regexp_parser 2.0.2 → 2.2.0

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

@@ -8,6 +8,7 @@ age=10.0: age=10.0
 age=11.0: age=11.0
 age=12.0: age=12.0
 age=12.1: age=12.1
+age=13.0: age=13.0
 age=2.0: age=2.0
 age=2.1: age=2.1
 age=3.0: age=3.0
@@ -64,6 +65,7 @@ changeswhenlowercased: changes_when_lowercased
 changeswhentitlecased: changes_when_titlecased
 changeswhenuppercased: changes_when_uppercased
 cherokee: cherokee
+chorasmian: chorasmian
 closepunctuation: close_punctuation
 cntrl: cntrl
 common: common
@@ -83,6 +85,7 @@ deseret: deseret
 devanagari: devanagari
 diacritic: diacritic
 digit: digit
+divesakuru: dives_akuru
 dogra: dogra
 duployan: duployan
 egyptianhieroglyphs: egyptian_hieroglyphs
@@ -167,6 +170,7 @@ incham: in_cham
 incherokee: in_cherokee
 incherokeesupplement: in_cherokee_supplement
 inchesssymbols: in_chess_symbols
+inchorasmian: in_chorasmian
 incjkcompatibility: in_cjk_compatibility
 incjkcompatibilityforms: in_cjk_compatibility_forms
 incjkcompatibilityideographs: in_cjk_compatibility_ideographs
@@ -181,6 +185,7 @@ incjkunifiedideographsextensionc: in_cjk_unified_ideographs_extension_c
 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
 incombiningdiacriticalmarks: in_combining_diacritical_marks
 incombiningdiacriticalmarksextended: in_combining_diacritical_marks_extended
 incombiningdiacriticalmarksforsymbols: in_combining_diacritical_marks_for_symbols
@@ -204,6 +209,7 @@ indeseret: in_deseret
 indevanagari: in_devanagari
 indevanagariextended: in_devanagari_extended
 indingbats: in_dingbats
+indivesakuru: in_dives_akuru
 indogra: in_dogra
 indominotiles: in_domino_tiles
 induployan: in_duployan
@@ -269,6 +275,7 @@ inkatakana: in_katakana
 inkatakanaphoneticextensions: in_katakana_phonetic_extensions
 inkayahli: in_kayah_li
 inkharoshthi: in_kharoshthi
+inkhitansmallscript: in_khitan_small_script
 inkhmer: in_khmer
 inkhmersymbols: in_khmer_symbols
 inkhojki: in_khojki
@@ -288,6 +295,7 @@ inlineara: in_linear_a
 inlinearbideograms: in_linear_b_ideograms
 inlinearbsyllabary: in_linear_b_syllabary
 inlisu: in_lisu
+inlisusupplement: in_lisu_supplement
 inlowsurrogates: in_low_surrogates
 inlycian: in_lycian
 inlydian: in_lydian
@@ -395,6 +403,7 @@ insupplementaryprivateuseareab: in_supplementary_private_use_area_b
 insuttonsignwriting: in_sutton_signwriting
 insylotinagri: in_syloti_nagri
 insymbolsandpictographsextendeda: in_symbols_and_pictographs_extended_a
+insymbolsforlegacycomputing: in_symbols_for_legacy_computing
 insyriac: in_syriac
 insyriacsupplement: in_syriac_supplement
 intagalog: in_tagalog
@@ -409,6 +418,7 @@ intamil: in_tamil
 intamilsupplement: in_tamil_supplement
 intangut: in_tangut
 intangutcomponents: in_tangut_components
+intangutsupplement: in_tangut_supplement
 intelugu: in_telugu
 inthaana: in_thaana
 inthai: in_thai
@@ -426,6 +436,7 @@ invedicextensions: in_vedic_extensions
 inverticalforms: in_vertical_forms
 inwancho: in_wancho
 inwarangciti: in_warang_citi
+inyezidi: in_yezidi
 inyijinghexagramsymbols: in_yijing_hexagram_symbols
 inyiradicals: in_yi_radicals
 inyisyllables: in_yi_syllables
@@ -437,6 +448,7 @@ kannada: kannada
 katakana: katakana
 kayahli: kayah_li
 kharoshthi: kharoshthi
+khitansmallscript: khitan_small_script
 khmer: khmer
 khojki: khojki
 khudawadi: khudawadi
@@ -590,5 +602,6 @@ xdigit: xdigit
 xidcontinue: xid_continue
 xidstart: xid_start
 xposixpunct: xposixpunct
+yezidi: yezidi
 yi: yi
 zanabazarsquare: zanabazar_square

