regexp_parser 2.7.0 → 2.9.2

data/lib/regexp_parser/scanner/errors/validation_error.rb

@@ -0,0 +1,63 @@
+class Regexp::Scanner
+  # Base for all scanner validation errors
+  class ValidationError < ScannerError
+    # Centralizes and unifies the handling of validation related errors.
+    def self.for(type, problem, reason = nil)
+      types.fetch(type).new(problem, reason)
+    end
+
+    def self.types
+      @types ||= {
+        backref:      InvalidBackrefError,
+        group:        InvalidGroupError,
+        group_option: InvalidGroupOption,
+        posix_class:  UnknownPosixClassError,
+        property:     UnknownUnicodePropertyError,
+        sequence:     InvalidSequenceError,
+      }
+    end
+  end
+
+  # Invalid sequence format. Used for escape sequences, mainly.
+  class InvalidSequenceError < ValidationError
+    def initialize(what = 'sequence', where = '')
+      super "Invalid #{what} at #{where}"
+    end
+  end
+
+  # Invalid group. Used for named groups.
+  class InvalidGroupError < ValidationError
+    def initialize(what, reason)
+      super "Invalid #{what}, #{reason}."
+    end
+  end
+
+  # Invalid groupOption. Used for inline options.
+  # TODO: should become InvalidGroupOptionError in v3.0.0 for consistency
+  class InvalidGroupOption < ValidationError
+    def initialize(option, text)
+      super "Invalid group option #{option} in #{text}"
+    end
+  end
+
+  # Invalid back reference. Used for name a number refs/calls.
+  class InvalidBackrefError < ValidationError
+    def initialize(what, reason)
+      super "Invalid back reference #{what}, #{reason}"
+    end
+  end
+
+  # The property name was not recognized by the scanner.
+  class UnknownUnicodePropertyError < ValidationError
+    def initialize(name, _)
+      super "Unknown unicode character property name #{name}"
+    end
+  end
+
+  # The POSIX class name was not recognized by the scanner.
+  class UnknownPosixClassError < ValidationError
+    def initialize(text, _)
+      super "Unknown POSIX class #{text}"
+    end
+  end
+end

