re2 1.5.0 → 1.6.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 8588a481a52f07a6c965094bb69c1aee177add80d071aee3a1771f97e19dc4f6
-  data.tar.gz: bb123e3a2d5352379548bf68f57fd195c8aa458e2da06f0864b7135682c19b1d
+  metadata.gz: 2151621115d04b197403c0f67276347c205928136405c05aac1fc70bf3a00dec
+  data.tar.gz: 8ba805a95c535ab7d30296a830448dea51da4f5f699a6aa3f858296b2590d188
 SHA512:
-  metadata.gz: b421f15ace868de905dece66db9e607636624cf2548eabe6be9979c397ac825035c0b24a5d8e8d6a351aa5809f6265d07f7b67a4c720a81e0b4e78f8d4599c67
-  data.tar.gz: fb7314faaf44a8847eadf207b0226ea71071ddb02ae41b7644ade69b69c06ff86f7c5d1a34042b4ef90633b98b6185bc07a6e2241bfa570089525e73eca6a61a
+  metadata.gz: 010e20dae629df302c35c6d5cc412c5a0add1cc3f0b6357114ce64c5d77e6c3bbca7401e57c0b86be5fe87acae17162cd7c860c079dc2c57898761f9fd8d4ce4
+  data.tar.gz: 6bfa2db432e91b87ab5d6fa21babd07a675df3b22904e20f1491322b1d34c8fe1c5814783049c33f84ab8f22e0b3c2ef9c2308161f8694dd6bcbd8ec4da278a4

data/README.md

@@ -4,8 +4,8 @@ re2 [![Build Status](https://github.com/mudge/re2/actions/workflows/tests.yml/ba
 A Ruby binding to [re2][], an "efficient, principled regular expression
 library".
 
-**Current version:** 1.5.0  
-**Supported Ruby versions:** 1.8.7, 1.9.3, 2.0, 2.1, 2.2, 2.3, 2.4, 2.5, 2.6, 2.7, 3.0  
+**Current version:** 1.6.0  
+**Supported Ruby versions:** 1.8.7, 1.9.3, 2.0, 2.1, 2.2, 2.3, 2.4, 2.5, 2.6, 2.7, 3.0, 3.1  
 **Supported re2 versions:** libre2.0 (< 2020-03-02), libre2.1 (2020-03-02), libre2.6 (2020-03-03), libre2.7 (2020-05-01), libre2.8 (2020-07-06), libre2.9 (2020-11-01)
 
 Installation
@@ -137,7 +137,7 @@ the pattern. After all patterns have been added, the set can be compiled using
 `RE2::Set#compile`, and then `RE2::Set#match` will return an `Array<Integer>`
 containing the indices of all the patterns that matched.
 
-``` ruby
+```ruby
 set = RE2::Set.new
 set.add("abc") #=> 0
 set.add("def") #=> 1
@@ -147,6 +147,27 @@ set.match("abcdefghi") #=> [0, 1, 2]
 set.match("ghidefabc") #=> [2, 1, 0]
 ```
 
+As of 1.6.0, you can use [Ruby's pattern matching](https://docs.ruby-lang.org/en/3.0/syntax/pattern_matching_rdoc.html) against `RE2::MatchData` with both array patterns and hash patterns:
+
+```ruby
+case RE2('(\w+) (\d+)').match("Alice 42")
+in [name, age]
+  puts "My name is #{name} and I am #{age} years old"
+else
+  puts "No match!"
+end
+# My name is Alice and I am 42 years old
+
+
+case RE2('(?P<name>\w+) (?P<age>\d+)').match("Alice 42")
+in {name:, age:}
+  puts "My name is #{name} and I am #{age} years old"
+else
+  puts "No match!"
+end
+# My name is Alice and I am 42 years old
+```
+
 Features
 --------
 
@@ -185,6 +206,8 @@ Features
   [`RE2.escape(unquoted)`](https://github.com/google/re2/blob/2016-02-01/re2/re2.h#L418) and
   `RE2.quote(unquoted)`
 
+* Pattern matching with `RE2::MatchData`
+
 Contributions
 -------------
 

data/ext/re2/re2.cc

@@ -687,6 +687,112 @@ static VALUE re2_matchdata_inspect(VALUE self) {
   return result;
 }
 
+/*
+ * Returns the array of submatches for pattern matching.