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

@@ -28,6 +28,7 @@ cari: carian
 cc: control
 cf: format
 cher: cherokee
+chrs: chorasmian
 ci: case_ignorable
 cn: unassigned
 co: private_use
@@ -45,12 +46,17 @@ dep: deprecated
 deva: devanagari
 di: default_ignorable_code_point
 dia: diacritic
+diak: dives_akuru
 dogr: dogra
 dsrt: deseret
 dupl: duployan
+ebase: emoji_modifier_base
+ecomp: emoji_component
 egyp: egyptian_hieroglyphs
 elba: elbasan
 elym: elymaic
+emod: emoji_modifier
+epres: emoji_presentation
 ethi: ethiopic
 ext: extender
 geor: georgian
@@ -89,6 +95,7 @@ kana: katakana
 khar: kharoshthi
 khmr: khmer
 khoj: khojki
+kits: khitan_small_script
 knda: kannada
 kthi: kaithi
 l: letter
@@ -127,7 +134,7 @@ mroo: mro
 mtei: meetei_mayek
 mult: multani
 mymr: myanmar
-n: number
+"n": number
 nand: nandinagari
 narb: old_north_arabian
 nbat: nabataean
@@ -226,6 +233,7 @@ xidc: xid_continue
 xids: xid_start
 xpeo: old_persian
 xsux: cuneiform
+yezi: yezidi
 yiii: yi
 z: separator
 zanb: zanabazar_square

data/lib/regexp_parser/scanner/scanner.rl

@@ -20,7 +20,7 @@
 
   group_open            = '(';
   group_close           = ')';
-  parantheses           = group_open | group_close;
+  parentheses           = group_open | group_close;
 
   set_open              = '[';
   set_close             = ']';
@@ -37,7 +37,7 @@
   class_posix           = ('[:' . '^'? . class_name_posix . ':]');
 
 
-  # these are not supported in ruby, and need verification
+  # these are not supported in ruby at the moment
   collating_sequence    = '[.' . (alpha | [\-])+ . '.]';
   character_equivalent  = '[=' . alpha . '=]';
 
@@ -58,6 +58,8 @@
 
   meta_sequence         = 'M-' . (backslash . ('c' | 'C-'))? . backslash? . any;
 
+  sequence_char         = [CMcux];
+
   zero_or_one           = '?' | '??' | '?+';
   zero_or_more          = '*' | '*?' | '*+';
   one_or_more           = '+' | '+?' | '++';
@@ -106,11 +108,15 @@
 
   group_named           = ('?' . group_name );
 
-  group_name_ref        = group_ref . (('<' . group_name_id_ab? . group_level? '>') |
-                                       ("'" . group_name_id_sq? . group_level? "'"));
+  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_number_ref      = group_ref . (('<' . group_number . group_level? '>') |
-                                       ("'" . 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_type            = group_atomic | group_passive | group_absence | group_named;
 
@@ -121,7 +127,7 @@
 
   # characters that 'break' a literal
   meta_char             = dot | backslash | alternation |
-                          curlies | parantheses | brackets |
+                          curlies | parentheses | brackets |
                           line_anchor | quantifier_greedy;
 
   literal_delimiters    = ']' | '}';
@@ -130,10 +136,12 @@
   ascii_nonprint        = (0x01..0x1f | 0x7f);
 
   non_literal_escape    = char_type_char | anchor_char | escaped_ascii |
-                          keep_mark | [xucCM];
+                          keep_mark | sequence_char;
+
+  # escapes that also work within a character set
+  set_escape            = backslash | brackets | escaped_ascii | property_char |
+                          sequence_char | single_codepoint_char_type;
 
-  non_set_escape        = (anchor_char - 'b') | group_ref | keep_mark |
-                          multi_codepoint_char_type | [0-9cCM];
 
   # EOF error, used where it can be detected
   action premature_end_error {
@@ -228,13 +236,13 @@
       emit(type, class_name.to_sym, text)
     };
 
-    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))
-    };
+    # 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))
@@ -249,16 +257,16 @@
   # set escapes scanner
   # --------------------------------------------------------------------------
   set_escape_sequence := |*
-    non_set_escape > (escaped_set_alpha, 2) {
-      emit(:escape, :literal, copy(data, ts-1, te))
-      fret;
-    };
-
-    any > (escaped_set_alpha, 1) {
+    set_escape > (escaped_set_alpha, 2) {
       fhold;
       fnext character_set;
       fcall escape_sequence;
     };
+
+    any > (escaped_set_alpha, 1) {
+      emit(:escape, :literal, copy(data, ts-1, te))
+      fret;
+    };
   *|;
 
