Table formulas: read the Panamanian balboa B/. in numbers (#3546)

* Table formulas: read the Panamanian balboa B/. in numbers

Word's table formula and sort number reader ignores a currency sign or short currency word around a number; the balboa's B/. was not in the list. Add it as a whitespace-delimited sign, as Word reads it: B/. 1,250.50 and 1,250.50 B/. are numbers, while B/.1,250.50 and B/ 1,250.50 are not. Tested in the locale reader and through =SUM(ABOVE); the localization parity doc records the behaviour.

* Table formulas: read the balboa cell in the document's language

The formula test summed a B/. cell through ev, which supplies no document
or language, so a regression in locale selection could pass. Evaluate
=SUM(ABOVE) with a document whose default text is es-PA and a pl-PL
editing language: 1,250.50 is a number under es-PA and not under pl-PL,
so the assertion fails if the formula stops consulting the cell's
language. The converse case reads B/. 1 250,50 from a pl-PL document with
an en-US fallback, which English alone cannot parse.

* docs: close the parenthesis in the localization-parity change note (#3546)

---------

Co-authored-by: Francisco Herrera <14956584+alosslessdev@users.noreply.github.com>
Co-authored-by: FredApps <FredApps@users.noreply.github.com>

crates/doc/src/locale_number.rs

@@ -22,10 +22,29 @@ pub fn decimal_separator(lang: Option<&str>) -> char {
 
 const SPACES: &[char] = &[' ', '\u{a0}', '\u{202f}', '\u{2009}'];
 const CURRENCY: &[char] = &['$', '€', '£', '¥', '¢', '₽', '₴', '₺', '₹', '₩', '₪', '₫', '฿', '₱', '₸', '₼', '₾'];
+/// Currency signs spelled with more than one character: the Panamanian balboa (`B/.`).
+const CURRENCY_SYMBOLS: &[&str] = &["B/."];
 
-/// Strip a currency symbol or a short currency word (`zł`, `kr`, `PLN`) from either end.
+/// Trim spaces and single-character currency signs from either end.
+fn trim_currency(s: &str) -> &str {
+    s.trim_matches(|c: char| c.is_whitespace() || CURRENCY.contains(&c))
+}
+
+/// Strip a currency symbol or a short currency word (`zł`, `kr`, `PLN`, `B/.`) from either end.
+/// As Word does, a multi-character sign is a word of its own (`B/. 1,250.50`, `1,250.50 B/.`);
+/// `B/.1,250.50` and `B/ 1,250.50` are not numbers.
 fn strip_currency(s: &str) -> &str {
-    let mut s = s.trim_matches(|c: char| c.is_whitespace() || CURRENCY.contains(&c));
+    let mut s = trim_currency(s);
+    if let Some(rest) =
+        CURRENCY_SYMBOLS.iter().find_map(|sym| s.strip_prefix(sym).filter(|rest| rest.chars().next().is_some_and(char::is_whitespace)))
+    {
+        s = trim_currency(rest);
+    }
+    if let Some(rest) =
+        CURRENCY_SYMBOLS.iter().find_map(|sym| s.strip_suffix(sym).filter(|rest| rest.chars().next_back().is_some_and(char::is_whitespace)))
+    {
+        s = trim_currency(rest);
+    }
     if let Some((head, word)) = s.rsplit_once(SPACES)
         && (1..=3).contains(&word.chars().count())
         && word.chars().all(|c| c.is_alphabetic() || c == '.')
@@ -137,6 +156,7 @@ mod tests {
         assert_eq!(parse_number("1.5", pl), Some(1.5));
         assert_eq!(parse_number("-2,25", pl), Some(-2.25));
         assert_eq!(parse_number("PLN 12", pl), Some(12.0));
+        assert_eq!(parse_number("-B/. 1 250,50", pl), Some(-1250.5));
         assert_eq!(parse_number("abc", pl), None);
         assert_eq!(parse_number("1,2,3", pl), None);
         // From review: ordinals, a leading zero, percentages and an en dash minus.
@@ -174,11 +194,40 @@ mod tests {
         assert_eq!(format_number(5.75, Some("es-ES")), "5,75");
     }
 