+ *
+ * @return [Array<String, nil>] the array of submatches
+ * @example
+ *   m = RE2::Regexp.new('(\d+)').match("bob 123")
+ *   m.deconstruct    #=> ["123"]
+ *
+ * @example pattern matching
+ *   case RE2::Regexp.new('(\d+) (\d+)').match("bob 123 456")
+ *   in x, y
+ *     puts "Matched #{x} #{y}"
+ *   else
+ *     puts "Unrecognised match"
+ *   end
+ */
+static VALUE re2_matchdata_deconstruct(VALUE self) {
+  int i;
+  re2_matchdata *m;
+  re2_pattern *p;
+  re2::StringPiece *match;
+  VALUE array;
+
+  Data_Get_Struct(self, re2_matchdata, m);
+  Data_Get_Struct(m->regexp, re2_pattern, p);
+
+  array = rb_ary_new2(m->number_of_matches - 1);
+  for (i = 1; i < m->number_of_matches; i++) {
+    match = &m->matches[i];
+
+    if (match->empty()) {
+      rb_ary_push(array, Qnil);
+    } else {
+      rb_ary_push(array, ENCODED_STR_NEW(match->data(), match->size(),
+            p->pattern->options().encoding() == RE2::Options::EncodingUTF8 ? "UTF-8" : "ISO-8859-1"));
+    }
+  }
+
+  return array;
+}
+
+/*
+ * Returns a hash of capturing group names to submatches for pattern matching.
+ *
+ * As this is used by Ruby's pattern matching, it will return an empty hash if given
+ * more keys than there are capturing groups. Given keys will populate the hash in
+ * order but an invalid name will cause the hash to be immediately returned.
+ *
+ * @return [Hash] a hash of capturing group names to submatches
+ * @param [Array<Symbol>, nil] keys an array of Symbol capturing group names or nil to return all names
+ * @example
+ *   m = RE2::Regexp.new('(?P<numbers>\d+) (?P<letters>[a-zA-Z]+)').match('123 abc')
+ *   m.deconstruct_keys(nil)                  #=> {:numbers => "123", :letters => "abc"}
+ *   m.deconstruct_keys([:numbers])           #=> {:numbers => "123"}
+ *   m.deconstruct_keys([:fruit])             #=> {}
+ *   m.deconstruct_keys([:letters, :fruit])   #=> {:letters => "abc"}
+ *
+ * @example pattern matching
+ *   case RE2::Regexp.new('(?P<numbers>\d+) (?P<letters>[a-zA-Z]+)').match('123 abc')
+ *   in numbers:, letters:
+ *     puts "Numbers: #{numbers}, letters: #{letters}"
+ *   else
+ *     puts "Unrecognised match"
+ *   end
+ */
+static VALUE re2_matchdata_deconstruct_keys(VALUE self, VALUE keys) {
+  int i;
+  VALUE capturing_groups, key;
+  re2_matchdata *m;
+  re2_pattern *p;
+  map<string, int> groups;
+  map<string, int>::iterator iterator;
+
+  Data_Get_Struct(self, re2_matchdata, m);
+  Data_Get_Struct(m->regexp, re2_pattern, p);
+
+  groups = p->pattern->NamedCapturingGroups();
+  capturing_groups = rb_hash_new();
+
+  if (NIL_P(keys)) {
+    for (iterator = groups.begin(); iterator != groups.end(); iterator++) {
+      rb_hash_aset(capturing_groups,
+          ID2SYM(rb_intern(iterator->first.data())),
+          re2_matchdata_nth_match(iterator->second, self));
+    }
+  } else {
+    Check_Type(keys, T_ARRAY);
+
+    if (p->pattern->NumberOfCapturingGroups() >= RARRAY_LEN(keys)) {
+      for (i = 0; i < RARRAY_LEN(keys); i++) {
+        key = rb_ary_entry(keys, i);
+        Check_Type(key, T_SYMBOL);
+        string name(rb_id2name(SYM2ID(key)));
+
+        if (groups.count(name) == 0) {
+          break;
+        }
+
+        rb_hash_aset(capturing_groups, key, re2_matchdata_nth_match(groups[name], self));
+      }
+    }
+  }
+
+  return capturing_groups;
+}
+
 /*
  * Returns a new RE2 object with a compiled version of
  * +pattern+ stored inside. Equivalent to +RE2.new+.
