fix: slugify filenames

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "epub2go"
-version = "2.2"
+version = "2.2.3"
 description = "EPUB converter using wget, pandoc and python glue"
 readme = "README.md"
 requires-python = ">=3.12"

src/epub2go/convert.py

@@ -7,7 +7,7 @@ from tqdm import tqdm
 from pyfzf.pyfzf import FzfPrompt
 import click
 
-import os, subprocess, shlex, logging
+import os, subprocess, shlex, logging, re
 import importlib.resources as pkg_resources
 from dataclasses import dataclass
 from typing import List
@@ -46,7 +46,7 @@ class GBConvert():
         cleanpages: bool = True,
     ):
         tocpage = os.path.dirname(url) # ToC website url
-        dir_output = self.getDir()
+        dir_output = self.getDir(url)
         logger.debug('Downloading to %s, expecting files in in %s', self.dir_download, dir_output)
         author = author
         title = title
@@ -105,8 +105,7 @@ class GBConvert():
     def create_epub(self, author, title, chapters, dir_output):
         #TODO --epub-cover-image
         #TODO toc if it isnt described by <h> tags, e.g. https://www.projekt-gutenberg.org/adlersfe/maskenba/
-        filename = f'{title} - {author}.epub'
+        filename = slugify(f'{title} - {author}.epub')
-        logger.debug('Creating epub as "%s"',filename)
         command = f'''pandoc -f html -t epub \
                     -o "{filename}" \
                     --reference-location=section \
@@ -115,6 +114,7 @@ class GBConvert():
                     --metadata author="{author}" \
                     --epub-title-page=false \
                     {" ".join(chapters)} '''
+        logger.debug('Calling "%s"', command)
         subprocess.run(shlex.split(command), cwd=dir_output, check=True)
         return os.path.abspath(os.path.join(dir_output,filename))
 
@@ -160,6 +160,13 @@ def get_all_books() -> List[Book]:
                 books.append(book)
     return books
 
+def slugify(value, replacement='_'):
+    value = re.sub(r'[<>:"/\\|?*\x00-\x1F]', replacement, value)
+    # Remove leading/trailing whitespace or dots
+    value = value.strip().strip(".")
+    # Optionally truncate to safe length (e.g. 255 chars for most filesystems)
+    return value[:255] or "untitled"
+
 # run main cli
 @click.command()
 #TODO include images flag