Tashfeer Banned Words Fixes #2

src/main/java/io/github/seen_arabic/arabic_services/ArabicServices.java

@@ -139,6 +139,9 @@ public static String wordToLetters(String word) {
      *         original word if no affix matches are found. {@link String}
      */
     public static String removeArabicAffixes(String word) {
+        if (word.isEmpty()) {
+            return word;
+        }
         if (Data.ARABIC_PREFIXES.contains(word.substring(0, 2))) {
             // For: ALEF & LAM
             word = word.substring(2);
@@ -155,6 +158,42 @@ public static String removeArabicAffixes(String word) {
         return word.trim();
     }
 
+    /**
+     * Performs tashfeer encryption on a given text, but only for words that are
+     * considered "banned" words.
+     * Banned words are determined based on a predefined similarity ratio.
+     *
+     * @param text The input text to be encrypted {@link String}.
+     * @return The encrypted text with tashfeer applied and an encryption level
+     *         is 2 to banned words {@link String}.
+     */
+    public static String tashfeerBannedWords(String text) {
+        return tashfeerBannedWords(text, 2);
+    }
+
+    /**
+     * Performs tashfeer encryption on a given text, but only for words that are
+     * considered "banned" words.
+     * Banned words are determined based on a predefined similarity ratio.
+     *
+     * @param text            The input text to be encrypted {@link String}.
+     * @param levelOfTashfeer The encryption level (default is 2).
+     * @return The encrypted text with tashfeer applied to banned words
+     *         {@link String}.
+     */
+    public static String tashfeerBannedWords(String text, int levelOfTashfeer) {
+        Objects.requireNonNull(text, Data.TEXT_NULL_MESSAGE);
+        StringBuilder newText = new StringBuilder();
+        for (String word : text.split(" ")) {
+            if (checkIfBannedWord(word)) {
+                newText.append(tashfeerHandler(word, levelOfTashfeer)).append(" ");
+            } else {
+                newText.append(word).append(" ");
+            }
+        }
+        return newText.toString().trim();
+    }
+
     private static String handleNoonIssue(String text) {
         String arabicLetters = String.join("", Data.LETTERS_DICT.keySet()) + "ـ";
         String regex = Data.NOON + "(" + "?=[^" + arabicLetters + "]" + ")|" + Data.NOON + "\\z";
@@ -259,4 +298,68 @@ private static char tashfeerCharacter(char character) {
         char replacementCharacter = replacementCharList[randomIndex].charAt(0);
         return replacementCharacter;
     }
+
+    private static double bannedSimilarityRatio(String string) {
+        double maximumSimilarity = -1;
+        for (String bannedWord : Data.BANNED_WORDS) {
+            double calculatedSimilarity = similarityScore(string, bannedWord);
+            if (calculatedSimilarity > maximumSimilarity) {
+                maximumSimilarity = calculatedSimilarity;
+            }
+        }
+        return maximumSimilarity * 100;
+    }
+
+    private static boolean checkIfBannedWord(String string) {
+        double stdRatio = 70;
+        return bannedSimilarityRatio(removeArabicAffixes(string)) >= stdRatio;
+    }
+
+    private static double similarityScore(String s1, String s2) {
+        String longer = s1;
+        String shorter = s2;
+
+        // swap them if s1 is bigger than s2
+        if (s1.length() < s2.length()) {
+            longer = s2;
+            shorter = s1;
+        }
+
+        int longerLength = longer.length();
+
+        // if both are empty strings return 1 (100% similarity)
+        if (longerLength == 0) {
+            return 1.0;
+        }
+
+        // calculate the similarity score
+        return (double) (longerLength - editDistance(longer, shorter)) / (double) longerLength;
+    }
+
+    private static int editDistance(String s1, String s2) {
+        s1 = s1.toLowerCase();
+        s2 = s2.toLowerCase();
+
+        int[] costs = new int[s2.length() + 1];
+
+        for (int i = 0; i <= s1.length(); i++) {
+            int lastValue = i;
+            for (int j = 0; j <= s2.length(); j++) {
+                if (i == 0) {
+                    costs[j] = j;
+                } else if (j > 0) {
+                    int newValue = costs[j - 1];
+                    if (s1.charAt(i - 1) != s2.charAt(j - 1)) {
+                        newValue = Math.min(Math.min(newValue, lastValue), costs[j]) + 1;
+                    }
+                    costs[j - 1] = lastValue;
+                    lastValue = newValue;
+                }
+            }
+            if (i > 0) {
+                costs[s2.length()] = lastValue;
+            }
+        }
+        return costs[s2.length()];
+    }
 }

