Adding Regex.EnumerateMatches

src/libraries/System.Text.RegularExpressions/ref/System.Text.RegularExpressions.cs

@@ -171,6 +171,10 @@ public static void CompileToAssembly(System.Text.RegularExpressions.RegexCompila
         public static int Count(System.ReadOnlySpan<char> input, [System.Diagnostics.CodeAnalysis.StringSyntax(System.Diagnostics.CodeAnalysis.StringSyntaxAttribute.Regex, "options")] string pattern, System.Text.RegularExpressions.RegexOptions options) { throw null; }
         public static int Count(System.ReadOnlySpan<char> input, [System.Diagnostics.CodeAnalysis.StringSyntax(System.Diagnostics.CodeAnalysis.StringSyntaxAttribute.Regex, "options")] string pattern, System.Text.RegularExpressions.RegexOptions options, System.TimeSpan matchTimeout) { throw null; }
         public static string Escape(string str) { throw null; }
+        public System.Text.RegularExpressions.Regex.ValueMatchEnumerator EnumerateMatches(System.ReadOnlySpan<char> input) { throw null; }
+        public static System.Text.RegularExpressions.Regex.ValueMatchEnumerator EnumerateMatches(System.ReadOnlySpan<char> input, [System.Diagnostics.CodeAnalysis.StringSyntaxAttribute("Regex")] string pattern) { throw null; }
+        public static System.Text.RegularExpressions.Regex.ValueMatchEnumerator EnumerateMatches(System.ReadOnlySpan<char> input, [System.Diagnostics.CodeAnalysis.StringSyntaxAttribute("Regex", new object[]{ "options"})] string pattern, System.Text.RegularExpressions.RegexOptions options) { throw null; }
+        public static System.Text.RegularExpressions.Regex.ValueMatchEnumerator EnumerateMatches(System.ReadOnlySpan<char> input, [System.Diagnostics.CodeAnalysis.StringSyntaxAttribute("Regex", new object[]{ "options"})] string pattern, System.Text.RegularExpressions.RegexOptions options, System.TimeSpan matchTimeout) { throw null; }
         public string[] GetGroupNames() { throw null; }
         public int[] GetGroupNumbers() { throw null; }
         public string GroupNameFromNumber(int i) { throw null; }
@@ -220,6 +224,14 @@ void System.Runtime.Serialization.ISerializable.GetObjectData(System.Runtime.Ser
         protected bool UseOptionC() { throw null; }
         protected internal bool UseOptionR() { throw null; }
         protected internal static void ValidateMatchTimeout(System.TimeSpan matchTimeout) { }
+        public ref partial struct ValueMatchEnumerator
+        {
+            private object _dummy;
+            private int _dummyPrimitive;
+            public readonly System.Text.RegularExpressions.ValueMatch Current { get { throw null; } }
+            public readonly System.Text.RegularExpressions.Regex.ValueMatchEnumerator GetEnumerator() { throw null; }
+            public bool MoveNext() { throw null; }
+        }
     }
     [System.ObsoleteAttribute("Regex.CompileToAssembly is obsolete and not supported. Use the RegexGeneratorAttribute with the regular expression source generator instead.", DiagnosticId = "SYSLIB0036", UrlFormat = "https://aka.ms/dotnet-warnings/{0}")]
     public partial class RegexCompilationInfo
@@ -359,4 +371,10 @@ public abstract partial class RegexRunnerFactory
         protected RegexRunnerFactory() { }
         protected internal abstract System.Text.RegularExpressions.RegexRunner CreateInstance();
     }
+    public readonly ref partial struct ValueMatch
+    {
+        private readonly int _dummyPrimitive;
+        public int Index { get { throw null; } }
+        public int Length { get { throw null; } }
+    }
 }

src/libraries/System.Text.RegularExpressions/src/System.Text.RegularExpressions.csproj

@@ -7,6 +7,7 @@
   <ItemGroup>
     <Compile Include="System\Collections\HashtableExtensions.cs" />
     <Compile Include="System\Collections\Generic\ValueListBuilder.Pop.cs" />
+    <Compile Include="System\Text\RegularExpressions\ValueMatch.cs" />
     <Compile Include="System\Threading\StackHelper.cs" />
     <Compile Include="System\Text\SegmentStringBuilder.cs" />
     <Compile Include="System\Text\RegularExpressions\Capture.cs" />
@@ -23,6 +24,7 @@
     <Compile Include="System\Text\RegularExpressions\Regex.Match.cs" />
     <Compile Include="System\Text\RegularExpressions\Regex.Replace.cs" />
     <Compile Include="System\Text\RegularExpressions\Regex.Split.cs" />
+    <Compile Include="System\Text\RegularExpressions\Regex.EnumerateMatches.cs" />
     <Compile Include="System\Text\RegularExpressions\Regex.Timeout.cs" />
     <Compile Include="System\Text\RegularExpressions\RegexCaseBehavior.cs" />
     <Compile Include="System\Text\RegularExpressions\RegexCaseEquivalences.Data.cs" />

src/libraries/System.Text.RegularExpressions/src/System/Text/RegularExpressions/Regex.EnumerateMatches.cs

