - 35.4).abs() < 0.1, "section {i}");
|
||
}
|
||
}
|
||
|
||
fn assert_section_breaks_and_layout(s: &[&craft_wordmodel::SectionProps]) {
|
||
use craft_wordmodel::section::{NumFormat, SectionStart};
|
||
assert_eq!(
|
||
s.iter().map(|s| s.start).collect::<Vec<_>>(),
|
||
[SectionStart::NextPage, SectionStart::NextPage, SectionStart::Continuous, SectionStart::OddPage]
|
||
);
|
||
// Columns.
|
||
assert_eq!(s[2].columns.count, 2);
|
||
assert!((s[2].columns.space - 28.35).abs() < 0.01 && s[2].columns.separator);
|
||
assert_eq!(s[0].columns.count, 1);
|
||
assert_eq!(s[3].columns.count, 1);
|
||
// Page numbering: restart at 1 (lower roman) in 2 and 3; section 4 carries on.
|
||
assert_eq!(s.iter().map(|s| s.page_num_start).collect::<Vec<_>>(), [None, Some(1), Some(1), None]);
|
||
assert_eq!(s[1].page_num_format, NumFormat::LowerRoman);
|
||
assert_eq!(s[0].page_num_format, NumFormat::Decimal);
|
||
assert!(!s[0].title_page && s[1].title_page && !s[3].title_page);
|
||
}
|
||
|
||
fn assert_section_headers_and_footers(d: &craft_wordmodel::Document, s: &[&craft_wordmodel::SectionProps]) {
|
||
let text = |id: Option<u32>| {
|
||
let p = id.and_then(|id| d.parts.get(&id)).unwrap();
|
||
p.blocks
|
||
.iter()
|
||
.filter_map(|b| if let craft_wordmodel::Block::Para(p) = &**b { Some(p.text.clone()) } else { None })
|
||
.collect::<Vec<_>>()
|
||
.join("|")
|
||
};
|
||
assert_eq!(text(s[0].headers.default), "One head");
|
||
assert!(text(s[0].footers.default).starts_with("Page"));
|
||
assert_eq!(text(s[1].headers.default), "Two head");
|
||
assert_eq!(text(s[1].headers.first), "Two first");
|
||
assert_eq!(text(s[1].footers.default), "Two foot");
|
||
assert!(s[0].headers.first.is_none());
|
||
for i in [2, 3] {
|
||
assert_eq!((s[i].headers, s[i].footers), Default::default(), "section {i}");
|
||
}
|
||
}
|
||
|
||
fn assert_section_body_boundaries(d: &craft_wordmodel::Document, secs: &[(usize, &craft_wordmodel::SectionProps)]) {
|
||
// Each section ends on its own paragraph, after the text it holds.
|
||
let body_text = |i: usize| match d.body.get(i).map(|b| &**b) {
|
||
Some(craft_wordmodel::Block::Para(p)) => p.text.clone(),
|
||
_ => "<table>".into(),
|
||
};
|
||
assert_eq!(body_text(secs[0].0 - 1), "First");
|
||
assert_eq!(body_text(secs[1].0 - 1), "<table>");
|
||
assert_eq!(body_text(secs[2].0), "Third");
|
||
assert_eq!(body_text(secs[3].0), "Fourth");
|
||
}
|
||
|
||
fn assert_section_page_defaults() {
|
||
// A landscape document default doesn't turn a section that gives its own portrait page.
