Very basic python tool that extracts resources from an MHTML file and saves them to a directory for viewing / further processing.
Usage:
python3 unmht.py example.mhtml
Default output directory : "/files". To change, add option "-o path/to/directory".
getCss.py can generate html tags to include all extracted css files, makes it easier to copy into the main html file.
Fixed:
HTML file without extension is added ".html" extension.
Resources with question marks, eg 'src="/path/to/script.js?v=3"', the question mark and everything that follows are removed.
Resources with special characters, eg 'src="/path/to/some%20image%20file.jpg"' or css files whose location is "cid:****@mthml.blink", the special characters are replaced with underscores.
Known issues:
Sometimes some links to style sheets are removed from the main html file, I don't know why. The getCss.py is created for manually adding the tags.
It seems that it's unable to parse "multipart/alternative" content type.
Current goal:
Converting "mht" and "mhtml" files into single html files, similar to what SingleFile extension does. It seems that SingleFile can't process local html file that links to local resources. A workaround is, run a python server in a directory, then use unmht.py to extract the html and resource files into that folder, open server in browser, then use SingleFile to convert it into a single html file.
Ultimate goal:
Convert "mht" or "mhtml" file directly into a single html file. Probably it can be implemented by using a dict to get all resource locations and data, then replace the links in the html data with the actual resource data.
- perl unmht (crude parsing, did not work for me)
- mpack (doesn’t preserve filenames)
- mht2htm (OMG pascal!)