data/lib/regexp_parser/scanner/properties/long.csv

@@ -7,6 +7,8 @@ age=12.0,age=12.0
 age=12.1,age=12.1
 age=13.0,age=13.0
 age=14.0,age=14.0
+age=15.0,age=15.0
+age=15.1,age=15.1
 age=2.0,age=2.0
 age=2.1,age=2.1
 age=3.0,age=3.0
@@ -97,6 +99,7 @@ emojimodifierbase,emoji_modifier_base
 emojipresentation,emoji_presentation
 enclosingmark,enclosing_mark
 ethiopic,ethiopic
+extendedpictographic,extended_pictographic
 extender,extender
 finalpunctuation,final_punctuation
 format,format
@@ -106,6 +109,19 @@ gothic,gothic
 grantha,grantha
 graph,graph
 graphemebase,grapheme_base
+graphemeclusterbreak=control,grapheme_cluster_break=control
+graphemeclusterbreak=cr,grapheme_cluster_break=cr
+graphemeclusterbreak=extend,grapheme_cluster_break=extend
+graphemeclusterbreak=l,grapheme_cluster_break=l
+graphemeclusterbreak=lf,grapheme_cluster_break=lf
+graphemeclusterbreak=lv,grapheme_cluster_break=lv
+graphemeclusterbreak=lvt,grapheme_cluster_break=lvt
+graphemeclusterbreak=prepend,grapheme_cluster_break=prepend
+graphemeclusterbreak=regionalindicator,grapheme_cluster_break=regional_indicator
+graphemeclusterbreak=spacingmark,grapheme_cluster_break=spacingmark
+graphemeclusterbreak=t,grapheme_cluster_break=t
+graphemeclusterbreak=v,grapheme_cluster_break=v
+graphemeclusterbreak=zwj,grapheme_cluster_break=zwj
 graphemeextend,grapheme_extend
 graphemelink,grapheme_link
 greek,greek
@@ -121,11 +137,14 @@ hebrew,hebrew
 hexdigit,hex_digit
 hiragana,hiragana
 hyphen,hyphen
+idcompatmathcontinue,id_compat_math_continue
+idcompatmathstart,id_compat_math_start
 idcontinue,id_continue
 ideographic,ideographic
 idsbinaryoperator,ids_binary_operator
 idstart,id_start
 idstrinaryoperator,ids_trinary_operator
+idsunaryoperator,ids_unary_operator
 imperialaramaic,imperial_aramaic
 inadlam,in_adlam
 inaegeannumbers,in_aegean_numbers
@@ -139,6 +158,7 @@ inancientsymbols,in_ancient_symbols
 inarabic,in_arabic
 inarabicextendeda,in_arabic_extended_a
 inarabicextendedb,in_arabic_extended_b
+inarabicextendedc,in_arabic_extended_c
 inarabicmathematicalalphabeticsymbols,in_arabic_mathematical_alphabetic_symbols
 inarabicpresentationformsa,in_arabic_presentation_forms_a
 inarabicpresentationformsb,in_arabic_presentation_forms_b
@@ -186,6 +206,8 @@ incjkunifiedideographsextensiond,in_cjk_unified_ideographs_extension_d
 incjkunifiedideographsextensione,in_cjk_unified_ideographs_extension_e
 incjkunifiedideographsextensionf,in_cjk_unified_ideographs_extension_f
 incjkunifiedideographsextensiong,in_cjk_unified_ideographs_extension_g
+incjkunifiedideographsextensionh,in_cjk_unified_ideographs_extension_h
+incjkunifiedideographsextensioni,in_cjk_unified_ideographs_extension_i
 incombiningdiacriticalmarks,in_combining_diacritical_marks
 incombiningdiacriticalmarksextended,in_combining_diacritical_marks_extended
 incombiningdiacriticalmarksforsymbols,in_combining_diacritical_marks_for_symbols
@@ -205,10 +227,12 @@ incyrillic,in_cyrillic
 incyrillicextendeda,in_cyrillic_extended_a
 incyrillicextendedb,in_cyrillic_extended_b
 incyrillicextendedc,in_cyrillic_extended_c
+incyrillicextendedd,in_cyrillic_extended_d
 incyrillicsupplement,in_cyrillic_supplement
 indeseret,in_deseret
 indevanagari,in_devanagari
 indevanagariextended,in_devanagari_extended
+indevanagariextendeda,in_devanagari_extended_a
 indingbats,in_dingbats
 indivesakuru,in_dives_akuru
 indogra,in_dogra
@@ -268,6 +292,7 @@ inipaextensions,in_ipa_extensions
 initialpunctuation,initial_punctuation
 injavanese,in_javanese
 inkaithi,in_kaithi
+inkaktoviknumerals,in_kaktovik_numerals
 inkanaextendeda,in_kana_extended_a
 inkanaextendedb,in_kana_extended_b
 inkanasupplement,in_kana_supplement
@@ -276,6 +301,7 @@ inkangxiradicals,in_kangxi_radicals
 inkannada,in_kannada
 inkatakana,in_katakana
 inkatakanaphoneticextensions,in_katakana_phonetic_extensions
+inkawi,in_kawi
 inkayahli,in_kayah_li
 inkharoshthi,in_kharoshthi
 inkhitansmallscript,in_khitan_small_script
@@ -339,6 +365,7 @@ inmyanmar,in_myanmar
 inmyanmarextendeda,in_myanmar_extended_a
 inmyanmarextendedb,in_myanmar_extended_b
 innabataean,in_nabataean
+innagmundari,in_nag_mundari
 innandinagari,in_nandinagari
 innewa,in_newa
 innewtailue,in_new_tai_lue
@@ -457,6 +484,7 @@ joincontrol,join_control
 kaithi,kaithi
 kannada,kannada
 katakana,katakana
+kawi,kawi
 kayahli,kayah_li
 kharoshthi,kharoshthi
 khitansmallscript,khitan_small_script
@@ -503,6 +531,7 @@ mro,mro
 multani,multani
 myanmar,myanmar
 nabataean,nabataean
+nagmundari,nag_mundari
 nandinagari,nandinagari
 newa,newa
 newline,newline

data/lib/regexp_parser/scanner/properties/short.csv

@@ -57,6 +57,7 @@ emod,emoji_modifier
 epres,emoji_presentation
 ethi,ethiopic
 ext,extender
+extpict,extended_pictographic
 geor,georgian
 glag,glagolitic
 gong,gunjala_gondi
@@ -85,6 +86,7 @@ ideo,ideographic
 ids,id_start
 idsb,ids_binary_operator
 idst,ids_trinary_operator
+idsu,ids_unary_operator
 ital,old_italic
 java,javanese
 joinc,join_control
@@ -133,6 +135,7 @@ mtei,meetei_mayek
 mult,multani
 mymr,myanmar
 n,number
+nagm,nag_mundari
 nand,nandinagari
 narb,old_north_arabian
 nbat,nabataean

data/lib/regexp_parser/scanner/property.rl

@@ -20,7 +20,7 @@
       name = text[3..-2].gsub(/[\^\s_\-]/, '').downcase
 
       token = self.class.short_prop_map[name] || self.class.long_prop_map[name]
-      validation_error(:property, name) unless token
+      raise ValidationError.for(:property, name) unless token
 
       self.emit(type, token.to_sym, text)
 

data/lib/regexp_parser/scanner/scanner.rl

@@ -30,11 +30,6 @@
 
   class_posix           = ('[:' . '^'? . [^\[\]]* . ':]');
 
