anil.recoil.org
Detect which human language a document uses from OCaml, from the Nu Html validator
languages
unicode
ocaml
Add langdetect-js package for browser language detection
New langdetect-js package provides:
- JavaScript API via js_of_ocaml (langdetect.detect, detectAll, etc.)
- Browser-based regression tests across 37 languages
- Interactive demo page (test.html)
Build with: opam exec -- dune build lib/js/
Open test.html in browser to run tests and try the demo.
API:
langdetect.detect(text) → lang code or null
langdetect.detectWithProb(text) → {lang, prob} or null
langdetect.detectAll(text) → [{lang, prob}, ...]
langdetect.languages() → array of supported language codes
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
+15
-1
dune-project
+15
-1
dune-project
···
16
16
(synopsis "Language detection library using n-gram frequency analysis")
17
17
(description
18
18
"An OCaml port of the Cybozu langdetect algorithm. Detects the natural \
19
-
language of text using n-gram frequency profiles. Supports 49 languages \
19
+
language of text using n-gram frequency profiles. Supports 47 languages \
20
20
including English, Chinese, Japanese, Arabic, and many European languages.")
21
21
(depends
22
22
(ocaml (>= 5.1.0))
23
23
(uutf (>= 1.0.0))
24
24
(odoc :with-doc)
25
25
(alcotest (and :with-test (>= 1.7.0)))))
26
+
27
+
(package
28
+
(name langdetect-js)
29
+
(synopsis "Language detection for browsers via js_of_ocaml/wasm_of_ocaml")
30
+
(description
31
+
"Browser-compatible language detection compiled with js_of_ocaml. \
32
+
Provides a JavaScript API for detecting languages in web applications. \
33
+
Supports both JS and WebAssembly output for optimal performance.")
34
+
(depends
35
+
(ocaml (>= 5.1.0))
36
+
(langdetect (= :version))
37
+
(brr (>= 0.0.6))
38
+
(js_of_ocaml (>= 5.0.0))
39
+
(js_of_ocaml-compiler (>= 5.0.0))))
+34
langdetect-js.opam
+34
langdetect-js.opam
···
1
+
# This file is generated by dune, edit dune-project instead
2
+
opam-version: "2.0"
3
+
synopsis: "Language detection for browsers via js_of_ocaml/wasm_of_ocaml"
4
+
description:
5
+
"Browser-compatible language detection compiled with js_of_ocaml. Provides a JavaScript API for detecting languages in web applications. Supports both JS and WebAssembly output for optimal performance."
6
+
maintainer: ["Anil Madhavapeddy <anil@recoil.org>"]
7
+
authors: ["Anil Madhavapeddy"]
8
+
license: "MIT"
9
+
homepage: "https://tangled.org/@anil.recoil.org/ocaml-langdetect"
10
+
bug-reports: "https://tangled.org/@anil.recoil.org/ocaml-langdetect/issues"
11
+
depends: [
12
+
"dune" {>= "3.20"}
13
+
"ocaml" {>= "5.1.0"}
14
+
"langdetect" {= version}
15
+
"brr" {>= "0.0.6"}
16
+
"js_of_ocaml" {>= "5.0.0"}
17
+
"js_of_ocaml-compiler" {>= "5.0.0"}
18
+
"odoc" {with-doc}
19
+
]
20
+
build: [
21
+
["dune" "subst"] {dev}
22
+
[
23
+
"dune"
24
+
"build"
25
+
"-p"
26
+
name
27
+
"-j"
28
+
jobs
29
+
"@install"
30
+
"@runtest" {with-test}
31
+
"@doc" {with-doc}
32
+
]
33
+
]
34
+
x-maintenance-intent: ["(latest)"]
+1
-1
langdetect.opam
+1
-1
langdetect.opam
···
2
2
opam-version: "2.0"
3
3
synopsis: "Language detection library using n-gram frequency analysis"
4
4
description:
5
-
"An OCaml port of the Cybozu langdetect algorithm. Detects the natural language of text using n-gram frequency profiles. Supports 49 languages including English, Chinese, Japanese, Arabic, and many European languages."
5
+
"An OCaml port of the Cybozu langdetect algorithm. Detects the natural language of text using n-gram frequency profiles. Supports 47 languages including English, Chinese, Japanese, Arabic, and many European languages."