@@ -0,0 +1,144 @@
+// Licensed to the .NET Foundation under one or more agreements.
+// The .NET Foundation licenses this file to you under the MIT license.
+
+using System.Diagnostics.CodeAnalysis;
+
+namespace System.Text.RegularExpressions
+{
+    public partial class Regex
+    {
+        /// <summary>
+        /// Searches an input span for all occurrences of a regular expression and returns a <see cref="ValueMatchEnumerator"/> to iterate over the matches.
+        /// </summary>
+        /// <remarks>
+        /// The enumerator returned by this method, as well as the structs returned by the enumerator that wrap each match found in the input are ref structs which
+        /// make this method be amortized allocation free.
+        /// </remarks>
+        /// <param name="input">The span to search for a match.</param>
+        /// <param name="pattern">The regular expression pattern to match.</param>
+        /// <returns>A <see cref="ValueMatchEnumerator"/> to iterate over the matches.</returns>
+        /// <exception cref="ArgumentNullException"><paramref name="pattern"/> is null.</exception>
+        /// <exception cref="RegexParseException">A regular expression parsing error occurred.</exception>
+        public static ValueMatchEnumerator EnumerateMatches(ReadOnlySpan<char> input, [StringSyntax(StringSyntaxAttribute.Regex)] string pattern) =>
+            RegexCache.GetOrAdd(pattern).EnumerateMatches(input);
+
+        /// <summary>
+        /// Searches an input span for all occurrences of a regular expression and returns a <see cref="ValueMatchEnumerator"/> to iterate over the matches.
+        /// </summary>
+        /// <remarks>
+        /// The enumerator returned by this method, as well as the structs returned by the enumerator that wrap each match found in the input are ref structs which
+        /// make this method be amortized allocation free.
+        /// </remarks>
+        /// <param name="input">The span to search for a match.</param>
+        /// <param name="pattern">The regular expression pattern to match.</param>
+        /// <param name="options">A bitwise combination of the enumeration values that specify options for matching.</param>
+        /// <returns>A <see cref="ValueMatchEnumerator"/> to iterate over the matches.</returns>
+        /// <exception cref="ArgumentNullException"><paramref name="pattern"/> is null.</exception>
+        /// <exception cref="ArgumentOutOfRangeException"><paramref name="options"/> is not a valid bitwise combination of RegexOptions values.</exception>
+        /// <exception cref="RegexParseException">A regular expression parsing error occurred.</exception>
+        public static ValueMatchEnumerator EnumerateMatches(ReadOnlySpan<char> input, [StringSyntax(StringSyntaxAttribute.Regex, "options")] string pattern, RegexOptions options) =>
+            RegexCache.GetOrAdd(pattern, options, s_defaultMatchTimeout).EnumerateMatches(input);
+
+        /// <summary>
+        /// Searches an input span for all occurrences of a regular expression and returns a <see cref="ValueMatchEnumerator"/> to iterate over the matches.
+        /// </summary>
+        /// <remarks>
+        /// The enumerator returned by this method, as well as the structs returned by the enumerator that wrap each match found in the input are ref structs which
+        /// make this method be amortized allocation free.
+        /// </remarks>
+        /// <param name="input">The span to search for a match.</param>
+        /// <param name="pattern">The regular expression pattern to match.</param>
+        /// <param name="options">A bitwise combination of the enumeration values that specify options for matching.</param>
+        /// <param name="matchTimeout">A time-out interval, or <see cref="InfiniteMatchTimeout"/> to indicate that the method should not time out.</param>
+        /// <returns>A <see cref="ValueMatchEnumerator"/> to iterate over the matches.</returns>
+        /// <exception cref="ArgumentNullException"><paramref name="pattern"/> is null.</exception>
+        /// <exception cref="ArgumentOutOfRangeException"><paramref name="options"/> is not a valid bitwise combination of RegexOptions values, or <paramref name="matchTimeout"/> is negative, zero, or greater than approximately 24 days.</exception>
+        /// <exception cref="RegexParseException">A regular expression parsing error occurred.</exception>
+        public static ValueMatchEnumerator EnumerateMatches(ReadOnlySpan<char> input, [StringSyntax(StringSyntaxAttribute.Regex, "options")] string pattern, RegexOptions options, TimeSpan matchTimeout) =>
+            RegexCache.GetOrAdd(pattern, options, matchTimeout).EnumerateMatches(input);
+
+        /// <summary>
