fix(txt): recover author for 【】-titled web-novel TXT imports, closes #4390 (#4423)

Chinese web-novel TXT files are commonly named 【书名】1-129 作者：起落.txt and
carry a noisy metadata block at the top of the file. Two author-recognition
failures resulted after importing them:

  1. Author missing — extractTxtFilenameMetadata only pulled an author from
     《》-wrapped names, so 【】-style names yielded no filename author, and when
     the file header had no clean "作者：X" line the author came out empty.

  2. Irrelevant content as author — the greedy file-header capture
     (/作者…(.+)\r?\n/) grabbed a publication blob like
     "2024/08/01发表于：是否首发：是 字数1023150字…" and surfaced it as the author.

Fix:
  - extractTxtFilenameMetadata now extracts the labeled "作者：X" form from any
    filename (title stays the full name; only the labeled form is safe so a
    leading 【title】 isn't mistaken for the author).
  - Validate the header-matched author (isPlausibleAuthorName) and fall back to
    the filename author when it looks like a metadata blob — embedded field
    separator, long digit run, or excessive length. Applied to both the small-
    and large-file conversion paths.

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

apps/readest-app/src/__tests__/utils/txt-converter.test.ts

@@ -492,4 +492,65 @@ describe('extractTxtFilenameMetadata', () => {
   it('returns empty object for empty input', () => {
     expect(extractTxtFilenameMetadata('')).toEqual({ title: '' });
   });
+
+  // Chinese web-novel TXT files are commonly named with a 【】 title and a
+  // labeled author tacked on, e.g. 【书名】1-129 作者：起落.txt. There are no 《》,
+  // so the whole name stays the title, but the labeled author must still be
+  // extracted. See issue #4390.
+  it('extracts a labeled author from a 【】-titled filename without 《》 (issue #4390)', () => {
+    expect(extractTxtFilenameMetadata('【细雨飘香】1-129 作者：起落.txt')).toEqual({
+      title: '【细雨飘香】1-129 作者：起落',
+      author: '起落',
+    });
+    expect(extractTxtFilenameMetadata('【月如无恨月长圆】（1-154）作者：陈西.txt')).toEqual({
+      title: '【月如无恨月长圆】（1-154）作者：陈西',
+      author: '陈西',
+    });
+  });
+
+  it('does not mistake a leading 【tag】 for the author when no 作者 label is present', () => {
+    expect(extractTxtFilenameMetadata('【完结】斗破苍穹.txt')).toEqual({
+      title: '【完结】斗破苍穹',
+    });
+  });
+});
+
+describe('author resolution during conversion (issue #4390)', () => {
+  const convertAndCaptureMetadata = async (name: string, content: string) => {
+    const converter = new TxtToEpubConverter() as unknown as TxtConverterFlowPrivateAPI;
+    let captured: TestMetadata | undefined;
+    converter.detectEncoding = () => 'utf-8';
+    converter.createEpub = async (_chapters, metadata) => {
+      captured = metadata;
+      return new Blob();
+    };
+    converter.extractChapters = () => [{ title: '第一章', content: '正文', isVolume: false }];
+    const file = new File([content], name);
+    await converter.convert({ file });
+    return captured;
+  };
+
+  it('falls back to the filename author when the header has none (missing author)', async () => {
+    const metadata = await convertAndCaptureMetadata(
+      '【细雨飘香】1-129 作者：起落.txt',
+      '第一章 初见\n正文内容……\n',
+    );
+    expect(metadata?.author).toBe('起落');
+  });
+
+  it('rejects a metadata-blob header author and uses the filename author (irrelevant content)', async () => {
+    const metadata = await convertAndCaptureMetadata(
+      '【月如无恨月长圆】（1-154）作者：陈西.txt',
+      '作者：2024/08/01发表于：是否首发：是字数1023150字116:01\n第一章 初见\n正文内容……\n',
+    );
+    expect(metadata?.author).toBe('陈西');
+  });
+
+  it('keeps a clean labeled author parsed from the file header', async () => {
+    const metadata = await convertAndCaptureMetadata(
+      '【幻灵幽火】1-23未完结 作者：月夜银狐.txt',
+      '作者：月夜银狐\n第一章 初见\n正文内容……\n',
+    );
+    expect(metadata?.author).toBe('月夜银狐');
+  });
 });

