Suggestion: Not parseable resources ->parseable resources

I took a peak at your source code. One source for crawling issues is that you currently define in the code not_parseable_ressources. Instead, if you define parseable resources and limit those to only truly parseable resources that are are supported in the sitemap and may contain plain text html links, you can limit issues with unknown extensions. Also you might take a look at using mime types instead of file extensions. I am not sure how that works in Python though.

[c4software]

The not parseable ressource is more like a self guard to avoid some nasty case. But, its not a bad idea i will take a look.