@@ -1245,7 +1351,7 @@ static VALUE re2_regexp_match(int argc, VALUE *argv, VALUE self) {
  *
  * @return [Boolean] whether the match was successful
  */
-static VALUE re2_regexp_match_query(VALUE self, VALUE text) {
+static VALUE re2_regexp_match_p(VALUE self, VALUE text) {
   VALUE argv[2];
   argv[0] = text;
   argv[1] = INT2FIX(0);
@@ -1514,6 +1620,7 @@ static VALUE re2_set_compile(VALUE self) {
  * @return [Bool] whether the underlying re2 outputs error information from Set matches
  */
 static VALUE re2_set_match_raises_errors_p(VALUE self) {
+  UNUSED(self);
 #ifdef HAVE_ERROR_INFO_ARGUMENT
   return Qtrue;
 #else
@@ -1526,18 +1633,40 @@ static VALUE re2_set_match_raises_errors_p(VALUE self) {
  * integer indices of the matching patterns if matched or an empty array if
  * there are no matches.
  *
- * @param [String] str the text to match against
- * @param [Hash] options the options with which to match
- * @option options [Boolean] :exception (true) whether to raise exceptions with re2's error information (not supported on ABI version 0 of re2)
- * @return [Array<Integer>] the indices of matching regexps
- * @raise [MatchError] if an error occurs while matching
- * @raise [UnsupportedError] if using the :exception option against a version of re2 that does not support it
- * @example
- *   set = RE2::Set.new
- *   set.add("abc")
- *   set.add("def")
- *   set.compile
- *   set.match("abcdef")    # => [0, 1]
+ * @return [Array<Integer>]
+ *
+ * @overload match(str)
+ *   Returns an array of integer indices of patterns matching the given string
+ *   (if any). Raises exceptions if there are any errors while matching.
+ *
+ *   @param [String] str the text to match against
+ *   @return [Array<Integer>] the indices of matching regexps
+ *   @raise [MatchError] if an error occurs while matching
+ *   @raise [UnsupportedError] if the underlying version of re2 does not output error information
+ *   @example
+ *     set = RE2::Set.new
+ *     set.add("abc")
+ *     set.add("def")
+ *     set.compile
+ *     set.match("abcdef")    # => [0, 1]
+ *
+ * @overload match(str, options)
+ *   Returns an array of integer indices of patterns matching the given string
+ *   (if any). Raises exceptions if there are any errors while matching and the
+ *   :exception option is set to true.
+ *
+ *   @param [String] str the text to match against
+ *   @param [Hash] options the options with which to match
+ *   @option options [Boolean] :exception (true) whether to raise exceptions with re2's error information (not supported on ABI version 0 of re2)
+ *   @return [Array<Integer>] the indices of matching regexps
+ *   @raise [MatchError] if an error occurs while matching
+ *   @raise [UnsupportedError] if the underlying version of re2 does not output error information
+ *   @example
+ *     set = RE2::Set.new
+ *     set.add("abc")
+ *     set.add("def")
+ *     set.compile
+ *     set.match("abcdef", :exception => true)    # => [0, 1]
  */
 static VALUE re2_set_match(int argc, VALUE *argv, VALUE self) {
   VALUE str, options, exception_option;
@@ -1644,6 +1773,10 @@ void Init_re2(void) {
         RUBY_METHOD_FUNC(re2_matchdata_to_s), 0);
   rb_define_method(re2_cMatchData, "inspect",
       RUBY_METHOD_FUNC(re2_matchdata_inspect), 0);
+  rb_define_method(re2_cMatchData, "deconstruct",
+      RUBY_METHOD_FUNC(re2_matchdata_deconstruct), 0);
+  rb_define_method(re2_cMatchData, "deconstruct_keys",
