Merge PR #1976: i18n: add Indonesian, Vietnamese, Hindi and Korean interfaces

Conflict resolutions: - crates/ui-egui/src/system_fonts.rs: Hangul fallback kept from #2872 (U+D7A3 probe, its candidate list); #1976's duplicate probe, hangul_fallback and hangul_candidates dropped; #1976's hangul_candidates_are_font_files_without_adobe_designs test kept (checks #2872's list). - crates/ui-egui/src/theme.rs: SYSTEM_HANGUL_FALLBACK and its fallback entry kept once, at #2872's position and doc wording. - crates/ui-egui/tests/fonts.rs: #2872's system_fallback_covers_hangul kept, #1976's duplicate dropped. - ATTRIBUTION.md: asset count 279 (both sides' rows).
2026-10-28 03:58:45 -06:00 · 2026-10-22 01:50:27 +00:00
parent 97c403fa17 756a2f6fe2
commit d52247c7d7
13 changed files with 8485 additions and 15 deletions
@@ -4,7 +4,7 @@

 Every asset PdfCraft includes, bundles or uses to build its published material, with its author, source and licence. The policy is in [AGENTS.md](AGENTS.md) §1. The machine-readable list, with SHA-256 hashes, is [ATTRIBUTION.toml](ATTRIBUTION.toml). Licence texts are kept beside the assets and summarised in [NOTICE](NOTICE).

-## In this repository (276)
+## In this repository (279)

 | Asset | Title | Author | Licence | Source | Used for |
 |---|---|---|---|---|---|
@@ -28,6 +28,9 @@ Every asset PdfCraft includes, bundles or uses to build its published material,
 | `crates/ui-egui/src/i18n/ar.tsv` | Arabic interface translation catalog | PdfCraft contributors | MIT OR Apache-2.0 | Contributor-original: clean-room translations of PdfCraft's English UI labels, no proprietary localisation resources | Arabic UI strings, compiled in by crates/ui-egui/src/i18n/mod.rs |
 | `crates/ui-egui/src/i18n/tr.tsv` | Turkish interface translation catalog | PdfCraft contributors | MIT OR Apache-2.0 | Contributor-original: clean-room translations of PdfCraft's English UI labels, no proprietary localisation resources | Turkish UI strings, compiled in by crates/ui-egui/src/i18n/mod.rs |
 | `crates/ui-egui/src/i18n/lt.tsv` | Lithuanian interface translation catalog | PdfCraft contributors | MIT OR Apache-2.0 | Contributor-original: clean-room translations of PdfCraft's English UI labels and user-facing messages; original catalog reviewed by a native Lithuanian-speaking contributor; 2026-10-15 coverage additions are AI-assisted and not separately native-reviewed | Lithuanian UI strings, compiled in by crates/ui-egui/src/i18n/mod.rs |
+| `crates/ui-egui/src/i18n/hi.tsv` | Hindi interface translation catalog | PdfCraft contributors | MIT OR Apache-2.0 | Contributor-original: clean-room translations of PdfCraft's English UI labels, no proprietary localisation resources | Hindi UI strings, compiled in by crates/ui-egui/src/i18n/mod.rs |
+| `crates/ui-egui/src/i18n/id.tsv` | Indonesian interface translation catalog | PdfCraft contributors | MIT OR Apache-2.0 | Contributor-original: clean-room translations of PdfCraft's English UI labels, no proprietary localisation resources | Indonesian UI strings, compiled in by crates/ui-egui/src/i18n/mod.rs |
+| `crates/ui-egui/src/i18n/ko.tsv` | Korean interface translation catalog | PdfCraft contributors | MIT OR Apache-2.0 | Contributor-original: clean-room translations of PdfCraft's English UI labels, no proprietary localisation resources | Korean UI strings, compiled in by crates/ui-egui/src/i18n/mod.rs |
 | `assets/icons/accessibility.svg` | Lucide icon "accessibility" | Lucide Contributors (portions Cole Bemis, Feather) | ISC | https://github.com/lucide-icons/lucide (npm lucide-static 1.49.0, icons/accessibility.svg) | UI icon, embedded via crates/ui-egui/src/icon_data.rs |
 | `assets/icons/align-left.svg` | Lucide icon "align-left" | Lucide Contributors (portions Cole Bemis, Feather) | ISC | https://github.com/lucide-icons/lucide (npm lucide-static 1.49.0, icons/align-left.svg) | UI icon, embedded via crates/ui-egui/src/icon_data.rs |
 | `assets/icons/align-center.svg` | Lucide icon "align-center" | Lucide Contributors (portions Cole Bemis, Feather) | ISC | https://github.com/lucide-icons/lucide (npm lucide-static 1.49.0, icons/align-center.svg) | UI icon, embedded via crates/ui-egui/src/icon_data.rs |
@@ -230,6 +230,37 @@ kind = "translation"
 usage = "Lithuanian UI strings, compiled in by crates/ui-egui/src/i18n/mod.rs"
 sha256 = "6ed886d00ad8cd39d8bf254f4733c8b8c90d5bf2c6868ff9c4ca42765a148786"

