403 Error because tweets are too long

[fakebenjay]

Hey,
I've been running the trackthenews script and when it encounters an article title that's too long, it returns a 403 error:

Traceback (most recent call last):
  File "/home/ubuntu/.pyenv/versions/3.6.5/bin/trackthenews", line 11, in <module>
    sys.exit(main())
  File "/home/ubuntu/.pyenv/versions/3.6.5/lib/python3.6/site-packages/trackthenews/core.py", line 441, in main
    article.tweet()
  File "/home/ubuntu/.pyenv/versions/3.6.5/lib/python3.6/site-packages/trackthenews/core.py", line 115, in tweet
    twitter.update_status(status=status, media_ids=media_ids)
  File "/home/ubuntu/.pyenv/versions/3.6.5/lib/python3.6/site-packages/twython/endpoints.py", line 114, in update_status
    return self.post('statuses/update', params=params)
  File "/home/ubuntu/.pyenv/versions/3.6.5/lib/python3.6/site-packages/twython/api.py", line 268, in post
    return self.request(endpoint, 'POST', params=params, version=version)
  File "/home/ubuntu/.pyenv/versions/3.6.5/lib/python3.6/site-packages/twython/api.py", line 258, in request
    api_call=url)
  File "/home/ubuntu/.pyenv/versions/3.6.5/lib/python3.6/site-packages/twython/api.py", line 194, in _request
    retry_after=response.headers.get('X-Rate-Limit-Reset'))
twython.exceptions.TwythonError: Twitter API returned a 403 (Forbidden), Tweet needs to be a bit shorter.

Is there a workaround for this?

[thisisparker]

Hm, no, I hadn't run into that! It's easy enough to truncate (probably with a …) which I should include in a new version. Out of curiosity, what story has such a long title?

[fakebenjay]

I lost the exact title, but it was a long FDA announcement about some technical drug-related thing. Very dense and dry gov't agency stuff.

[fakebenjay]

Also, while I'm on this issue, I've also gotten a KeyError: title traceback error. Any idea what that might be?

Traceback (most recent call last):
  File "/home/ubuntu/.pyenv/versions/3.6.5/bin/trackthenews", line 11, in <module>
    sys.exit(main())
  File "/home/ubuntu/.pyenv/versions/3.6.5/lib/python3.6/site-packages/trackthenews/core.py", line 420, in main
    articles = parse_feed(outlet, url, delicate, redirects)
  File "/home/ubuntu/.pyenv/versions/3.6.5/lib/python3.6/site-packages/trackthenews/core.py", line 185, in parse_feed
    title = entry['title']
  File "/home/ubuntu/.pyenv/versions/3.6.5/lib/python3.6/site-packages/feedparser.py", line 356, in __getitem__
    return dict.__getitem__(self, key)
KeyError: 'title'

[thisisparker]

Huh, that would happen if an RSS feed entry doesn't have the (required) title field, probably. It doesn't happen much (I haven't seen it in the wild!) but I guess people put all kinds of RSS feeds out there, huh? This should probably fail a little more gracefully, though! Thanks for the report!

[fakebenjay]

Thank you! Will try running without that problem feed temporarily...

[fakebenjay]

Just a heads up about that KeyError, this is the problem feed: https://www.michigan.gov/rss/0,2348,7-154-11472--53088-,00.xml

In theory, I'd imagine you wouldn't want the entire program to crap out because of one faulty article. Is there a way to make it continue, even if one blog post was untweetable?

[thisisparker]

Yes, and honestly it should probably tweet without a title just as it does without an outlet. I'll plan to fix that, thank you!

[matthewmcvickar]

I've submitted a PR that I think addresses this: #22

[matthewmcvickar]

And another that addresses potentially empty title/URL: #23