6
6
maintainer: ["Anil Madhavapeddy <anil@recoil.org>"]
7
7
authors: ["Anil Madhavapeddy"]
8
8
license: "MIT"
+38
lib/js/dune
+38
lib/js/dune
···
1
+
; Langdetect JavaScript Library
2
+
; Compiled with js_of_ocaml for browser use
3
+
4
+
(library
5
+
(name langdetect_js)
6
+
(public_name langdetect-js)
7
+
(libraries langdetect brr)
8
+
(modes byte) ; js_of_ocaml requires bytecode
9
+
(modules langdetect_js))
10
+
11
+
; Standalone JavaScript file for direct browser use
12
+
(executable
13
+
(name langdetect_js_main)
14
+
(libraries langdetect_js)
15
+
(js_of_ocaml
16
+
(javascript_files))
17
+
(modes js)
18
+
(modules langdetect_js_main))
19
+
20
+
; Browser-based test runner
21
+
(executable
22
+
(name langdetect_js_tests)
23
+
(libraries langdetect_js)
24
+
(js_of_ocaml
25
+
(javascript_files))
26
+
(modes js)
27
+
(modules langdetect_js_tests))
28
+
29
+
; Copy to nice filenames (JS)
30
+
(rule
31
+
(targets langdetect.js)
32
+
(deps langdetect_js_main.bc.js)
33
+
(action (copy %{deps} %{targets})))
34
+
35
+
(rule
36
+
(targets langdetect-tests.js)
37
+
(deps langdetect_js_tests.bc.js)
38
+
(action (copy %{deps} %{targets})))
+95
lib/js/langdetect_js.ml
+95
lib/js/langdetect_js.ml
···
1
+
(*---------------------------------------------------------------------------
2
+
Copyright (c) 2025 Anil Madhavapeddy <anil@recoil.org>
3
+
SPDX-License-Identifier: MIT
4
+
---------------------------------------------------------------------------*)
5
+
6
+
(** JavaScript bindings for langdetect.
7
+
8
+
This module provides browser-compatible language detection via js_of_ocaml.
9
+
It exposes a simple API on window.langdetect for detecting languages in text. *)
10
+
11
+
12
+
(** The detector instance, lazily initialized *)
13
+
let detector = lazy (Langdetect.create_default ())
14
+
15
+
(** Detect the language of text, returning the best match or null *)
16
+
let detect_best text =
17
+
let t = Lazy.force detector in
18
+
Langdetect.detect_best t text
19
+
20
+
(** Detect language with probability score *)
21
+
let detect_with_prob text =
22
+
let t = Lazy.force detector in
23
+
Langdetect.detect_with_prob t text
24
+
25
+
(** Detect all matching languages above threshold *)
26
+
let detect_all text =
27
+
let t = Lazy.force detector in
28
+
Langdetect.detect t text
29
+
30
+
(** Get list of supported languages *)
31
+
let supported_languages () =
32
+
let t = Lazy.force detector in
33
+
Langdetect.supported_languages t
34
+
35
+
(** Console logging helper *)
36
+
let console_log msg =
37
+
ignore (Jv.call (Jv.get Jv.global "console") "log" [| Jv.of_string msg |])
38
+
39
+
(** Convert a detection result to a JavaScript object *)
40
+
let result_to_jv (r : Langdetect.result) =
41
+
Jv.obj [|
42
+
"lang", Jv.of_string r.lang;
43
+
"prob", Jv.of_float r.prob;
44
+
|]
45
+
46
+
(** Register the API on a JavaScript object *)
47
+
let register_api_on obj =
48
+
(* detect(text) -> string | null *)
49
+
Jv.set obj "detect" (Jv.callback ~arity:1 (fun text_jv ->
50
+
let text = Jv.to_string text_jv in
51
+
match detect_best text with
52
+
| Some lang -> Jv.of_string lang
53
+
| None -> Jv.null
54
+
));
55
+
56
+
(* detectWithProb(text) -> {lang, prob} | null *)
57
+
Jv.set obj "detectWithProb" (Jv.callback ~arity:1 (fun text_jv ->
58
+
let text = Jv.to_string text_jv in
59
+
match detect_with_prob text with
60
+
| Some (lang, prob) ->
61
+
Jv.obj [|
62
+
"lang", Jv.of_string lang;
63
+
"prob", Jv.of_float prob;
64
+
|]
65
+
| None -> Jv.null
66
+
));
67
+
68
+
(* detectAll(text) -> [{lang, prob}, ...] *)
69
+
Jv.set obj "detectAll" (Jv.callback ~arity:1 (fun text_jv ->
70
+
let text = Jv.to_string text_jv in
71
+
let results = detect_all text in
72
+
Jv.of_list result_to_jv results
73
+
));
74
+
75
+
(* languages() -> string[] *)
76
+
Jv.set obj "languages" (Jv.callback ~arity:0 (fun () ->
77
+
let langs = supported_languages () in
78
+
Jv.of_array Jv.of_string langs
79
+
));
80
+
81
+
(* version *)
82
+
Jv.set obj "version" (Jv.of_string "1.0.0")
83
+
84
+
(** Register the global API on window.langdetect *)
85
+
let register_global_api () =
86
+
let api = Jv.obj [||] in
87
+
register_api_on api;
88
+
Jv.set Jv.global "langdetect" api;
89
+
90
+
(* Dispatch 'langdetectReady' event for async loaders *)
91
+
let document = Jv.get Jv.global "document" in
92
+
let event_class = Jv.get Jv.global "CustomEvent" in
93
+
let event = Jv.new' event_class [| Jv.of_string "langdetectReady" |] in
94
+
ignore (Jv.call document "dispatchEvent" [| event |]);
95
+
console_log "[langdetect] API ready - 47 languages loaded"
+9
lib/js/langdetect_js_main.ml
+9
lib/js/langdetect_js_main.ml
···
1
+
(*---------------------------------------------------------------------------
2
+
Copyright (c) 2025 Anil Madhavapeddy <anil@recoil.org>
3
+
SPDX-License-Identifier: MIT
4
+
---------------------------------------------------------------------------*)
5
+
6
+
(** Entry point for the standalone JavaScript build.