|
||
let d = import(br"{\rtf1\ansi\paperw16838\paperh11906\landscape\sectd a\sect\sectd\pgwsxn11906\pghsxn16838 b\par}").unwrap();
|
||
let secs = d.sections();
|
||
assert_eq!(secs.len(), 2);
|
||
assert!(secs[0].1.landscape && secs[0].1.page_w > secs[0].1.page_h, "{:?}", secs[0].1);
|
||
assert!(!secs[1].1.landscape && secs[1].1.page_w < secs[1].1.page_h, "{:?}", secs[1].1);
|
||
}
|
||
|
||
#[test]
|
||
fn export_writes_every_section() {
|
||
use craft_wordmodel::section::{NumFormat, SectionStart};
|
||
use craft_wordmodel::{Block, SectionProps};
|
||
let mut d = craft_wordmodel::Document::from_text("one\ntwo\nthree");
|
||
let s1 = SectionProps { margin_left: 90.0, header: 30.0, footer: 40.0, ..Default::default() };
|
||
let mut s2 = SectionProps { start: SectionStart::Continuous, page_w: 792.0, page_h: 612.0, landscape: true, ..Default::default() };
|
||
s2.columns.count = 2;
|
||
s2.columns.space = 18.0;
|
||
s2.columns.separator = true;
|
||
s2.page_num_start = Some(3);
|
||
s2.page_num_format = NumFormat::LowerRoman;
|
||
let mut s3 = SectionProps { start: SectionStart::OddPage, title_page: true, ..Default::default() };
|
||
s3.page_num_format = NumFormat::UpperLetter;
|
||
for (i, s) in [(0usize, s1.clone()), (1, s2.clone())] {
|
||
if let Some(Block::Para(p)) = d.body.get_mut(i).map(std::sync::Arc::make_mut) {
|
||
p.section = Some(Box::new(s));
|
||
}
|
||
}
|
||
d.last_section = s3.clone();
|
||
let rtf = export(&d);
|
||
assert_eq!(rtf.matches("\\sect\n").count() + rtf.matches("\\sect\\").count(), 2, "{rtf}");
|
||
let back = import(rtf.as_bytes()).unwrap();
|
||
let secs = back.sections();
|
||
assert_eq!(secs.len(), 3, "{rtf}");
|
||
let got: Vec<&SectionProps> = secs.iter().map(|(_, s)| *s).collect();
|
||
for (g, want) in got.iter().zip([&s1, &s2, &s3]) {
|
||
assert_eq!(g.start, want.start);
|
||
assert_eq!(g.landscape, want.landscape);
|
||
assert!((g.page_w - want.page_w).abs() < 0.1 && (g.page_h - want.page_h).abs() < 0.1);
|
||
for (a, b) in [
|
||
(g.margin_left, want.margin_left),
|
||
(g.margin_right, want.margin_right),
|
||
(g.margin_top, want.margin_top),
|
||
(g.margin_bottom, want.margin_bottom),
|
||
] {
|
||
assert!((a - b).abs() < 0.1);
|
||
}
|
||
assert!((g.header - want.header).abs() < 0.1 && (g.footer - want.footer).abs() < 0.1);
|
||
assert_eq!(g.columns.count, want.columns.count);
|
||
assert_eq!(g.columns.separator, want.columns.separator);
|
||
assert_eq!(g.page_num_start, want.page_num_start);
|
||
assert_eq!(g.page_num_format, want.page_num_format);
|
||
assert_eq!(g.title_page, want.title_page);
|
||
}
|
||
assert!((got[1].columns.space - 18.0).abs() < 0.1);
|
||
// Each section keeps its text.
|
||
assert_eq!(secs.iter().map(|(i, _)| *i).collect::<Vec<_>>(), [0, 1, 2]);
|
||
// One section: no section break.
