javascript - Data Structure for storing partial urls where search speed is the priority -

we have large database of partial urls (strings) such as:

• "example1.com"

• "example2.com/test.js"

• "/foo.js"

our software listens http requests , tries find 1 of our database's partial urls in http request's full url.

so getting full urls (i.e.: http://www.example.com/blah.js?foo=bar") , trying match 1 of our database's partial patterns on it.

which best data structure store our partial urls database on if care search speed?

---

right now, do:

• iterating through entire database of partial urls (strings) , using indexof (in javascript) see if full url contains each partial string.

update:

this software extension firefox written in javascript on firefox's addon sdk.

assuming partial strings domain names and/or page names try generate possible combinations url starting end:

 http://www.example.com/blah.js?foo=bar  blaj.js  example.com/blah.js  www.example.com/blah.js 

then hash combinations, store them in array , try find of them in array contains hashes of partial strings database.

note:

in case want match string in url, ample in example.com becomes little complicated in terms of storage, because random combinations of strings in url

where n length of url , k length of string find. according this question maximum reasonable length of url 2000 characters. , assuming want match random string you'd have k vary 1 2000 result in large amount of hashes generated url - sum of n on k each k 1 2000. or more precisely - 2000! / (k!*(2000-k)!) different hashes