Here are the available Rails validation filters: before_validation_on_update before_validation_on_create validate validate_on_update validate_on_create after_validation after_validation_on_update after_validation_on_create before_save before_update before_create after_update after_create after_save These filters are callbacks that are invoked automatically. The difference between update and create is really with respect to the database. If the record you are saving is a new record in the db, then it means you are creating. If the item already exists but you are making a change, then it means you are updating. In other words, it does not matter, whether you called obj.save or obj.create, before_update filter will be invoked. The same is true for obj.save vs. obj.update_attributes. So no matter what Active record method you invoke the filter is decided based on the database record. How about before_save, before_after_save? They will always be invoked whether you are creating or updating....
Given that: TestABC.java is located at: test/x/y/z/TestABC.java testdata.txt is located at: resources/x/y/z/TestABC/testdata.txt classpath includes resources folder We are trying to read the testdata.txt from TestABC.java. We need a location independent way to point to these files. Therefore, we would like to use getResource() method of a Class or ClassLoader. Here are the correct ways: 1: this.getClass().getClassLoader().getResource("x/y/z/TestABC/testdata.txt"); 2: this.getClass().getResource("/x/y/z/TestABC/testdata.txt")); 3: this.getClass().getResource("TestABC/testdata.txt")); Explanation: 1: works because ClassLoader takes the path as it is (no manipulation) and search in the classpath folders. In this case, it will go to the resources folder and it will find the testdata.txt file in the given path. 2: works because Class.getResource takes the "absolute" path, removes its starting / and t...
Let's say you have a HTTP streaming server such as the one below where the server sends the HTTP response slowly or may it even holds on the request and does not close the connection and keep sending data (so called "HTTP Streaming"). How would you write a client that receives the response segments as soon as they arrive? (we do not want to wait till the whole response ends or we do not have a choice since the server is streaming "infinite" data). require 'rubygems' require 'mongrel' class StreamHandler def process(request, response) response.status = 200 response.send_status(nil) response.header['Content-Type'] = "text/plain" response.send_header 10.times do response.write("Message \r\n") sleep 5 end response.done end end Mongrel::Configurator.new do listener :port => 3000 do uri "/", :handler => StreamHandler.new end run; join end The client code...
Comments