Test end to end #30
Test end to end #30
Conversation
| st.lists(st.text(ascii_letters + digits + punctuation, min_size=1), min_size=1) | ||
| ) | ||
|
|
||
| frequencies = draw( |
There was a problem hiding this comment.
ℹ️ This is nice - just a thought from reading this: It might be useful to also support supplying an already-created Counter object, especially if people have already processed huge lists of words.
There was a problem hiding this comment.
good idea, I'll make an issue
| ): | ||
| if seed is not None: |
There was a problem hiding this comment.
ℹ️ Suggestion: We should not set the global seed, but we can keep our own local Random instance. That way, people can get determinism from wrdcld without upsetting the randomness in the rest of their script (whatever it may be).
There was a problem hiding this comment.
but we don't touch the random state unless a seed is supplied, no? If the user is controlling it globally then they just wouldn't touch this?
There was a problem hiding this comment.
The point is we do touch the (global) random state. If the user wanted to do that, they could just call random.seed themselves before calling make_word_cloud. If we instantiate our own random generator and only seed that one, then this argument only affects randomness within wrdcld, and nothing else in a user's library or script. They may want "true" randomness elsewhere but a handy way to get determinism from wrdcld.
(But we can do this in a separate PR for sure)
| @@ -86,7 +86,7 @@ def fill_next_word(word, available_rectangles, image, font, frequency): | |||
| options.append("vertical") | |||
|
|
|||
| if not options: | |||
| print(f"skipping word '{word}', couldn't find a good rectangle") | |||
| # print(f"skipping word '{word}', couldn't find a good rectangle") | |||
There was a problem hiding this comment.
ℹ️ Idea (not for this PR perhaps): We move this to a logger, and add a silent: bool flag to the main function that allows people to toggle this output on or off.
There was a problem hiding this comment.
Agreed! I would also like to give some more thought to how silently we fail, when there is a word which is high up in the frequency list, but that we can't fit in the diagram for whatever reason (assuming this happens at all)
|
the tests and stylechecks didn't run for some reason, I'll investigate later... |
No description provided.