+    #[test]
+    fn balboa_numbers() {
+        let en = Some("en-US");
+        // Word 16.0 reads `B/.` as a word of its own, before or after the number.
+        assert_eq!(parse_number("B/. 1,234.50", en), Some(1234.5));
+        assert_eq!(parse_number("1,234.50 B/.", en), Some(1234.5));
+        assert_eq!(parse_number("B/.\u{a0}1,234.50", en), Some(1234.5), "a no-break space after the sign");
+        assert_eq!(parse_number("B/. 1,234.50 B/.", en), Some(1234.5), "the sign at both ends");
+        assert_eq!(parse_number("-B/. 1,234.50", en), Some(-1234.5), "a minus before the sign");
+        assert_eq!(parse_number("B/. 1,234.50", Some("es-PA")), Some(1234.5), "the balboa's own region");
+        // The sign has no space of its own in Word: touching the number is text.
+        assert_eq!(parse_number("B/.1,234.50", en), None);
+        assert_eq!(parse_number("1,234.50B/.", en), None);
+        assert_eq!(parse_number("B/ 1,234.50", en), None, "the balboa is B/. in Word");
+        assert_eq!(parse_number("B/.", en), None);
+        // The slash and the full stop make `B/.` ambiguous with a fraction: the sign is matched at
+        // the ends of the text only, so a fraction is never read as a balboa amount and a sign in
+        // the middle is not a sign.
+        assert_eq!(parse_number("1/.5", en), None, "a fraction is not a price");
+        assert_eq!(parse_number("1/.5", Some("es-PA")), None);
+        assert_eq!(parse_number("1/.5 B/.", en), None, "the sign at the end does not rescue a fraction");
+        assert_eq!(parse_number("B/. 1/.5", en), None);
+        assert_eq!(parse_number("2 B/. 1,234.50", en), None, "the sign sits before the number");
+        assert_eq!(parse_number("1,234.50 B/. 1,234.50", en), None, "the sign sits between two numbers");
+        assert_eq!(parse_number("1,234.50/2", en), None, "a division is not a balboa amount");
+        assert_eq!(parse_number("B/. 1,234.50,50", en), None, "a second decimal separator");
+    }
+
     #[test]
     fn english_numbers() {
         let en = Some("en-US");
         assert_eq!(parse_number("1,250.50", en), Some(1250.5));
         assert_eq!(parse_number("$1,250", en), Some(1250.0));
+        assert_eq!(parse_number("PAB 1,250.50", en), Some(1250.5), "the ISO code");
         assert_eq!(parse_number("1.5", en), Some(1.5));
         assert_eq!(parse_number("12", None), Some(12.0));
         assert_eq!(parse_number(".5", None), Some(0.5));

crates/engine/src/cmd/table_formula.rs

@@ -924,6 +924,33 @@ mod tests {
     #[test]
     fn numbers_read_currency_and_separators_but_not_letters() {
         assert_eq!(parse_number("$1,200", None), Some(1200.0));
+        assert_eq!(parse_number("B/. 1,234.50", None), Some(1234.5), "the Panamanian balboa");
+        assert_eq!(parse_number("1,234.50 B/.", None), Some(1234.5));
+        assert_eq!(parse_number("(B/. 1,234.50)", None), Some(-1234.5), "an accounting negative");
+        assert_eq!(parse_number("B/.1,234.50", None), None, "Word needs a space after the sign");
+        assert_eq!(parse_number("1/.5", None), None, "a fraction is not a balboa amount");
+        // The formula reads the cell in the document's language. With the document's default text
+        // as es-PA and the editing language as pl-PL the choice is observable: `1,250.50` is a
+        // number under es-PA and not under pl-PL (whose decimal separator is `,`).
+        let balboa = table(&[&["B/. 1,234.50"], &["(B/. 1,234.50)"], &[""]]);
+        let mut doc = wordcraft_doc::Document::new();
+        std::sync::Arc::make_mut(&mut doc.styles).default_chr.lang = Some("es-PA".into());
+        assert_eq!(
+            evaluate_with(&balboa, Some(&doc), Some("pl-PL"), 2, 0, "=SUM(ABOVE)"),
+            Ok(0.0),
+            "a balboa cell sums in the document's language, and its negative in parentheses cancels it",
+        );
+        assert_eq!(evaluate_with(&balboa, Some(&doc), Some("pl-PL"), 2, 0, "=COUNT(ABOVE)"), Ok(2.0), "both balboa cells are numbers",);
+        // And the other way around: a balboa cell written with Polish separators is read only when
+        // the document language reaches the parser, not when it falls back to English.
+        let mut polish = wordcraft_doc::Document::new();
+        std::sync::Arc::make_mut(&mut polish.styles).default_chr.lang = Some("pl-PL".into());
+        let balboa_pl = table(&[&["B/. 1 250,50"], &[""]]);
+        assert_eq!(
+            evaluate_with(&balboa_pl, Some(&polish), Some("en-US"), 1, 0, "=SUM(ABOVE)"),
+            Ok(1250.5),
+            "the cell's language selects the separators; English does not read `1 250,50`",
+        );
         assert_eq!(parse_number(" 12% ", None), Some(12.0));
         assert_eq!(parse_number("(5)", None), Some(-5.0));
         assert_eq!(parse_number("x9", None), None);