+[[asset]]
+path = "crates/ui-egui/src/i18n/hi.tsv"
+title = "Hindi interface translation catalog"
+author = "PdfCraft contributors"
+source = "Contributor-original: clean-room translations of PdfCraft's English UI labels, no proprietary localisation resources"
+licence = "MIT OR Apache-2.0"
+licence_file = "LICENSE-MIT"
+kind = "translation"
+usage = "Hindi UI strings, compiled in by crates/ui-egui/src/i18n/mod.rs"
+sha256 = "72951e5f6ba5f1808c991db5c6ba32f7b2d210b8cc9f20da7634532dc5294e27"
+[[asset]]
+path = "crates/ui-egui/src/i18n/id.tsv"
+title = "Indonesian interface translation catalog"
+author = "PdfCraft contributors"
+source = "Contributor-original: clean-room translations of PdfCraft's English UI labels, no proprietary localisation resources"
+licence = "MIT OR Apache-2.0"
+licence_file = "LICENSE-MIT"
+kind = "translation"
+usage = "Indonesian UI strings, compiled in by crates/ui-egui/src/i18n/mod.rs"
+sha256 = "fa4e8dbd5c81e71000507fcde6bb7f4d702171e162b005c9c4b5e7be1f66b20a"
+[[asset]]
+path = "crates/ui-egui/src/i18n/ko.tsv"
+title = "Korean interface translation catalog"
+author = "PdfCraft contributors"
+source = "Contributor-original: clean-room translations of PdfCraft's English UI labels, no proprietary localisation resources"
+licence = "MIT OR Apache-2.0"
+licence_file = "LICENSE-MIT"
+kind = "translation"
+usage = "Korean UI strings, compiled in by crates/ui-egui/src/i18n/mod.rs"
+sha256 = "408e777db94b4d5807dba807f0ed82aee4fee6f99166c8fc2dc1b7dea0027f85"
+

 [[asset]]
 path = "assets/icons/accessibility.svg"
@@ -56,7 +56,7 @@ const AR_DAYS: [&str; 7] = ["الأحد", "الاثنين", "الثلاثاء",

 /// The languages month and weekday names can be written in (Preferences ▸ Date format ▸ Language):
 /// the interface languages, in the same order and with the same names.
-pub static DATE_LANGUAGES: [DateLanguage; 23] = [
+pub static DATE_LANGUAGES: [DateLanguage; 26] = [
    DateLanguage {
        code: "en",
        name: "English",
@@ -360,6 +360,17 @@ pub static DATE_LANGUAGES: [DateLanguage; 23] = [
        days: ["sekmadienis", "pirmadienis", "antradienis", "trečiadienis", "ketvirtadienis", "penktadienis", "šeštadienis"],
        days_short: ["sk", "pr", "an", "tr", "kt", "pn", "št"],
    },
+    // Indonesian. Month and weekday names are the same with and after a day number, so
+    // `months_with_day` is `None`.
+    DateLanguage {
+        code: "id",
+        name: "Bahasa Indonesia",
+        months: ["Januari", "Februari", "Maret", "April", "Mei", "Juni", "Juli", "Agustus", "September", "Oktober", "November", "Desember"],
+        months_with_day: None,
+        months_short: ["Jan", "Feb", "Mar", "Apr", "Mei", "Jun", "Jul", "Agu", "Sep", "Okt", "Nov", "Des"],
+        days: ["Minggu", "Senin", "Selasa", "Rabu", "Kamis", "Jumat", "Sabtu"],
+        days_short: ["Min", "Sen", "Sel", "Rab", "Kam", "Jum", "Sab"],
+    },
    // Vietnamese: numeric month forms (`tháng 1` with a day, `Tháng 1` standing alone).
    // Fill & Sign still refuses any format whose final text fails the existing WinAnsi check
    // (for example weekday names); `dd.mm.yyyy` remains writable.
@@ -398,6 +409,26 @@ pub static DATE_LANGUAGES: [DateLanguage; 23] = [
        days: ["Chủ Nhật", "Thứ Hai", "Thứ Ba", "Thứ Tư", "Thứ Năm", "Thứ Sáu", "Thứ Bảy"],
        days_short: ["CN", "T2", "T3", "T4", "T5", "T6", "T7"],
    },
+    // Hindi (Devanagari). Month and weekday names are the same with and after a day number.
+    DateLanguage {
+        code: "hi",
+        name: "हिन्दी",
+        months: ["जनवरी", "फ़रवरी", "मार्च", "अप्रैल", "मई", "जून", "जुलाई", "अगस्त", "सितंबर", "अक्तूबर", "नवंबर", "दिसंबर"],
+        months_with_day: None,
+        months_short: ["जन॰", "फ़र॰", "मार्च", "अप्रैल", "मई", "जून", "जुला॰", "अग॰", "सित॰", "अक्तू॰", "नव॰", "दिस॰"],
+        days: ["रविवार", "सोमवार", "मंगलवार", "बुधवार", "गुरुवार", "शुक्रवार", "शनिवार"],
+        days_short: ["रवि", "सोम", "मंगल", "बुध", "गुरु", "शुक्र", "शनि"],
+    },
+    // Korean: months are numbered (`3월`) and weekdays end in 요일; no names change after a day.
+    DateLanguage {
+        code: "ko",
+        name: "한국어",
+        months: ["1월", "2월", "3월", "4월", "5월", "6월", "7월", "8월", "9월", "10월", "11월", "12월"],
+        months_with_day: None,
+        months_short: ["1월", "2월", "3월", "4월", "5월", "6월", "7월", "8월", "9월", "10월", "11월", "12월"],
+        days: ["일요일", "월요일", "화요일", "수요일", "목요일", "금요일", "토요일"],
+        days_short: ["일", "월", "화", "수", "목", "금", "토"],
+    },
 ];

