add rule "duplicated diacritics on the same letter"

aborazmeh · Jul 9, 2024 · 0aebfac · 0aebfac
1 parent dd2f86a
commit 0aebfac
Show file tree

Hide file tree

Showing 3 changed files with 66 additions and 1 deletion.
diff --git a/README.md b/README.md
@@ -26,6 +26,14 @@ Only diacritics will be removed and there likely to be an extra space
 
     أهلا  وسهلا  بكم
 
+### No Duplicated Diacritics on the Same Letter
+
+    يونََُُِِس: قال أبو عبيدة، «يقال:يونس بضم النون وكسرها». والمشهور في القراءة يونُس برفع النون من غير همز.
+
+Duplicated diacritics will be removed
+
+    يونَُِس: قال أبو عبيدة، «يقال:يونس بضم النون وكسرها». والمشهور في القراءة يونُس برفع النون من غير همز.
+
 ### No Shadda With Madda
 
 Shadda can't be combined with Madda
@@ -51,6 +59,7 @@ These are default options, you can change them in your .textlintrc file
           "remove_loose_diacritics": true,
           "no_shadda_with_madda": true,
           "no_shadda_with_sukun": true,
+          "no_duplicated_diacritics": true,
         }
     }
 }
@@ -86,7 +95,6 @@ Test textlint rule by [textlint-tester](https://github.com/textlint/textlint-tes
 - No Tanween *and* Haraka on the same letter
 - No Haraka *and* Sukun on the same letter
 - No Sukun on the first letter of the word
-- Duplicated diacritics on the same letter
 - Option for no combining diacritics
 - No Shadda with Sukun
 - Normalize diacritics: like `U+FE7C` and `U+FE7D` to `U+0651`, and `U+FCF2` to `U+0651` and `U+064E`

diff --git a/src/index.ts b/src/index.ts
@@ -5,6 +5,7 @@ export interface Options {
     remove_loose_diacritics?: boolean;
     no_shadda_with_madda?: boolean;
     no_shadda_with_sukun?: boolean;
+    no_duplicated_diacritics?: boolean;
 }
 
 const regex = {
@@ -73,13 +74,42 @@ function noShaddaWithSukun(node: TxtStrNode, text: string, context: Readonly<Tex
     }
 }
 
+function noDuplicatedDiacritics(node: TxtStrNode, text: string, context: Readonly<TextlintRuleContext>) {
+    const { report, fixer, locator, RuleError } = context;
+
+    let match;
+    let matches = [];
+    const doubleRegex = new RegExp(`(${regex.diacritics})${regex.diacritics}*\\1`, "g");
+
+    for (let i = 0; i < text.length; i++) {
+        doubleRegex.lastIndex = i; // Start matching from the current position
+        match = doubleRegex.exec(text);
+        if (match) {
+            matches.push(match);
+            i = match.index; // Move to the start of the match for the next iteration
+        }
+    }
+
+    for (const match of matches) {
+        const index = match.index ?? 0;
+        const matchRange = [index, index + match[0].length] as const;
+        const remove = fixer.removeRange([index + match[0].length - 1, index + match[0].length]);
+        const ruleError = new RuleError("Found duplicated Arabic diacritic on the same letter.", {
+            padding: locator.range(matchRange),
+            fix: remove
+        });
+        report(node, ruleError);
+    }
+}
+
 const report: TextlintRuleModule<Options> = (context, options = {}) => {
     const { getSource, Syntax } = context;
     return {
         [Syntax.Str](node) {
             const removeLooseDiacritics = options.remove_loose_diacritics ?? true;
             const shaddaWithMaddaOpt = options.no_shadda_with_madda ?? true;
             const shaddaWithSukunOpt = options.no_shadda_with_sukun ?? true;
+            const duplicatedDiacriticsOpt = options.no_duplicated_diacritics ?? true;
 
             const text = getSource(node); // Get text
             noLooseDiacritics(node, text, context, removeLooseDiacritics);
@@ -91,6 +121,10 @@ const report: TextlintRuleModule<Options> = (context, options = {}) => {
             if (shaddaWithSukunOpt) {
                 noShaddaWithSukun(node, text, context);
             }
+
+            if (duplicatedDiacriticsOpt) {
+                noDuplicatedDiacritics(node, text, context);
+            }
         }
     };
 };

diff --git a/test/index-test.ts b/test/index-test.ts
@@ -8,6 +8,7 @@ tester.run("rule", rule, {
         "أهلاً وسهلاً",
         "الآن",
         "ضيّق",
+        "يونَُِس: قال أبو عبيدة، «يقال:يونس بضم النون وكسرها». والمشهور في القراءة يونُس برفع النون من غير همز.",
         {
             text: "ضيّْق",
             options: {
@@ -81,6 +82,28 @@ tester.run("rule", rule, {
                     range: [2, 4]
                 }
             ]
+        },
+        {
+            text: "يونََُُِِس: قال أبو عبيدة، «يقال:يونس بضم النون وكسرها». والمشهور في القراءة يونُس برفع النون من غير همز.",
+            output: "يونَُِس: قال أبو عبيدة، «يقال:يونس بضم النون وكسرها». والمشهور في القراءة يونُس برفع النون من غير همز.",
+            errors: [
+                {
+                    message: "Found duplicated Arabic diacritic on the same letter.",
+                    range: [3, 7]
+                },
+                { message: "Found duplicated Arabic diacritic on the same letter.", range: [4, 8] },
+                { message: "Found duplicated Arabic diacritic on the same letter.", range: [5, 9] }
+            ]
+        },
+        {
+            text: "تجربة لشدَّّتين على نفس الحرف",
+            output: "تجربة لشدَّتين على نفس الحرف",
+            errors: [
+                {
+                    message: "Found duplicated Arabic diacritic on the same letter.",
+                    range: [9, 12]
+                }
+            ]
         }
     ]
 });