skrifa/
string.rs

1//! Localized strings describing font names and other metadata.
2//!
3//! This provides higher level interfaces for accessing the data in the
4//! OpenType [name](https://learn.microsoft.com/en-us/typography/opentype/spec/name)
5//! table.
6//!
7//! # Example
8//! The following function will print all localized strings from the set
9//! of predefined identifiers in a font:
10//! ```
11//! use skrifa::{string::StringId, MetadataProvider};
12//!
13//! fn print_well_known_strings<'a>(font: &impl MetadataProvider<'a>) {
14//!     for id in StringId::predefined() {
15//!         let strings = font.localized_strings(id);
16//!         if strings.clone().next().is_some() {
17//!             println!("[{:?}]", id);
18//!             for string in font.localized_strings(id) {
19//!                 println!("{:?} {}", string.language(), string.to_string());
20//!             }
21//!         }
22//!     }
23//! }
24//! ```
25
26use read_fonts::{
27    tables::name::{CharIter, Name, NameRecord, NameString},
28    FontRef, TableProvider,
29};
30
31use core::fmt;
32
33#[doc(inline)]
34pub use read_fonts::types::NameId as StringId;
35
36/// Iterator over the characters of a string.
37#[derive(Clone)]
38pub struct Chars<'a> {
39    inner: Option<CharIter<'a>>,
40}
41
42impl Iterator for Chars<'_> {
43    type Item = char;
44
45    fn next(&mut self) -> Option<Self::Item> {
46        self.inner.as_mut()?.next()
47    }
48}
49
50/// Iterator over a collection of localized strings for a specific identifier.
51#[derive(Clone)]
52pub struct LocalizedStrings<'a> {
53    name: Option<Name<'a>>,
54    records: core::slice::Iter<'a, NameRecord>,
55    id: StringId,
56}
57
58impl<'a> LocalizedStrings<'a> {
59    /// Creates a new localized string iterator from the given font and string identifier.
60    pub fn new(font: &FontRef<'a>, id: StringId) -> Self {
61        let name = font.name().ok();
62        let records = name
63            .as_ref()
64            .map(|name| name.name_record().iter())
65            .unwrap_or([].iter());
66        Self { name, records, id }
67    }
68
69    /// Returns the informational string identifier for this iterator.
70    pub fn id(&self) -> StringId {
71        self.id
72    }
73
74    /// Returns the best available English string or the first string in the sequence.
75    ///
76    /// This prefers the following languages, in order: "en-US", "en",
77    /// "" (empty, for bare Unicode platform strings which don't have an associated
78    /// language).
79    ///
80    /// If none of these are found, returns the first string, or `None` if the sequence
81    /// is empty.
82    pub fn english_or_first(self) -> Option<LocalizedString<'a>> {
83        let mut best_rank = -1;
84        let mut best_string = None;
85        for (i, string) in self.enumerate() {
86            let rank = match (i, string.language()) {
87                (_, Some("en-US")) => return Some(string),
88                (_, Some("en")) => 2,
89                (_, None) => 1,
90                (0, _) => 0,
91                _ => continue,
92            };
93            if rank > best_rank {
94                best_rank = rank;
95                best_string = Some(string);
96            }
97        }
98        best_string
99    }
100}
101
102impl<'a> Iterator for LocalizedStrings<'a> {
103    type Item = LocalizedString<'a>;
104
105    fn next(&mut self) -> Option<Self::Item> {
106        let name = self.name.as_ref()?;
107        loop {
108            let record = self.records.next()?;
109            if record.name_id() == self.id {
110                return Some(LocalizedString::new(name, record));
111            }
112        }
113    }
114}
115
116impl Default for LocalizedStrings<'_> {
117    fn default() -> Self {
118        Self {
119            name: None,
120            records: [].iter(),
121            id: StringId::default(),
122        }
123    }
124}
125
126/// String containing a name or other font metadata in a specific language.
127#[derive(Clone, Debug)]
128pub struct LocalizedString<'a> {
129    language: Option<Language>,
130    value: Option<NameString<'a>>,
131}
132
133impl<'a> LocalizedString<'a> {
134    pub fn new(name: &Name<'a>, record: &NameRecord) -> Self {
135        let language = Language::new(name, record);
136        let value = record.string(name.string_data()).ok();
137        Self { language, value }
138    }
139
140    /// Returns the BCP-47 language identifier for the localized string.