7
+
Registers the API on window.langdetect when the script loads. *)
8
+
9
+
let () = Langdetect_js.register_global_api ()
+267
lib/js/langdetect_js_tests.ml
+267
lib/js/langdetect_js_tests.ml
···
1
+
(*---------------------------------------------------------------------------
2
+
Copyright (c) 2025 Anil Madhavapeddy <anil@recoil.org>
3
+
SPDX-License-Identifier: MIT
4
+
---------------------------------------------------------------------------*)
5
+
6
+
(** Browser-based test runner for langdetect.
7
+
8
+
This module runs regression tests in the browser and displays results
9
+
in the DOM. It demonstrates language detection across multiple languages. *)
10
+
11
+
open Brr
12
+
13
+
(** Test case definition *)
14
+
type test_case = {
15
+
name : string;
16
+
text : string;
17
+
expected : string;
18
+
}
19
+
20
+
(** Test results *)
21
+
type test_result = {
22
+
test : test_case;
23
+
detected : string option;
24
+
prob : float option;
25
+
passed : bool;
26
+
time_ms : float;
27
+
}
28
+
29
+
(** Sample texts for testing various languages *)
30
+
let test_cases = [|
31
+
(* European languages *)
32
+
{ name = "English"; text = "The quick brown fox jumps over the lazy dog. This is a sample of English text for language detection testing."; expected = "en" };
33
+
{ name = "French"; text = "Bonjour le monde. La langue française est belle et mélodieuse. J'aime beaucoup la culture française."; expected = "fr" };
34
+
{ name = "German"; text = "Guten Tag! Die deutsche Sprache hat viele interessante Eigenschaften. Ich lerne gerne Deutsch."; expected = "de" };
35
+
{ name = "Spanish"; text = "Hola mundo. El español es un idioma muy hablado en todo el mundo. Me gusta mucho la cultura española."; expected = "es" };
36
+
{ name = "Italian"; text = "Ciao mondo! L'italiano è una lingua bellissima. Mi piace molto la cultura italiana e il cibo."; expected = "it" };
37
+
{ name = "Portuguese"; text = "Olá mundo! O português é uma língua muito bonita. Eu gosto muito da cultura portuguesa."; expected = "pt" };
38
+
{ name = "Dutch"; text = "Hallo wereld! De Nederlandse taal is interessant. Ik hou van de Nederlandse cultuur."; expected = "nl" };
39
+
{ name = "Swedish"; text = "Hej världen! Svenska är ett vackert språk. Jag tycker om svensk kultur och mat."; expected = "sv" };
40
+
{ name = "Norwegian"; text = "Hei verden! Norsk er et fint språk. Jeg liker norsk kultur og natur veldig godt."; expected = "no" };
41
+
{ name = "Danish"; text = "Hej verden! Dansk er et interessant sprog. Jeg kan godt lide dansk kultur."; expected = "da" };
42
+
{ name = "Finnish"; text = "Hei maailma! Suomen kieli on erittäin mielenkiintoinen. Pidän suomalaisesta kulttuurista."; expected = "fi" };
43
+
{ name = "Polish"; text = "Witaj świecie! Język polski jest bardzo piękny. Lubię polską kulturę i historię."; expected = "pl" };
44
+
{ name = "Russian"; text = "Привет мир! Русский язык очень красивый и богатый. Я люблю русскую литературу."; expected = "ru" };
45
+
{ name = "Ukrainian"; text = "Привіт світ! Українська мова дуже гарна. Я люблю українську культуру."; expected = "uk" };
46
+
{ name = "Czech"; text = "Ahoj světe! Čeština je zajímavý jazyk. Mám rád českou kulturu a historii."; expected = "cs" };
47
+
{ name = "Romanian"; text = "Salut lume! Limba română este foarte frumoasă. Îmi place cultura românească."; expected = "ro" };
48
+
{ name = "Hungarian"; text = "Szia világ! A magyar nyelv nagyon érdekes. Szeretem a magyar kultúrát."; expected = "hu" };
49
+
{ name = "Greek"; text = "Γειά σου κόσμε! Η ελληνική γλώσσα είναι πολύ όμορφη. Μου αρέσει ο ελληνικός πολιτισμός."; expected = "el" };