+        /// Searches an input span for all occurrences of a regular expression and returns a <see cref="ValueMatchEnumerator"/> to iterate over the matches.
+        /// </summary>
+        /// <remarks>
+        /// The enumerator returned by this method, as well as the structs returned by the enumerator that wrap each match found in the input are ref structs which
+        /// make this method be amortized allocation free.
+        /// </remarks>
+        /// <param name="input">The span to search for a match.</param>
+        /// <returns>A <see cref="ValueMatchEnumerator"/> to iterate over the matches.</returns>
+        public ValueMatchEnumerator EnumerateMatches(ReadOnlySpan<char> input) =>
+            new ValueMatchEnumerator(this, input, RightToLeft ? input.Length : 0);
+
+        /// <summary>
+        /// Represents an enumerator containing the set of successful matches found by iteratively applying a regular expression pattern to the input span. The
+        /// enumerator has no public constructor. The <see cref="Regex.EnumerateMatches(ReadOnlySpan{char})"/> method returns a <see cref="Regex.ValueMatchEnumerator"/>
+        /// object.
+        /// </summary>
+        /// <remarks>
+        /// The enumerator will lazily iterate over zero or more <see cref="ValueMatch"/> objects. If there is at least one successful match in the span, then
+        /// <see cref="MoveNext"/> returns <see langword="true"/> and <see cref="Current"/> will contain the first <see cref="ValueMatch"/>. If there are no successful matches,
+        /// then <see cref="MoveNext"/> returns <see langword="false"/> and <see cref="Current"/> throws an <see cref="InvalidOperationException"/>.
+        ///
+        /// This type is a ref struct since it stores the input span as a field in order to be able to lazily iterate over it.
+        /// </remarks>
+        public ref struct ValueMatchEnumerator
+        {
+            private readonly Regex _regex;
+            private readonly ReadOnlySpan<char> _input;
+            private ValueMatch _matchReference;
+            private int _startAt;
+            private int _prevLen;
+            private bool _hasValidValue;
+
+            /// <summary>
+            /// Creates an instance of the <see cref="ValueMatchEnumerator"/> for the passed in <paramref name="regex"/> which iterates over <paramref name="input"/>.
+            /// </summary>
+            /// <param name="regex">The <see cref="Regex"/> to use for finding matches.</param>
+            /// <param name="input">The input span to iterate over.</param>
+            /// <param name="startAt">The position where the engine should start looking for matches from.</param>
+            internal ValueMatchEnumerator(Regex regex, ReadOnlySpan<char> input, int startAt)
+            {
+                _regex = regex;
+                _input = input;
+                _matchReference = default;
+                _startAt = startAt;
+                _prevLen = -1;
+                _hasValidValue = false;
+            }
+
+            /// <summary>
+            /// Provides an enumerator that iterates through the matches in the input span.
+            /// </summary>
+            /// <returns>A copy of this enumerator.</returns>
+            public readonly ValueMatchEnumerator GetEnumerator() => this;
+
+            /// <summary>
+            /// Advances the enumerator to the next match in the span.
+            /// </summary>
+            /// <returns>
+            /// <see langword="true"/> if the enumerator was successfully advanced to the next element; <see langword="false"/> if the enumerator cannot find additional matches.
+            /// </returns>
+            public bool MoveNext()
+            {
+                Match? match = _regex.RunSingleMatch(quick: false, _prevLen, _input, _startAt);
+                if (match is not null && match != RegularExpressions.Match.Empty)
+                {
+                    _matchReference = new ValueMatch(match);
+                    _hasValidValue = true;
+                    _startAt = match._textpos;
+                    _prevLen = match.Length;
+                    return true;
+                }
+                _hasValidValue = false;
+                _matchReference = default;
+                return false;
+            }
+
+            /// <summary>
+            /// Gets the <see cref="ValueMatch"/> element at the current position of the enumerator.
+            /// </summary>
+            /// <exception cref="InvalidOperationException">Enumeration has either not started or has already finished.</exception>
+            public readonly ValueMatch Current => _hasValidValue ? _matchReference : throw new InvalidOperationException(SR.EnumNotStarted);
+        }
+    }
+}