|
||
let one = export(&craft_wordmodel::Document::from_text("a\nb"));
|
||
assert!(!one.contains("\\sect\n") && !one.contains("\\sect\\"), "{one}");
|
||
}
|
||
|
||
#[test]
|
||
fn export_keeps_a_single_sections_formatting() {
|
||
use craft_wordmodel::SectionProps;
|
||
let mut d = craft_wordmodel::Document::from_text("a\nb");
|
||
d.last_section = SectionProps { title_page: true, ..Default::default() };
|
||
d.last_section.columns.count = 2;
|
||
d.last_section.columns.space = 20.0;
|
||
let back = import(export(&d).as_bytes()).unwrap();
|
||
assert_eq!(back.sections().len(), 1);
|
||
assert_eq!(back.last_section.columns.count, 2);
|
||
assert!(back.last_section.title_page);
|
||
}
|
||
|
||
#[test]
|
||
fn export_keeps_a_section_that_ends_the_body() {
|
||
use craft_wordmodel::{Block, SectionProps};
|
||
let mut d = craft_wordmodel::Document::from_text("a\nb");
|
||
let s1 = SectionProps { landscape: true, page_w: 792.0, page_h: 612.0, ..Default::default() };
|
||
if let Some(Block::Para(p)) = d.body.get_mut(1).map(std::sync::Arc::make_mut) {
|
||
p.section = Some(Box::new(s1));
|
||
}
|
||
d.last_section.columns.count = 3;
|
||
let back = import(export(&d).as_bytes()).unwrap();
|
||
let secs = back.sections();
|
||
assert_eq!(secs.len(), 2);
|
||
assert!(secs[0].1.landscape);
|
||
assert_eq!(secs[1].1.columns.count, 3);
|
||
}
|
||
|
||
#[test]
|
||
fn export_writes_no_section_break_inside_a_table() {
|
||
use craft_wordmodel::{Block, SectionProps};
|
||
let mut d = craft_wordmodel::Document::from_text("a");
|
||
let mut t = craft_wordmodel::Table::new(1, 1, 400.0);
|
||
if let Some(Block::Para(p)) =
|
||
t.rows.first_mut().and_then(|r| r.cells.first_mut()).and_then(|c| c.blocks.first_mut()).map(std::sync::Arc::make_mut)
|
||
{
|
||
p.section = Some(Box::new(SectionProps::default()));
|
||
}
|
||
d.body.push(std::sync::Arc::new(Block::Table(t)));
|
||
let rtf = export(&d);
|
||
assert!(!rtf.contains("\\sect\n") && !rtf.contains("\\sect\\") && rtf.matches("\\sectd").count() == 1, "{rtf}");
|
||
assert_eq!(import(rtf.as_bytes()).unwrap().sections().len(), 1);
|
||
}
|
||
|
||
#[test]
|
||
fn sections_are_capped_and_a_trailing_sect_adds_no_empty_section() {
|
||
// A file that ends with `\sect` has no empty section after it.
|
||
let d = import(br"{\rtf1\ansi\sectd\sbknone a\sect\sectd\lndscpsxn b\sect}").unwrap();
|
||
let secs = d.sections();
|
||
assert_eq!(secs.len(), 2, "{secs:?}");
|
||
assert!(secs[1].1.landscape && !secs[0].1.landscape);
|
||
// Hostile: a flood of section breaks with headers is capped and never panics.
|
||
let mut rtf = b"{\\rtf1\\ansi".to_vec();
|
||
for _ in 0..(MAX_SECTIONS + 50) {
|
||
rtf.extend_from_slice(b"\\sectd\\cols99999\\colsx-5\\pgnrestart\\pgnstarts-7{\\header h\\par}{\\footerl f\\par}x\\sect");
|
||
}
|
||
rtf.push(b'}');
|
||
let d = import(&rtf).unwrap();
|
||
let secs = d.sections();
|
||
assert_eq!(secs.len(), MAX_SECTIONS);
|
||
// Each section keeps its own header and footer; the ones past the cap replace the last's.
|
||
assert!(d.parts.len() <= 2 * MAX_SECTIONS, "{}", d.parts.len());
|
||
assert!(secs.iter().all(|(_, s)| s.columns.count <= 45 && s.columns.space >= 0.0 && s.page_num_start == Some(0)));
|
||
}
|
||
|
||
#[test]
|
||
fn rejects_non_rtf() {
|
||
assert!(parse(b"hello").is_err());
|
||
assert!(parse(b"{\\rtf1").is_ok());
|
||
}
|
||
|
||
#[test]
|
||
fn escapes() {
|
||
assert_eq!(esc("a{b}\\c\té😀"), "a\\{b\\}\\\\c\\tab \\u233?\\u-10179?\\u-8704?");
|
||
assert_eq!(esc("col1\u{000E}col2"), "col1\\column col2");
|
||
}
|
||
|
||
#[test]
|
||
fn column_break_import() {
|
||
let f = parse(br"{\rtf1\ansi \pard a\column b\par}").unwrap();
|
||
assert_eq!(texts(&f), ["Normal:a\u{000E}b"]);
|
||
}
|
||
|
||
#[test]
|
||
fn hyperlink_field_targets() {
|
||
assert_eq!(hyperlink(r#"HYPERLINK "http://example.com" \l "heading""#), Some("http://example.com#heading".into()));
|
||
assert_eq!(hyperlink(r#"HYPERLINK \l "heading""#), Some("#heading".into()));
|
||
assert_eq!(hyperlink(r#"Hyperlink "http://example.com""#), Some("http://example.com".into()));
|
||
assert_eq!(hyperlink(r#"HYPERLINK "http://example.com/path\latest""#), Some("http://example.com/pathlatest".into()));
|
||
assert_eq!(hyperlink(r#"HYPERLINK "http://example.com" \o "Help text""#), Some("http://example.com".into()));
|
||
// A tooltip or frame before the address, a longer field name, escapes and the writer's `%22`.
