summaryrefslogtreecommitdiffhomepage
path: root/d.html.markdown
diff options
context:
space:
mode:
authorNickPapanastasiou <nickpap9411@gmail.com>2015-06-16 17:46:58 -0400
committerNickPapanastasiou <nickpap9411@gmail.com>2015-06-16 17:46:58 -0400
commita54c5b580b9a2900f2f6eb1b2b5f07d0c1dd2e9d (patch)
tree4aacdc6301d8eee682754cbd09b4240e8ccb8427 /d.html.markdown
parent5f0d8c28cc8c8f33db45c82a79699a9c1ff6fbbc (diff)
parallel stuff
Diffstat (limited to 'd.html.markdown')
-rw-r--r--d.html.markdown24
1 files changed, 23 insertions, 1 deletions
diff --git a/d.html.markdown b/d.html.markdown
index 7356174d..88c7e37f 100644
--- a/d.html.markdown
+++ b/d.html.markdown
@@ -219,7 +219,29 @@ void main() {
Notice how we got to build a nice Haskellian pipeline to compute num?
That's thanks to a D innovation know as Uniform Function Call Syntax.
With UFCS, we can choose whether to write a function call as a method
-or free function call! Walter wrote a nice article on this [here.](http://www.drdobbs.com/cpp/uniform-function-call-syntax/232700394) In short, you can call functions whose first parameter
+or free function call! Walter wrote a nice article on this
+[here.](http://www.drdobbs.com/cpp/uniform-function-call-syntax/232700394)
+In short, you can call functions whose first parameter
is of some type A on any expression of type A as a method.
+I like parallelism. Anyone else like parallelism? Sure you do. Let's do some!
+```d
+import std.stdio;
+import std.parallelism : parallel;
+import std.math : sqrt;
+
+void main() {
+ // We want take the square root every number in our array,
+ // and take advantage of as many cores as we have available.
+ auto arr = new double[1_000_000];
+
+ // Use an index, and an array element by referece,
+ // and just call parallel on the array!
+ foreach(i, ref elem; parallel(arr)) {
+ ref = sqrt(i + 1.0);
+ }
+}
+
+
+```