aerkalov/ebooklib

AttributeError: 'bytes' object has no attribute 'encode'. Did you mean: 'decode'?

pjz opened this issue · 2 comments

pjz commented

I'm working on a CLI for editing ebook metadata (called ebmeta, available on pypi if you care), and I found this:

Traceback (most recent call last):
  File "/home/pj/.local/pipx/venvs/ebmeta/lib/python3.10/site-packages/ebooklib/utils.py", line 35, in parse_string
    tree = etree.parse(io.BytesIO(s.encode('utf-8')))
AttributeError: 'bytes' object has no attribute 'encode'. Did you mean: 'decode'?

It was in a long logfile and among a bunch of other errors, but it looks fairly easy to fix.

pjz commented

This was me misunderstanding the way the code works. Malformed XML threw this as part of the error due to the re-tried call in the except: block. Maybe parse_string () starting at line 35 could be rewritten like:

if isinstance(s, str):
    s = s.encode('utf-8')
return etree.parse(io.BytesIO(s) , parser=parser)

to avoid confusing nested exceptions?

c1924959470 commented