Skip to main content
The M440 handler provides a lightweight extraction method for M440.in and Mangas.in using simple HTML parsing without AI assistance.

​
Supported domains

The handler supports multiple domains: 
@staticmethod
def get_supported_domains() -> list:
    return ["m440.in", "mangas.in"]
Location: ~/workspace/source/core/sites/m440.py:18-20

​
URL patterns

Single chapter: 
https://m440.in/manga/[series-name]/[chapter-id]
https://mangas.in/manga/[series-name]/[chapter-id]
Series cover: 
https://m440.in/manga/[series-name]
https://mangas.in/manga/[series-name]

​
Extraction technology

​
Simple Crawl4AI

M440 uses Crawl4AI without AI assistance, relying on straightforward regex extraction: 
async with AsyncWebCrawler(verbose=True) as crawler:
    result = await crawler.arun(url=url, bypass_cache=True)
    if not result.success:
        log_callback(f"[ERROR] Page load failed: {result.error_message}")
        return
    
    html = result.html
    matches = re.findall(r'data-src=["\']https://[^"\']+)["\']', html)
Location: ~/workspace/source/core/sites/m440.py:32-36,131 This approach is:
  • Faster than AI extraction
  • More reliable for simple HTML structures
  • Requires no API keys
  • Uses less CPU and memory

​
Cover detection

The handler intelligently detects whether a URL is a cover page or single chapter: 
clean_url = url.split("?")[0].rstrip("/")
is_cover_page = bool(re.search(r'/manga/[^/]+$', clean_url))

if not is_cover_page:
    # Check if there are many chapter links
    potential_chapters = re.findall(
        r'href=["\']https://m440.in/manga/[^/]+/[^"\']+)["\']', 
        html
    )
    if len(set(potential_chapters)) > 3: 
        is_cover_page = True
Location: ~/workspace/source/core/sites/m440.py:39-44 A page is considered a cover if:
  1. URL ends with /manga/[series-name] (no chapter path)
  2. OR it contains more than 3 unique chapter links

​
Series download

When a cover page is detected, the handler downloads all chapters:

​
Title extraction

manga_title = "Manga_M440"
title_match = re.search(
    r'<h2[^>]*class=["\']widget-title["\'][^>]*>(.*?)</h2>', 
    html
)
if title_match: 
    manga_title = clean_filename(title_match.group(1).strip())
Location: ~/workspace/source/core/sites/m440.py:48-51

​
Chapter extraction

links = re.findall(
    r'href=["\']https://m440.in/manga/[^/]+/[^"\']+)["\']', 
    html
)

seen = set()
clean_links = []
for l in links:
    if l not in seen and "/manga/" in l and l != url:
        seen.add(l)
        clean_links.append(l)

clean_links.reverse()  # Oldest to newest
Location: ~/workspace/source/core/sites/m440.py:53-60 The handler:
  1. Extracts all chapter URLs
  2. Removes duplicates while preserving order
  3. Filters out the current URL
  4. Reverses to start with chapter 1

​
Batch processing

pdf_dir = os.path.join(os.getcwd(), config.PDF_FOLDER_NAME, manga_title)
os.makedirs(pdf_dir, exist_ok=True)

for i, chap_url in enumerate(clean_links):
    if check_cancel and check_cancel(): 
        break
    if progress_callback: 
        progress_callback(i + 1, len(clean_links))
    
    log_callback(f"Processing Cap {i+1}/{len(clean_links)}")
    
    pdf_name = f"{manga_title} - {chap_url.split('/')[-1]}.pdf"
    full_pdf_path = os.path.join(pdf_dir, pdf_name)
    
    if os.path.exists(full_pdf_path): 
        continue
    
    await self._process_chapter(
        chap_url, 
        full_pdf_path, 
        crawler, 
        log_callback, 
        check_cancel, 
        None
    )
Location: ~/workspace/source/core/sites/m440.py:68-81 Features:
  • Creates series-specific folder
  • Skips already-downloaded chapters
  • Respects cancellation requests
  • Reports progress per chapter

​
Image extraction

M440 uses data-src attributes for lazy-loaded images: 
async def _process_chapter(
    self, 
    url: str, 
    output_pdf_path: str, 
    crawler: AsyncWebCrawler, 
    log_callback: Callable[[str], None], 
    check_cancel: Callable[[], bool], 
    progress_callback: Optional[Callable[[int, int], None]] = None
) -> None:
    result = await crawler.arun(url=url, bypass_cache=True)
    if not result.success: 
        return
    
    html = result.html
    matches = re.findall(r'data-src=["\']https://[^"\']+)["\']', html)
    
    if matches:
        images = list(dict.fromkeys(matches))  # Remove duplicates
        log_callback(f"[INFO] Downloading {len(images)} images...")
        
        await download_and_make_pdf(
            images, 
            output_pdf_path, 
            config.HEADERS_M440, 
            log_callback, 
            check_cancel, 
            progress_callback, 
            is_path=True,
            open_result=config.OPEN_RESULT_ON_FINISH
        )
Location: ~/workspace/source/core/sites/m440.py:111-144

​
Deduplication

The handler uses dict.fromkeys() to remove duplicate URLs while preserving order: 
images = list(dict.fromkeys(matches))
This is more efficient than: 
images = list(set(matches))  # Order not preserved

​
Headers and configuration

M440 requires specific headers for image downloads: 
HEADERS_M440 = {
    "User-Agent": "Mozilla/5.0 ...",
    "Referer": "https://m440.in/"
}
Defined in config.py and passed to the download utility.