141    pub fn language(&self) -> Option<&str> {
142        self.language.as_ref().map(|language| language.as_str())
143    }
144
145    /// Returns an iterator over the characters of the localized string.
146    pub fn chars(&self) -> Chars<'a> {
147        Chars {
148            inner: self.value.map(|value| value.chars()),
149        }
150    }
151}
152
153impl fmt::Display for LocalizedString<'_> {
154    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
155        for ch in self.chars() {
156            ch.fmt(f)?;
157        }
158        Ok(())
159    }
160}
161
162/// This value is chosen arbitrarily to accommodate common language tags that
163/// are almost always <= 11 bytes (LLL-SSSS-RR where L is primary language, S
164/// is script and R is region) and to keep the Language enum at a reasonable
165/// 32 bytes in size.
166const MAX_INLINE_LANGUAGE_LEN: usize = 30;
167
168#[derive(Copy, Clone, Debug)]
169#[repr(u8)]
170enum Language {
171    Inline {
172        buf: [u8; MAX_INLINE_LANGUAGE_LEN],
173        len: u8,
174    },
175    Static(&'static str),
176}
177
178impl Language {
179    fn new(name: &Name, record: &NameRecord) -> Option<Self> {
180        let language_id = record.language_id();
181        // For version 1 name tables, prefer language tags:
182        // https://learn.microsoft.com/en-us/typography/opentype/spec/name#naming-table-version-1
183        const BASE_LANGUAGE_TAG_ID: u16 = 0x8000;
184        if name.version() == 1 && language_id >= BASE_LANGUAGE_TAG_ID {
185            let index = (language_id - BASE_LANGUAGE_TAG_ID) as usize;
186            let language_string = name
187                .lang_tag_record()?
188                .get(index)?
189                .lang_tag(name.string_data())
190                .ok()?;
191            Self::from_name_string(&language_string)
192        } else {
193            match record.platform_id() {
194                // We only match Macintosh and Windows language ids.
195                1 | 3 => Self::from_language_id(language_id),
196                _ => None,
197            }
198        }
199    }
200
201    /// Decodes a language tag string into an inline ASCII byte sequence.
202    fn from_name_string(s: &NameString) -> Option<Self> {
203        let mut buf = [0u8; MAX_INLINE_LANGUAGE_LEN];
204        let mut len = 0;
205        for ch in s.chars() {
206            // From "Tags for Identifying Languages" <https://www.rfc-editor.org/rfc/rfc5646.html#page-6>:
207            // "Although [RFC5234] refers to octets, the language tags described in
208            // this document are sequences of characters from the US-ASCII [ISO646]
209            // repertoire"
210            // Therefore we assume that non-ASCII characters signal an invalid language tag.
211            if !ch.is_ascii() || len == MAX_INLINE_LANGUAGE_LEN {
212                return None;
213            }
214            buf[len] = ch as u8;
215            len += 1;
216        }
217        Some(Self::Inline {
218            buf,
219            len: len as u8,
220        })
221    }
222
223    fn from_language_id(language_id: u16) -> Option<Self> {
224        Some(Self::Static(language_id_to_bcp47(language_id)?))
225    }
226
227    fn as_str(&self) -> &str {
228        match self {
229            Self::Inline { buf: data, len } => {
230                let data = &data[..*len as usize];
231                core::str::from_utf8(data).unwrap_or_default()
232            }
233            Self::Static(str) => str,
234        }
235    }
236}
237
238/// Converts an OpenType language identifier to a BCP-47 language tag.
239fn language_id_to_bcp47(language_id: u16) -> Option<&'static str> {
240    match LANGUAGE_ID_TO_BCP47.binary_search_by(|entry| entry.0.cmp(&language_id)) {
241        Ok(ix) => LANGUAGE_ID_TO_BCP47.get(ix).map(|entry| entry.1),
242        _ => None,
243    }
244}
245
246/// Mapping of OpenType name table language identifier to BCP-47 language tag.
247/// Borrowed from Skia: <https://skia.googlesource.com/skia/+/refs/heads/main/src/sfnt/SkOTTable_name.cpp#98>
248const LANGUAGE_ID_TO_BCP47: &[(u16, &str)] = &[
249    /* A mapping from Mac Language Designators to BCP 47 codes.
