mirror of
https://github.com/ruby/ruby.git
synced 2026-08-01 02:20:48 +08:00
effac0a485
Improve performance of String#scrub by skipping ascii characters after finding a multi byte character using search_nonascii. Create benchmark for String#scrub to show performance difference made by including the new ascii skips.
48 lines
1.5 KiB
YAML
48 lines
1.5 KiB
YAML
prelude: |
|
|
|
|
STRING_SIZE = 1024
|
|
def duplicate_to_length(str, target_length)
|
|
return "" if target_length <= 0
|
|
return str[0, target_length] if str.length >= target_length
|
|
|
|
(str * ((target_length / str.length) + 1))[0, target_length]
|
|
end
|
|
base = "Hello \u{1f600} world! \u{00e9}\u{00f1}"
|
|
padding = duplicate_to_length(base, STRING_SIZE)
|
|
|
|
valid_utf8 = (padding.b + "OK".b).force_encoding("UTF-8")
|
|
valid_utf8.valid_encoding?
|
|
unknown_but_valid_utf8 = valid_utf8.dup.b.force_encoding("UTF-8")
|
|
invalid_utf8 = (padding.b + "\x80\xFF".b).force_encoding("UTF-8")
|
|
invalid_utf8.valid_encoding?
|
|
unknown_but_invalid_utf8 = (padding.b + "\x80\xFF".b).force_encoding("UTF-8")
|
|
|
|
worst_case_utf8 = duplicate_to_length("\u{1f600}\u{00e9}\u{00f1}", STRING_SIZE).b.force_encoding("UTF-8")
|
|
|
|
unknown_but_valid_utf8_worst_case = worst_case_utf8.dup.b.force_encoding("UTF-8")
|
|
unknown_but_invalid_utf8_worst_case = (worst_case_utf8.b + "\x80\xFF".b).force_encoding("UTF-8")
|
|
|
|
benchmark:
|
|
scrub_known_valid: |
|
|
string = valid_utf8.dup
|
|
string.scrub!
|
|
|
|
scrub_known_invalid: |
|
|
string = invalid_utf8.dup
|
|
string.scrub!
|
|
|
|
scrub_unknown_but_valid_coderange: |
|
|
string = unknown_but_valid_utf8.dup
|
|
string.scrub!
|
|
|
|
scrub_unknown_and_invalid_coderange: |
|
|
string = unknown_but_invalid_utf8.dup
|
|
string.scrub!
|
|
|
|
scrub_unknown_but_valid_coderange_worst_case: |
|
|
string = unknown_but_valid_utf8_worst_case.dup
|
|
string.scrub!
|
|
|
|
scrub_unknown_and_invalid_coderange_worst_case: |
|
|
string = unknown_but_invalid_utf8_worst_case.dup
|
|
string.scrub! |