50
+
{ name = "Bulgarian"; text = "Здравей свят! Българският език е много красив. Обичам българската култура."; expected = "bg" };
51
+
52
+
(* Asian languages *)
53
+
{ name = "Chinese (Simplified)"; text = "你好世界！中文是一种非常古老而美丽的语言。我喜欢学习中国文化和历史。"; expected = "zh-cn" };
54
+
{ name = "Chinese (Traditional)"; text = "你好世界！中文是一種非常古老而美麗的語言。我喜歡學習中國文化和歷史。"; expected = "zh-tw" };
55
+
{ name = "Japanese"; text = "こんにちは世界！日本語はとても美しい言語です。日本の文化が大好きです。"; expected = "ja" };
56
+
{ name = "Korean"; text = "안녕하세요 세계! 한국어는 매우 아름다운 언어입니다. 저는 한국 문화를 좋아합니다."; expected = "ko" };
57
+
{ name = "Vietnamese"; text = "Xin chào thế giới! Tiếng Việt là một ngôn ngữ rất đẹp. Tôi yêu văn hóa Việt Nam."; expected = "vi" };
58
+
{ name = "Thai"; text = "สวัสดีโลก! ภาษาไทยเป็นภาษาที่สวยงาม ฉันชอบวัฒนธรรมไทยมาก"; expected = "th" };
59
+
{ name = "Indonesian"; text = "Halo dunia! Bahasa Indonesia adalah bahasa yang indah. Saya suka budaya Indonesia."; expected = "id" };
60
+
61
+
(* Middle Eastern languages *)
62
+
{ name = "Arabic"; text = "مرحبا بالعالم! اللغة العربية جميلة جدا. أنا أحب الثقافة العربية والتاريخ."; expected = "ar" };
63
+
{ name = "Hebrew"; text = "שלום עולם! השפה העברית היא שפה יפה מאוד. אני אוהב את התרבות העברית."; expected = "he" };
64
+
{ name = "Persian"; text = "سلام دنیا! زبان فارسی بسیار زیباست. من فرهنگ ایرانی را دوست دارم."; expected = "fa" };
65
+
{ name = "Turkish"; text = "Merhaba dünya! Türkçe çok güzel bir dil. Türk kültürünü ve tarihini seviyorum."; expected = "tr" };
66
+
67
+
(* South Asian languages *)
68
+
{ name = "Hindi"; text = "नमस्ते दुनिया! हिंदी एक बहुत सुंदर भाषा है। मुझे भारतीय संस्कृति बहुत पसंद है।"; expected = "hi" };
69
+
{ name = "Bengali"; text = "হ্যালো বিশ্ব! বাংলা একটি অত্যন্ত সুন্দর ভাষা। আমি বাঙালি সংস্কৃতি পছন্দ করি।"; expected = "bn" };
70
+
{ name = "Tamil"; text = "வணக்கம் உலகம்! தமிழ் ஒரு மிக அழகான மொழி. நான் தமிழ் கலாச்சாரத்தை விரும்புகிறேன்."; expected = "ta" };
71
+
{ name = "Telugu"; text = "హలో ప్రపంచం! తెలుగు చాలా అందమైన భాష. నాకు తెలుగు సంస్కృతి చాలా ఇష్టం."; expected = "te" };
72
+
{ name = "Gujarati"; text = "હેલો વિશ્વ! ગુજરાતી એક સુંદર ભાષા છે. મને ગુજરાતી સંસ્કૃતિ ગમે છે."; expected = "gu" };
73
+
{ name = "Urdu"; text = "ہیلو دنیا! اردو ایک بہت خوبصورت زبان ہے۔ مجھے اردو ادب پسند ہے۔"; expected = "ur" };
74
+
|]
75
+
76
+
(** Get current time in milliseconds *)
77
+
let now_ms () =
78
+
Jv.to_float (Jv.call (Jv.get Jv.global "performance") "now" [||])
79
+
80
+
(** Run a single test *)
81
+
let run_test detector test =
82
+
let start = now_ms () in
83
+
let result = Langdetect.detect_with_prob detector test.text in
84
+
let time_ms = now_ms () -. start in
85
+
let detected, prob = match result with
86
+
| Some (lang, p) -> Some lang, Some p
87
+
| None -> None, None
88
+
in
89
+
let passed = match detected with
90
+
| Some lang -> lang = test.expected
91
+
| None -> false
92
+
in
93
+
{ test; detected; prob; passed; time_ms }
94
+
95
+
(** Run all tests and return results *)
96
+
let run_all_tests () =
97
+
let detector = Langdetect.create_default () in
98
+
Array.map (run_test detector) test_cases
99
+
100
+
(** Create a result row element *)
101
+
let create_result_row result =
102
+
let status_class = if result.passed then "pass" else "fail" in
103
+
let status_text = if result.passed then "✓" else "✗" in
104
+
let detected_text = match result.detected with
105
+
| Some lang -> lang
106
+
| None -> "(none)"
107
+
in
108
+
let prob_text = match result.prob with
109
+
| Some p -> Printf.sprintf "%.1f%%" (p *. 100.0)