​
Usage examples

​
Single chapter

from core.handler import process_url

await process_url(
    "https://m440.in/manga/one-piece/chapter-1000",
    log_callback=print,
    check_cancel=lambda: False
)
Output: PDF/m440_chapter.pdf

​
Full series

await process_url(
    "https://m440.in/manga/one-piece",
    log_callback=print,
    check_cancel=lambda: False,
    progress_callback=lambda current, total: print(f"Chapter {current}/{total}")
)
Output: 
PDF/
└── One Piece/
    ├── One Piece - chapter-1.pdf
    ├── One Piece - chapter-2.pdf
    └── One Piece - chapter-3.pdf

​
Via web interface

  1. Launch: START_WEB_VERSION.bat
  2. Open: http://localhost:3000
  3. Paste URL: https://m440.in/manga/my-manga
  4. Watch progress in real-time

​
Implementation details

​
Class structure

class M440Handler(BaseSiteHandler):
    @staticmethod
    def get_supported_domains() -> list:
        return ["m440.in", "mangas.in"]
    
    async def process(self, url, log_callback, check_cancel, progress_callback):
        """Process M440.in URL."""
        ...
    
    async def _process_chapter(self, url, output_pdf_path, crawler, ...):
        """Helper to process a single chapter."""
        ...
Location: ~/workspace/source/core/sites/m440.py:15-145

​
Reusing crawler instance

For series downloads, the handler reuses the same AsyncWebCrawler instance: 
async with AsyncWebCrawler(verbose=True) as crawler:
    result = await crawler.arun(url=url, bypass_cache=True)
    # ... cover page processing ...
    
    for i, chap_url in enumerate(clean_links):
        await self._process_chapter(
            chap_url, 
            full_pdf_path, 
            crawler,  # Reuse instance
            ...
        )
Location: ~/workspace/source/core/sites/m440.py:32,81 This improves performance by avoiding repeated browser initialization.

​
Known limitations

​
No lazy loading script

Unlike ZonaTMO and TMO-H, M440 doesn’t execute JavaScript to trigger lazy loading. This works because M440’s data-src attributes are present in the initial HTML.

​
Chapter naming

Chapters are named using the URL slug: 
pdf_name = f"{manga_title} - {chap_url.split('/')[-1]}.pdf"
Location: ~/workspace/source/core/sites/m440.py:76 For example:
  • URL: https://m440.in/manga/one-piece/chapter-1000
  • Output: One Piece - chapter-1000.pdf
This preserves the site’s chapter naming convention.

​
Skip existing files

The handler skips already-downloaded chapters: 
if os.path.exists(full_pdf_path): 
    continue
Location: ~/workspace/source/core/sites/m440.py:79
This allows you to resume interrupted series downloads without re-downloading existing chapters.

​
Performance characteristics

  • Speed: Very fast (no AI processing)
  • Reliability: High (simple regex)
  • Resource usage: Low (minimal CPU/memory)
  • Best for: Large series downloads

​
Comparison with other handlers

HandlerExtractionSpeedAPI Required
M440RegexFastNo
ZonaTMOAI + RegexMediumYes
TMO-HAI + RegexMediumYes
H2RJSONFastestNo

​
Troubleshooting

​
No chapters found

If cover detection fails:
  1. Verify the URL is correct
  2. Check if the site structure has changed
  3. Manually inspect the page HTML
  4. Try a different manga series

​
Missing images

If some images don’t download:
  1. Check if headers are correct in config.py
  2. Verify the regex pattern matches data-src format
  3. Some images may require additional authentication

​
Auto-open not working

On Linux/Mac, os.startfile() is not available: 
if config.OPEN_RESULT_ON_FINISH:
    try: 
        os.startfile(pdf_dir)
    except: 
        pass
Location: ~/workspace/source/core/sites/m440.py:83-85 You’ll need to manually open the output folder.

​
Next steps

Hentai2Read

Even faster JSON-based extraction

ZonaTMO

Compare with AI-powered approach

Configuration

Configure headers and paths

Utils

Learn about download_and_make_pdf

Build docs developers (and LLMs) love

Get started for freeTalk to us