250     *  The following list was constructed more or less manually.
251     *  Apple now uses BCP 47 (post OSX10.4), so there will be no new entries.
252     */
253    (0, "en"),        //English
254    (1, "fr"),        //French
255    (2, "de"),        //German
256    (3, "it"),        //Italian
257    (4, "nl"),        //Dutch
258    (5, "sv"),        //Swedish
259    (6, "es"),        //Spanish
260    (7, "da"),        //Danish
261    (8, "pt"),        //Portuguese
262    (9, "nb"),        //Norwegian
263    (10, "he"),       //Hebrew
264    (11, "ja"),       //Japanese
265    (12, "ar"),       //Arabic
266    (13, "fi"),       //Finnish
267    (14, "el"),       //Greek
268    (15, "is"),       //Icelandic
269    (16, "mt"),       //Maltese
270    (17, "tr"),       //Turkish
271    (18, "hr"),       //Croatian
272    (19, "zh-Hant"),  //Chinese (Traditional)
273    (20, "ur"),       //Urdu
274    (21, "hi"),       //Hindi
275    (22, "th"),       //Thai
276    (23, "ko"),       //Korean
277    (24, "lt"),       //Lithuanian
278    (25, "pl"),       //Polish
279    (26, "hu"),       //Hungarian
280    (27, "et"),       //Estonian
281    (28, "lv"),       //Latvian
282    (29, "se"),       //Sami
283    (30, "fo"),       //Faroese
284    (31, "fa"),       //Farsi (Persian)
285    (32, "ru"),       //Russian
286    (33, "zh-Hans"),  //Chinese (Simplified)
287    (34, "nl"),       //Dutch
288    (35, "ga"),       //Irish(Gaelic)
289    (36, "sq"),       //Albanian
290    (37, "ro"),       //Romanian
291    (38, "cs"),       //Czech
292    (39, "sk"),       //Slovak
293    (40, "sl"),       //Slovenian
294    (41, "yi"),       //Yiddish
295    (42, "sr"),       //Serbian
296    (43, "mk"),       //Macedonian
297    (44, "bg"),       //Bulgarian
298    (45, "uk"),       //Ukrainian
299    (46, "be"),       //Byelorussian
300    (47, "uz"),       //Uzbek
301    (48, "kk"),       //Kazakh
302    (49, "az-Cyrl"),  //Azerbaijani (Cyrillic)
303    (50, "az-Arab"),  //Azerbaijani (Arabic)
304    (51, "hy"),       //Armenian
305    (52, "ka"),       //Georgian
306    (53, "mo"),       //Moldavian
307    (54, "ky"),       //Kirghiz
308    (55, "tg"),       //Tajiki
309    (56, "tk"),       //Turkmen
310    (57, "mn-Mong"),  //Mongolian (Traditional)
311    (58, "mn-Cyrl"),  //Mongolian (Cyrillic)
312    (59, "ps"),       //Pashto
313    (60, "ku"),       //Kurdish
314    (61, "ks"),       //Kashmiri
315    (62, "sd"),       //Sindhi
316    (63, "bo"),       //Tibetan
317    (64, "ne"),       //Nepali
318    (65, "sa"),       //Sanskrit
319    (66, "mr"),       //Marathi
320    (67, "bn"),       //Bengali
321    (68, "as"),       //Assamese
322    (69, "gu"),       //Gujarati
323    (70, "pa"),       //Punjabi
324    (71, "or"),       //Oriya
325    (72, "ml"),       //Malayalam
326    (73, "kn"),       //Kannada
327    (74, "ta"),       //Tamil
328    (75, "te"),       //Telugu
329    (76, "si"),       //Sinhalese
330    (77, "my"),       //Burmese
331    (78, "km"),       //Khmer
332    (79, "lo"),       //Lao
333    (80, "vi"),       //Vietnamese
334    (81, "id"),       //Indonesian
335    (82, "tl"),       //Tagalog
336    (83, "ms-Latn"),  //Malay (Roman)
337    (84, "ms-Arab"),  //Malay (Arabic)
338    (85, "am"),       //Amharic
339    (86, "ti"),       //Tigrinya
