deac/instiki
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

Efficiency

Browse source 
This version of String#purify
is 12% faster, under Ruby 1.9,
than before.
This commit is contained in:
Jacques Distler 2009-12-05 10:50:58 -06:00
parent f7044ecbb4
commit 171c12d2c1
1 changed files with 23 additions and 10 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

33
lib/stringsupport.rb
View file

@ -6,20 +6,30 @@ class String
# #
# Under 1.8, this is a NOOP. Under 1.9, it sets the encoding to "ASCII-8BIT" # Under 1.8, this is a NOOP. Under 1.9, it sets the encoding to "ASCII-8BIT"
#-- #--
if "".respond_to?(:force_encoding)
def as_bytes
force_encoding("ASCII-8BIT")
end
else
def as_bytes def as_bytes
force_encoding("ASCII-8BIT") if self.respond_to?(:force_encoding)
self self
end end
end
#++ #++
# A method to allow string-oriented operations in both Ruby 1.8 and Ruby 1.9 # A method to allow string-oriented operations in both Ruby 1.8 and Ruby 1.9
# #
# Under 1.8, this is a NOOP. Under 1.9, it sets the encoding to "UTF-8" # Under 1.8, this is a NOOP. Under 1.9, it sets the encoding to "UTF-8"
#-- #--
if "".respond_to?(:force_encoding)
def as_utf8
force_encoding("UTF-8")
end
else
def as_utf8 def as_utf8
force_encoding("UTF-8") if self.respond_to?(:force_encoding)
self self
end end
end
#++ #++
# Take a string, and remove any invalid substrings, returning a valid utf-8 string. # Take a string, and remove any invalid substrings, returning a valid utf-8 string.
@ -29,14 +39,17 @@ class String
# #
# returns a valid utf-8 string, purged of any subsequences of illegal bytes. # returns a valid utf-8 string, purged of any subsequences of illegal bytes.
#-- #--
def purify if "".respond_to?(:force_encoding)
text = self.dup.check_ncrs.as_utf8 def purify
if text.respond_to?(:force_encoding) text = check_ncrs.as_utf8
text.chars.collect{|c| c.as_bytes}.grep(UTF8_REGEX).join.as_utf8 text.chars.collect{|c| c.as_bytes}.grep(UTF8_REGEX).join.as_utf8
else end
text.split(//u).grep(UTF8_REGEX).join else
end def purify
end text = check_ncrs
text.split(//u).grep(UTF8_REGEX).join
end
end
def check_ncrs def check_ncrs
text = gsub(/&#[xX]([a-fA-F0-9]+);/) { |m| [$1.hex].pack('U*').as_bytes =~ UTF8_REGEX ? m : '' } text = gsub(/&#[xX]([a-fA-F0-9]+);/) { |m| [$1.hex].pack('U*').as_bytes =~ UTF8_REGEX ? m : '' }