-
-  # these are not supported in ruby at the moment
-  collating_sequence    = '[.' . (alpha | [\-])+ . '.]';
-  character_equivalent  = '[=' . alpha . '=]';
-
   line_anchor           = beginning_of_line | end_of_line;
   anchor_char           = [AbBzZG];
 
@@ -83,10 +78,9 @@
   # try to treat every other group head as options group, like Ruby
   group_options         = '?' . ( [^!#'():<=>~]+ . ':'? ) ?;
 
-  group_ref             = [gk];
   group_name_id_ab      = ([^!0-9\->] | utf8_multibyte) . ([^>] | utf8_multibyte)*;
   group_name_id_sq      = ([^0-9\-']  | utf8_multibyte) . ([^'] | utf8_multibyte)*;
-  group_number          = '-'? . [1-9] . [0-9]*;
+  group_number          = '-'? . [0-9]+;
   group_level           = [+\-] . [0-9]+;
 
   group_name            = ('<' . group_name_id_ab? . '>') |
@@ -95,15 +89,11 @@
 
   group_named           = ('?' . group_name );
 
-  group_name_backref    = 'k' . (('<' . group_name_id_ab? . group_level? '>') |
-                                 ("'" . group_name_id_sq? . group_level? "'"));
-  group_name_call       = 'g' . (('<' . group_name_id_ab? . group_level? '>') |
-                                 ("'" . group_name_id_sq? . group_level? "'"));
+  group_ref_body        = (('<' . (group_name_id_ab? | group_number) . group_level? '>') |
+                           ("'" . (group_name_id_sq? | group_number) . group_level? "'"));
 
-  group_number_backref  = 'k' . (('<' . group_number . group_level? '>') |
-                                 ("'" . group_number . group_level? "'"));
-  group_number_call     = 'g' . (('<' . ((group_number . group_level?) | '0') '>') |
-                                 ("'" . ((group_number . group_level?) | '0') "'"));
+  group_ref             = 'k' . group_ref_body;
+  group_call            = 'g' . group_ref_body;
 
   group_type            = group_atomic | group_passive | group_absence | group_named;
 
@@ -134,13 +124,13 @@
   # EOF error, used where it can be detected
   action premature_end_error {
     text = copy(data, ts ? ts-1 : 0, -1)
-    raise PrematureEndError.new( text )
+    raise PrematureEndError.new(text)
   }
 
   # Invalid sequence error, used from sequences, like escapes and sets
   action invalid_sequence_error {
     text = copy(data, ts ? ts-1 : 0, -1)
-    validation_error(:sequence, 'sequence', text)
+    raise ValidationError.for(:sequence, 'sequence', text)
   }
 
   # group (nesting) and set open/close actions
@@ -221,20 +211,12 @@
       end
 
       unless self.class.posix_classes.include?(class_name)
-        validation_error(:posix_class, text)
+        raise ValidationError.for(:posix_class, text)
       end
 
       emit(type, class_name.to_sym, text)
     };
 
-    # These are not supported in ruby at the moment. Enable them if they are.
-    # collating_sequence >(open_bracket, 1) @set_closed @eof(premature_end_error) {
-    #   emit(:set, :collation, copy(data, ts, te))
-    # };
-    # character_equivalent >(open_bracket, 1) @set_closed @eof(premature_end_error) {
-    #   emit(:set, :equivalent, copy(data, ts, te))
-    # };
-
     meta_char > (set_meta, 1) {
       emit(:literal, :literal, copy(data, ts, te))
     };
@@ -285,6 +267,13 @@
       fret;
     };
 
+    [8-9] . [0-9] { # special case, emits two tokens
+      text = copy(data, ts-1, te)
+      emit(:escape, :literal, text[0, 2])
+      emit(:literal, :literal, text[2])
+      fret;
+    };
+
     meta_char {
       case text = copy(data, ts-1, te)
       when '\.';  emit(:escape, :dot,               text)
@@ -375,6 +364,7 @@
   conditional_expression := |*
     group_lookup . ')' {
       text = copy(data, ts, te-1)
+      text =~ /[^0]/ or raise ValidationError.for(:backref, 'condition', 'invalid ref ID')
       emit(:conditional, :condition, text)
       emit(:conditional, :condition_close, ')')
     };
@@ -457,10 +447,9 @@
 
