Can the performance be improved? defun is 4x slower than defn.

[ghost]

I ran the benchmark and similar to your results, the performance is not good enough. My test results:

Evaluation count : 92520 in 60 samples of 1542 calls.
      Execution time sample mean : 645.081223 µs
             Execution time mean : 645.082481 µs
Execution time sample std-deviation : 5.005565 µs
    Execution time std-deviation : 5.117810 µs
   Execution time lower quantile : 637.577837 µs ( 2.5%)
   Execution time upper quantile : 655.627332 µs (97.5%)
                   Overhead used : 9.233123 ns

Found 1 outliers in 60 samples (1.6667 %)
  low-severe   1 (1.6667 %)
 Variance from outliers : 1.6389 % Variance is slightly inflated by outliers

 Evaluation count : 20880 in 60 samples of 348 calls.
      Execution time sample mean : 2.919145 ms
             Execution time mean : 2.920135 ms
Execution time sample std-deviation : 40.550684 µs
    Execution time std-deviation : 41.397063 µs
   Execution time lower quantile : 2.879026 ms ( 2.5%)
   Execution time upper quantile : 2.964349 ms (97.5%)
                   Overhead used : 9.233123 ns

Found 4 outliers in 60 samples (6.6667 %)
  low-severe   1 (1.6667 %)
  low-mild   2 (3.3333 %)
  high-mild  1 (1.6667 %)
 Variance from outliers : 1.6389 % Variance is slightly inflated by outliers

As you can see defn takes 645.082481 µs whereas defun is taking 2.920135 ms (2920.135 µs), which is 4.5x slower. Love the library but I am not sure I can use it for any serious development. Please improve the performance.

[killme2008]

hi, defun is just a thin wrapper on core.match, and the performance bottleneck is also on it.

We are using defun in our production with instaparse to implement a DSL, it works well. But you should balance the performance and expressive in your development by yourself.