Explain allocation in examples/simple

examples/simple.zig

@@ -44,7 +44,8 @@ fn sub3(r: *cli.AppRunner) !cli.Command {
 }
 
 fn parseArgs() cli.AppRunner.Error!cli.ExecFn {
-    var r = try cli.AppRunner.init(std.heap.page_allocator);
+    // This allocator will be used to allocate config.ip and config.arg2.
+    var r = try cli.AppRunner.init(allocator);
 
     const sub2 = cli.Command{
         .name = "sub2",
@@ -115,7 +116,25 @@ fn parseArgs() cli.AppRunner.Error!cli.ExecFn {
 
 pub fn main() anyerror!void {
     const action = try parseArgs();
-    return action();
+    const r = action();
+    freeConfig();
+    return r;
+}
+
+// Usually, you just use an arena allocator to free all allocated resources in a batch.
+// This only illustrates the fact that the config data are allocated with the allocator
+// you pass to cli.AppRunner.init(allocator).
+fn freeConfig() void {
+    allocator.free(config.ip);
+    if (config.arg2.len > 0) {
+        for (config.arg2) |item| {
+            allocator.free(item);
+        }
+        allocator.free(config.arg2);
+    }
+    if (gpa.deinit() == .leak) {
+        @panic("config leaked");
+    }
 }
 
 fn run_sub3() !void {