     # (?#...) comments: parsed as a single expression, without introducing a
     # new nesting level. Comments may not include parentheses, escaped or not.
-    # special case for close, action performed on all transitions to get the
-    # correct closing count.
+    # special case for close to get the correct closing count.
     # ------------------------------------------------------------------------
-    group_open . group_comment $group_closed {
+    (group_open . group_comment) @group_closed {
       emit(:group, :comment, copy(data, ts, te))
     };
 
@@ -475,10 +464,10 @@
     #
     #   (?imxdau-imx:subexp)  option on/off for subexp
     # ------------------------------------------------------------------------
-    group_open . group_options >group_opened {
+    (group_open . group_options) >group_opened {
       text = copy(data, ts, te)
       if text[2..-1] =~ /([^\-mixdau:]|^$)|-.*([dau])/
-        validation_error(:group_option, $1 || "-#{$2}", text)
+        raise ValidationError.for(:group_option, $1 || "-#{$2}", text)
       end
       emit_options(text)
     };
@@ -489,7 +478,7 @@
     #   (?<=subexp)         look-behind
     #   (?<!subexp)         negative look-behind
     # ------------------------------------------------------------------------
-    group_open . assertion_type >group_opened {
+    (group_open . assertion_type) >group_opened {
       case text = copy(data, ts, te)
       when '(?=';  emit(:assertion, :lookahead,    text)
       when '(?!';  emit(:assertion, :nlookahead,   text)
@@ -506,14 +495,14 @@
     #   (?'name'subexp)     named group (single quoted version)
     #   (subexp)            captured group
     # ------------------------------------------------------------------------
-    group_open . group_type >group_opened {
+    (group_open . group_type) >group_opened {
       case text = copy(data, ts, te)
       when '(?:';  emit(:group, :passive,      text)
       when '(?>';  emit(:group, :atomic,       text)
       when '(?~';  emit(:group, :absence,      text)
 
       when /^\(\?(?:<>|'')/
-        validation_error(:group, 'named group', 'name is empty')
+        raise ValidationError.for(:group, 'named group', 'name is empty')
 
       when /^\(\?<[^>]+>/
         emit(:group, :named_ab,  text)
@@ -533,7 +522,7 @@
       if conditional_stack.last == group_depth + 1
         conditional_stack.pop
         emit(:conditional, :close, ')')
-      else
+      elsif group_depth >= 0
         if spacing_stack.length > 1 &&
            spacing_stack.last[:depth] == group_depth + 1
           spacing_stack.pop
@@ -541,41 +530,43 @@
         end
 
         emit(:group, :close, ')')
+      else
+        raise ValidationError.for(:group, 'group', 'unmatched close parenthesis')
       end
     };
 
 
     # Group backreference, named and numbered
     # ------------------------------------------------------------------------
-    backslash . (group_name_backref | group_number_backref) > (backslashed, 4) {
+    backslash . (group_ref) > (backslashed, 4) {
       case text = copy(data, ts, te)
-      when /^\\k(<>|'')/
-        validation_error(:backref, 'backreference', 'ref ID is empty')
-      when /^\\k(.)[^\p{digit}\-][^+\-]*\D$/
+      when /^\\k(.)[^0-9\-][^+\-]*['>]$/
         emit(:backref, $1 == '<' ? :name_ref_ab : :name_ref_sq, text)
-      when /^\\k(.)\d+\D$/
+      when /^\\k(.)0*[1-9]\d*['>]$/
         emit(:backref, $1 == '<' ? :number_ref_ab : :number_ref_sq, text)
-      when /^\\k(.)-\d+\D$/
+      when /^\\k(.)-0*[1-9]\d*['>]$/
         emit(:backref, $1 == '<' ? :number_rel_ref_ab : :number_rel_ref_sq, text)
-      when /^\\k(.)[^\p{digit}\-].*[+\-]\d+\D$/
+      when /^\\k(.)[^0-9\-].*[+\-]\d+['>]$/
         emit(:backref, $1 == '<' ? :name_recursion_ref_ab : :name_recursion_ref_sq, text)
-      when /^\\k(.)-?\d+[+\-]\d+\D$/
+      when /^\\k(.)-?0*[1-9]\d*[+\-]\d+['>]$/
         emit(:backref, $1 == '<' ? :number_recursion_ref_ab : :number_recursion_ref_sq, text)
+      else
+        raise ValidationError.for(:backref, 'backreference', 'invalid ref ID')
       end
     };
 