src/main/java/io/github/seen_arabic/arabic_services/Data.java

@@ -131,7 +131,61 @@ class Data {
             'ئ');
 
     static List<Character> ALONE_LETTERS = Arrays.asList('د', 'ذ', 'ر', 'ز', 'و', 'ا', 'أ', 'إ', 'ء', 'ؤ', 'آ');
-
+    static final String[] BANNED_WORDS = {
+            "فلسطين",
+            "عرب",
+            "هود",
+            "صهيون",
+            "سرائيل",
+            "دول",
+            "كيان",
+            "حتل",
+            "هتلر",
+            "خنازير",
+            "حي",
+            "شيخ",
+            "جراح",
+            "سقط",
+            "قدس",
+            "قصى",
+            "طبع",
+            "قتل",
+            "خان",
+            "كتائب",
+            "عز",
+            "دين",
+            "قسام",
+            "جهاد",
+            "جاهد",
+            "سلام",
+            "خوارزم",
+            "لوغاريزم",
+            "كتاب",
+            "بدون",
+            "نقط",
+            "مارك",
+            "لعن",
+            "حقير",
+            "موت",
+            "قاوم",
+            "زوربير",
+            "عاصم",
+            "حر",
+            "رهاب",
+            "قرد",
+            "دعم",
+            "غز",
+            "نقذ",
+            "نتهك",
+            "معايير",
+            "صل",
+            "بان",
+            "دون",
+            "جيش",
+            "عدو",
+            "حماس",
+            "كر"
+    };
     static {
         LETTERS_DICT = new HashMap<>();
         LETTERS_DICT.put("ا", "ا");

src/test/java/io/github/seen_arabic/arabic_services/ArabicServicesTest.java

@@ -1,8 +1,10 @@
 package io.github.seen_arabic.arabic_services;
 
 import static org.junit.Assert.assertEquals;
+import static org.junit.Assert.assertFalse;
 import static org.junit.Assert.assertNotEquals;
 import static org.junit.Assert.assertNotNull;
+import static org.junit.Assert.assertTrue;
 
 import org.junit.Test;
 
@@ -241,4 +243,28 @@ private void itShouldRemoveHumSuffixFromAWord() {
         assertEquals("طلاب", result);
     }
 
+    @Test
+    public void tashfeerBannedWordsShouldPerformTashfeerEncryptionOnBannedWordsOnly() {
+        String sentence = "جيش العدو يقتل الأطفال";
+        String result = ArabicServices.tashfeerBannedWords(sentence);
+        assertNotEquals(sentence, result);
+        assertTrue(result.contains("الأطفال"));
+        assertFalse(result.contains("جيش"));
+        assertFalse(result.contains("العدو"));
+        assertFalse(result.contains("يقتل"));
+    }
+
+    @Test
+    public void tashfeerBannedWordsShouldNotPerformTashfeerEncryptionOnNonBannedWords() {
+        String sentence = "هذه جملة غير مشفرة";
+        String result = ArabicServices.tashfeerBannedWords(sentence);
+        assertEquals(sentence, result);
+    }
+
+    @Test
+    public void tashfeerBannedWordsShouldHandleEmptyInput() {
+        String sentence = "";
+        String result = ArabicServices.tashfeerBannedWords(sentence);
+        assertEquals("", result);
+    }
 }