fix: improve title and episode parsing regex for sub-episodes, seasons, and clean formatting

src/utils/titleParser.ts

@@ -7,22 +7,52 @@ export interface ParsedTitle {
 export function parseTitleAndEpisode(rawTitle: string | null): ParsedTitle {
   if (!rawTitle) return { title: '', episode: null, fullCleanTitle: '' };
 
-  // Remove site name suffixes like "- 마나토키 330", "- 뉴토끼", "- newtoki"
+  let clean = rawTitle.trim();
-  let clean = rawTitle
+
-    .replace(/\s*[-|_]\s*(마나토키|뉴토끼|newtoki|manatoki|웹툰|만화).*$/i, '')
+  // 1. Remove site name / brand suffixes & domain noise
+  // Matches: "- 마나토키 330", "> 뉴토끼 340", ":: manatoki 100", "| 뉴토끼", " - 웹툰", " - 만화", etc.
+  clean = clean
+    .replace(/\s*[-|_|>|:|=|~|/|\\]+\s*(마나토키|뉴토끼|newtoki|manatoki|웹툰|만화|toki|토끼)\s*\d*\s*$/i, '')
+    .replace(/\s*[-|_|>|:|=|~|/|\\]+\s*(https?:\/\/)?[\w.-]+\.(com|org|net|xyz|me|info|live)\s*$/i, '')
+    .replace(/\s*[\(\[]\s*(마나토키|뉴토끼|newtoki|manatoki|웹툰|만화)\s*\d*\s*[\)\]]\s*$/i, '')
+    .replace(/\s+(마나토키|뉴토끼|newtoki|manatoki)\s*\d+\s*$/i, '')
     .trim();
 
-  // 1. Match explicit episode format like "150화", "150.5화", "EP.15"
+  // 2. Main Episode Pattern:
-  const epMatch = clean.match(/(\d+(?:\.\d+)?\s*화|ep\.?\s*\d+)/i);
+  // Group 1 (optional season/part): (시즌\s*\d+|[1-9]기|[1-9]부|[Pp]art\s*\d+)
+  // Group 2 (episode number with optional sub-part or comma): (\d{1,4}(?:,\d{3})*(?:[._-]\d+)?)
+  // Group 3 (unit): (화|회|화차|강|話)
+  const mainEpRegex = /(?:((?:시즌\s*\d+|[1-9]기|[1-9]부|[Pp]art\s*\d+))\s+)?(?:제\s*)?(\d{1,4}(?:,\d{3})*(?:[._-]\d+)?)\s*(화|회|화차|강|話)/i;
+
+  const epMatch = clean.match(mainEpRegex);
+
   if (epMatch) {
-    let episode = epMatch[1].replace(/\s+/g, '');
+    const season = epMatch[1] ? epMatch[1].trim() + ' ' : '';
-    if (!episode.endsWith('화') && !episode.toLowerCase().startsWith('ep')) {
+    const num = epMatch[2].replace(/\s+/g, '');
-      episode += '화';
+    const unit = epMatch[3] || '화';
+    const fullEpString = epMatch[0]; // e.g. "2기 150화" or "150-1화"
+
+    const episode = `${season}${num}${unit}`;
+
+    // Remove the episode portion from clean to extract remaining title
+    let title = clean.replace(fullEpString, ' ');
+    title = cleanUpTitleString(title);
+
+    return {
+      title: title || clean,
+      episode,
+      fullCleanTitle: clean,
+    };
   }
-    // Remove the matched episode part from the title
+
-    let title = clean.replace(epMatch[0], '').trim();
+  // 3. "EP.150", "Ep 150", "#150", "No.150" pattern
-    // Clean up trailing dashes/colons/dots left over
+  const epPrefixRegex = /(?:ep\.?|#|no\.?)\s*(\d+(?:[.-_]\d+)?)/i;
-    title = title.replace(/[-|_:]\s*$/, '').trim();
+  const prefixMatch = clean.match(epPrefixRegex);
+
+  if (prefixMatch) {
+    const episode = `EP.${prefixMatch[1]}`;
+    let title = clean.replace(prefixMatch[0], ' ');
+    title = cleanUpTitleString(title);
 
     return {
       title: title || clean,
@@ -31,12 +61,17 @@ export function parseTitleAndEpisode(rawTitle: string | null): ParsedTitle {
     };
   }
 
-  // 2. Match trailing numbers like "제목 150"
+  // 4. Trailing number without '화' (e.g. "원피스 1100", "원피스 1100-1")
-  const numMatch = clean.match(/(\d+(?:\.\d+)?)\s*$/);
+  const trailingNumRegex = /(?:((?:시즌\s*\d+|[1-9]기|[1-9]부|[Pp]art\s*\d+))\s+)?(\d{1,4}(?:[.-_]\d+)?)\s*$/i;
-  if (numMatch) {
+  const trailingMatch = clean.match(trailingNumRegex);
-    const episode = `${numMatch[1]}화`;
+
-    let title = clean.slice(0, numMatch.index).trim();
+  if (trailingMatch) {
-    title = title.replace(/[-|_:]\s*$/, '').trim();
+    const season = trailingMatch[1] ? trailingMatch[1].trim() + ' ' : '';
+    const num = trailingMatch[2];
+    const episode = `${season}${num}화`;
+
+    let title = clean.slice(0, trailingMatch.index).trim();
+    title = cleanUpTitleString(title);
 
     return {
       title: title || clean,
@@ -47,3 +82,12 @@ export function parseTitleAndEpisode(rawTitle: string | null): ParsedTitle {
 
   return { title: clean, episode: null, fullCleanTitle: clean };
 }
+
+function cleanUpTitleString(str: string): string {
+  let s = str.replace(/\s+/g, ' ').trim();
+  s = s
+    .replace(/^[\s\-_|:;=><~,/\\\\]+/, '')
+    .replace(/[\s\-_|:;=><~,/\\\\]+$/, '')
+    .trim();
+  return s;
+}