g0dc0der / click-count Goto Github PK

Since the logic layer should be unaware of the rest layer.

http://www.google.com/recaptcha/intro/index.html
https://github.com/VividCortex/angular-recaptcha

FileServlet should start with "/pages" and ApplicationPath should be "/"

For example, we getting 404, redirect to error page with a key query paramteter, which is fetched with an ajax request to get the full data of the error message.

This means updating all entities and DAOs.

link -> sourceURL
url -> alias

Basically, if something is already taken, let the db throw exception and catch it in the facade. From there, forward the error to the rest service which in turn return an appropriate response.

• How much logging do we really need?
• A way to report user?
• Banning users?
• Stop curling by random people?
• Can I use port 9000(non-80) for my server?
• What do I need to do to protect myself from hackers and other saboteurs?

• The requests need to be logged.
• Hibernate logging must be turned to INFO or SEVERE.

Rename

• Move to new class
• New annotation to enable captcha
• New filter to handle the semantics

For example, amazon rejects it when the page is actually working.
I think we have to download the entire page to actually check.

Click-Count will work with unique clicks. So one IP address can only increase the counter for a given URL once every hour.

Always respond with URL, just not increase counter if disallowed.
This logic should be done asynchronously to keep fast redirects.