added utils folder; added several suplamentary modules

4 files changed, 223 insertions, 0 deletions

diff --git a/utils/kb-util.js b/utils/kb-util.js
new file mode 100644
index 0000000..aac88f9
--- /dev/null
+++ b/utils/kb-util.js
@@ -0,0 +1,88 @@
+/**
+ * fast string manipulation utilities
+ * for bi-lingual (EL/EN) words/phrases
+ * based on the keyboard layout
+ */
+
+// suplamentary arrays (mostly for cache)
+// --- -- -- - - -
+
+var ORiGiNal = 'ςερτυθιοπασδφγηξκλζχψωβνμΕΡΤΥΘΙΟΠΑΣΔΦΓΗΞΚΛΖΧΨΩΒΝΜάέήίόύώϊΐϋΆΈΉΊΌΎΏΪΫQWERTYUIOPASDFGHJKLZXCVBNMqwertyuiopasdfghjklzxcvbnm0123456789- '.split('');
+
+var kbKeyZed = 'sertyuiopasdfghjklzxcvbnmertyuiopasdfghjklzxcvbnmaehioyviiyaehioyviyqwertyuiopasdfghjklzxcvbnmqwertyuiopasdfghjklzxcvbnm0123456789- '.split('');
+
+var map = new Map();
+for (var i=0; i<ORiGiNal.length; i++) map.set(ORiGiNal[i], kbKeyZed[i]);
+
+
+// cache (=create a global array)
+// of accended to non-accended vowels mapping
+// --- -- -- - - -
+accented_vowels = [];   
+[
+    'ά α', 'έ ε', 'ή η', 'ί ι', 'ϊ ι', 'ΐ ι', 'ό ο', 'ύ υ', 'ϋ υ', 'ώ ω',
+    'Ά Α', 'Έ Ε', 'Ή Η', 'Ί Ι', 'Ϊ Ι', 'Ό Ο', 'Ύ Υ', 'Ϋ Υ', 'Ώ Ω'
+].forEach( pair => {
+    ap = pair.split(' ');
+    accented_vowels.push({
+        a: ap[0],   // accented
+        p: ap[1]    // pure = non accended
+    });
+});
+
+
+// translates string to keyboard-latin keys
+// (the ones that used whan typing each letter of the word)
+const keyboardize = (str) => {
+    str = str.replace('\'','');
+    var out = '';
+    // [map]'s implementation is 40x faster than [for]'s
+    for (var i=0 ; i< str.length; i++)  out += map.get(str[i]);
+    return out;
+}
+
+// keyboardize an array of strings
+const keyb_array = (arr) => {
+    kb_arr = [];
+    arr.forEach( w => {
+        kb_arr.push(keyboardize(w));
+    });
+    return kb_arr;
+}
+
+
+// transforms to lowercase; handles sigma-teliko
+const sanitizeGR = (str) => {
+    str = str.toLowerCase();
+
+    // replace accended vowels with pure ones
+    accented_vowels.forEach( v => {
+        str = str.replaceAll(v.a, v.p);
+    });
+
+    // replace sigma on the end of words
+    str = str + ' ';
+    str = str.replaceAll('σ-', 'ς-');
+    str = str.replaceAll('σ ', 'ς ');
+    
+    return str;
+}
+
+// removes non keyword characters [+ . , !] and internal multiple-spaces
+// @param txt (string): product description
+const clean = (txt) => {
+    return txt.replace('+',' ').replace('.',' ').replace(',',' ')   // change to space
+        .replace('!','').replace('\"', '')                  // remove character
+        .replace('   ',' ').replace('  ',' ');          // remove multiple spaces
+}
+
+
+// exports
+// --- -- -- - - -
+
+module.exports = {
+    keyboardize,
+    keyb_array,
+    sanitizeGR,
+    clean
+};
\ No newline at end of file
diff --git a/utils/match-util.js b/utils/match-util.js
new file mode 100644
index 0000000..a13fddb
--- /dev/null
+++ b/utils/match-util.js
@@ -0,0 +1,92 @@
+/** Ngram fuzzy match algorithm
+ * (simple and fast)
+ */
+const createNgram = (word, n) => {          // Ngram creation
+    if (word.length <3)     return word;
+    const vector = [];
+    for (let i = 0; i < word.length-n+1; ++i) {
+        vector.push(word.slice(i, i + n));
+    }
+    return vector;
+};
+
+/** similarity
+ * rates similarity between 2 words
+ * based on Ngram matches of N = n letters;
+ * implements a 2-dim check (all a-Ngrams vs all all b-Ngrams)
+ * 
+ * @param {string} a : first word
+ * @param {string} b : second word
+ * @param {int} n : Ngram base
