This is a web parser for Registro Publico de Panama to look for companies registered (useful for any newspaper).
- Using Akka 2.0 in Master/Worker classs
- Pimp My Library in JSoupExtensions
- Cake Pattern, which I'm not sure is the correct implementation, for the Storage trait
- jSoup for HTML parsing
- Have a trait for either ScalaJ or Dispatch. Using ScalaJ because Dispatch requires to set the correct charset
- MongoDB for storage
- Java wrapper for creting runnable jar
Take note that this is not using SBT , which I tried but is failing to work properly.
##Libraries
- Akka 2.0
- Casbah commons, core and query
- Commons logging
- Dispatch core and http
- Http Client and Core (Dispatch dependencies)
- Joda Time (MongoDB dependency)
- jSoup
- MongoDB 2.7.3 and Driver
- ScalaJ collection and http
- SF4LJ (MongoDB dependency)