HN user

dtreacy

21 karma
Posts3
Comments5
View on HN

If you're able to input invalid URL's I guess I didn't test the regex thoroughly enough ;) I'll look into it. The prepending "http" is just a lazy hack to handle the case where they enter "google.com" instead of "http://www.google.com, which would fail the regex, but is a valid URL. This would be deprecated by improving the regex validation. Keep in mind I threw this together in an afternoon.... But thanks for the critique ;)

Good point. Yes, in theory they could be duplicated. I chose 6 characters firstly to keep the url as short as possible (I thought 5 chars was too short, 7 too long).

I am planning on introducing a background process to clean up shortened URL's that haven't been accessed in a given time frame (still deciding but I think around 7 days). While it still wouldn't solve the problem you mentioned at a certain scale, it would help. But true, I very much doubt it will ever reach that scale.

Thanks for the critique