 
@@ -538,67 +546,35 @@
 
     # Group backreference, named and numbered
     # ------------------------------------------------------------------------
-    backslash . (group_name_ref | group_number_ref) > (backslashed, 4) {
+    backslash . (group_name_backref | group_number_backref) > (backslashed, 4) {
       case text = copy(data, ts, te)
-      when /^\\([gk])(<>|'')/ # angle brackets
-        validation_error(:backref, 'ref/call', 'ref ID is empty')
-
-      # TODO: finer quirks of choosing recursive or non-recursive refs/calls.
-      # e.g.: `a-1` is a valid group id: 'aa'[/(?<a-1>a)\g<a-1>/] # => 'aa'
-      when /^\\([gk])<[^\p{digit}+\->][^>+\-]*>/ # angle-brackets
-        if $1 == 'k'
-          emit(:backref, :name_ref_ab, text)
-        else
-          emit(:backref, :name_call_ab, text)
-        end
-
-      when /^\\([gk])'[^\p{digit}+\-'][^'+\-]*'/ # single quotes
-        if $1 == 'k'
-          emit(:backref, :name_ref_sq, text)
-        else
-          emit(:backref, :name_call_sq, text)
-        end
-
-      when /^\\([gk])<\d+>/ # angle-brackets
-        if $1 == 'k'
-          emit(:backref, :number_ref_ab, text)
-        else
-          emit(:backref, :number_call_ab, text)
-        end
-
-      when /^\\([gk])'\d+'/ # single quotes
-        if $1 == 'k'
-          emit(:backref, :number_ref_sq, text)
-        else
-          emit(:backref, :number_call_sq, text)
-        end
-
-      when /^\\(?:g<\+|g<-|(k)<-)\d+>/ # angle-brackets
-        if $1 == 'k'
-          emit(:backref, :number_rel_ref_ab, text)
-        else
-          emit(:backref, :number_rel_call_ab, text)
-        end
-
-      when /^\\(?:g'\+|g'-|(k)'-)\d+'/ # single quotes
-        if $1 == 'k'
-          emit(:backref, :number_rel_ref_sq, text)
-        else
-          emit(:backref, :number_rel_call_sq, text)
-        end
-
-      when /^\\k<[^\p{digit}+\->][^>]*[+\-]\d+>/ # angle-brackets
-        emit(:backref, :name_recursion_ref_ab, text)
-
-      when /^\\k'[^\p{digit}+\-'][^']*[+\-]\d+'/ # single-quotes
-        emit(:backref, :name_recursion_ref_sq, text)
-
-      when /^\\([gk])<[+\-]?\d+[+\-]\d+>/ # angle-brackets
-        emit(:backref, :number_recursion_ref_ab, text)
-
-      when /^\\([gk])'[+\-]?\d+[+\-]\d+'/ # single-quotes
-        emit(:backref, :number_recursion_ref_sq, text)
+      when /^\\k(<>|'')/
+        validation_error(:backref, 'backreference', 'ref ID is empty')
+      when /^\\k(.)[^\p{digit}\-][^+\-]*\D$/
+        emit(:backref, $1 == '<' ? :name_ref_ab : :name_ref_sq, text)
+      when /^\\k(.)\d+\D$/
+        emit(:backref, $1 == '<' ? :number_ref_ab : :number_ref_sq, text)
+      when /^\\k(.)-\d+\D$/
+        emit(:backref, $1 == '<' ? :number_rel_ref_ab : :number_rel_ref_sq, text)
+      when /^\\k(.)[^\p{digit}\-].*[+\-]\d+\D$/
+        emit(:backref, $1 == '<' ? :name_recursion_ref_ab : :name_recursion_ref_sq, text)
+      when /^\\k(.)-?\d+[+\-]\d+\D$/
+        emit(:backref, $1 == '<' ? :number_recursion_ref_ab : :number_recursion_ref_sq, text)
+      end
+    };
 
+    # Group call, named and numbered
+    # ------------------------------------------------------------------------
+    backslash . (group_name_call | group_number_call) > (backslashed, 4) {
+      case text = copy(data, ts, te)
+      when /^\\g(<>|'')/
+        validation_error(:backref, 'subexpression call', 'ref ID is empty')
+      when /^\\g(.)[^\p{digit}+\->][^+\-]*/
+        emit(:backref, $1 == '<' ? :name_call_ab : :name_call_sq, text)
+      when /^\\g(.)\d+\D$/
+        emit(:backref, $1 == '<' ? :number_call_ab : :number_call_sq, text)
+      when /^\\g(.)[+-]\d+/
+        emit(:backref, $1 == '<' ? :number_rel_call_ab : :number_rel_call_sq, text)
       end
     };
 
@@ -675,12 +651,14 @@
 # THIS IS A GENERATED FILE, DO NOT EDIT DIRECTLY
 # This file was generated from lib/regexp_parser/scanner/scanner.rl
 
+require 'regexp_parser/error'
+
 class Regexp::Scanner
   # General scanner error (catch all)
-  class ScannerError < StandardError; end
+  class ScannerError < Regexp::Parser::Error; end
 
   # Base for all scanner validation errors
-  class ValidationError < StandardError
+  class ValidationError < Regexp::Parser::Error
     def initialize(reason)
       super reason
     end
@@ -782,14 +760,13 @@ class Regexp::Scanner
 
   # lazy-load property maps when first needed
   require 'yaml'
-  PROP_MAPS_DIR = File.expand_path('../scanner/properties', __FILE__)
 
   def self.short_prop_map
-    @short_prop_map ||= YAML.load_file("#{PROP_MAPS_DIR}/short.yml")
+    @short_prop_map ||= YAML.load_file("#{__dir__}/scanner/properties/short.yml")
   end
 
   def self.long_prop_map
-    @long_prop_map ||= YAML.load_file("#{PROP_MAPS_DIR}/long.yml")
+    @long_prop_map ||= YAML.load_file("#{__dir__}/scanner/properties/long.yml")
   end
 
   # Emits an array with the details of the scanned pattern