added exception handing & query parameter parsing for comics where fi…

[hon1nbo]

Fix PR for Issue #110

Whilst adding "Quantum Vibe" I found a bug in the URL parser implementation within comic.py

parsed_filepath = urlparse(url).path
file_extension = parsed_filepath[parsed_filepath.rindex(".") :]

This will throw a ValueError exception when the image is identified within a query parameter rather than a URL path.
Example that throws exception from "Quantum Vibe"
https://quantumvibe.com/disppageV3?story=qv&file=/simages/qv/qv1-001.jpg

To remedy this, I implemented exception handling and query parsing to identify where the filename/extension lives within the full URL. If none can be identified, it will gracefully save the file with a .unknown extension to at least preserve the data and allow later recovery through a rename operation when the file type is known. I could probably implement some MIME detection here in that scenario, but have not yet.

Sample config for Quantum Vibe for testing:

    "QuantumVibe": {
        "name": "QuantumVibe",
        "start_url": "https://quantumvibe.com/strip?page=1",
        "comic_image_selector": "//a[contains(@href, 'strip?page=')]//img[contains(@src, 'disppage')]/@src",
        "next_page_selector": "//a[img[contains(@src, 'nav/NextStrip2.gif')]]/@href",
    },

Limitations with this draft PR:

• Does not gracefully handle an edge case where the path has an extension but it's not the image extension which is located deeper in the URL ( example: something.php?comic=somecomic&page=pagexyz.jpg)
• Despite logic changes leaving the version number change to package maintainer.

[hon1nbo]

Looks like this is failing on JackRabbit again; I swapped it to the line you had edited in the last PR.

Testing locally I couldn't get it running still after you proposed the change. I opened a separate PR to with a change to jackrabbit as a hard comparison of @rel='Next' vs the contains function. Building with poetry and running the downloader with that seems to be running smoothly.

[coveralls]

coverage: 92.951% (-1.5%) from 94.439%
when pulling b02e708 on hon1nbo:queryparsing
into 28bed86 on J-CPelletier:master.

[J-CPelletier]

1. We'd probably want to extract this into a different method (like get_file_path that returns both file path and extension)
2. We could also use the parse_qs(...) method available in the urllib.parse module. Here's an example of what the results would look like:

>>> from urllib.parse import parse_qs
>>> parts = urlparse("https://quantumvibe.com/disppageV3?story=qv&file=/simages/qv/qv1-001.jpg")
>>> parse_qs(parts.query)
{'story': ['qv'], 'file': ['/simages/qv/qv1-001.jpg']}

3. We should also add unit tests for the extracted method

[hon1nbo]

ah yeah that parse_qs would probably be more efficient, I didn't catch that in urlparse documentation.

I'll break this out in the next couple days due to #job

[hon1nbo]

Quick comment, I haven't forgotten about this life stuff just happened. I still plan on a refactor