Merge pull request #18 from MarvNC:remove-punctuation-readings

Strip Punctuation From Ruby Text
MarvNC · Jan 21, 2024 · 95d44f4 · 95d44f4
2 parents 6f589ff + 4635f94
commit 95d44f4
Show file tree

Hide file tree

Showing 2 changed files with 23 additions and 15 deletions.
diff --git a/src/test/parseCantoneseReadings.test.js b/src/test/parseCantoneseReadings.test.js
@@ -44,7 +44,7 @@ const testCases = [
       { text: '我', reading: 'ngo5' },
       { text: '講', reading: 'gong2' },
       { text: '咩', reading: 'me1' },
-      { text: '？', reading: '?' },
+      { text: '？', reading: '' },
     ],
   },
   {
@@ -61,7 +61,7 @@ const testCases = [
       { text: '斷', reading: 'tyun5' },
       { text: '韌', reading: 'jan6' },
       { text: '帶', reading: 'daai2' },
-      { text: '。', reading: '.' },
+      { text: '。', reading: '' },
     ],
   },
   {
@@ -118,15 +118,15 @@ const testCases = [
       { text: '啲', reading: 'di1' },
       { text: '巴', reading: 'baa1' },
       { text: '士', reading: 'si2' },
-      { text: '，', reading: ',' },
+      { text: '，', reading: '' },
       { text: '一', reading: 'jat1' },
       { text: '定', reading: 'ding6' },
       { text: '經', reading: 'ging1' },
       { text: '地', reading: 'dei6' },
       { text: '鐵', reading: 'tit3' },
       { text: '站', reading: 'zaam6' },
       { text: '㗎', reading: 'gaa3' },
-      { text: '。', reading: '.' },
+      { text: '。', reading: '' },
     ],
   },
   {
@@ -161,14 +161,14 @@ const testCases = [
       { text: '經', reading: 'ging1' },
       { text: '之', reading: 'zi1' },
       { text: '路', reading: 'lou6' },
-      { text: '，', reading: ',' },
+      { text: '，', reading: '' },
       { text: '有', reading: 'jau5' },
       { text: '好', reading: 'hou2' },
       { text: '有', reading: 'jau5' },
       { text: '唔', reading: 'm4' },
       { text: '好', reading: 'hou2' },
       { text: '囉', reading: 'lo1' },
-      { text: '。', reading: '.' },
+      { text: '。', reading: '' },
     ],
   },
   {
@@ -190,7 +190,7 @@ const testCases = [
       { text: '白', reading: 'baak6' },
       { text: '兔', reading: 'tou3' },
       { text: '會', reading: 'wui6' },
-      { text: '》', reading: '.' },
+      { text: '》', reading: '' },
       { text: '。', reading: '' },
     ],
   },

diff --git a/src/util/textHandling/parseCantoneseReadings.js b/src/util/textHandling/parseCantoneseReadings.js
@@ -32,31 +32,39 @@ function parseCantoneseReadings(rawText, readings) {
     const isTextHanzi = isHanzi(text);
     const isTextAlphanumeric = isJyuutping(text);
     const isTextPunctuation = isPunctuation(text);
-    const isReadingJyuutping = isJyuutping(reading);
+    const isReadingJyutping = isJyuutping(reading);
     const isReadingPunctuation = isPunctuation(reading);
     // Ideal case
     if (
       !!text &&
       !!reading &&
-      ((isTextHanzi && isReadingJyuutping) ||
-        (isTextPunctuation && isReadingPunctuation) ||
+      ((isTextHanzi && isReadingJyutping) ||
         // Case where for example text is 'bu' and reading is 'bu4'
-        (isTextAlphanumeric && isReadingJyuutping) ||
-        // Where both are special characters
-        (!isTextAlphanumeric && !isTextHanzi && !isReadingJyuutping))
+        (isTextAlphanumeric && isReadingJyutping))
     ) {
       resultArray.push({ text, reading });
       textIndex++;
       readingIndex++;
     } else if (
       !!text &&
-      ((isTextPunctuation && isReadingJyuutping) ||
+      ((isTextPunctuation && isReadingJyutping) ||
         (!!text && reading === undefined) ||
-        (!isTextAlphanumeric && !isTextHanzi && isReadingJyuutping))
+        (!isTextAlphanumeric && !isTextHanzi && isReadingJyutping))
     ) {
       // Send empty string to reading
       resultArray.push({ text, reading: '' });
       textIndex++;
+    } else if (
+      !!text &&
+      !!reading &&
+      ((isTextPunctuation && isReadingPunctuation) ||
+        // Where both are special characters
+        (!isTextAlphanumeric && !isTextHanzi && !isReadingJyutping))
+    ) {
+      // Don't add the punctuation but consume it
+      resultArray.push({ text, reading: '' });
+      textIndex++;
+      readingIndex++;
     } else {
       throw new Error(
         `Unexpected text "${text}" and reading "${reading}" at index ${i} in ${rawText}: ${readings}`