|
||
assert_eq!(hyperlink(r#"HYPERLINK \o "tip" \t "_blank" "http://example.com""#), Some("http://example.com".into()));
|
||
assert_eq!(hyperlink(r#"HYPERLINKX "http://example.com""#), None);
|
||
assert_eq!(hyperlink(r#"HYPERLINK "C:\\Docs\\f.rtf""#), Some(r"C:\Docs\f.rtf".into()));
|
||
assert_eq!(hyperlink(r#"HYPERLINK "http://example.com/a%22b""#), Some(r#"http://example.com/a"b"#.into()));
|
||
}
|
||
|
||
/// A link with a quote in it survives export and import: the writer stores it as `%22`.
|
||
#[test]
|
||
fn a_quote_in_a_link_round_trips() {
|
||
let rtf = br#"{\rtf1\ansi{\field{\*\fldinst{HYPERLINK "http://a.b/x%22y"}}{\fldrslt{link}}}\par}"#;
|
||
let link = |d: &Document| d.body.iter().find_map(|b| b.as_para()?.runs.iter().find_map(|r| r.props.link.clone()));
|
||
let d = import(rtf).unwrap();
|
||
assert_eq!(link(&d).as_deref(), Some(r#"http://a.b/x"y"#));
|
||
let exported = export(&d);
|
||
assert!(exported.contains("x%22y"), "{exported}");
|
||
assert_eq!(link(&import(exported.as_bytes()).unwrap()).as_deref(), Some(r#"http://a.b/x"y"#));
|
||
}
|
||
|
||
/// Backslashes in a link survive export and import, a trailing one and two in a row too.
|
||
#[test]
|
||
fn backslashes_in_a_link_round_trip() {
|
||
let link = |d: &Document| d.body.iter().find_map(|b| b.as_para()?.runs.iter().find_map(|r| r.props.link.clone()));
|
||
for target in [r"docs\", r"a\\b", r"..\up\f.doc"] {
|
||
let mut d = import(br#"{\rtf1\ansi{\field{\*\fldinst{HYPERLINK "x"}}{\fldrslt{link}}}\par}"#).unwrap();
|
||
for b in d.body.iter_mut() {
|
||
if let craft_wordmodel::Block::Para(p) = std::sync::Arc::make_mut(b) {
|
||
p.runs.iter_mut().filter(|r| r.props.link.is_some()).for_each(|r| r.props.link = Some(target.into()));
|
||
}
|
||
}
|
||
assert_eq!(link(&d).as_deref(), Some(target));
|
||
let exported = export(&d);
|
||
assert_eq!(link(&import(exported.as_bytes()).unwrap()).as_deref(), Some(target), "{exported}");
|
||
}
|
||
}
|
||
|
||
#[test]
|
||
fn distributed_alignment_round_trips() {
|
||
let rtf = br"{\rtf1\ansi\pard\qd Distributed\par\pard\qj Justified\par}";
|
||
let flow = parse(rtf).unwrap();
|
||
let p0 = match &flow.blocks[0] {
|
||
FBlock::Para(p) => p.align,
|
||
_ => panic!("expected para"),
|
||
};
|
||
let p1 = match &flow.blocks[1] {
|
||
FBlock::Para(p) => p.align,
|
||
_ => panic!("expected para"),
|
||
};
|
||
assert_eq!(p0, Some(Align::Distribute));
|
||
assert_eq!(p1, Some(Align::Justify));
|
||
|
||
let doc = import(rtf).unwrap();
|
||
let exported = export(&doc);