110
+
| None -> "-"
111
+
in
112
+
let time_text = Printf.sprintf "%.1fms" result.time_ms in
113
+
114
+
let tr = El.tr [] in
115
+
El.set_children tr [
116
+
El.td [El.txt' result.test.name];
117
+
El.td ~at:[At.class' (Jstr.v "code")] [El.txt' result.test.expected];
118
+
El.td ~at:[At.class' (Jstr.v "code")] [El.txt' detected_text];
119
+
El.td [El.txt' prob_text];
120
+
El.td [El.txt' time_text];
121
+
El.td ~at:[At.class' (Jstr.v status_class)] [El.txt' status_text];
122
+
];
123
+
tr
124
+
125
+
(** Create summary stats *)
126
+
let create_summary results =
127
+
let total = Array.length results in
128
+
let passed = Array.fold_left (fun acc r -> if r.passed then acc + 1 else acc) 0 results in
129
+
let failed = total - passed in
130
+
let total_time = Array.fold_left (fun acc r -> acc +. r.time_ms) 0.0 results in
131
+
let avg_time = total_time /. float_of_int total in
132
+
133
+
El.div ~at:[At.class' (Jstr.v "summary")] [
134
+
El.h2 [El.txt' "Test Results"];
135
+
El.p [
136
+
El.strong [El.txt' (Printf.sprintf "%d/%d tests passed" passed total)];
137
+
El.txt' (Printf.sprintf " (%d failed)" failed);
138
+
];
139
+
El.p [
140
+
El.txt' (Printf.sprintf "Total time: %.1fms (avg %.2fms per test)" total_time avg_time);
141
+
];
142
+
]
143
+
144
+
(** Main test runner *)
145
+
let run_tests_ui () =
146
+
(* Find or create output container *)
147
+
let container = match El.find_first_by_selector (Jstr.v "#test-results") ~root:(Document.body G.document) with
148
+
| Some el -> el
149
+
| None ->
150
+
let el = El.div ~at:[At.id (Jstr.v "test-results")] [] in
151
+
El.append_children (Document.body G.document) [el];
152
+
el
153
+
in
154
+
155
+
(* Show loading message *)
156
+
El.set_children container [
157
+
El.p [El.txt' "Running tests..."]
158
+
];
159
+
160
+
(* Run tests (async to allow UI update) *)
161
+
let _ = Jv.call Jv.global "setTimeout" [|
162
+
Jv.callback ~arity:0 (fun () ->
163
+
let results = run_all_tests () in
164
+
165
+
(* Build results table *)
166
+
let thead = El.thead [
167
+
El.tr [
168
+
El.th [El.txt' "Language"];
169
+
El.th [El.txt' "Expected"];
170
+
El.th [El.txt' "Detected"];
171
+
El.th [El.txt' "Confidence"];
172
+
El.th [El.txt' "Time"];
173
+
El.th [El.txt' "Status"];
174
+
]
175
+
] in
176
+
177
+
let tbody = El.tbody [] in
178
+
Array.iter (fun result ->
179
+
El.append_children tbody [create_result_row result]
180
+
) results;
181
+
182
+
let table = El.table ~at:[At.class' (Jstr.v "results-table")] [thead; tbody] in
183
+
184
+
(* Update container *)
185
+
El.set_children container [
186
+
create_summary results;
187
+
table;
188
+
]
189
+
);
190
+
Jv.of_int 10
191
+
|] in
192
+
()
193
+
194
+
(** Interactive demo section *)
195
+
let setup_demo () =
196
+
let demo_container = match El.find_first_by_selector (Jstr.v "#demo") ~root:(Document.body G.document) with
197
+
| Some el -> el
198
+
| None -> Document.body G.document
199
+
in
200
+
201
+
let textarea = El.textarea ~at:[
202
+
At.id (Jstr.v "demo-input");
203
+
At.v (Jstr.v "rows") (Jstr.v "4");
204
+
At.v (Jstr.v "placeholder") (Jstr.v "Enter text to detect language...");
205
+
] [] in
206
+
207
+
let result_div = El.div ~at:[At.id (Jstr.v "demo-result")] [
208
+
El.txt' "Enter text above and click Detect"
209
+
] in
210
+
211
+
let detect_button = El.button ~at:[At.id (Jstr.v "demo-button")] [El.txt' "Detect Language"] in
212
+
213
+
(* Set up click handler *)
214
+
let detector = Langdetect.create_default () in
215
+
ignore (Ev.listen Ev.click (fun _ ->
216
+
let text = Jstr.to_string (El.prop El.Prop.value textarea) in