+ * @returns {float} : match percentage as a float in [0, 1]
+ */
+const similarity = (a, b, n) => {         // Ngram match score
+    if (a.length > 0 && b.length > 0) {
+        const aNgram = createNgram(a, n);
+        const bNgram = createNgram(b, n);
+        let hits = 0;
+        for (let x = 0; x < aNgram.length; ++x) {
+            for (let y = 0; y < bNgram.length; ++y) {
+                if (aNgram[x] === bNgram[y]) {
+                    hits += 1;
+                }
+            }
+        }
+        if (hits > 0) {
+            const union = aNgram.length + bNgram.length;
+            return (2.0 * hits) / union;
+        }
+    }
+    return 0;
+};
+
+/** resemblance
+ * is an alternative similarity rating;
+ * implements an 1-dim Ngram similarity check
+ * and it's much faster than similarity()
+ */
+const resemblance = (a, b, n) => {
+    if (a.length > n && b.length >= a.length) {
+        const aNgram = createNgram(a, n);
+        let hits = 0;
+        for (let i = 0; i < aNgram.length; ++i) {
+            if (b.includes(aNgram[i])) {
+                hits++;
+            }
+        }
+        if (hits > 0) {
+            // rate resemblance based on hits and length-similarity
+            return (hits / aNgram.length) * (a.length / b.length);
+        }
+    }
+    return 0;
+}
+
+
+/** is_exact_match
+ * 
+ * check if a searching string -> query (string/latin in kb-format)
+ * matches exactly an item of the array of synonyms -> chkArr (array of utf-8/strings)
+ * 
+ * @param query (string): searching string; string/latin in kb-format
+ * @param chkArr (array): array of synonyms; (array of utf-8/strings)
+ * @return (boolean): true|false
+ */
+function exact( query, chkArr ) {
+    found = false;
+    chkArr.forEach( w => { if (w == query) found = true });
+    return found;
+}
+
+function partial( query, chkArr ) {
+    found = false;
+    chkArr.forEach( w => { if (w.includes(query)) found = true });
+    return found;
+}
+
+module.exports = {
+    exact,
+    partial,
+    similarity,
+    resemblance
+}
diff --git a/utils/mem-usage.js b/utils/mem-usage.js
new file mode 100644
index 0000000..766a93e
--- /dev/null
+++ b/utils/mem-usage.js
@@ -0,0 +1,20 @@
+/**
+ * memory usage report utility
+ */
+
+const formatMemoryUsage = (data) => `${Math.round(data / 1024 / 1024 * 100) / 100} MB`;
+
+function report() {
+    let memoryData = process.memoryUsage();
+
+    let memoryUsage = {
+      rss: `${formatMemoryUsage(memoryData.rss)} -> Resident Set Size - total memory allocated for the process execution`,
+      heapTotal: `${formatMemoryUsage(memoryData.heapTotal)} -> total size of the allocated heap`,
+      heapUsed: `${formatMemoryUsage(memoryData.heapUsed)} -> actual memory used during the execution`,
+      external: `${formatMemoryUsage(memoryData.external)} -> V8 external memory`,
+    };
+    
+    console.log(memoryUsage);
+}
+
+module.exports = { report }
diff --git a/utils/url-util.js b/utils/url-util.js
new file mode 100644
index 0000000..e86ad9b
--- /dev/null
+++ b/utils/url-util.js
@@ -0,0 +1,23 @@
+/**
+ * url utility
+ */
+
+const querystring = require('querystring');
+
+function struct(req, url) {
+    let url_parts = url.split('?');
+    let query = (url_parts.length > 1)
+        ? querystring.decode(url_parts[1])
+        : {};
+    return {
+        method: req.method,
+        host: req.host,
+        path: url_parts[0],
+        query: query
+    }
+}
+
+/**
+ * exports
+ */
+module.exports = { struct }
\ No newline at end of file