Commit

1 files changed, 140 insertions, 0 deletions

diff --git a/src/detect.js b/src/detect.js
new file mode 100644
index 0000000..1319fff
--- /dev/null
+++ b/src/detect.js
@@ -0,0 +1,140 @@
+splangs = require('./languages.json');

+

+function reverseObject(object) {

+    var newObject = {};

+    var keys = [];

+

+    for (var key in object) {

+        keys.push(key);

+    }

+

+    for (var i = keys.length - 1; i >= 0; i--) {

+        var value = object[keys[i]];

+        newObject[keys[i]]= value;

+    }

+

+    return newObject;

+}

+

+try {

+    old = new Date();

+    lold = new Date();

+

+    const diacritics = require('./diacritics');

+    const count = require('./counter');

+    const percentages = require('./percentages');

+    const differenciate = require('./differenciate');

+    const bestLanguage = require('./better');

+    const prettylang = require('./prettylang');

+    const wordsDetect = require('./wdetect');

+

+    fs = require('fs');

+    chalk = require('chalk');

+

+    class LangdetectError extends Error {

+        constructor(...params) {

+            super(...params)

+            // We're spreading `params` as a way to bring all of `Error`'s functionality in.

+        }

+    }

+

+    if (typeof process.argv[2] == "undefined") {

+        throw new LangdetectError("args: Missing operand");

+    }

+

+    console.log("Let's try to detect which language this is!");

+

+    lrec = new Date();

+

+    require("./status");

+

+    told = new Date();

+    console.log("Decomposing text... This will take a while for long texts!");

+    clean = diacritics(process.argv[2].toLowerCase());

+    cleaner = clean.replace(/[^a-zA-Z]/g, "");

+    cleaner2 = clean.replace(/[^a-zA-Z ]/g, "");

+    wcwords = cleaner2.split(/[, ]+/).length;

+    letters = cleaner.split("");

+    console.log("Text contains " + letters.length + " letters");

+    console.log("Calculating percentage for all letters...");

+    qty = count(letters);

+    if (qty.total != letters.length) {

+        console.log(chalk.red(chalk.bold("WARNING: ") + (letters.length - qty.total) + " characters were not included while counting all letters. This is probably a bug and should be reported."));

+    }

+    percs = percentages(qty);

+    trec = new Date();

+    tdiff = trec - told;

+    mold = new Date();

+    console.log("Getting matches for each letter in each supported language...");

+    diffs = differenciate(percs, langs);

+    mrec = new Date();

+    mdiff = mrec - mold

+

+    bold = new Date();

+    console.log("Finding the best matches between all supported languages...");

+    precounts = bestLanguage(diffs);

+    brec = new Date();

+    bdiff = brec - bold;

+

+    wold = new Date();

+    console.log("Including proportion of recognized words...");

+    newcounts = wordsDetect(precounts, cleaner2, langs);

+    wcount = newcounts["_words"];

+    delete newcounts["_words"];

+    preprecounts = Object.entries(newcounts)

+        .sort(([,a],[,b]) => a-b)

+        .reduce((r, [k, v]) => ({ ...r, [k]: v }), {});

+    counts = reverseObject(preprecounts);

+    wrec = new Date();

+    wdiff = wrec - wold;

+

+    rec = new Date();

+    ddiff = rec - old;

+    ldiff = lrec - lold;

+    console.log(chalk.green("Done in " + Math.ceil(ddiff) + " ms:\n * " + Math.ceil(ldiff) + " ms loading required files\n * " + Math.ceil(tdiff) + " ms decomposing text\n * " + Math.ceil(mdiff) + " ms gathering matches\n * " + Math.ceil(bdiff) + " ms finding the best matches\n * " + Math.ceil(wdiff) + " ms including known words\n * " + Math.ceil(ddiff - (ldiff + tdiff + mdiff + bdiff + wdiff)) + " ms doing other things"));

+

+    if (ent < 600) {

+        console.log(chalk.red(chalk.bold("WARNING: ") + "Not enough entries to have accurate results: " + ent + "/600 (" + Math.round((ent/600)*100) + "%)"));

+    }

+

+    keys = Object.keys(counts);

+    total = 0;

+    keys.forEach((key) => {

+        total = total + counts[key];

+    })

+

+    prewcount = Object.entries(wcount)

+        .sort(([,a],[,b]) => a-b)

+        .reduce((r, [k, v]) => ({ ...r, [k]: v }), {});

+    wcount = reverseObject(prewcount);

+

+    if (wcount[keys[0]] < 1) {

+        console.log(chalk.red(chalk.bold("WARNING: ") + "No words detected in text, results may not be very accurate"));

+    }

+

+    if (letters.length < 16) {

+        console.log(chalk.red(chalk.bold("WARNING: ") + "Not enough letters to get accurate results, consider adding a whole sentence instead of isolated words."));

+    }

+    if (letters.length > 1000) {

+        console.log(chalk.red(chalk.bold("WARNING: ") + "Too much letters, consider detecting only a single sentence instead of a whole document."));

+    }

+

+    console.log("\nResults:");

+    console.log(chalk.bold(" 1. " + prettylang(keys[0]) + " (" + Math.round((counts[keys[0]] / total) * 100) + "%, " + wcount[keys[0]] + "/" + wcwords + " words)"));

+    if (counts[keys[1]] > 0) {

+        console.log(" 2. " + prettylang(keys[1]) + " (" + Math.round((counts[keys[1]] / total) * 100) + "%, " + wcount[keys[1]] + "/" + wcwords + " words)");

+    } else {

+        console.log(chalk.gray(" 2. " + prettylang(keys[1]) + " (" + Math.round((counts[keys[1]] / total) * 100) + "%, " + wcount[keys[1]] + "/" + wcwords + " words)"));

+    }

+    if (counts[keys[2]] > 0) {

+        console.log(" 3. " + prettylang(keys[2]) + " (" + Math.round((counts[keys[2]] / total) * 100) + "%, " + wcount[keys[2]] + "/" + wcwords + " words)");

+    } else {

+        console.log(chalk.gray(" 3. " + prettylang(keys[2]) + " (" + Math.round((counts[keys[2]] / total) * 100) + "%, " + wcount[keys[2]] + "/" + wcwords + " words)"));

+    }

+} catch (e) {

+    if (e.message == "Unexpected end of JSON input") {

+        console.log(require('chalk').red("Unable to open database file. Is the databased opened by another program? Or is it corrupted?"));

+    } else {

+        throw e;

+    }

+}
\ No newline at end of file