217
+
if String.length text > 0 then begin
218
+
let start = now_ms () in
219
+
let results = Langdetect.detect detector text in
220
+
let time_ms = now_ms () -. start in
221
+
222
+
let result_html = match results with
223
+
| [] ->
224
+
[El.txt' "No language detected (text may be too short)"]
225
+
| _ ->
226
+
let items = List.map (fun (r : Langdetect.result) ->
227
+
El.li [
228
+
El.strong [El.txt' r.lang];
229
+
El.txt' (Printf.sprintf " — %.1f%% confidence" (r.prob *. 100.0))
230
+
]
231
+
) results in
232
+
[
233
+
El.p [El.txt' (Printf.sprintf "Detected in %.1fms:" time_ms)];
234
+
El.ul items
235
+
]
236
+
in
237
+
El.set_children result_div result_html
238
+
end
239
+
) (El.as_target detect_button));
240
+
241
+
(* Only add demo section if container exists *)
242
+
if El.has_tag_name (Jstr.v "DIV") demo_container then
243
+
El.set_children demo_container [
244
+
El.h2 [El.txt' "Try It"];
245
+
El.div ~at:[At.class' (Jstr.v "demo-area")] [
246
+
textarea;
247
+
detect_button;
248
+
result_div;
249
+
]
250
+
]
251
+
252
+
(** Entry point *)
253
+
let () =
254
+
(* Wait for DOM to be ready *)
255
+
let ready_state = Jv.get (Jv.get Jv.global "document") "readyState" |> Jv.to_string in
256
+
if ready_state = "loading" then
257
+
ignore (Jv.call Jv.global "addEventListener" [|
258
+
Jv.of_string "DOMContentLoaded";
259
+
Jv.callback ~arity:1 (fun _ ->
260
+
run_tests_ui ();
261
+
setup_demo ()
262
+
)
263
+
|])
264
+
else begin
265
+
run_tests_ui ();
266
+
setup_demo ()
267
+
end
+2
lib/langdetect.ml
+2
lib/langdetect.ml
+4
lib/langdetect.mli
+4
lib/langdetect.mli
···
63
63
val detect_with_prob : t -> string -> (string * float) option
64
64
(** [detect_with_prob t text] returns the best matching language code with its
65
65
probability, or [None] if no language could be detected. *)
66
+
67
+
val supported_languages : t -> string array
68
+
(** [supported_languages t] returns an array of language codes that this
69
+
detector supports. *)
+299
test.html
+299
test.html
···
1
+
<!DOCTYPE html>
2
+
<html lang="en">
3
+
<head>
4
+
<meta charset="UTF-8">
5
+
<meta name="viewport" content="width=device-width, initial-scale=1.0">
6
+
<title>Langdetect - Language Detection Demo</title>
7
+
<style>
8
+
* {
9
+
box-sizing: border-box;
10
+
}
11
+
body {
12
+
font-family: system-ui, -apple-system, BlinkMacSystemFont, 'Segoe UI', sans-serif;
13
+
line-height: 1.6;
14
+
max-width: 1000px;
15
+
margin: 0 auto;
16
+
padding: 2rem;
17
+
background: #f5f5f5;
18
+
color: #333;
19
+
}
20
+
h1 {
21
+
color: #2563eb;
22
+
border-bottom: 3px solid #3b82f6;
23
+
padding-bottom: 0.5rem;
24
+
margin-bottom: 0.5rem;
25
+
}
26
+
.subtitle {
27
+
color: #666;
28
+
margin-top: 0;
29
+
margin-bottom: 2rem;
30
+
}
31
+
.section {
32
+
background: white;
33
+
border: 1px solid #e0e0e0;
34
+
border-radius: 12px;
35
+
padding: 1.5rem;
36
+
margin: 1.5rem 0;
37
+
box-shadow: 0 2px 4px rgba(0,0,0,0.05);
38
+
}
39
+
.section h2 {
40
+
margin-top: 0;
41
+
color: #1e40af;
42
+
}
43
+
.demo-area {
44
+
display: flex;
45
+
flex-direction: column;
46
+
gap: 1rem;
47
+
}
48
+
textarea {
49
+
width: 100%;
50
+
padding: 1rem;
51
+
font-size: 1rem;
52
+
font-family: inherit;
53
+
border: 2px solid #e0e0e0;
54
+
border-radius: 8px;
55
+
resize: vertical;
56
+
transition: border-color 0.2s;
57
+
}
58
+
textarea:focus {
59
+
outline: none;
60
+
border-color: #3b82f6;
61
+
}
62
+
button {
63
+
padding: 0.75rem 2rem;
64
+
font-size: 1rem;
65
+
font-weight: 600;
66
+
cursor: pointer;
67
+
border: none;
68
+
border-radius: 8px;
69
+
background: #2563eb;