340    (87, "om"),       //Oromo
341    (88, "so"),       //Somali
342    (89, "sw"),       //Swahili
343    (90, "rw"),       //Kinyarwanda/Ruanda
344    (91, "rn"),       //Rundi
345    (92, "ny"),       //Nyanja/Chewa
346    (93, "mg"),       //Malagasy
347    (94, "eo"),       //Esperanto
348    (128, "cy"),      //Welsh
349    (129, "eu"),      //Basque
350    (130, "ca"),      //Catalan
351    (131, "la"),      //Latin
352    (132, "qu"),      //Quechua
353    (133, "gn"),      //Guarani
354    (134, "ay"),      //Aymara
355    (135, "tt"),      //Tatar
356    (136, "ug"),      //Uighur
357    (137, "dz"),      //Dzongkha
358    (138, "jv-Latn"), //Javanese (Roman)
359    (139, "su-Latn"), //Sundanese (Roman)
360    (140, "gl"),      //Galician
361    (141, "af"),      //Afrikaans
362    (142, "br"),      //Breton
363    (143, "iu"),      //Inuktitut
364    (144, "gd"),      //Scottish (Gaelic)
365    (145, "gv"),      //Manx (Gaelic)
366    (146, "ga"),      //Irish (Gaelic with Lenition)
367    (147, "to"),      //Tongan
368    (148, "el"),      //Greek (Polytonic) Note: ISO 15924 does not have an equivalent script name.
369    (149, "kl"),      //Greenlandic
370    (150, "az-Latn"), //Azerbaijani (Roman)
371    (151, "nn"),      //Nynorsk
372    /* A mapping from Windows LCID to BCP 47 codes.
373     *  This list is the sorted, curated output of tools/win_lcid.cpp.
374     *  Note that these are sorted by value for quick binary lookup, and not logically by lsb.
375     *  The 'bare' language ids (e.g. 0x0001 for Arabic) are omitted
376     *  as they do not appear as valid language ids in the OpenType specification.
377     */
378    (0x0401, "ar-SA"),        //Arabic
379    (0x0402, "bg-BG"),        //Bulgarian
380    (0x0403, "ca-ES"),        //Catalan
381    (0x0404, "zh-TW"),        //Chinese (Traditional)
382    (0x0405, "cs-CZ"),        //Czech
383    (0x0406, "da-DK"),        //Danish
384    (0x0407, "de-DE"),        //German
385    (0x0408, "el-GR"),        //Greek
386    (0x0409, "en-US"),        //English
387    (0x040a, "es-ES_tradnl"), //Spanish
388    (0x040b, "fi-FI"),        //Finnish
389    (0x040c, "fr-FR"),        //French
390    (0x040d, "he-IL"),        //Hebrew
391    (0x040d, "he"),           //Hebrew
392    (0x040e, "hu-HU"),        //Hungarian
393    (0x040e, "hu"),           //Hungarian
394    (0x040f, "is-IS"),        //Icelandic
395    (0x0410, "it-IT"),        //Italian
396    (0x0411, "ja-JP"),        //Japanese
397    (0x0412, "ko-KR"),        //Korean
398    (0x0413, "nl-NL"),        //Dutch
399    (0x0414, "nb-NO"),        //Norwegian (Bokmål)
400    (0x0415, "pl-PL"),        //Polish
401    (0x0416, "pt-BR"),        //Portuguese
402    (0x0417, "rm-CH"),        //Romansh
403    (0x0418, "ro-RO"),        //Romanian
404    (0x0419, "ru-RU"),        //Russian
405    (0x041a, "hr-HR"),        //Croatian
406    (0x041b, "sk-SK"),        //Slovak
407    (0x041c, "sq-AL"),        //Albanian
408    (0x041d, "sv-SE"),        //Swedish
409    (0x041e, "th-TH"),        //Thai
410    (0x041f, "tr-TR"),        //Turkish
411    (0x0420, "ur-PK"),        //Urdu
412    (0x0421, "id-ID"),        //Indonesian
413    (0x0422, "uk-UA"),        //Ukrainian
414    (0x0423, "be-BY"),        //Belarusian
415    (0x0424, "sl-SI"),        //Slovenian
