fix: slugify filenames

errors from wget and pandoc are thrown up

README.md

@@ -1,18 +1,39 @@
 # epub2go.py
-web to epub converter for https://projekt-gutenberg.org.
+Web to ePUB Converter for [projekt-gutenberg.org](https://projekt-gutenberg.org)  developed in conjunction with a [web interface](https://github.com/eneller/epub2go-web).
+
+## Installation
 Requires:
 - [pandoc](https://pandoc.org/)
 - [wget](https://www.gnu.org/software/wget/)
-- [fzf](https://github.com/junegunn/fzf) (only for interactive mode)
+- [fzf](https://github.com/junegunn/fzf) (optional, only for interactive mode)
-- python (duh)
+- [python](https://www.python.org/) (duh)
-## Usage
+
-Invoke the script using the url of any page of the book you would like to download:
-``` 
-epub2go https://www.projekt-gutenberg.org/ibsen/solness/
-```
-## Installation
 Assuming you have a recent version of python installed, run
+
 ```
 pip install git+https://github.com/eneller/epub2go.py
 ```
-   This will provide the 'epub2go' command.
+This will provide the `epub2go` command.
+
+## Usage
+```
+Usage: epub2go [OPTIONS] [ARGS]...
+
+  Download ePUBs from https://www.projekt-gutenberg.org/
+
+  Provide either 0 arguments to enter interactive mode or an arbitrary number
+  of URLs to download from
+
+Options:
+  -d, --debug      Set the log level to DEBUG
+  -s, --silent     Disable the progress bar
+  -p, --path TEXT  The path to which files are saved
+  --no-clean       Do not parse html files with blocklist
+  --help           Show this message and exit.
+```
+
+Examples:
+```bash
+epub2go https://www.projekt-gutenberg.org/ibsen/solness/
+epub2go # will enter interactive mode
+```

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "epub2go"
-version = "1.3"
+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
@@ -23,75 +23,100 @@ class Book():
     url: str
 class GBConvert():
     def __init__(self,
-        url:str,
+        downloaddir,
-        author:str = None,
-        title:str = None,
-        downloaddir = './',
-        showprogress:bool = False,
         ):
         # NOTE move non-code files to data folder
         self.style_path_drama = pkg_resources.files('epub2go').joinpath("drama.css")
         with open(pkg_resources.files('epub2go').joinpath('blocklist.txt')) as blocklist:
             self.blocklist = blocklist.read().splitlines()
-        self.tocpage = os.path.dirname(url) # ToC website url
-        url = urlparse(self.tocpage)
         self.dir_download = downloaddir
-        self.dir_output = os.path.join(self.dir_download, url.netloc + url.path )# directories created by wget recreating the URL
-        logger.debug('Downloading in %s, expecting files in in %s', self.dir_download, self.dir_output)
-        self.showprogress = showprogress
-        self.author = author
-        self.title = title
-        self.chapters = []
 
-        self.parse_meta()
+    def getDir(self, url):
+        tocpage = os.path.dirname(url) # ToC website url
+        parsed_url = urlparse(tocpage)
+        # directories created by wget recreating the URL
+        dir_output = os.path.join(self.dir_download, parsed_url.netloc + parsed_url.path )
+        return dir_output
         
-    def parse_meta(self):
+    def download(self,
-        response = requests.get(self.tocpage)
+        url:str,
+        author:str = None,
+        title:str = None,
+        showprogress: bool = False,
+        cleanpages: bool = True,
+    ):
+        tocpage = os.path.dirname(url) # ToC website url
+        dir_output = self.getDir(url)
+        logger.debug('Downloading to %s, expecting files in in %s', self.dir_download, dir_output)
+        author = author
+        title = title
+
+        #parse_meta
+        response = requests.get(tocpage)
         response.raise_for_status()
         soup = BeautifulSoup(response.content, 'html.parser')
         # TODO allow setting these from interactive mode where those parameters are figured out from the list
-        if not self.author:
+        if not author:
             try:
-                self.author = soup.find('meta', {'name': 'author'})['content']
+                author = soup.find('meta', {'name': 'author'})['content']
             except:
-                self.author = "UnknownAuthor"
+                author = "UnknownAuthor"
-        if not self.title:
+        if not title:
             try:
-                self.title = soup.find('meta', {'name': 'title'})['content']
+                title = soup.find('meta', {'name': 'title'})['content']
             except:
-                self.title = "UnknownTitle"
+                title = "UnknownTitle"
-        self.toc = soup.find('ul').find_all('a')
+        chapter_urls = soup.find('ul').find_all('a')
-        logger.debug('Found ToC with %d entries', len(self.toc))
+        logger.debug('Found ToC with %d entries', len(chapter_urls))
 