70
+
color: white;
71
+
transition: all 0.2s;
72
+
align-self: flex-start;
73
+
}
74
+
button:hover {
75
+
background: #1d4ed8;
76
+
transform: translateY(-1px);
77
+
}
78
+
button:active {
79
+
transform: translateY(0);
80
+
}
81
+
#demo-result {
82
+
padding: 1rem;
83
+
background: #f8fafc;
84
+
border-radius: 8px;
85
+
border: 1px solid #e2e8f0;
86
+
}
87
+
#demo-result ul {
88
+
margin: 0.5rem 0;
89
+
padding-left: 1.5rem;
90
+
}
91
+
#demo-result li {
92
+
margin: 0.25rem 0;
93
+
}
94
+
.summary {
95
+
background: #eff6ff;
96
+
padding: 1rem 1.5rem;
97
+
border-radius: 8px;
98
+
margin-bottom: 1rem;
99
+
}
100
+
.summary h2 {
101
+
margin: 0 0 0.5rem 0;
102
+
}
103
+
.summary p {
104
+
margin: 0.25rem 0;
105
+
}
106
+
.results-table {
107
+
width: 100%;
108
+
border-collapse: collapse;
109
+
font-size: 0.9rem;
110
+
}
111
+
.results-table th,
112
+
.results-table td {
113
+
padding: 0.75rem 1rem;
114
+
text-align: left;
115
+
border-bottom: 1px solid #e0e0e0;
116
+
}
117
+
.results-table th {
118
+
background: #f8fafc;
119
+
font-weight: 600;
120
+
color: #475569;
121
+
}
122
+
.results-table tr:hover {
123
+
background: #f8fafc;
124
+
}
125
+
.results-table .code {
126
+
font-family: 'SF Mono', Monaco, 'Cascadia Code', monospace;
127
+
font-size: 0.85rem;
128
+
background: #f1f5f9;
129
+
padding: 0.2rem 0.4rem;
130
+
border-radius: 4px;
131
+
}
132
+
.results-table .pass {
133
+
color: #16a34a;
134
+
font-weight: bold;
135
+
font-size: 1.1rem;
136
+
}
137
+
.results-table .fail {
138
+
color: #dc2626;
139
+
font-weight: bold;
140
+
font-size: 1.1rem;
141
+
}
142
+
.loading {
143
+
text-align: center;
144
+
padding: 2rem;
145
+
color: #666;
146
+
}
147
+
.api-docs {
148
+
background: #1e293b;
149
+
color: #e2e8f0;
150
+
padding: 1rem;
151
+
border-radius: 8px;
152
+
overflow-x: auto;
153
+
}
154
+
.api-docs code {
155
+
color: #7dd3fc;
156
+
}
157
+
.api-docs .comment {
158
+
color: #94a3b8;
159
+
}
160
+
.sample-texts {
161
+
display: grid;
162
+
grid-template-columns: repeat(auto-fill, minmax(200px, 1fr));
163
+
gap: 0.5rem;
164
+
margin-top: 1rem;
165
+
}
166
+
.sample-text {
167
+
padding: 0.5rem 1rem;
168
+
font-size: 0.85rem;
169
+
background: #f1f5f9;
170
+
border: 1px solid #e2e8f0;
171
+
border-radius: 6px;
172
+
cursor: pointer;
173
+
transition: all 0.2s;
174
+
}
175
+
.sample-text:hover {
176
+
background: #e2e8f0;
177
+
border-color: #cbd5e1;
178
+
}
179
+
.sample-text .lang {
180
+
font-weight: 600;
181
+
color: #1e40af;
182
+
}
183
+
</style>
184
+
</head>
185
+
<body>
186
+
187
+
<h1>🌍 Langdetect</h1>
188
+
<p class="subtitle">Language detection for the browser using n-gram frequency analysis</p>
189
+
190
+
<!-- Interactive Demo -->
191
+
<div class="section" id="demo">
192
+
<h2>Try It</h2>
193
+
<div class="demo-area">
194
+
<textarea id="demo-input" rows="4" placeholder="Enter or paste text to detect its language..."></textarea>
195
+
<button id="demo-button">Detect Language</button>
196
+
<div id="demo-result">Enter text above and click Detect</div>
197
+
</div>
198
+
199
+
<p style="margin-top: 1rem; margin-bottom: 0.5rem; color: #666; font-size: 0.9rem;">
200
+
Click a sample to try:
201
+
</p>
202
+
<div class="sample-texts">
203
+
<div class="sample-text" data-text="The quick brown fox jumps over the lazy dog.">
204
+
<span class="lang">🇬🇧 English</span>
205
+
</div>
206
+
<div class="sample-text" data-text="Bonjour le monde! Comment allez-vous aujourd'hui?">
207
+
<span class="lang">🇫🇷 French</span>
208
+
</div>
209
+
<div class="sample-text" data-text="Guten Tag! Wie geht es Ihnen heute?">
