In some cases you may want to re-attempt a block of code, if it encounters a failure for the first time.
A typical example would be if you were trying to do a SOAP/REST request to a web service that is known to be occasionally intermittent.
One way to handle this is with the use of loops:
def soap_request 1.upto (3) do |attempt| begin retrieve_soap_object = API.request(www.WorldWeather.com) break # this exit the re-attempting loop as soon as there is a success. rescue RunetimeError => error_object puts error_object.message if attempt == 3 puts "request failed." raise end # end of if-statment check. end # end of begin statement end # end of 1..3 loop end # end of method.
However there is a special syntax you can use just for this purpose, which is using the “retry” statement:
def soap_request retrieve_soap_object = API.request(www.WorldWeather.com) # .. # .. # .. rescue RunetimeError => error_object attempts ||= 0 # this creates variable if it doesn't exist yet. attempts += 1 if attempts < 3 puts error_object.message + ". Reattempting SOAP/REST request again." retry # this instructs ruby to return to the first line of the method again. # i.e. it will try to recreate retrieve_soap_object again. else puts "Request failed." raise end # end of if-statement end # end of method.
As you can see, this approach has a lot less nesting and is easer to read.