+      RUBY_METHOD_FUNC(re2_matchdata_deconstruct_keys), 1);
 
   rb_define_method(re2_cScanner, "string",
       RUBY_METHOD_FUNC(re2_scanner_string), 0);
@@ -1674,11 +1807,11 @@ void Init_re2(void) {
   rb_define_method(re2_cRegexp, "match", RUBY_METHOD_FUNC(re2_regexp_match),
       -1);
   rb_define_method(re2_cRegexp, "match?",
-      RUBY_METHOD_FUNC(re2_regexp_match_query), 1);
+      RUBY_METHOD_FUNC(re2_regexp_match_p), 1);
   rb_define_method(re2_cRegexp, "=~",
-      RUBY_METHOD_FUNC(re2_regexp_match_query), 1);
+      RUBY_METHOD_FUNC(re2_regexp_match_p), 1);
   rb_define_method(re2_cRegexp, "===",
-      RUBY_METHOD_FUNC(re2_regexp_match_query), 1);
+      RUBY_METHOD_FUNC(re2_regexp_match_p), 1);
   rb_define_method(re2_cRegexp, "scan",
       RUBY_METHOD_FUNC(re2_regexp_scan), 1);
   rb_define_method(re2_cRegexp, "to_s", RUBY_METHOD_FUNC(re2_regexp_to_s), 0);

data/spec/re2/match_data_spec.rb

@@ -241,4 +241,62 @@ RSpec.describe RE2::MatchData do
       expect(md.end(:foo)).to be_nil
     end
   end
+
+  describe "#deconstruct" do
+    it "returns all capturing groups" do
+      md = RE2::Regexp.new('w(o)(o)').match('woo')
+
+      expect(md.deconstruct).to eq(['o', 'o'])
+    end
+
+    it "includes optional capturing groups as nil" do
+      md = RE2::Regexp.new('w(.)(.)(.)?').match('woo')
+
+      expect(md.deconstruct).to eq(['o', 'o', nil])
+    end
+  end
+
+  describe "#deconstruct_keys" do
+    it "returns all named captures if given nil" do
+      md = RE2::Regexp.new('(?P<numbers>\d+) (?P<letters>[a-zA-Z]+)').match('123 abc')
+
+      expect(md.deconstruct_keys(nil)).to eq(:numbers => '123', :letters => 'abc')
+    end
+
+    it "returns only named captures if given names" do
+      md = RE2::Regexp.new('(?P<numbers>\d+) (?P<letters>[a-zA-Z]+)').match('123 abc')
+
+      expect(md.deconstruct_keys([:numbers])).to eq(:numbers => '123')
+    end
+
+    it "returns named captures up until an invalid name is given" do
+      md = RE2::Regexp.new('(?P<numbers>\d+) (?P<letters>[a-zA-Z]+)').match('123 abc')
+
+      expect(md.deconstruct_keys([:numbers, :punctuation])).to eq(:numbers => '123')
+    end
+
+    it "returns an empty hash if given more capture names than exist" do
+      md = RE2::Regexp.new('(?P<numbers>\d+) (?P<letters>[a-zA-Z]+)').match('123 abc')
+
+      expect(md.deconstruct_keys([:numbers, :letters, :punctuation])).to eq({})
+    end
+
+    it "returns an empty hash if there are no named capturing groups" do
+      md = RE2::Regexp.new('(\d+) ([a-zA-Z]+)').match('123 abc')
+
+      expect(md.deconstruct_keys(nil)).to eq({})
+    end
+
+    it "raises an error if given a non-array of keys" do
+      md = RE2::Regexp.new('(?P<numbers>\d+) (?P<letters>[a-zA-Z]+)').match('123 abc')
+
+      expect { md.deconstruct_keys(0) }.to raise_error(TypeError)
+    end
+
+    it "raises an error if given keys as non-symbols" do
+      md = RE2::Regexp.new('(?P<numbers>\d+) (?P<letters>[a-zA-Z]+)').match('123 abc')
+
+      expect { md.deconstruct_keys([0]) }.to raise_error(TypeError)
+    end
+  end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: re2
 version: !ruby/object:Gem::Version
-  version: 1.5.0
+  version: 1.6.0
 platform: ruby
 authors:
 - Paul Mucur
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2022-10-16 00:00:00.000000000 Z
+date: 2022-10-22 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake-compiler