 /// The date language for `code` (any case), or `None` when there's none.
@@ -718,6 +749,35 @@ mod tests {
        assert_eq!(in_lang("dddd, mmmm", "cs"), "sobota, březen");
    }

+    /// Indonesian month and weekday names are the same with and after a day number. (Vietnamese
+    /// has its own test upstream: its numeric month forms are writable, weekday names are not.)
+    #[test]
+    fn indonesian_month_and_weekday_names() {
+        assert_eq!(in_lang("d mmmm yyyy", "id"), "7 Maret 2026");
+        assert_eq!(in_lang("mmmm yyyy", "id"), "Maret 2026");
+        assert_eq!(in_lang("dddd, d mmmm yyyy", "id"), "Sabtu, 7 Maret 2026");
+        // Indonesian month names are Latin, so they reach a PDF.
+        let mut s = Session::new().with_clock(|| 1_700_000_000);
+        s.set_date_language(Some("id")).unwrap();
+        assert_eq!(s.today_text_for_pdf(Some("d mmmm yyyy"), None).as_deref(), Ok("14 November 2023"));
+    }
+
+    /// Hindi and Korean names are the same with and after a day number. Neither is Western
+    /// European, so the date-text path refuses them with what to use instead.
+    #[test]
+    fn hindi_and_korean_month_and_weekday_names() {
+        assert_eq!(in_lang("d mmmm yyyy", "hi"), "7 मार्च 2026");
+        assert_eq!(in_lang("dddd, mmmm", "hi"), "शनिवार, मार्च");
+        assert_eq!(in_lang("ddd, d mmm", "hi"), "शनि, 7 मार्च");
+        assert_eq!(in_lang("yyyy\\년 mmmm d\\일 dddd", "ko"), "2026년 3월 7일 토요일");
+        assert_eq!(in_lang("ddd", "ko"), "토");
+        let mut s = Session::new().with_clock(|| 1_700_000_000);
+        for code in ["hi", "ko"] {
+            s.set_date_language(Some(code)).unwrap();
+            assert!(s.today_text_for_pdf(Some("d mmmm yyyy"), None).unwrap_err().contains("dd.mm.yyyy"), "{code}");
+        }
+    }
+
    #[test]
    fn the_session_writes_names_in_its_date_language() {
        let mut s = Session::new().with_clock(|| 1_700_000_000);
@@ -1,8 +1,8 @@
 //! Embeds the fonts of the optional craft-fonts build input (`CRAFT_FONTS_DIR`) as `CRAFT_FONTS`.
 //! The recipe is craft-fonts' `docs/integration.md`; unset, `CRAFT_FONTS` is empty. It only reads
 //! the local checkout (no network). On wasm32 it embeds only the UI faces (BIZ UDPGothic Regular
-//! and any `Arab`, `Telu`, `Hebr` or `Thai` face), to keep the web build within hosting limits (Cloudflare Pages:
-//! 25 MiB per file).
+//! and any `Arab`, `Telu`, `Hebr`, `Thai` or `Deva` face), to keep the web build within hosting
+//! limits (Cloudflare Pages: 25 MiB per file).

 use std::fmt::Write as _;
 use std::path::PathBuf;
@@ -55,8 +55,14 @@ fn craft_fonts(dir: &std::path::Path) -> Result<String, String> {
        let telugu = scripts.split(',').any(|s| s.trim() == "Telu");
        // And the Hebrew ones (small), for Hebrew text and Edit text on the web.
        let hebrew = scripts.split(',').any(|s| s.trim() == "Hebr");
-        let web_face =
-            (*family == "BIZ UDPGothic" && *style == "Regular") || arabic || telugu || hebrew || scripts.split(',').any(|s| s.trim() == "Thai");
+        // Devanagari too, for the Hindi interface (one small face).
+        let devanagari = scripts.split(',').any(|s| s.trim() == "Deva");
+        let web_face = (*family == "BIZ UDPGothic" && *style == "Regular")
+            || arabic
+            || telugu
+            || hebrew
+            || scripts.split(',').any(|s| s.trim() == "Thai")
+            || devanagari;
        if wasm && !web_face {
            continue;
        }
@@ -75,6 +75,20 @@ pub fn document_thai_font() -> Option<&'static CraftFont> {
    ui_thai_fonts().into_iter().next()
 }