src/libraries/System.Text.RegularExpressions/src/System/Text/RegularExpressions/Regex.Match.cs

@@ -87,7 +87,7 @@ public bool IsMatch(string input)
         /// <returns><see langword="true"/> if the regular expression finds a match; otherwise, <see langword="false"/>.</returns>
         /// <exception cref="RegexMatchTimeoutException">A time-out ocurred.</exception>
         public bool IsMatch(ReadOnlySpan<char> input) =>
-            RunSingleMatch(input, RightToLeft ? input.Length : 0) is null;
+            RunSingleMatch(quick: true, -1, input, RightToLeft ? input.Length : 0) is null;
 
         /// <summary>
         /// Searches the input string for one or more matches using the previous pattern and options,

src/libraries/System.Text.RegularExpressions/src/System/Text/RegularExpressions/Regex.cs

@@ -403,7 +403,7 @@ protected void InitializeReferences()
         }
 
         /// <summary>Internal worker which will scan the passed in span <paramref name="input"/> for a match. Used by public APIs.</summary>
-        internal Match? RunSingleMatch(ReadOnlySpan<char> input, int startat)
+        internal Match? RunSingleMatch(bool quick, int prevlen, ReadOnlySpan<char> input, int startat)
         {
             // startat parameter is always either 0 or input.Length since public API for IsMatch doesn't have an overload
             // that takes in startat.
@@ -413,13 +413,36 @@ protected void InitializeReferences()
             try
             {
                 runner.InitializeTimeout(internalMatchTimeout);
-                runner.InitializeForScan(this, input, startat, quick: true);
+                runner.InitializeForScan(this, input, startat, quick);
+
+                int stoppos = RightToLeft ? 0 : input.Length;
+
+                // If previous match was empty or failed, advance by one before matching.
+                if (prevlen == 0)
+                {
+                    if (runner.runtextstart == stoppos)
+                    {
+                        return RegularExpressions.Match.Empty;
+                    }
+
+                    runner.runtextpos += RightToLeft ? -1 : 1;
+                }
 
                 runner.Scan(input);
 
                 // If runmatch is null it means that an override of Scan didn't implement it correctly, so we will
                 // let this null ref since there are lots of ways where you can end up in a erroneous state.
-                return runner.runmatch!.FoundMatch ? null : RegularExpressions.Match.Empty;
+                if (runner.runmatch!.FoundMatch)
+                {
+                    if (quick)
+                    {
+                        return null;
+                    }
+                    runner.runmatch.Tidy(runner.runtextpos, 0);
+                    return runner.runmatch;
+                }
+
+                return RegularExpressions.Match.Empty;
             }
             finally
             {

src/libraries/System.Text.RegularExpressions/src/System/Text/RegularExpressions/ValueMatch.cs

@@ -0,0 +1,38 @@
+// Licensed to the .NET Foundation under one or more agreements.
+// The .NET Foundation licenses this file to you under the MIT license.
+
+namespace System.Text.RegularExpressions
+{
+    /// <summary>
+    /// Represents the results from a single regular expression match.
+    /// </summary>
+    /// <remarks>
+    /// The <see cref="ValueMatch"/> type is immutable and has no public constructor. An instance of the <see cref="ValueMatch"/> struct is returned by the
+    /// <see cref="Regex.ValueMatchEnumerator.Current"/> method when iterating over the results from calling <see cref="Regex.EnumerateMatches(ReadOnlySpan{char})"/>.
+    /// </remarks>
+    public readonly ref struct ValueMatch
+    {
+        private readonly int _index;
+        private readonly int _length;
+
+        /// <summary>
+        /// Crates an instance of the <see cref="ValueMatch"/> type based on the passed in <paramref name="match"/>.
+        /// </summary>
+        /// <param name="match">The <see cref="Match"/> object represented by this ValueMatch.</param>
+        internal ValueMatch(Match match)
+        {
+            _index = match.Index;
+            _length = match.Length;
+        }
+
+        /// <summary>
+        /// Gets the position in the original span where the first character of the captured sliced span is found.
+        /// </summary>
+        public int Index => _index;
+
+        /// <summary>
+        /// Gets the length of the captured sliced span.
+        /// </summary>
+        public int Length => _length;
+    }
+}

src/libraries/System.Text.RegularExpressions/tests/FunctionalTests/Regex.Count.Tests.cs

@@ -8,7 +8,7 @@
 
 namespace System.Text.RegularExpressions.Tests
 {
-    public class RegexCountTests
+    public partial class RegexCountTests
     {
         [Theory]
         [MemberData(nameof(Count_ReturnsExpectedCount_TestData))]