Sha256: e9e8491e061e20e8236117a3c0ef28a7932a6759ca9beee1754704495f4df218
Contents?: true
Size: 1.82 KB
Versions: 18
Compression:
Stored size: 1.82 KB
Contents
module Recog
class Fingerprint
#
# @example
# r = RegexpFactory.build("^Apache[ -]Coyote/(\d\.\d)$", "REG_ICASE")
# r.match("Apache-Coyote/1.1")
#
module RegexpFactory
# Currently, only options relating to case insensitivity and
# multiline/newline are supported. Because Recog's data is used by tools
# written in different languages like Ruby and Java, we currently support
# specifying them in a variety of ways. This map controls how they can
# be specified.
#
# TODO: consider supporting only a simpler variant and require that tools
# that use Recog data translate accordingly
FLAG_MAP = {
# multiline variations
'REG_DOT_NEWLINE' => Regexp::MULTILINE,
'REG_LINE_ANY_CRLF' => Regexp::MULTILINE,
'MULTILINE' => Regexp::MULTILINE,
# case variations
'REG_ICASE' => Regexp::IGNORECASE,
'IGNORECASE' => Regexp::IGNORECASE
}
# @return [Regexp]
def self.build(pattern, flags)
options = build_options(flags)
Regexp.new(pattern, options)
end
# Convert string flag names as used in Recog XML into a Fixnum suitable for
# passing as the `options` parameter to `Regexp.new`
#
# @see FLAG_MAP
# @param flags [Array<String>]
# @return [Fixnum] Flags for creating a regular expression object
def self.build_options(flags)
unsupported_flags = flags.select { |flag| !FLAG_MAP.key?(flag) }
unless unsupported_flags.empty?
fail "Unsupported regular expression flags found: #{unsupported_flags.join(',')}. Must be one of: #{FLAG_MAP.keys.join(',')}"
end
flags.reduce(Regexp::NOENCODING) do |sum, flag|
sum |= (FLAG_MAP[flag] || 0)
end
end
end
end
end
Version data entries
18 entries across 18 versions & 1 rubygems