|
||
assert!(exported.contains("\\qd"), "expected \\qd in: {exported}");
|
||
assert!(exported.contains("\\qj"), "expected \\qj in: {exported}");
|
||
}
|
||
|
||
#[test]
|
||
fn table_cell_vertical_alignment() {
|
||
let rtf = br"{\rtf1\ansi\trowd\clvertalt\cellx1000\clvertalc\cellx2000\clvertalb\cellx3000\intbl top\cell center\cell bottom\cell\row}";
|
||
let doc = import(rtf).unwrap();
|
||
let craft_wordmodel::Block::Table(t) = &*doc.body[0] else { panic!() };
|
||
assert_eq!(t.rows[0].cells[0].props.valign, Some(VAlign::Top));
|
||
assert_eq!(t.rows[0].cells[1].props.valign, Some(VAlign::Center));
|
||
assert_eq!(t.rows[0].cells[2].props.valign, Some(VAlign::Bottom));
|
||
let exported = export(&doc);
|
||
assert!(exported.contains("\\clvertalt"));
|
||
assert!(exported.contains("\\clvertalc"));
|
||
assert!(exported.contains("\\clvertalb"));
|
||
}
|
||
|
||
#[test]
|
||
fn rtf_backslash_space_and_special_breaks() {
|
||
let rtf = br"{\rtf1\ansi A\ B\~C\zwbo D\softline E\softpage F\par}";
|
||
let doc = import(rtf).unwrap();
|
||
let p = doc.body.first().unwrap().as_para().unwrap();
|
||
assert_eq!(p.text, "A\u{00A0}B\u{00A0}C\u{200B}D\nE\u{000C}F");
|
||
}
|
||
|
||
#[test]
|
||
fn compound_borders_round_trip() {
|
||
use craft_wordmodel::props::BorderStyle;
|
||
let lines = [
|
||
("brdrtnthsg", BorderStyle::ThinThickSmallGap),
|
||
("brdrthtnsg", BorderStyle::ThickThinSmallGap),
|
||
("brdrtnthtnsg", BorderStyle::ThinThickThinSmallGap),
|
||
("brdrtnthmg", BorderStyle::ThinThickMediumGap),
|
||
("brdrthtnmg", BorderStyle::ThickThinMediumGap),
|
||
("brdrtnthtnmg", BorderStyle::ThinThickThinMediumGap),
|
||
("brdrtnthlg", BorderStyle::ThinThickLargeGap),
|
||
("brdrthtnlg", BorderStyle::ThickThinLargeGap),
|
||
("brdrtnthtnlg", BorderStyle::ThinThickThinLargeGap),
|
||
];
|
||
let body: String = lines.iter().map(|(w, _)| format!("\\pard\\brdrt\\{w}\\brdrw30 {w}\\par\n")).collect();
|
||
let rtf = format!("{{\\rtf1\\ansi\n{body}}}");
|
||
let tops = |d: &craft_wordmodel::Document| -> Vec<Option<BorderStyle>> {
|
||
d.body
|
||
.iter()
|
||
.filter_map(|b| b.as_para())
|
||
.filter(|p| !p.plain_text().is_empty())
|
||
.map(|p| p.props.borders.and_then(|b| b.top).map(|t| t.style))
|
||
.collect()
|
||
};
|
||
let want: Vec<Option<BorderStyle>> = lines.iter().map(|(_, s)| Some(*s)).collect();
|
||
let d = import(rtf.as_bytes()).unwrap();
|
||
assert_eq!(tops(&d), want);
|
||
let back = import(export(&d).as_bytes()).unwrap();
|
||
assert_eq!(tops(&back), want);
|
||
}
|