-    def parse_toc_entry(self, entry):
+        #run
-        url = os.path.join(self.tocpage, entry['href'])
+        #TODO include images flag
+        # download all files in toc (chapters)
+        chapter_files = []
+        for item in (tqdm(chapter_urls) if showprogress else chapter_urls):
+            item_url = self.parse_toc_entry(tocpage, item)
+            parsed_url = urlparse(item_url)
+            filepath = os.path.join(self.dir_download, parsed_url.netloc + parsed_url.path)
+            if cleanpages: self.parse_page(filepath)
+            chapter_files.append(os.path.basename(item_url))
+        
+        return self.create_epub(author,title,chapter_files,dir_output)
+        
+    def parse_toc_entry(self, tocpage, entry):
+        url = os.path.join(tocpage, entry['href'])
         self.save_page(url)
         return url
 
     # apply blocklist to file
     def parse_page(self,file_path):
         #TODO clean up file opening, mmap?
-        logger.debug('Parsing page at %s', file_path)
+        count=0
         with open(file_path, 'r+') as f:
             soup = BeautifulSoup(f.read(), 'html.parser')
             for blocker in self.blocklist:
                 for item in soup.select(blocker):
                     item.decompose()
+                    count+=1
+            f.seek(0)
+            f.truncate()
             f.write(str(soup))
+        logger.debug('Removed %d tags from page %s during parsing', count, file_path)
 
-    def create_epub(self,  filename='out.epub')-> int:
+    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/
-        logger.debug('Creating epub as "%s"',filename)
+        filename = slugify(f'{title} - {author}.epub')
         command = f'''pandoc -f html -t epub \
                     -o "{filename}" \
                     --reference-location=section \
                     --css="{self.style_path_drama}" \
-                    --metadata title="{self.title}" \
+                    --metadata title="{title}" \
-                    --metadata author="{self.author}" \
+                    --metadata author="{author}" \
                     --epub-title-page=false \
-                    {" ".join(self.chapters)} '''
+                    {" ".join(chapters)} '''
-        return subprocess.run(shlex.split(command), cwd=self.dir_output).returncode
+        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))
 
     def save_page(self, url):
         logger.debug('Saving page at %s', url)
@@ -103,19 +128,7 @@ class GBConvert():
                     --tries=5 \
                     --quiet \
                     {url}'''
-        return subprocess.run(shlex.split(command), cwd=self.dir_download).returncode
+        subprocess.run(shlex.split(command), cwd=self.dir_download, check=True)
-    def run(self):
-        #TODO include images flag
-
-        # download all files in toc (chapters)
-        for item in (tqdm(self.toc) if self.showprogress else self.toc):
-            item_url = self.parse_toc_entry(item)
-            parsed_url = urlparse(item_url)
-            filepath = os.path.join(self.dir_download, parsed_url.netloc + parsed_url.path)
-            self.parse_page(filepath)
-            self.chapters.append(os.path.basename(item_url))
-        
-        return self.create_epub(f'{self.title} - {self.author}.epub')
 
 # get a list of all books for interactive selection or scraping
 def get_all_books() -> List[Book]:
@@ -147,14 +160,24 @@ 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
 @click.option('--debug', '-d', is_flag=True, help='Set the log level to DEBUG')
 @click.option('--silent', '-s', is_flag=True, help='Disable the progress bar')
+@click.option('--path','-p',type=str,default='./', help='The path to which files are saved' )
+@click.option('--no-clean',is_flag=True,help='Do not parse html files with blocklist')
 @click.argument('args', nargs=-1)
-def main(args, debug, silent):
+def main(args, debug, silent, path, no_clean):
     '''
-    Download ePUBs from https://www.projekt-gutenberg.org/
+    Download ePUBs from https://www.projekt-gutenberg.org/ \n
     Provide either 0 arguments to enter interactive mode or an arbitrary number of URLs to download from
     '''
     logging.basicConfig(level=logging.ERROR,format='%(asctime)s - %(levelname)s - %(message)s')
@@ -173,10 +196,12 @@ def main(args, debug, silent):
         books = [item.split(';')[1].strip() for item in selection]
 
     logger.debug('Attempting to download from %d URL(s)', len(books))
+    converter = GBConvert(path)
     if len(books)==1:
-        GBConvert(books[0], showprogress=not silent).run()
+        converter.download(books[0], showprogress=not silent, cleanpages= not no_clean)
     else:
         for book in (tqdm(books) if not silent else books):
-                GBConvert(book).run()
+            converter.download(book, cleanpages= not no_clean)
+
 if __name__ == "__main__":
     main()

src/epub2go/crawl.py

@@ -11,9 +11,10 @@ from convert import GBConvert, allbooks_url, get_all_books, Book
 def main():
     books = get_all_books()
     # NOTE consider making this a map()
+    converter = GBConvert('./')
     for book in tqdm(books):
         if book.url is not None:
-            GBConvert(book.url).run()
+            converter.download(book.url)
 
 
 if __name__ == "__main__":

uv.lock

@@ -81,7 +81,7 @@ wheels = [
 
 [[package]]
 name = "epub2go"
-version = "1.2"
+version = "2.2"
 source = { editable = "." }
 dependencies = [
     { name = "beautifulsoup4" },