416    (0x0425, "et-EE"),        //Estonian
417    (0x0426, "lv-LV"),        //Latvian
418    (0x0427, "lt-LT"),        //Lithuanian
419    (0x0428, "tg-Cyrl-TJ"),   //Tajik (Cyrillic)
420    (0x0429, "fa-IR"),        //Persian
421    (0x042a, "vi-VN"),        //Vietnamese
422    (0x042b, "hy-AM"),        //Armenian
423    (0x042c, "az-Latn-AZ"),   //Azeri (Latin)
424    (0x042d, "eu-ES"),        //Basque
425    (0x042e, "hsb-DE"),       //Upper Sorbian
426    (0x042f, "mk-MK"),        //Macedonian (FYROM)
427    (0x0432, "tn-ZA"),        //Setswana
428    (0x0434, "xh-ZA"),        //isiXhosa
429    (0x0435, "zu-ZA"),        //isiZulu
430    (0x0436, "af-ZA"),        //Afrikaans
431    (0x0437, "ka-GE"),        //Georgian
432    (0x0438, "fo-FO"),        //Faroese
433    (0x0439, "hi-IN"),        //Hindi
434    (0x043a, "mt-MT"),        //Maltese
435    (0x043b, "se-NO"),        //Sami (Northern)
436    (0x043e, "ms-MY"),        //Malay
437    (0x043f, "kk-KZ"),        //Kazakh
438    (0x0440, "ky-KG"),        //Kyrgyz
439    (0x0441, "sw-KE"),        //Kiswahili
440    (0x0442, "tk-TM"),        //Turkmen
441    (0x0443, "uz-Latn-UZ"),   //Uzbek (Latin)
442    (0x0443, "uz"),           //Uzbek
443    (0x0444, "tt-RU"),        //Tatar
444    (0x0445, "bn-IN"),        //Bengali
445    (0x0446, "pa-IN"),        //Punjabi
446    (0x0447, "gu-IN"),        //Gujarati
447    (0x0448, "or-IN"),        //Oriya
448    (0x0449, "ta-IN"),        //Tamil
449    (0x044a, "te-IN"),        //Telugu
450    (0x044b, "kn-IN"),        //Kannada
451    (0x044c, "ml-IN"),        //Malayalam
452    (0x044d, "as-IN"),        //Assamese
453    (0x044e, "mr-IN"),        //Marathi
454    (0x044f, "sa-IN"),        //Sanskrit
455    (0x0450, "mn-Cyrl"),      //Mongolian (Cyrillic)
456    (0x0451, "bo-CN"),        //Tibetan
457    (0x0452, "cy-GB"),        //Welsh
458    (0x0453, "km-KH"),        //Khmer
459    (0x0454, "lo-LA"),        //Lao
460    (0x0456, "gl-ES"),        //Galician
461    (0x0457, "kok-IN"),       //Konkani
462    (0x045a, "syr-SY"),       //Syriac
463    (0x045b, "si-LK"),        //Sinhala
464    (0x045d, "iu-Cans-CA"),   //Inuktitut (Syllabics)
465    (0x045e, "am-ET"),        //Amharic
466    (0x0461, "ne-NP"),        //Nepali
467    (0x0462, "fy-NL"),        //Frisian
468    (0x0463, "ps-AF"),        //Pashto
469    (0x0464, "fil-PH"),       //Filipino
470    (0x0465, "dv-MV"),        //Divehi
471    (0x0468, "ha-Latn-NG"),   //Hausa (Latin)
472    (0x046a, "yo-NG"),        //Yoruba
473    (0x046b, "quz-BO"),       //Quechua
474    (0x046c, "nso-ZA"),       //Sesotho sa Leboa
475    (0x046d, "ba-RU"),        //Bashkir
476    (0x046e, "lb-LU"),        //Luxembourgish
477    (0x046f, "kl-GL"),        //Greenlandic
478    (0x0470, "ig-NG"),        //Igbo
479    (0x0478, "ii-CN"),        //Yi
480    (0x047a, "arn-CL"),       //Mapudungun
481    (0x047c, "moh-CA"),       //Mohawk
482    (0x047e, "br-FR"),        //Breton
483    (0x0480, "ug-CN"),        //Uyghur
484    (0x0481, "mi-NZ"),        //Maori
485    (0x0482, "oc-FR"),        //Occitan
486    (0x0483, "co-FR"),        //Corsican
487    (0x0484, "gsw-FR"),       //Alsatian
488    (0x0485, "sah-RU"),       //Yakut
489    (0x0486, "qut-GT"),       //K'iche
490    (0x0487, "rw-RW"),        //Kinyarwanda