+/// The `Deva` craft-fonts faces for Devanagari interface text (the Hindi catalog, file names,
+/// document titles), in manifest order. Empty when built without craft-fonts or when its revision
+/// has no Devanagari face; a desktop build then falls back to one font already installed on the
+/// machine (`SYSTEM_DEVANAGARI_FALLBACK`), and the web build shows replacement boxes.
+pub fn ui_devanagari_fonts() -> Vec<&'static CraftFont> {
+    CRAFT_FONTS.iter().filter(|f| f.covers("Deva")).collect()
+}
+
+/// The face for Devanagari text written into PDFs: the first `Deva` face. `None` when built
+/// without craft-fonts or when its revision has no Devanagari face.
+pub fn document_devanagari_font() -> Option<&'static CraftFont> {
+    ui_devanagari_fonts().into_iter().next()
+}
+
 /// The `Hebr` craft-fonts faces for Hebrew interface text (file names, document titles, text
 /// being edited), in manifest order. Empty when built without craft-fonts or when it has no
 /// Hebrew face.
@@ -18,9 +18,9 @@ pub mod source_shaping;
 mod std14;
 mod std14_metrics;
 pub use craft::{
-    CRAFT_FONTS, CraftFont, SHIPPORI_MINCHO, document_arabic_font, document_hebrew_font, document_japanese_font, document_japanese_font_for_style,
-    document_japanese_fonts_for_style, document_thai_font, ui_arabic_fonts, ui_chinese_fonts, ui_cjk_fonts, ui_hebrew_fonts, ui_japanese_fonts,
-    ui_telugu_fonts, ui_thai_fonts,
+    CRAFT_FONTS, CraftFont, SHIPPORI_MINCHO, document_arabic_font, document_devanagari_font, document_hebrew_font, document_japanese_font,
+    document_japanese_font_for_style, document_japanese_fonts_for_style, document_thai_font, ui_arabic_fonts, ui_chinese_fonts, ui_cjk_fonts,
+    ui_devanagari_fonts, ui_hebrew_fonts, ui_japanese_fonts, ui_telugu_fonts, ui_thai_fonts,
 };
 pub use script::{
    GlyphError, GlyphOutline, INTER_REGULAR, MAX_SIGNATURE_CHARS, ScriptOutline, japanese_glyph, japanese_glyph_from, script_missing, script_outline,
@@ -147,7 +147,7 @@ fn plural_sl(n: u64) -> usize {
 }

 /// The registry. English first: it is the fallback and the source language.
-pub static LANGUAGES: [LangInfo; 23] = [
+pub static LANGUAGES: [LangInfo; 26] = [
    LangInfo { code: "en", name: "English", source: "", plural: plural_one_other, rtl: false, catalog: OnceLock::new() },
    LangInfo { code: "ja", name: "日本語", source: include_str!("ja.tsv"), plural: plural_none, rtl: false, catalog: OnceLock::new() },
    // Simplified Chinese; `zh`, `zh-CN`, `zh-SG` and `zh-Hans-*` locales resolve here (see `candidates`).
@@ -207,9 +207,19 @@ pub static LANGUAGES: [LangInfo; 23] = [
    LangInfo { code: "tr", name: "Türkçe", source: include_str!("tr.tsv"), plural: plural_none, rtl: false, catalog: OnceLock::new() },
    // Lithuanian; every `lt-*` locale (`lt-LT`, `lt_LT.UTF-8`) resolves here. Three plural forms (one, few, other).
    LangInfo { code: "lt", name: "Lietuvių", source: include_str!("lt.tsv"), plural: plural_lithuanian, rtl: false, catalog: OnceLock::new() },
+    // Indonesian; every `id-*` locale (`id-ID`, `id_ID.UTF-8`) resolves here. CLDR `id` counts have
+    // one form ("other"), so `@plural` entries list a single form.
+    LangInfo { code: "id", name: "Bahasa Indonesia", source: include_str!("id.tsv"), plural: plural_none, rtl: false, catalog: OnceLock::new() },
    // Vietnamese; every `vi-*` locale (`vi-VN`, `vi_VN.UTF-8`) resolves here. CLDR `vi` counts have
    // one form ("other"), so `@plural` entries list a single form.
    LangInfo { code: "vi", name: "Tiếng Việt", source: include_str!("vi.tsv"), plural: plural_none, rtl: false, catalog: OnceLock::new() },
+    // Hindi; every `hi-*` locale (`hi-IN`, `hi_IN.UTF-8`) resolves here. CLDR `hi` has "one"
+    // (0 and 1) and "other", but the catalog's nouns are the same in both ("{n} पृष्ठ"), so
+    // `@plural` entries list a single form.
+    LangInfo { code: "hi", name: "हिन्दी", source: include_str!("hi.tsv"), plural: plural_none, rtl: false, catalog: OnceLock::new() },
+    // Korean; every `ko-*` locale (`ko-KR`, `ko_KR.UTF-8`) resolves here. CLDR `ko` counts have
+    // one form ("other"), so `@plural` entries list a single form.
+    LangInfo { code: "ko", name: "한국어", source: include_str!("ko.tsv"), plural: plural_none, rtl: false, catalog: OnceLock::new() },
 ];

 impl LangInfo {
@@ -2732,6 +2742,271 @@ mod tests {
        assert!(missing.is_empty(), "untranslated Slovene UI literals: {missing:#?}");
    }

