jthompson18 / comp330 Goto Github PK

To start, I will sort of nitpick at your directory naming because nitpicking means looking for the small (allegedly unimportant) issues. It makes sense on how you got to 'messager' (vs. messenger) but something tells me that it could have been a typo. However, we are all artists in our own way so I will accept that this was actually your intention and in the grand scale of things, is not a big deal either way.

Regarding the re lib, we certainly agreed on some things #teamfindall() but I would argue that you should compile your regexs when you declare them. This is for a couple reasons:

1. From my understanding it is not very time/data consuming to do so, especially on a project of this size.
2. It allows for your regex to be readily available to use, especially if it is used in more that one case.

I'm not sure if I am interpreting the script correctly but from what I can see all the URI's being tested are checking for http://. If this is the case then would the tests fail for links that are automatically shortened by twitter?
For example:
LIVE: President Obama is speaking in Springfield, Illinois. ofa.bo/z096

If I am incorrect in this then the issue can be closed.

Hi. I know you did this very intensely and carefully.

So. frick you, because I have to double check everything to find somewhere to give you some constructive criticism.

I did! Actually. But it was ruff.

Your test cases cover everything. They're also super sterile. Im of the idea you should have some complex dirty test that still passes.

Example:

#message% is a super sterile test
#message%@ Hi shouldnotparse #hello is a more... rough treatment of your code

I also didn't see any tests for empty strings. But I might have missed them, python isn't my thing.

Finally, a joke:
QA Engineer walks into a bar. Orders a beer. Orders 0 beers. Orders 999999999 beers. Orders a lizard. Orders -1 beers. Orders a sfdeljknesv.

applicable. test for some lizards, homie. make your tests designed to fail.