491    (0x0488, "wo-SN"),        //Wolof
492    (0x048c, "prs-AF"),       //Dari
493    (0x0491, "gd-GB"),        //Scottish Gaelic
494    (0x0801, "ar-IQ"),        //Arabic
495    (0x0804, "zh-Hans"),      //Chinese (Simplified)
496    (0x0807, "de-CH"),        //German
497    (0x0809, "en-GB"),        //English
498    (0x080a, "es-MX"),        //Spanish
499    (0x080c, "fr-BE"),        //French
500    (0x0810, "it-CH"),        //Italian
501    (0x0813, "nl-BE"),        //Dutch
502    (0x0814, "nn-NO"),        //Norwegian (Nynorsk)
503    (0x0816, "pt-PT"),        //Portuguese
504    (0x081a, "sr-Latn-CS"),   //Serbian (Latin)
505    (0x081d, "sv-FI"),        //Swedish
506    (0x082c, "az-Cyrl-AZ"),   //Azeri (Cyrillic)
507    (0x082e, "dsb-DE"),       //Lower Sorbian
508    (0x082e, "dsb"),          //Lower Sorbian
509    (0x083b, "se-SE"),        //Sami (Northern)
510    (0x083c, "ga-IE"),        //Irish
511    (0x083e, "ms-BN"),        //Malay
512    (0x0843, "uz-Cyrl-UZ"),   //Uzbek (Cyrillic)
513    (0x0845, "bn-BD"),        //Bengali
514    (0x0850, "mn-Mong-CN"),   //Mongolian (Traditional Mongolian)
515    (0x085d, "iu-Latn-CA"),   //Inuktitut (Latin)
516    (0x085f, "tzm-Latn-DZ"),  //Tamazight (Latin)
517    (0x086b, "quz-EC"),       //Quechua
518    (0x0c01, "ar-EG"),        //Arabic
519    (0x0c04, "zh-Hant"),      //Chinese (Traditional)
520    (0x0c07, "de-AT"),        //German
521    (0x0c09, "en-AU"),        //English
522    (0x0c0a, "es-ES"),        //Spanish
523    (0x0c0c, "fr-CA"),        //French
524    (0x0c1a, "sr-Cyrl-CS"),   //Serbian (Cyrillic)
525    (0x0c3b, "se-FI"),        //Sami (Northern)
526    (0x0c6b, "quz-PE"),       //Quechua
527    (0x1001, "ar-LY"),        //Arabic
528    (0x1004, "zh-SG"),        //Chinese (Simplified)
529    (0x1007, "de-LU"),        //German
530    (0x1009, "en-CA"),        //English
531    (0x100a, "es-GT"),        //Spanish
532    (0x100c, "fr-CH"),        //French
533    (0x101a, "hr-BA"),        //Croatian (Latin)
534    (0x103b, "smj-NO"),       //Sami (Lule)
535    (0x1401, "ar-DZ"),        //Arabic
536    (0x1404, "zh-MO"),        //Chinese (Traditional)
537    (0x1407, "de-LI"),        //German
538    (0x1409, "en-NZ"),        //English
539    (0x140a, "es-CR"),        //Spanish
540    (0x140c, "fr-LU"),        //French
541    (0x141a, "bs-Latn-BA"),   //Bosnian (Latin)
542    (0x141a, "bs"),           //Bosnian
543    (0x143b, "smj-SE"),       //Sami (Lule)
544    (0x143b, "smj"),          //Sami (Lule)
545    (0x1801, "ar-MA"),        //Arabic
546    (0x1809, "en-IE"),        //English
547    (0x180a, "es-PA"),        //Spanish
548    (0x180c, "fr-MC"),        //French
549    (0x181a, "sr-Latn-BA"),   //Serbian (Latin)
550    (0x183b, "sma-NO"),       //Sami (Southern)
551    (0x1c01, "ar-TN"),        //Arabic
552    (0x1c09, "en-ZA"),        //English
553    (0x1c0a, "es-DO"),        //Spanish
554    (0x1c1a, "sr-Cyrl-BA"),   //Serbian (Cyrillic)
555    (0x1c3b, "sma-SE"),       //Sami (Southern)
556    (0x1c3b, "sma"),          //Sami (Southern)
557    (0x2001, "ar-OM"),        //Arabic
558    (0x2009, "en-JM"),        //English
559    (0x200a, "es-VE"),        //Spanish
560    (0x201a, "bs-Cyrl-BA"),   //Bosnian (Cyrillic)
561    (0x201a, "bs-Cyrl"),      //Bosnian (Cyrillic)
562    (0x203b, "sms-FI"),       //Sami (Skolt)