+    #[test]
+    fn indonesian_and_vietnamese_are_registered_and_persist() {
+        let id = Lang::from_code("id").expect("id registered");
+        let vi = Lang::from_code("vi").expect("vi registered");
+        assert_eq!(id.name(), "Bahasa Indonesia");
+        assert_eq!(vi.name(), "Tiếng Việt");
+        assert_eq!(normalize_pref("ID"), Some("id"));
+        assert_eq!(normalize_pref("Vi"), Some("vi"));
+        for tag in ["id", "id-ID", "id_ID.UTF-8", "ID-id"] {
+            assert_eq!(lang_from_tag(tag), Some(id), "{tag}");
+        }
+        for tag in ["vi", "vi-VN", "vi_VN.UTF-8", "VI-vn"] {
+            assert_eq!(lang_from_tag(tag), Some(vi), "{tag}");
+        }
+        assert_eq!(first_supported("id-ID\r\nen-US"), Some(id));
+        assert_eq!(first_supported("vi-VN\r\nen-US"), Some(vi));
+        assert_eq!(tr(id, "Save"), "Simpan");
+        assert_eq!(tr(id, "Save as…"), "Simpan sebagai…");
+        assert_eq!(tr(id, "Bookmarks"), "Bookmark");
+        assert_eq!(tr(id, "Fill & Sign"), "Isi & Tanda Tangani");
+        assert_eq!(tr(vi, "Save"), "Lưu");
+        assert_eq!(tr(vi, "Save as…"), "Lưu thành…");
+        assert_eq!(tr(vi, "Bookmarks"), "Dấu trang");
+        assert_eq!(tr(vi, "Fill & Sign"), "Điền & Ký");
+        assert_eq!(tr(vi, "Interface language"), "Ngôn ngữ giao diện");
+        assert_eq!(tr(id, "Interface language"), "Bahasa antarmuka");
+        // One CLDR plural form (CLDR `id`, `vi`): the message reads the same for every count.
+        assert_eq!((0..=3).map(|n| (id.0.plural)(n)).collect::<Vec<_>>(), [0, 0, 0, 0]);
+        assert_eq!((0..=3).map(|n| (vi.0.plural)(n)).collect::<Vec<_>>(), [0, 0, 0, 0]);
+        for n in [0, 1, 2, 5] {
+            assert_eq!(trn(id, n, "{n} page", "{n} pages"), format!("{n} halaman"));
+            assert_eq!(trn(vi, n, "{n} page", "{n} pages"), format!("{n} trang"));
+            assert_eq!(trn(id, n, "{n} field", "{n} fields"), format!("{n} kolom"));
+            assert_eq!(trn(vi, n, "{n} field", "{n} fields"), format!("{n} trường"));
+        }
+        let mut app = crate::PdfCraftApp::default();
+        app.set_option("language", "ID").unwrap();
+        assert_eq!(app.language, "id");
+        let mut restored = crate::PdfCraftApp::default();
+        restored.restore(&app.persist());
+        assert_eq!(restored.language, "id");
+        let mut app = crate::PdfCraftApp::default();
+        app.set_option("language", "vi").unwrap();
+        assert_eq!(app.language, "vi");
+        let mut restored = crate::PdfCraftApp::default();
+        restored.restore(&app.persist());
+        assert_eq!(restored.language, "vi");
+        assert!(restored.set_option("language", "xx").is_err());
+        assert_eq!(restored.language, "vi");
+    }
+
+    /// Indonesian and Vietnamese translate every registered command and every All tools group,
+    /// section and item.
+    #[test]
+    fn indonesian_and_vietnamese_cover_commands_and_catalogue() {
+        for lang in [Lang::from_code("id").expect("id registered"), Lang::from_code("vi").expect("vi registered")] {
+            for command in pdfcraft_engine::commands::COMMANDS {
+                assert!(has(lang, command.label), "{} missing command: {}", lang.code(), command.label);
+                if let Some(menu) = command.menu {
+                    assert!(has(lang, menu), "{} missing menu: {menu}", lang.code());
+                }
+            }
+            for group in pdfcraft_engine::catalog::TOOL_GROUPS {
+                assert!(has(lang, group.label), "{} missing group: {}", lang.code(), group.label);
+                for section in group.sections {
+                    assert!(has(lang, section.title), "{} missing section: {}", lang.code(), section.title);
+                    for item in section.items {
+                        assert!(has(lang, item.label), "{} missing item: {}", lang.code(), item.label);
+                    }
+                }
+            }
+        }
+    }
+
+    /// Every `tl!("…")` literal in the UI source is translated in Indonesian and Vietnamese, so a
+    /// new label cannot silently fall back to English in either.
+    #[test]
+    fn indonesian_and_vietnamese_cover_ui_literals() {
+        let mut stack = vec![std::path::Path::new(env!("CARGO_MANIFEST_DIR")).join("src")];
+        let mut literals = std::collections::BTreeSet::new();
+        while let Some(dir) = stack.pop() {
+            for entry in std::fs::read_dir(dir).expect("UI source directory") {
+                let path = entry.expect("UI source entry").path();
+                if path.is_dir() {
+                    if path.file_name().is_some_and(|name| name != "i18n") {
+                        stack.push(path);
+                    }
+                } else if path.extension().is_some_and(|ext| ext == "rs") {
+                    let source = std::fs::read_to_string(path).expect("UI source file").replace("\r\n", "\n");
+                    let mut rest = source.split("#[cfg(test)]\nmod ").next().unwrap_or_default();
+                    while let Some((_, after)) = rest.split_once("tl!(\"") {
+                        let mut escaped = false;
+                        let end = after
+                            .char_indices()
+                            .find_map(|(i, c)| {
+                                if c == '"' && !escaped {
+                                    return Some(i);
+                                }
+                                escaped = c == '\\' && !escaped;
+                                None
+                            })
+                            .expect("closed tl! literal");
+                        let (raw, tail) = after.split_at(end);
+                        if tail.starts_with("\")") {
+                            let label: String = serde_json::from_str(&format!("\"{raw}\"")).expect("UI literal escapes");
+                            literals.insert(label);
+                        }
+                        rest = tail.strip_prefix('"').expect("closing quote");
+                    }
+                }
+            }
+        }
+        assert!(literals.len() > 900, "source scan found only {} literals", literals.len());
+        for lang in [Lang::from_code("id").expect("id registered"), Lang::from_code("vi").expect("vi registered")] {
+            let missing: Vec<_> = literals.iter().filter(|label| !has(lang, label)).collect();
+            assert!(missing.is_empty(), "untranslated {} UI literals: {missing:#?}", lang.code());
+        }
+    }
+
+    #[test]
+    fn hindi_and_korean_are_registered_and_persist() {
+        let hi = Lang::from_code("hi").expect("hi registered");
+        let ko = Lang::from_code("ko").expect("ko registered");
+        assert_eq!(hi.name(), "हिन्दी");
+        assert_eq!(ko.name(), "한국어");
+        assert_eq!(normalize_pref("HI"), Some("hi"));
+        assert_eq!(normalize_pref("Ko"), Some("ko"));
+        for tag in ["hi", "hi-IN", "hi_IN.UTF-8", "HI-in"] {
+            assert_eq!(lang_from_tag(tag), Some(hi), "{tag}");
+        }
+        for tag in ["ko", "ko-KR", "ko_KR.UTF-8", "KO-kr"] {
+            assert_eq!(lang_from_tag(tag), Some(ko), "{tag}");
+        }
+        assert_eq!(first_supported("hi-IN\r\nen-US"), Some(hi));
+        assert_eq!(first_supported("ko-KR\r\nen-US"), Some(ko));
+        assert_eq!(tr(hi, "Save"), "सहेजें");
+        assert_eq!(tr(hi, "Save as…"), "इस रूप में सहेजें…");
+        assert_eq!(tr(hi, "Bookmarks"), "बुकमार्क");
+        assert_eq!(tr(hi, "Home"), "होम");
+        assert_eq!(tr(hi, "Interface language"), "इंटरफ़ेस भाषा");
+        assert_eq!(tr(ko, "Save"), "저장");
+        assert_eq!(tr(ko, "Save as…"), "다른 이름으로 저장…");
+        assert_eq!(tr(ko, "Bookmarks"), "책갈피");
+        assert_eq!(tr(ko, "Home"), "홈");
+        assert_eq!(tr(ko, "Interface language"), "인터페이스 언어");
+        // One plural form per catalog: the message reads the same for every count.
+        assert_eq!((0..=3).map(|n| (hi.0.plural)(n)).collect::<Vec<_>>(), [0, 0, 0, 0]);
+        assert_eq!((0..=3).map(|n| (ko.0.plural)(n)).collect::<Vec<_>>(), [0, 0, 0, 0]);
+        for n in [0, 1, 2, 5] {
+            assert_eq!(trn(hi, n, "{n} page", "{n} pages"), format!("{n} पृष्ठ"));
+            assert_eq!(trn(ko, n, "{n} page", "{n} pages"), format!("{n}페이지"));
+            assert_eq!(trn(ko, n, "{n} field", "{n} fields"), format!("필드 {n}개"));
+        }
+        // The broken-line regression: a `\n` escape stays one catalog line.
+        assert_eq!(tr(ko, "This page couldn't be displayed.\n{e}"), "이 페이지를 표시할 수 없습니다.\n{e}");
+        for code in ["hi", "ko"] {
+            let mut app = crate::PdfCraftApp::default();
+            app.set_option("language", &code.to_uppercase()).unwrap();
+            assert_eq!(app.language, code);
+            let mut restored = crate::PdfCraftApp::default();
+            restored.restore(&app.persist());
+            assert_eq!(restored.language, code);
+            assert!(restored.set_option("language", "xx").is_err());
+            assert_eq!(restored.language, code);
+        }
+    }
+
+    /// Redaction first marks a region; applying it irreversibly removes marked content.
+    /// Korean must not present either stage as censorship ("검열") or imply that
+    /// clearing un-applied marks deletes PDF content.
+    #[test]
+    fn korean_redaction_labels_distinguish_marks_and_permanent_removal() {
+        let ko = Lang::from_code("ko").expect("ko registered");
+        for english in [
+            "Redact",
+            "Redact a PDF",
+            "Redact text and images",
+            "Redact pages…",
+            "Mark for redaction",
+            "Apply redactions",
+            "Apply redactions…",
+            "Clear redaction marks",
+            "Redact all",
+        ] {
+            assert!(!tr(ko, english).contains("검열"), "Korean label misstates redaction: {english}");
+        }
+        assert_eq!(tr(ko, "Mark for redaction"), "제거할 영역 표시");
+        assert_eq!(tr(ko, "Apply redactions"), "표시된 정보 영구 제거");
+        assert_eq!(tr(ko, "Clear redaction marks"), "제거 대상 표시 지우기");
+        let warning = concat!(
+            "You are about to apply 1 redaction mark. ",
+            "Text, images and drawings under the marks, and comments ",
+            "and form fields that overlap them, are removed permanently."
+        );
+        assert!(tr(ko, warning).contains("영구적으로 제거됩니다"));
+    }
+
+    /// Hindi and Korean translate every registered command and every All tools group, section
+    /// and item.
+    #[test]
+    fn hindi_and_korean_cover_commands_and_catalogue() {
+        for lang in [Lang::from_code("hi").expect("hi registered"), Lang::from_code("ko").expect("ko registered")] {
+            for command in pdfcraft_engine::commands::COMMANDS {
+                assert!(has(lang, command.label), "{} missing command: {}", lang.code(), command.label);
+                if let Some(menu) = command.menu {
+                    assert!(has(lang, menu), "{} missing menu: {menu}", lang.code());
+                }
+            }
+            for group in pdfcraft_engine::catalog::TOOL_GROUPS {
+                assert!(has(lang, group.label), "{} missing group: {}", lang.code(), group.label);
+                for section in group.sections {
+                    assert!(has(lang, section.title), "{} missing section: {}", lang.code(), section.title);
+                    for item in section.items {
+                        assert!(has(lang, item.label), "{} missing item: {}", lang.code(), item.label);
+                    }
+                }
+            }
+        }
+    }
+
+    /// Every `tl!("…")` literal in the UI source is translated in Hindi and Korean, so a new
+    /// label cannot silently fall back to English in either.
+    #[test]
+    fn hindi_and_korean_cover_ui_literals() {
+        let mut stack = vec![std::path::Path::new(env!("CARGO_MANIFEST_DIR")).join("src")];
+        let mut literals = std::collections::BTreeSet::new();
+        while let Some(dir) = stack.pop() {
+            for entry in std::fs::read_dir(dir).expect("UI source directory") {
+                let path = entry.expect("UI source entry").path();
+                if path.is_dir() {
+                    if path.file_name().is_some_and(|name| name != "i18n") {
+                        stack.push(path);
+                    }
+                } else if path.extension().is_some_and(|ext| ext == "rs") {
+                    let source = std::fs::read_to_string(path).expect("UI source file").replace("\r\n", "\n");
+                    let mut rest = source.split("#[cfg(test)]\nmod ").next().unwrap_or_default();
+                    while let Some((_, after)) = rest.split_once("tl!(\"") {
+                        let mut escaped = false;
+                        let end = after
+                            .char_indices()
+                            .find_map(|(i, c)| {
+                                if c == '"' && !escaped {
+                                    return Some(i);
+                                }
+                                escaped = c == '\\' && !escaped;
+                                None
+                            })
+                            .expect("closed tl! literal");
+                        let (raw, tail) = after.split_at(end);
+                        if tail.starts_with("\")") {
+                            let label: String = serde_json::from_str(&format!("\"{raw}\"")).expect("UI literal escapes");
+                            literals.insert(label);
+                        }
+                        rest = tail.strip_prefix('"').expect("closing quote");
+                    }
+                }
+            }
+        }
+        assert!(literals.len() > 900, "source scan found only {} literals", literals.len());
+        for lang in [Lang::from_code("hi").expect("hi registered"), Lang::from_code("ko").expect("ko registered")] {
+            let missing: Vec<_> = literals.iter().filter(|label| !has(lang, label)).collect();
+            assert!(missing.is_empty(), "untranslated {} UI literals: {missing:#?}", lang.code());
+        }
+    }
+
    #[test]
    fn bulgarian_is_registered() {
        let bg = Lang::from_code("bg").expect("bg registered");
@@ -440,6 +440,7 @@ mod tests {
        assert!(script_fallback(&candidates, 'A').is_some());
        assert!(script_fallback(&candidates, TELUGU_PROBE).is_none());
        assert!(script_fallback(&candidates, DEVANAGARI_PROBE).is_none());
+        assert!(script_fallback(&candidates, HANGUL_PROBE).is_none());
        assert!(script_fallback(&candidates, SINHALA_PROBE).is_none());
        assert!(script_fallback(&candidates, TAMIL_PROBE).is_none());
    }
@@ -453,6 +454,20 @@ mod tests {
        }
    }