     # Group call, named and numbered
     # ------------------------------------------------------------------------
-    backslash . (group_name_call | group_number_call) > (backslashed, 4) {
+    backslash . (group_call) > (backslashed, 4) {
       case text = copy(data, ts, te)
-      when /^\\g(<>|'')/
-        validation_error(:backref, 'subexpression call', 'ref ID is empty')
-      when /^\\g(.)[^\p{digit}+\->][^+\-]*/
+      when /^\\g(.)[^0-9+\-].*['>]$/
         emit(:backref, $1 == '<' ? :name_call_ab : :name_call_sq, text)
-      when /^\\g(.)\d+\D$/
+      when /^\\g(.)(?:0|0*[1-9]\d*)['>]$/
         emit(:backref, $1 == '<' ? :number_call_ab : :number_call_sq, text)
-      when /^\\g(.)[+-]\d+/
+      when /^\\g(.)[+-]0*[1-9]\d*/
         emit(:backref, $1 == '<' ? :number_rel_call_ab : :number_rel_call_sq, text)
+      else
+        raise ValidationError.for(:backref, 'subexpression call', 'invalid ref ID')
       end
     };
 
@@ -649,72 +640,11 @@
   *|;
 }%%
 
-# THIS IS A GENERATED FILE, DO NOT EDIT DIRECTLY
-# This file was generated from lib/regexp_parser/scanner/scanner.rl
-
-require 'regexp_parser/error'
+require_relative 'scanner/errors/scanner_error'
+require_relative 'scanner/errors/premature_end_error'
+require_relative 'scanner/errors/validation_error'
 
 class Regexp::Scanner
-  # General scanner error (catch all)
-  class ScannerError < Regexp::Parser::Error; end
-
-  # Base for all scanner validation errors
-  class ValidationError < Regexp::Parser::Error
-    def initialize(reason)
-      super reason
-    end
-  end
-
-  # Unexpected end of pattern
-  class PrematureEndError < ScannerError
-    def initialize(where = '')
-      super "Premature end of pattern at #{where}"
-    end
-  end
-
-  # Invalid sequence format. Used for escape sequences, mainly.
-  class InvalidSequenceError < ValidationError
-    def initialize(what = 'sequence', where = '')
-      super "Invalid #{what} at #{where}"
-    end
-  end
-
-  # Invalid group. Used for named groups.
-  class InvalidGroupError < ValidationError
-    def initialize(what, reason)
-      super "Invalid #{what}, #{reason}."
-    end
-  end
-
-  # Invalid groupOption. Used for inline options.
-  # TODO: should become InvalidGroupOptionError in v3.0.0 for consistency
-  class InvalidGroupOption < ValidationError
-    def initialize(option, text)
-      super "Invalid group option #{option} in #{text}"
-    end
-  end
-
-  # Invalid back reference. Used for name a number refs/calls.
-  class InvalidBackrefError < ValidationError
-    def initialize(what, reason)
-      super "Invalid back reference #{what}, #{reason}"
-    end
-  end
-
-  # The property name was not recognized by the scanner.
-  class UnknownUnicodePropertyError < ValidationError
-    def initialize(name)
-      super "Unknown unicode character property name #{name}"
-    end
-  end
-
-  # The POSIX class name was not recognized by the scanner.
-  class UnknownPosixClassError < ValidationError
-    def initialize(text)
-      super "Unknown POSIX class #{text}"
-    end
-  end
-
   # Scans the given regular expression text, or Regexp object and collects the
   # emitted token into an array that gets returned at the end. If a block is
   # given, it gets called for each emitted token.
@@ -891,24 +821,8 @@ class Regexp::Scanner
 
   def emit_meta_control_sequence(data, ts, te, token)
     if data.last < 0x00 || data.last > 0x7F
-      validation_error(:sequence, 'escape', token.to_s)
+      raise ValidationError.for(:sequence, 'escape', token.to_s)
     end
     emit(:escape, token, copy(data, ts-1, te))
   end
-
-  # Centralizes and unifies the handling of validation related
-  # errors.
-  def validation_error(type, what, reason = nil)
-    error =
-      case type
-      when :backref      then InvalidBackrefError.new(what, reason)
-      when :group        then InvalidGroupError.new(what, reason)
-      when :group_option then InvalidGroupOption.new(what, reason)
-      when :posix_class  then UnknownPosixClassError.new(what)
-      when :property     then UnknownUnicodePropertyError.new(what)
-      when :sequence     then InvalidSequenceError.new(what, reason)
-      end
-
-    raise error # unless @@config.validation_ignore
-  end
 end # module Regexp::Scanner