remove_tags won't remove

[bubak]

Hi all,

I'm trying to set up a recipe for mandiner.hu, so far with mixed success. First of all, the builtin recipe does not work at all. When I fix it, I get several problems:

- the output has a lot of errors "Could not fetch image /assets/images..." and "AttributeError: 'str' object has no attribute 'get_full_url'", even when I put dict(name='img') into remove_tags

- other tags in remove_tags also have no effect, the tags are still present in the html output created by ebook-convert "$@" .epub --test -vv --debug-pipeline debug, so I have no means to remove garbage from the html output.

There is probably something obvious I'm missing, but I cannot find any reason why remove_tags would not work. Should I use preprocess_html or even preprocess_raw_html instead?