+    #[test]
+    fn hangul_candidates_are_font_files_without_adobe_designs() {
+        let list = hangul_candidates();
+        assert!(!list.is_empty());
+        for path in list {
+            assert!(path.extension().is_some_and(|e| e == "ttf" || e == "ttc"), "{path:?}");
+            let name = path.to_string_lossy().to_lowercase();
+            assert!(!["notosanscjk", "notoserifcjk", "sourcehan", "source-han"].iter().any(|b| name.contains(b)), "{name}");
+        }
+        // Inter has no Hangul.
+        let inter = include_bytes!("../../../assets/fonts/Inter-Regular.ttf");
+        assert_eq!(face_with(inter, HANGUL_PROBE), None);
+    }
+
    #[test]
    fn han_candidates_skip_adobe_designs() {
        for need in [Need::Hans, Need::Hant] {
@@ -385,13 +385,14 @@ fn font_definitions_for_platform(prefer_hans: bool, platform: FontPolicyPlatform
        }
    }
    // Arabic-script faces (file names, document titles), Telugu faces (the Telugu interface,
-    // file names), Hebrew faces (file names, text being edited) and Thai faces after the CJK ones; the
-    // ranges don't overlap.
+    // file names), Hebrew faces (file names, text being edited), Thai faces and Devanagari faces
+    // (the Hindi interface) after the CJK ones; the ranges don't overlap.
    for face in pdfcraft_fonts::ui_arabic_fonts()
        .into_iter()
        .chain(pdfcraft_fonts::ui_telugu_fonts())
        .chain(pdfcraft_fonts::ui_hebrew_fonts())
        .chain(pdfcraft_fonts::ui_thai_fonts())
+        .chain(pdfcraft_fonts::ui_devanagari_fonts())
    {
        let name = face.name();
        if !fonts.font_data.contains_key(&name) {
@@ -8,6 +8,7 @@ const JAPANESE: &str = "日本語の文字";
 const CHINESE: &str = "简体中文欢迎";
 const ARABIC: &str = "واحد اثنين";
 const TELUGU: &str = "తెలుగు ఫైల్";
+const DEVANAGARI: &str = "हिंदी पाठ सम्पादन";

