Taking a crack at issue #14 #17
Conversation
| """ | ||
|
|
||
| # via Martijn Pieters, http://stackoverflow.com/questions/26568722/remove-unicode-emoji-using-re-in-python | ||
| if PY2: |
There was a problem hiding this comment.
@BrendanAdkins The u"string" syntax is supported by both Python 2 and 3 so you can drop the check here and the else below.
|
@BrendanAdkins 👍 Looks good so far! I like the @carpedm20 Any comments? How do you feel about the |
|
@geowurster Good call on the unit tests! You'll be glad to know they fail spectacularly on adjacent emoji. This is going to be a hex math thing, isn't it? I too could use some help with the regex here. I have to say I don't grasp the rationale for adding aliases to the demojize function--I can't see a use case for it, and it seems like it would be violating least-astonishment just for the sake of symmetry. |
|
Looks like the regex doesn't include some character ranges that are treated as emoji but not technically defined as emoji--working on tracking those down. |
|
@BrendanAdkins Good point about supporting aliases. The user gets an arbitrary emoji back so there really isn't a need to support the aliases. Unfortunately I can't offer any help with the adjacent emoji issue, but I'll ask some of my co-workers. |
|
These flag emoji are going to be the death of me. |
|
@BrendanAdkins Thanks for doing the initial R&D on this! |
I'm pretty new to Python, but this seems to work. Having trouble safely ensuring that a given string sent to demojize is interpreted as Unicode, so right now I've just added that as a requirement going in.