txt2ebook 0.1.151__tar.gz → 0.1.152__tar.gz

{txt2ebook-0.1.151 → txt2ebook-0.1.152}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: txt2ebook
-Version: 0.1.151
+Version: 0.1.152
 Summary: CLI tool to convert txt file to ebook format
 Keywords: cjk,ebook,epub,gmi,latex,md,pdf,txt,typst
 Author-email: Kian-Meng Ang <kianmeng@cpan.org>

{txt2ebook-0.1.151 → txt2ebook-0.1.152}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "txt2ebook"
-version = "0.1.151"
+version = "0.1.152"
 description = "CLI tool to convert txt file to ebook format"
 authors = [{ name = "Kian-Meng Ang", email = "kianmeng@cpan.org" }]
 requires-python = "~=3.9"

{txt2ebook-0.1.151 → txt2ebook-0.1.152}/src/txt2ebook/languages/zh_cn.py

@@ -45,6 +45,7 @@ RE_CHAPTERS = [
     "作者[介绍自介].*",
     "正文",
     "人物谱",
+    "作者按",
 ]
 
 DEFAULT_RE_AUTHOR = r"作者：(.*)"

{txt2ebook-0.1.151 → txt2ebook-0.1.152}/src/txt2ebook/subcommands/massage.py

@@ -22,9 +22,11 @@ from importlib import import_module
 from pathlib import Path
 
 import cjkwrap
+import jieba.analyse
 import regex as re
 from bs4 import UnicodeDammit
 
+from txt2ebook import detect_and_expect_language
 from txt2ebook.exceptions import EmptyFileError
 from txt2ebook.models.book import Book
 from txt2ebook.zh_utils import zh_halfwidth_to_fullwidth, zh_words_to_numbers
@@ -70,7 +72,22 @@ def build_subparser(subparsers) -> None:
         default=False,
         action="store_true",
         dest="fullwidth",
-        help="use fullwidth character (only for zh-cn and zh-tw)",
+        help=(
+            "use fullwidth character (only for zh-cn and zh-tw) "
+            "(default: %(default)r)"
+        ),
+    )
+
+    massage_parser.add_argument(
+        "-ri",
+        "--reindent",
+        default=False,
+        action="store_true",
+        dest="reindent",
+        help=(
+            "reindent each paragraph (only for zh-cn and zh-tw) "
+            "(default: %(default)r)"
+        ),
     )
 
     massage_parser.add_argument(
@@ -318,12 +335,18 @@ def massage_txt(args: argparse.Namespace) -> str:
 
     content = to_unix_newline(content)
 
+    args.language = detect_and_expect_language(content, args.language)
+
     (metadata, body) = extract_metadata_and_body(args, content)
 
     if args.fullwidth and args.language in ("zh-cn", "zh-tw"):
         logger.info("Convert halfwidth ASCII characters to fullwidth")
         body = zh_halfwidth_to_fullwidth(body)
 
+    if args.reindent and args.language in ("zh-cn", "zh-tw"):
+        logger.info("Reindent paragraph")
+        body = do_reindent_paragraph(args, body)
+
     if args.re_delete:
         body = do_delete_regex(args, body)
 
@@ -339,7 +362,7 @@ def massage_txt(args: argparse.Namespace) -> str:
     if args.width:
         body = do_wrapping(args, body)
 
-    return f"{metadata}\n\n{body}"
+    return f"{metadata}{body}"
 
 
 def to_unix_newline(content: str) -> str:
@@ -353,6 +376,30 @@ def to_unix_newline(content: str) -> str:
     """
     return content.replace("\r\n", "\n").replace("\r", "\n")
 
+def do_reindent_paragraph(args, content: str) -> str:
+    """Reindent each paragraph.
+
+    Args:
+        content(str): The formatted book content.
+
+    Returns:
+        str: The formatted book content.
+    """
+    paragraphs = re.split(r'\n\s*\n+', content)
+    reindented_paragraphs = []
+    for paragraph in paragraphs:
+        lines = paragraph.split('\n')
+        reindented_lines = []
+        for line in lines:
+            stripped_line = line.strip()
+            reindented_lines.append(stripped_line)
+
+        reindented_paragraph = '\n'.join(reindented_lines)
+        reindented_paragraph = "　　" + reindented_paragraph
+        reindented_paragraphs.append(reindented_paragraph)
+
+    return args.paragraph_separator.join(reindented_paragraphs)
+
 
 def do_delete_regex(args, content: str) -> str:
     """Remove words/phrases based on regex.
@@ -425,7 +472,24 @@ def extract_metadata_and_body(_args, content: str) -> tuple:
         metadata = match.group(0).strip()
         body = content.replace(metadata, "", 1)
 
-    return (metadata, body)
+
+    metadata_block = metadata.split("---")[1]
+
+    metadata_dict = {}
+    for line in metadata_block.strip().splitlines():
+        key, value = line.split("：", 1)
+        metadata_dict[key.strip()] = value.strip()
+
+    tags = jieba.analyse.extract_tags(content, topK=100)
+    metadata_tags = " ".join(tags)
+    logger.info("tags: %s", metadata_tags)
+    metadata_dict["索引"] = metadata_tags
+
+    meta_lines = [f"{key}：{value}" for key, value in metadata_dict.items()]
+    meta_body = "\n".join(meta_lines)
+    meta_str = f"---\n{meta_body}\n---"
+
+    return (meta_str, body)
 
 
 def do_single_newline(args, content: str) -> str:

{txt2ebook-0.1.151 → txt2ebook-0.1.152}/src/txt2ebook/subcommands/typ.py

@@ -72,7 +72,7 @@ def build_subparser(subparsers) -> None:
 
     typ_parser.add_argument(
         "--toc",
-        default=False,
+        default=True,
         action=argparse.BooleanOptionalAction,
         dest="with_toc",
         help="add table of content",