Last night I was working on an import script in Groovy to bring data back from a web service and into a MongoDB database. There’s a lot of data, and I wanted to know how much time the whole process was taking, as well as how much time certain pieces of the script were taking. As such I built a little closure that I could wrap around each code block I wished to time and it would print to the console how many milliseconds, seconds, or minutes that block took to execute. Here’s a sample of code demonstrating this.
package com.adampresley.timing
class Main {
static main(args) {
/*
* Closure for timing stuff
*/
def timeMe = { Closure codeBlock ->
def start = new Date().getTime()
codeBlock()
def end = new Date().getTime()
def totalTime = end - start
def verbage = ((totalTime > 60000) ? "${(totalTime / 1000) / 60}m" : ((totalTime > 1000) ? "${totalTime / 1000}s" : "${totalTime}ms"))
println "Executed in ${verbage}"
}
/*
* Time something simple, like inserting 1000 hash maps
* into an array.
*/
timeMe {
def iterations = 1000
print "Inserting ${iterations} hash maps into an array... "
def result = []
(1..1000).each {
result << [ firstName: "Adam", lastName: "Presley", age: 32 ]
}
}
/*
* Do something bigger, like recurse over a directory
* structure and put all file names that start with
* 'A' into an array.
*/
timeMe {
print "Get all file names that start with 'A'... "
def result = []
new File("C:/").eachFileRecurse { currentFile ->
if (!currentFile.isDirectory()) {
if (currentFile.name =~ /(?i)^a.*/) result << currentFile.path
}
}
print "${result.size()} match(es) found. "
}
}
}As you can see I wrap each code block I want to time with the timeMe() closure. In this example I provide two tests. The first test is quick so I can show that the closure displays the time executed in milliseconds. The second test is slightly longer so I can show the closure is still smart enough to show the execution time in minutes.
And there you have it. That’s some groovy coding yo! Happy coding!