apps/readest-app/src/utils/txt.ts

@@ -21,7 +21,13 @@ export const extractTxtFilenameMetadata = (
   const base = getBaseFilename(filename);
   const cjkMatch = base.match(/《([^》]+)》(.*)/);
   if (!cjkMatch) {
-    return { title: base };
+    // No 《》 wrapper: keep the whole filename as the title (web-novel files use
+    // 【】 brackets for the title and tack the author on, e.g.
+    // 【书名】1-129 作者：起落.txt). Only the labeled "作者：X" form is safe to pull
+    // here — a bracketed/bare fallback would mistake a leading 【title】 for the
+    // author. See issue #4390.
+    const author = parseLabeledAuthor(base);
+    return author ? { title: base, author } : { title: base };
   }
   const title = cjkMatch[1]!.trim();
   const rest = (cjkMatch[2] ?? '').trim();
@@ -29,11 +35,17 @@ export const extractTxtFilenameMetadata = (
   return author ? { title, author } : { title };
 };
 
+// 作者：X / 作者:X / 作者 X — a labeled author. Returns '' when absent.
+const parseLabeledAuthor = (text: string): string => {
+  const labeled = text.match(/作者\s*[：:\s]\s*(.+)$/);
+  return labeled ? stripWrappingPunctuation(labeled[1]!) : '';
+};
+
 const parseAuthorFragment = (text: string): string => {
   if (!text) return '';
   // 作者：X / 作者:X / 作者 X — labeled author wins
-  const labeled = text.match(/作者\s*[：:\s]\s*(.+)$/);
-  if (labeled) return stripWrappingPunctuation(labeled[1]!);
+  const labeled = parseLabeledAuthor(text);
+  if (labeled) return labeled;
   // [X] (X) 【X】 （X）［X］ — bracketed author
   const bracketed = text.match(/[[(（【［]\s*([^\])）】］]+?)\s*[\])）】］]/);
   if (bracketed) return stripWrappingPunctuation(bracketed[1]!);
@@ -50,6 +62,15 @@ const stripWrappingPunctuation = (text: string): string => {
   }
 };
 
+// A header line like "作者：X" is meant to yield a short personal/pen name. Some
+// web-novel TXT files instead carry a metadata blob there (e.g.
+// "作者：2024/08/01发表于：是否首发：是 字数1023150字…") that the greedy capture would
+// otherwise surface as the author. Reject values that look like such a blob —
+// an embedded field separator (a second colon), a long digit run, or excessive
+// length — so callers fall back to the filename's labeled author. See #4390.
+const isPlausibleAuthorName = (name: string): boolean =>
+  name.length > 0 && name.length <= 20 && !/[:：]/.test(name) && !/\d{4,}/.test(name);
+
 interface Chapter {
   title: string;
   content: string;
@@ -130,7 +151,8 @@ export class TxtToEpubConverter {
     try {
       matchedAuthor = matchedAuthor.replace(/^[\p{P}\p{S}]+|[\p{P}\p{S}]+$/gu, '');
     } catch {}
-    const author = matchedAuthor || filenameMeta.author || providedAuthor || '';
+    const headerAuthor = isPlausibleAuthorName(matchedAuthor) ? matchedAuthor : '';
+    const author = headerAuthor || filenameMeta.author || providedAuthor || '';
     const language = providedLanguage || detectLanguage(fileHeader);
     // console.log(`Detected language: ${language}`);
     const identifier = await partialMD5(txtFile);
@@ -523,11 +545,12 @@ export class TxtToEpubConverter {
     const authorMatch =
       fileHeader.match(/[【\[]?作者[】\]]?[:：\s]\s*(.+)\r?\n/) ||
       fileHeader.match(/[【\[]?\s*(.+)\s+著\s*[】\]]?\r?\n/);
-    let matchedAuthor = authorMatch ? authorMatch[1]!.trim() : providedAuthor || '';
+    let matchedAuthor = authorMatch ? authorMatch[1]!.trim() : '';
     try {
       matchedAuthor = matchedAuthor.replace(/^[\p{P}\p{S}]+|[\p{P}\p{S}]+$/gu, '');
     } catch {}
-    const author = matchedAuthor || providedAuthor || '';
+    const headerAuthor = isPlausibleAuthorName(matchedAuthor) ? matchedAuthor : '';
+    const author = headerAuthor || providedAuthor || '';
     const language = providedLanguage || detectLanguage(fileHeader);
     return { author, language };
   }