Commit 66018e1
committed
Performance: improve .trim performance for large strings contains lots of whitespaces
The old implementation took O(N^2) time to trim the string when multiple adjacent spaces
were present.
For instance, consider the string "A B" (10 spaces between A and B).
Then old regexp /[\s]+$/ would take 10 steps until it realizes
the regexp does not match at position 1.
Then it would try to match the regexp at position 2, and it would take 9 steps.
Then 8 steps for position 3, ... so it would take 10*10/2 steps until
it figures out the regexp does not match.
The new approach is to require "non-whitespace" char before the whitespace run,
so it spends just one step for each space in the string.1 parent 0f6c3d9 commit 66018e1
1 file changed
+24
-6
lines changed| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
15 | 15 | | |
16 | 16 | | |
17 | 17 | | |
18 | | - | |
| 18 | + | |
| 19 | + | |
| 20 | + | |
| 21 | + | |
| 22 | + | |
| 23 | + | |
| 24 | + | |
19 | 25 | | |
20 | 26 | | |
21 | 27 | | |
| |||
79 | 85 | | |
80 | 86 | | |
81 | 87 | | |
82 | | - | |
83 | | - | |
84 | | - | |
85 | | - | |
86 | | - | |
| 88 | + | |
| 89 | + | |
| 90 | + | |
| 91 | + | |
| 92 | + | |
| 93 | + | |
| 94 | + | |
| 95 | + | |
| 96 | + | |
| 97 | + | |
| 98 | + | |
| 99 | + | |
| 100 | + | |
| 101 | + | |
| 102 | + | |
| 103 | + | |
| 104 | + | |
87 | 105 | | |
0 commit comments