563    (0x203b, "sms"),          //Sami (Skolt)
564    (0x2401, "ar-YE"),        //Arabic
565    (0x2409, "en-029"),       //English
566    (0x240a, "es-CO"),        //Spanish
567    (0x241a, "sr-Latn-RS"),   //Serbian (Latin)
568    (0x243b, "smn-FI"),       //Sami (Inari)
569    (0x2801, "ar-SY"),        //Arabic
570    (0x2809, "en-BZ"),        //English
571    (0x280a, "es-PE"),        //Spanish
572    (0x281a, "sr-Cyrl-RS"),   //Serbian (Cyrillic)
573    (0x2c01, "ar-JO"),        //Arabic
574    (0x2c09, "en-TT"),        //English
575    (0x2c0a, "es-AR"),        //Spanish
576    (0x2c1a, "sr-Latn-ME"),   //Serbian (Latin)
577    (0x3001, "ar-LB"),        //Arabic
578    (0x3009, "en-ZW"),        //English
579    (0x300a, "es-EC"),        //Spanish
580    (0x301a, "sr-Cyrl-ME"),   //Serbian (Cyrillic)
581    (0x3401, "ar-KW"),        //Arabic
582    (0x3409, "en-PH"),        //English
583    (0x340a, "es-CL"),        //Spanish
584    (0x3801, "ar-AE"),        //Arabic
585    (0x380a, "es-UY"),        //Spanish
586    (0x3c01, "ar-BH"),        //Arabic
587    (0x3c0a, "es-PY"),        //Spanish
588    (0x4001, "ar-QA"),        //Arabic
589    (0x4009, "en-IN"),        //English
590    (0x400a, "es-BO"),        //Spanish
591    (0x4409, "en-MY"),        //English
592    (0x440a, "es-SV"),        //Spanish
593    (0x4809, "en-SG"),        //English
594    (0x480a, "es-HN"),        //Spanish
595    (0x4c0a, "es-NI"),        //Spanish
596    (0x500a, "es-PR"),        //Spanish
597    (0x540a, "es-US"),        //Spanish
598];
599
600#[cfg(test)]
601mod tests {
602    use crate::MetadataProvider;
603
604    use super::*;
605    use read_fonts::FontRef;
606
607    #[test]
608    fn localized() {
609        let font = FontRef::new(font_test_data::NAMES_ONLY).unwrap();
610        let mut subfamily_names = font
611            .localized_strings(StringId::SUBFAMILY_NAME)
612            .map(|s| (s.language().unwrap().to_string(), s.to_string()))
613            .collect::<Vec<_>>();
614        subfamily_names.sort_by(|a, b| a.0.cmp(&b.0));
615        let expected = [
616            (String::from("ar-SA"), String::from("عادي")),
617            (String::from("el-GR"), String::from("Κανονικά")),
618            (String::from("en"), String::from("Regular")),
619            (String::from("eu-ES"), String::from("Arrunta")),
620            (String::from("pl-PL"), String::from("Normalny")),
621            (String::from("zh-Hans"), String::from("正常")),
622        ];
623        assert_eq!(subfamily_names.as_slice(), expected);
624    }
625
626    #[test]
627    fn find_by_language() {
628        let font = FontRef::new(font_test_data::NAMES_ONLY).unwrap();
629        assert_eq!(
630            font.localized_strings(StringId::SUBFAMILY_NAME)
631                .find(|s| s.language() == Some("pl-PL"))
632                .unwrap()
633                .to_string(),
634            "Normalny"
635        );
636    }
637
638    #[test]
639    fn english_or_first() {
640        let font = FontRef::new(font_test_data::NAMES_ONLY).unwrap();
641        assert_eq!(
642            font.localized_strings(StringId::SUBFAMILY_NAME)
643                .english_or_first()
644                .unwrap()
645                .to_string(),
646            "Regular"
647        );
648    }
649}