docs/localization-parity.md

@@ -1,6 +1,6 @@
 # Localization parity with Microsoft Word
 
-> **Last reviewed:** 2026-10-10 · **Last updated:** 2026-11-04 · **Change:** trivial (Arabic: the dialogs of the mirrored Arabic interface read right to left: label-and-field rows, tables (list columns, the Models table), tile galleries and the Add/Remove arrows follow the reading direction, through one `rtl::Table` helper that draws exactly as before in English; no ranked gap closed, readiness estimates unchanged); previously minor (Options › Language redesign: Your languages, Add a language…, Set as default; the Dictionaries folder moved from Proofing); previously trivial (Language packs: the missing-pack bar works on a fresh install and counts the caret's language); previously trivial (Language lists: every installed and available Language pack's language can be chosen in Set Proofing Language, installed ones in Review › Language and the status bar; craft-i18n.rev at 9d6d54d: region names and the Language (Region) template in 35 locales, craft-i18n#145); previously trivial (craft-i18n.rev at cde4900: rows for Options › Language, the missing-pack bar and 28 Language pack language names in 35 locales, craft-i18n#136); previously trivial (craft-proof pin: French elision suggestions first, ca-ES-valencia reads the ca-valencia pack); previously minor (Arabic `ar` interface: craft-i18n catalog, mirrored shell, font fallback, #1195; 28 languages); previously trivial (Language packs: language-packs-2 is published and pinned: 75 packs, adding Danish, German (Germany, Austria, Switzerland), Esperanto, Estonian, Persian, Croatian, Hungarian, Norwegian (Bokmål, Nynorsk), Dutch and Swedish (Sweden, Finland)); previously trivial (Language packs: language-packs-1 is published and pinned, 61 packs; the language requests it covers); previously minor (Language packs: languages beyond English install their spelling dictionary and hyphenation as a pack from File › Options › Language or the missing-pack bar; the default build embeds English Language data only) · **Target:** Microsoft Word (Microsoft 365) for Mac 16.113.4
+> **Last reviewed:** 2026-10-10 · **Last updated:** 2026-11-04 · **Change:** trivial (table sort and formulas read the Panamanian balboa's `B/.` around a number (`B/. 1,250.50`, `1,250.50 B/.`), as Word does, matched only at the ends of the text so `1/.5` stays a fraction); previously trivial (Arabic: the dialogs of the mirrored Arabic interface read right to left: label-and-field rows, tables (list columns, the Models table), tile galleries and the Add/Remove arrows follow the reading direction, through one `rtl::Table` helper that draws exactly as before in English; no ranked gap closed, readiness estimates unchanged); previously minor (Options › Language redesign: Your languages, Add a language…, Set as default; the Dictionaries folder moved from Proofing); previously trivial (Language packs: the missing-pack bar works on a fresh install and counts the caret's language); previously trivial (Language lists: every installed and available Language pack's language can be chosen in Set Proofing Language, installed ones in Review › Language and the status bar; craft-i18n.rev at 9d6d54d: region names and the Language (Region) template in 35 locales, craft-i18n#145); previously trivial (craft-i18n.rev at cde4900: rows for Options › Language, the missing-pack bar and 28 Language pack language names in 35 locales, craft-i18n#136); previously trivial (craft-proof pin: French elision suggestions first, ca-ES-valencia reads the ca-valencia pack); previously minor (Arabic `ar` interface: craft-i18n catalog, mirrored shell, font fallback, #1195; 28 languages); previously trivial (Language packs: language-packs-2 is published and pinned: 75 packs, adding Danish, German (Germany, Austria, Switzerland), Esperanto, Estonian, Persian, Croatian, Hungarian, Norwegian (Bokmål, Nynorsk), Dutch and Swedish (Sweden, Finland)); previously trivial (Language packs: language-packs-1 is published and pinned, 61 packs; the language requests it covers); previously minor (Language packs: languages beyond English install their spelling dictionary and hyphenation as a pack from File › Options › Language or the missing-pack bar; the default build embeds English Language data only) · **Target:** Microsoft Word (Microsoft 365) for Mac 16.113.4
 