 fn families() -> Vec<FontId> {
    vec![FontId::proportional(13.0), FontId::monospace(13.0), theme::medium(13.0), theme::semibold(17.0)]
@@ -163,7 +164,31 @@ fn system_fallback_covers_devanagari() {
        assert_eq!(stack.iter().filter(|n| *n == theme::SYSTEM_DEVANAGARI_FALLBACK).count(), 1, "{family:?}");
    }
    let mut fonts = Fonts::new(TextOptions::default(), defs);
-    const DEVANAGARI: &str = "हिंदी पाठ सम्पादन";
+    for id in families() {
+        assert!(fonts.has_glyphs(&id, DEVANAGARI), "{id:?} lacks {DEVANAGARI}");
+    }
+    assert!(layout_widths(&mut fonts, DEVANAGARI).iter().all(|w| w.is_finite() && *w > 0.0));
+}
+
+/// Built with craft-fonts, Hindi text renders with real glyphs (no tofu) in every family: a
+/// `Deva` face is a fallback after the app's own fonts. Without craft-fonts it is the installed
+/// Devanagari fallback's job (see `system_fallback_covers_devanagari`), and the web build has
+/// neither.
+#[test]
+fn devanagari_ui_text_uses_craft_fonts() {
+    if pdfcraft_fonts::ui_devanagari_fonts().is_empty() {
+        eprintln!("skipping devanagari_ui_text_uses_craft_fonts: built without craft-fonts (set CRAFT_FONTS_DIR to run it)");
+        return;
+    }
+    let defs = theme::font_definitions();
+    for family in [FontFamily::Proportional, FontFamily::Monospace] {
+        let stack = &defs.families[&family];
+        let first =
+            stack.iter().position(|n| pdfcraft_fonts::ui_devanagari_fonts().iter().any(|f| f.name() == *n)).expect("a Deva face is a fallback");
+        let own = stack.iter().position(|n| n == "Inter" || n == "JetBrainsMono").expect("the app's own font");
+        assert!(own < first, "{family:?}: {stack:?}");
+    }
+    let mut fonts = Fonts::new(TextOptions::default(), defs);
    for id in families() {
        assert!(fonts.has_glyphs(&id, DEVANAGARI), "{id:?} lacks {DEVANAGARI}");
    }
@@ -290,7 +315,7 @@ fn ui_fonts_work_without_craft_fonts() {
    theme::install_fonts(&ctx);
 }

