From d061df81ce4c7ee5d15388b5db48b7d1c24444e4 Mon Sep 17 00:00:00 2001
From: Brian Adkins <brian@lojic.com>
Date: Tue, 23 Mar 2021 12:16:07 -0400
Subject: [PATCH] Optimize splitting strings on whitespace

Use an optimized function, instead of regexp-split, when string-split
is called with the default separator and trim setting.
---
 .../racket-test-core/tests/racket/string.rktl |  2 +-
 racket/collects/racket/string.rkt             | 27 ++++++++++++++++++-
 2 files changed, 27 insertions(+), 2 deletions(-)

diff --git a/pkgs/racket-test-core/tests/racket/string.rktl b/pkgs/racket-test-core/tests/racket/string.rktl
index 868f0e7d9f..e8ddacf12f 100644
--- a/pkgs/racket-test-core/tests/racket/string.rktl
+++ b/pkgs/racket-test-core/tests/racket/string.rktl
@@ -482,7 +482,7 @@
 
 ;; ---------- string-split ----------
 (let ()
-  (for ([s (in-list '("x y z" " x y z "  "\nx y z" "  \t x\r\r\ry    z\n"))])
+  (for ([s (in-list '("x y z" " x y z "  "\nx y z" "  \t x\r\r\ry \n\f\n   z\n"))])
     (test '("x" "y" "z") string-split s))
   (for ([s (in-list '(" " "   "  "\n\t\r"))])
     (test '() string-split s))
diff --git a/racket/collects/racket/string.rkt b/racket/collects/racket/string.rkt
index c7622c5bd0..c22fe11bc5 100644
--- a/racket/collects/racket/string.rkt
+++ b/racket/collects/racket/string.rkt
@@ -124,8 +124,33 @@
   ;; don't do that for all nulls).)
   (if (equal? strs '("")) '() strs))
 
+;; A faster split implementation when splitting on whitespace. The
+;; string will also be trimmed.
+(define (internal-split-whitespace str)
+  (define (is-whitespace? c)
+    (or (eq? c #\space)
+        (eq? c #\tab)
+        (eq? c #\newline)
+        (eq? c #\return)
+        (eq? c #\page)))
+  (define len (string-length str))
+  (let loop ([beg 0] [end 0])
+    (cond [(= end len)
+           (if (= beg end)
+               '()
+               (list (substring str beg end)))]
+          [(is-whitespace? (string-ref str end))
+           (let ([pos (add1 end)])
+             (if (= beg end)
+                 (loop pos pos)
+                 (cons (substring str beg end)
+                       (loop pos pos))))]
+          [else (loop beg (add1 end))])))
+
 (define (string-split str [sep none] #:trim? [trim? #t] #:repeat? [+? #f])
-  (internal-split 'string-split str sep trim? +?))
+  (if (and (eq? sep none) trim?)
+      (internal-split-whitespace str)
+      (internal-split 'string-split str sep trim? +?)))
 
 (define (string-normalize-spaces str [sep none] [space " "]
                                  #:trim? [trim? #t] #:repeat? [+? #f])