 Interface languages, script support and proofing per language. How translation works (adding a
 language, the embedded catalogs): `crates/ui-egui/src/i18n.rs`. The catalogs, their format and
@@ -219,7 +219,12 @@ Polish, `1.250,50` in German, `1,250.50` in English) and write formula results w
 separator (`crates/doc/src/locale_number.rs`). Cuban and Peruvian Spanish (`es-CU`, `es-PE`) use
 decimal points (`1,234.50`), matching the existing region settings and CLDR's Latin American
 Spanish parent locale; `cuban_and_peruvian_numbers_use_decimal_points` covers parsing and result
-formatting. Interface lists (styles, fonts) still sort by code point.
+formatting. Currency signs and short currency words around the number are ignored, as Word
+ignores them: a single-character sign may touch the number (`$1,250`, `1 250,00 zł`, `PLN 12`)
+while the Panamanian balboa's `B/.` is a word of its own (`B/. 1,250.50`, `1,250.50 B/.`, as
+observed in Word; `B/.1,250.50` and `B/ 1,250.50` are not numbers). Because `B/.` carries a slash
+and a full stop, it is matched only at the ends of the text, so `1/.5` and `1,250.50/2` are not
+read as balboa amounts. Interface lists (styles, fonts) still sort by code point.
 
 Russian (`ru`) sorting keeps ё with е (a secondary diaeresis difference), before ж. Composed and decomposed ё and й have the same keys and index headings; the stored text is unchanged. Tests cover Russian locale aliases, case, representative words and these two canonical pairs. This is a narrow Cyrillic correction, not full Unicode normalization or CLDR collation. Ordering reference: [Unicode CLDR Russian collation chart](https://unicode.org/cldr/charts/42/collation/ru.html).
 
@@ -270,6 +275,7 @@ WordCraft keeps 2,54 cm).
 
 | Date | Change | Summary |
 |---|---|---|
+| 2026-11-04 | trivial | Table sort and formulas read the Panamanian balboa's `B/.` around a number (`B/. 1,250.50`, `1,250.50 B/.`), as Word does; `B/.1,250.50` and `B/ 1,250.50` stay text, matching Word 16.0, and the sign is matched only at the ends of the text, so `1/.5` is unchanged |
 | 2026-11-04 | trivial | craft-i18n.rev at 9d6d54d (craft-i18n#145): 69 region names and the `{language} ({region})` template (full-width in ja, ko, zh) in 35 locales, AI-assisted without native-speaker review |
 | 2026-11-04 | trivial | craft-i18n.rev at cde4900 (craft-i18n#136): translated rows for Options › Language (Your languages, Add a language…, Set as default, Dictionary files), the missing-pack bar and 28 Language pack language names in 35 locales, AI-assisted without native-speaker review |
 | 2026-11-03 | trivial | craft-proof pin moved to craft-libs main with #589 and #590: French suggestions offer the elided form first (`Jadore` → `J'adore` before `J adore`; English keeps plural fixes first), and Valencian (`ca-ES-valencia`) text reads the `ca-valencia` pack |