-/// Latin- and Cyrillic-script catalogs (Czech, Brazilian Portuguese, German, Spanish, French, Russian, Bulgarian, Hungarian, Ukrainian, Italian, Slovene) are drawn entirely by the app's own faces
+/// Latin- and Cyrillic-script catalogs (Czech, Brazilian Portuguese, German, Spanish, French, Russian, Bulgarian, Kazakh, Hungarian, Ukrainian, Italian, Slovene, Lithuanian, Greek, Turkish, Indonesian, Vietnamese) are drawn entirely by the app's own faces
 /// (Inter, JetBrains Mono with Inter fallback): no letter falls through to egui's defaults or a CJK fallback. (egui's
 /// `has_glyphs` can't answer this: with only the primary face it is also the replacement face.)
 #[test]
@@ -317,9 +342,14 @@ fn primary_ui_fonts_cover_latin_and_cyrillic_catalogs() {
        ("tr", include_str!("../src/i18n/tr.tsv")),
        ("sl", include_str!("../src/i18n/sl.tsv")),
        ("lt", include_str!("../src/i18n/lt.tsv")),
+        ("id", include_str!("../src/i18n/id.tsv")),
        ("vi", include_str!("../src/i18n/vi.tsv")),
    ] {
-        let mut text = String::from("áčďéěíňóřšťúůýžÁČĎÉĚÍŇÓŘŠŤÚŮÝŽãõçâêôàÃÕÇÂÊÔÀñÑüÜäÄöÖőűß„“¿¡«»…ґҐєЄіІїЇәғқңөұүһӘҒҚҢӨҰҮҺ");
+        // Vietnamese stacks its diacritics: the precomposed letters and the base letters they
+        // combine with, including `đ` and the tone marks `ạ ả ấ ầ ẩ ẫ ậ ắ ằ ẳ ẵ ặ ế ề ể ễ ệ`.
+        let mut text = String::from(
+            "áčďéěíňóřšťúůýžÁČĎÉĚÍŇÓŘŠŤÚŮÝŽãõçâêôàÃÕÇÂÊÔÀñÑüÜäÄöÖőűß„“¿¡«»…ґҐєЄіІїЇәғқңөұүһӘҒҚҢӨҰҮҺđĐạảấầẩẫậắằẳẵặếềểễệọỏốồổỗộớờởỡợụủứừửữựỳỵỷỹ",
+        );
        for line in catalog.lines().filter(|l| !l.starts_with('#')) {
            if let Some(translation) = line.split('\t').nth(2) {
                text.extend(translation.chars().filter(|c| !c.is_whitespace()));