refactor Lexer to be static class function

This commit is contained in:
Michael Go
2025-01-07 14:32:07 -04:00
parent 3c16c27ee1
commit 002e4caea7
4 changed files with 88 additions and 92 deletions
+83 -87
View File
@@ -25,38 +25,37 @@ module Liquid
COMPARISON_OPERATOR = /==|!=|<>|<=?|>=?|contains(?=\s)/ COMPARISON_OPERATOR = /==|!=|<>|<=?|>=?|contains(?=\s)/
WHITESPACE_OR_NOTHING = /\s*/ WHITESPACE_OR_NOTHING = /\s*/
def initialize(input) class << self
@ss = StringScanner.new(input) def tokenize(input)
end ss = StringScanner.new(input)
output = []
def tokenize until ss.eos?
@output = [] ss.skip(WHITESPACE_OR_NOTHING)
break if ss.eos?
until @ss.eos? tok = if (t = ss.scan(COMPARISON_OPERATOR))
@ss.skip(WHITESPACE_OR_NOTHING) [:comparison, t]
break if @ss.eos? elsif (t = ss.scan(STRING_LITERAL))
tok = if (t = @ss.scan(COMPARISON_OPERATOR)) [:string, t]
[:comparison, t] elsif (t = ss.scan(NUMBER_LITERAL))
elsif (t = @ss.scan(STRING_LITERAL)) [:number, t]
[:string, t] elsif (t = ss.scan(IDENTIFIER))
elsif (t = @ss.scan(NUMBER_LITERAL)) [:id, t]
[:number, t] elsif (t = ss.scan(DOTDOT))
elsif (t = @ss.scan(IDENTIFIER)) [:dotdot, t]
[:id, t]
elsif (t = @ss.scan(DOTDOT))
[:dotdot, t]
else
c = @ss.getch
if (s = SPECIALS[c])
[s, c]
else else
raise SyntaxError, "Unexpected character #{c}" c = ss.getch
if (s = SPECIALS[c])
[s, c]
else
raise SyntaxError, "Unexpected character #{c}"
end
end end
output << tok
end end
@output << tok
end
@output << [:end_of_string] output << [:end_of_string]
end
end end
end end
@@ -157,82 +156,79 @@ module Liquid
table.freeze table.freeze
end end
def initialize(input)
@input = input
end
# rubocop:disable Metrics/BlockNesting # rubocop:disable Metrics/BlockNesting
def tokenize class << self
ss = StringScannerPool.pop(@input) def tokenize(input)
@output = [] ss = StringScannerPool.pop(input)
output = []
until ss.eos? until ss.eos?
ss.skip(WHITESPACE_OR_NOTHING) ss.skip(WHITESPACE_OR_NOTHING)
break if ss.eos? break if ss.eos?
start_pos = ss.pos start_pos = ss.pos
peeked = ss.peek_byte peeked = ss.peek_byte
if (special = SPECIAL_TABLE[peeked]) if (special = SPECIAL_TABLE[peeked])
ss.scan_byte
# Special case for ".."
if special == DOT && ss.peek_byte == DOT_ORD
ss.scan_byte ss.scan_byte
@output << DOTDOT # Special case for ".."
elsif special == DASH if special == DOT && ss.peek_byte == DOT_ORD
# Special case for negative numbers ss.scan_byte
if (peeked_byte = ss.peek_byte) && NUMBER_TABLE[peeked_byte] output << DOTDOT
ss.pos -= 1 elsif special == DASH
@output << [:number, ss.scan(NUMBER_LITERAL)] # Special case for negative numbers
if (peeked_byte = ss.peek_byte) && NUMBER_TABLE[peeked_byte]
ss.pos -= 1
output << [:number, ss.scan(NUMBER_LITERAL)]
else
output << special
end
else else
@output << special output << special
end
elsif (sub_table = TWO_CHARS_COMPARISON_JUMP_TABLE[peeked])
ss.scan_byte
if (peeked_byte = ss.peek_byte) && (found = sub_table[peeked_byte])
output << found
ss.scan_byte
else
raise_syntax_error(start_pos, ss)
end
elsif (sub_table = COMPARISON_JUMP_TABLE[peeked])
ss.scan_byte
if (peeked_byte = ss.peek_byte) && (found = sub_table[peeked_byte])
output << found
ss.scan_byte
else
output << SINGLE_COMPARISON_TOKENS[peeked]
end end
else else
@output << special type, pattern = NEXT_MATCHER_JUMP_TABLE[peeked]
end
elsif (sub_table = TWO_CHARS_COMPARISON_JUMP_TABLE[peeked])
ss.scan_byte
if (peeked_byte = ss.peek_byte) && (found = sub_table[peeked_byte])
@output << found
ss.scan_byte
else
raise_syntax_error(start_pos, ss)
end
elsif (sub_table = COMPARISON_JUMP_TABLE[peeked])
ss.scan_byte
if (peeked_byte = ss.peek_byte) && (found = sub_table[peeked_byte])
@output << found
ss.scan_byte
else
@output << SINGLE_COMPARISON_TOKENS[peeked]
end
else
type, pattern = NEXT_MATCHER_JUMP_TABLE[peeked]
if type && (t = ss.scan(pattern)) if type && (t = ss.scan(pattern))
# Special case for "contains" # Special case for "contains"
@output << if type == :id && t == "contains" && @output.last&.first != :dot output << if type == :id && t == "contains" && output.last&.first != :dot
COMPARISON_CONTAINS COMPARISON_CONTAINS
else
[type, t]
end
else else
[type, t] raise_syntax_error(start_pos, ss)
end end
else
raise_syntax_error(start_pos, ss)
end end
end end
# rubocop:enable Metrics/BlockNesting
output << EOS
ensure
StringScannerPool.release(ss)
end end
# rubocop:enable Metrics/BlockNesting
@output << EOS def raise_syntax_error(start_pos, ss)
ensure ss.pos = start_pos
StringScannerPool.release(ss) # the character could be a UTF-8 character, use getch to get all the bytes
end raise SyntaxError, "Unexpected character #{ss.getch}"
end
def raise_syntax_error(start_pos, ss)
ss.pos = start_pos
# the character could be a UTF-8 character, use getch to get all the bytes
raise SyntaxError, "Unexpected character #{ss.getch}"
end end
end end
+1 -2
View File
@@ -3,8 +3,7 @@
module Liquid module Liquid
class Parser class Parser
def initialize(input) def initialize(input)
l = Lexer.new(input) @tokens = Lexer.tokenize(input)
@tokens = l.tokenize
@p = 0 # pointer to current location @p = 0 # pointer to current location
end end
+3 -2
View File
@@ -1,8 +1,10 @@
# frozen_string_literal: true
module Liquid module Liquid
class StringScannerPool class StringScannerPool
class << self class << self
def pop(input) def pop(input)
@ss_pool ||= [StringScanner.new("")] * 5 @ss_pool ||= 5.times.each_with_object([]) { |_i, arr| arr << StringScanner.new("") }
if @ss_pool.empty? if @ss_pool.empty?
StringScanner.new(input) StringScanner.new(input)
@@ -14,7 +16,6 @@ module Liquid
end end
def release(ss) def release(ss)
binding.irb if ss.nil?
@ss_pool ||= [] @ss_pool ||= []
@ss_pool << ss @ss_pool << ss
end end
+1 -1
View File
@@ -134,6 +134,6 @@ class LexerUnitTest < Minitest::Test
private private
def tokenize(input) def tokenize(input)
Lexer.new(input).tokenize Lexer.tokenize(input)
end end
end end