210
+
<span class="lang">🇩🇪 German</span>
211
+
</div>
212
+
<div class="sample-text" data-text="¡Hola mundo! ¿Cómo estás hoy?">
213
+
<span class="lang">🇪🇸 Spanish</span>
214
+
</div>
215
+
<div class="sample-text" data-text="你好世界！今天你好吗？">
216
+
<span class="lang">🇨🇳 Chinese</span>
217
+
</div>
218
+
<div class="sample-text" data-text="こんにちは世界！今日はお元気ですか？">
219
+
<span class="lang">🇯🇵 Japanese</span>
220
+
</div>
221
+
<div class="sample-text" data-text="مرحبا بالعالم! كيف حالك اليوم؟">
222
+
<span class="lang">🇸🇦 Arabic</span>
223
+
</div>
224
+
<div class="sample-text" data-text="Привет мир! Как дела сегодня?">
225
+
<span class="lang">🇷🇺 Russian</span>
226
+
</div>
227
+
</div>
228
+
</div>
229
+
230
+
<!-- Test Results -->
231
+
<div class="section">
232
+
<div id="test-results">
233
+
<p class="loading">Loading tests...</p>
234
+
</div>
235
+
</div>
236
+
237
+
<!-- API Documentation -->
238
+
<div class="section">
239
+
<h2>JavaScript API</h2>
240
+
<div class="api-docs">
241
+
<pre><span class="comment">// Detect the most likely language</span>
242
+
<code>langdetect.detect</code>("Hello, world!")
243
+
<span class="comment">// → "en"</span>
244
+
245
+
<span class="comment">// Get detection with confidence score</span>
246
+
<code>langdetect.detectWithProb</code>("Bonjour le monde!")
247
+
<span class="comment">// → { lang: "fr", prob: 0.9999 }</span>
248
+
249
+
<span class="comment">// Get all matching languages</span>
250
+
<code>langdetect.detectAll</code>("Hello world")
251
+
<span class="comment">// → [{ lang: "en", prob: 0.85 }, { lang: "de", prob: 0.10 }, ...]</span>
252
+
253
+
<span class="comment">// List supported languages</span>
254
+
<code>langdetect.languages</code>()
255
+
<span class="comment">// → ["ar", "bg", "bn", "ca", "cs", "da", "de", "el", ...]</span></pre>
256
+
</div>
257
+
</div>
258
+
259
+
<!-- Supported Languages -->
260
+
<div class="section">
261
+
<h2>Supported Languages (47)</h2>
262
+
<p>
263
+
Arabic, Bengali, Bulgarian, Catalan, Chinese (Simplified), Chinese (Traditional),
264
+
Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Greek,
265
+
Gujarati, Hebrew, Hindi, Hungarian, Indonesian, Italian, Japanese, Korean, Latvian,
266
+
Lithuanian, Macedonian, Malayalam, Norwegian, Panjabi, Persian, Polish, Portuguese,
267
+
Romanian, Russian, Sinhala, Slovak, Spanish, Swedish, Tagalog, Tamil, Telugu, Thai,
268
+
Turkish, Ukrainian, Urdu, Vietnamese
269
+
</p>
270
+
</div>
271
+
272
+
<!-- Load the compiled library -->
273
+
<script src="_build/default/lib/js/langdetect-tests.js"></script>
274
+
275
+
<script>
276
+
// Handle sample text clicks
277
+
document.querySelectorAll('.sample-text').forEach(el => {
278
+
el.addEventListener('click', () => {
279
+
const text = el.getAttribute('data-text');
280
+
document.getElementById('demo-input').value = text;
281
+
document.getElementById('demo-button').click();
282
+
});
283
+
});
284
+
285
+
// Check if library loaded
286
+
window.addEventListener('load', () => {
287
+
if (typeof langdetect === 'undefined') {
288
+
document.getElementById('demo-result').innerHTML =
289
+
'<strong style="color: #dc2626;">Library not loaded</strong><br>' +
290
+
'Run <code>opam exec -- dune build lib/js/langdetect-tests.js</code> first.';
291
+
document.getElementById('test-results').innerHTML =
292
+
'<p style="color: #dc2626;"><strong>Tests cannot run:</strong> langdetect.js not found.<br>' +
293
+
'Build with: <code>opam exec -- dune build lib/js/</code></p>';
294
+
}
295
+
});
296
+
</script>
297
+
298
+
</body>
299
+
</html>