Closed ealashwali closed 5 years ago
First, there is a canonicalization step, which ensures that URLs have at least a path of "/" (even if none was supplied, as in the google.com case
After canonicalization, the given url is turned into a variety of prefix/suffix expression combinations as described here: https://developers.google.com/safe-browsing/v4/urls-hashing#suffixprefix-expressions
Those expressions (which are what is actually hashed), do not have the scheme (http://) in them.
So am I understanding correctly that if I supplied a list with https://
leading the check will be ok. I guess so, ut I will appreciate confirming.
In the example of the API HTTP request URL entries in the query body, the URLs are represented in two different format (e.g. the second entry is with
http://
and ends with/
, while the first one withouthttp://
,https://
and does not end with/
.Does adding
http://
,https://
or ending the URL with/
make difference? I tested with and without them and there seems no difference. But I wonder, how this is the case while the hashes with or without them is different. Can you plz briefly clarify?Here is the example in your home page that I am referring to: