[automated] Merge branch 'main' => 'main-vs-deps' (#78838)

I detected changes in the main branch which have not been merged yet to
main-vs-deps. I'm a robot and am configured to help you automatically
keep main-vs-deps up to date, so I've opened this PR.

This PR merges commits made on main by the following committers:

* jjonescz
* JoeRobich
* CyrusNajmabadi
* dibarbet
* tmat

## Instructions for merging from UI

This PR will not be auto-merged. When pull request checks pass, complete
this PR by creating a merge commit, *not* a squash or rebase commit.

<img alt="merge button instructions"
src="https://i.imgur.com/GepcNJV.png" width="300" />

If this repo does not allow creating merge commits from the GitHub UI,
use command line instructions.

## Instructions for merging via command line

Run these commands to merge this pull request from the command line.

``` sh
git fetch
git checkout main
git pull --ff-only
git checkout main-vs-deps
git pull --ff-only
git merge --no-ff main

# If there are merge conflicts, resolve them and then run git merge --continue to complete the merge
# Pushing the changes to the PR branch will re-trigger PR validation.
git push https://github.com/dotnet/roslyn HEAD:merge/main-to-main-vs-deps
```

<details>
<summary>or if you are using SSH</summary>

```
git push [email protected]:dotnet/roslyn HEAD:merge/main-to-main-vs-deps
```

</details>


After PR checks are complete push the branch
```
git push
```

## Instructions for resolving conflicts

:warning: If there are merge conflicts, you will need to resolve them
manually before merging. You can do this [using GitHub][resolve-github]
or using the [command line][resolve-cli].

[resolve-github]:
https://help.github.com/articles/resolving-a-merge-conflict-on-github/
[resolve-cli]:
https://help.github.com/articles/resolving-a-merge-conflict-using-the-command-line/

## Instructions for updating this pull request

Contributors to this repo have permission update this pull request by
pushing to the branch 'merge/main-to-main-vs-deps'. This can be done to
resolve conflicts or make other changes to this pull request before it
is merged.
The provided examples assume that the remote is named 'origin'. If you
have a different remote name, please replace 'origin' with the name of
your remote.

```
git fetch
git checkout -b merge/main-to-main-vs-deps origin/main-vs-deps
git pull https://github.com/dotnet/roslyn merge/main-to-main-vs-deps
(make changes)
git commit -m "Updated PR with my changes"
git push https://github.com/dotnet/roslyn HEAD:merge/main-to-main-vs-deps
```

<details>
    <summary>or if you are using SSH</summary>

```
git fetch
git checkout -b merge/main-to-main-vs-deps origin/main-vs-deps
git pull [email protected]:dotnet/roslyn merge/main-to-main-vs-deps
(make changes)
git commit -m "Updated PR with my changes"
git push [email protected]:dotnet/roslyn HEAD:merge/main-to-main-vs-deps
```

</details>

Contact .NET Core Engineering (dotnet/dnceng) if you have questions or
issues.
Also, if this PR was generated incorrectly, help us fix it. See
https://github.com/dotnet/arcade/blob/main/.github/workflows/scripts/inter-branch-merge.ps1.

Author: dotnet-policy-service[bot]

src/Compilers/CSharp/Portable/Parser/DirectiveParser.cs

@@ -695,19 +695,22 @@ private DirectiveTriviaSyntax ParseShebangDirective(SyntaxToken hash, SyntaxToke
 
         private DirectiveTriviaSyntax ParseIgnoredDirective(SyntaxToken hash, SyntaxToken colon, bool isActive, bool isFollowingToken)
         {
-            if (!lexer.Options.FileBasedProgram)
+            if (isActive)
             {
-                colon = this.AddError(colon, ErrorCode.ERR_PPIgnoredNeedsFileBasedProgram);
-            }
+                if (!lexer.Options.FileBasedProgram)
+                {
+                    colon = this.AddError(colon, ErrorCode.ERR_PPIgnoredNeedsFileBasedProgram);
+                }
 
-            if (isFollowingToken)
-            {
-                colon = this.AddError(colon, ErrorCode.ERR_PPIgnoredFollowsToken);
-            }
+                if (isFollowingToken)
+                {
+                    colon = this.AddError(colon, ErrorCode.ERR_PPIgnoredFollowsToken);
+                }
 
-            if (_context.SeenAnyIfDirectives)
-            {
-                colon = this.AddError(colon, ErrorCode.ERR_PPIgnoredFollowsIf);
+                if (_context.SeenAnyIfDirectives)
+                {
+                    colon = this.AddError(colon, ErrorCode.ERR_PPIgnoredFollowsIf);
+                }
             }
 
             SyntaxToken endOfDirective = this.lexer.LexEndOfDirectiveWithOptionalContent(out SyntaxToken content);

src/Compilers/CSharp/Portable/Parser/Lexer.cs

@@ -475,12 +475,7 @@ private void ScanSyntaxToken(ref TokenInfo info)
                             // dots followed by an integer (which will be treated as a range expression).
                             //
                             // Move back one space to see what's before this dot and adjust accordingly.
-
-                            this.TextWindow.Reset(atDotPosition - 1);
-                            var priorCharacterIsDot = this.TextWindow.PeekChar() is '.';
-                            this.TextWindow.Reset(atDotPosition);
-
-                            if (priorCharacterIsDot)
+                            if (this.TextWindow.PreviousChar() is '.')
                             {
                                 // We have two dots in a row.  Treat the second dot as a dot, not the start of a number literal.
                                 TextWindow.AdvanceChar();

src/Compilers/CSharp/Portable/Parser/SlidingTextWindow.cs

@@ -3,12 +3,8 @@
 // See the LICENSE file in the project root for more information.
 
 using System;
-using System.Collections.Concurrent;
 using System.Diagnostics;
 using System.Text;
-using Microsoft.CodeAnalysis.Collections;
-using Microsoft.CodeAnalysis.CSharp.Symbols;
-using Microsoft.CodeAnalysis.CSharp.Syntax;
 using Microsoft.CodeAnalysis.PooledObjects;
 using Microsoft.CodeAnalysis.Text;
 using Roslyn.Utilities;
@@ -374,6 +370,25 @@ public char PeekChar(int delta)
             return ch;
         }
 
+        public char PreviousChar()
+        {
+            Debug.Assert(this.Position > 0);
+            if (_offset > 0)
+            {
+                // The allowed region of the window that can be read is from 0 to _characterWindowCount (which _offset
+                // is in between).  So as long as _offset is greater than 0, we can read the previous character directly
+                // from the current chunk of characters in the window.
+                return this.CharacterWindow[_offset - 1];
+            }
+
+            // The prior character isn't in the window (trying to read the current character caused us to
+            // read in the next chunk of text into the window, throwing out the preceding characters).
+            // Just go back to the source text to find this character.  While more expensive, this should
+            // be rare given that most of the time we won't be calling this right after loading a new text
+            // chunk.
+            return this.Text[this.Position - 1];
+        }
+
         /// <summary>
         /// If the next characters in the window match the given string,
         /// then advance past those characters.  Otherwise, do nothing.
@@ -480,5 +495,16 @@ internal static char GetCharsFromUtf32(uint codepoint, out char lowSurrogate)
                 return (char)((codepoint - 0x00010000) / 0x0400 + 0xD800);
             }
         }
+
+        internal TestAccessor GetTestAccessor()
+            => new TestAccessor(this);
+
+        internal readonly struct TestAccessor(SlidingTextWindow window)
+        {
+            private readonly SlidingTextWindow _window = window;
+
+            internal void SetDefaultCharacterWindow()
+                => _window._characterWindow = new char[DefaultWindowLength];
+        }
     }
 }

src/Compilers/CSharp/Test/Syntax/LexicalAndXml/LexicalTests.cs

@@ -8,6 +8,7 @@
 using System.Collections.Generic;
 using System.Globalization;
 using System.Linq;
+using System.Threading;
 using Microsoft.CodeAnalysis.CSharp.Syntax.InternalSyntax;
 using Microsoft.CodeAnalysis.CSharp.Test.Utilities;
 using Microsoft.CodeAnalysis.Text;
@@ -4572,5 +4573,90 @@ disabled text 2
             Assert.True(trivia.ContainsDiagnostics);
             Assert.Equal((int)ErrorCode.ERR_Merge_conflict_marker_encountered, trivia.Errors().Single().Code);
         }
+
+        [Fact, WorkItem("https://github.com/dotnet/roslyn/issues/78593")]
+        public void TestDotPrefixedNumberStartingAtStartOfSlidingTextWindow()
+        {
+            // This test depends on the line endings for the file being \r\n to ensure the right contents lines up at
+            // the right locations.
+            //
+            // It specifically validates what happens when we see `.0` at the start of the
+            // sliding text window, where the lexer tries to peek back one char to see if this
+            // is actually `..0` (a range expr) or `.0` (a floating point number).
+            var code = Resources.DotPrefixedNumberStartingAtStartOfSlidingTextWindow;
+            if (!code.Contains("\r\n"))
+                code = code.Replace("\n", "\r\n");
+
+            var sourceText = SourceText.From(code);
+
+            {
+                // Run a full parse, and validate the tree returned).
+
+                using var lexer = new Lexer(sourceText, CSharpParseOptions.Default);
+
+                // Ensure we have a normal window size, not some larger array that another test created and cached in
+                // the window pool
+                lexer.TextWindow.GetTestAccessor().SetDefaultCharacterWindow();
+
+                using var parser = new LanguageParser(lexer, oldTree: null, changes: null);
+
+                Microsoft.CodeAnalysis.SyntaxTreeExtensions.VerifySource(
+                    sourceText, parser.ParseCompilationUnit().CreateRed());
+            }
+
+            {
+                // Now, replicate the same conditions that hte parser runs through by driving the a new lexer here
+                // directly.  That ensures that we are actually validating exactly the conditions that led to the bug
+                // (a dot token starting a number, right at the start of the character window).
+                var lexer = new Lexer(sourceText, CSharpParseOptions.Default);
+
+                // Ensure we have a normal window size, not some larger array that another test created and cached in
+                // the window pool
+                lexer.TextWindow.GetTestAccessor().SetDefaultCharacterWindow();
+
+                var mode = LexerMode.Syntax;
+                for (var i = 0; i < 1326; i++)
+                    lexer.Lex(ref mode);
+
+                // Lexer will read from index 0 in the arrray.
+                Assert.Equal(0, lexer.TextWindow.Offset);
+
+                // We have 205 real chars in the window
+                Assert.Equal(205, lexer.TextWindow.CharacterWindowCount);
+
+                // The lexer is at position 10199 in the file.
+                Assert.Equal(10199, lexer.TextWindow.Position);
+
+                /// The 205 characters represent the final part of the doc
+                Assert.Equal(lexer.TextWindow.Text.Length, lexer.TextWindow.Position + lexer.TextWindow.CharacterWindowCount);
+
+                // We're at the start of a token.
+                Assert.Equal(lexer.TextWindow.LexemeStartPosition, lexer.TextWindow.Position);
+
+                // Ensure that the lexer's window is starting with the next FP number (".03") right at
+                // the start of the window.
+                Assert.True(lexer.TextWindow.CharacterWindow is ['.', '0', '3', ',', ..], $"Start of window was '{new string(lexer.TextWindow.CharacterWindow, 0, 4)}'");
+
+                var token = lexer.Lex(ref mode);
+                Assert.Equal(SyntaxKind.NumericLiteralToken, token.Kind);
+                Assert.Equal(3, token.FullWidth);
+                Assert.Equal(".03", token.ToString());
+
+                // But we moved 3 characters forward.
+                Assert.Equal(3, lexer.TextWindow.Offset);
+
+                // We still have 205 real chars in the window
+                Assert.Equal(205, lexer.TextWindow.CharacterWindowCount);
+
+                // The lexer position has moved 3 characters forward as well.
+                Assert.Equal(10202, lexer.TextWindow.Position);
+
+                // We're at the start of a token.
+                Assert.Equal(lexer.TextWindow.LexemeStartPosition, lexer.TextWindow.Position);
+
+                // Character window didn't changee.
+                Assert.True(lexer.TextWindow.CharacterWindow is ['.', '0', '3', ',', ..], $"Start of window was '{new string(lexer.TextWindow.CharacterWindow, 0, 4)}'");
+            }
+        }
     }
 }

src/Compilers/CSharp/Test/Syntax/Parsing/IgnoredDirectiveParsingTests.cs

@@ -287,7 +287,7 @@ public void AfterToken()
         EOF();
     }
 
-    [Fact]
+    [Fact, WorkItem("https://github.com/dotnet/roslyn/issues/78157")]
     public void AfterIf()
     {
         var source = """
@@ -299,7 +299,7 @@ public void AfterIf()
             """;
 
         VerifyTrivia();
-        UsingTree(source, TestOptions.Regular.WithFeature(FeatureName),
+        UsingTree(source, TestOptions.Regular.WithFeature(FeatureName).WithPreprocessorSymbols("X"),
             // (3,2): error CS9283: '#:' directives cannot be after '#if' directive
             // #:y
             Diagnostic(ErrorCode.ERR_PPIgnoredFollowsIf, ":").WithLocation(3, 2),
@@ -366,6 +366,71 @@ public void AfterIf()
             }
         }
         EOF();
+
+        UsingTree(source, TestOptions.Regular.WithFeature(FeatureName),
+            // (5,2): error CS9299: '#:' directives cannot be after '#if' directive
+            // #:z
+            Diagnostic(ErrorCode.ERR_PPIgnoredFollowsIf, ":").WithLocation(5, 2));
+
+        N(SyntaxKind.CompilationUnit);
+        {
+            N(SyntaxKind.EndOfFileToken);
+            {
+                L(SyntaxKind.IgnoredDirectiveTrivia);
+                {
+                    N(SyntaxKind.HashToken);
+                    N(SyntaxKind.ColonToken);
+                    N(SyntaxKind.StringLiteralToken, "x");
+                    N(SyntaxKind.EndOfDirectiveToken);
+                    {
+                        T(SyntaxKind.EndOfLineTrivia, "\n");
+                    }
+                }
+                L(SyntaxKind.IfDirectiveTrivia);
+                {
+                    N(SyntaxKind.HashToken);
+                    N(SyntaxKind.IfKeyword);
+                    {
+                        T(SyntaxKind.WhitespaceTrivia, " ");
+                    }
+                    N(SyntaxKind.IdentifierName);
+                    {
+                        N(SyntaxKind.IdentifierToken, "X");
+                    }
+                    N(SyntaxKind.EndOfDirectiveToken);
+                    {
+                        T(SyntaxKind.EndOfLineTrivia, "\n");
+                    }
+                }
+                L(SyntaxKind.IgnoredDirectiveTrivia);
+                {
+                    N(SyntaxKind.HashToken);
+                    N(SyntaxKind.ColonToken);
+                    N(SyntaxKind.StringLiteralToken, "y");
+                    N(SyntaxKind.EndOfDirectiveToken);
+                    {
+                        T(SyntaxKind.EndOfLineTrivia, "\n");
+                    }
+                }
+                L(SyntaxKind.EndIfDirectiveTrivia);
+                {
+                    N(SyntaxKind.HashToken);
+                    N(SyntaxKind.EndIfKeyword);
+                    N(SyntaxKind.EndOfDirectiveToken);
+                    {
+                        T(SyntaxKind.EndOfLineTrivia, "\n");
+                    }
+                }
+                L(SyntaxKind.IgnoredDirectiveTrivia);
+                {
+                    N(SyntaxKind.HashToken);
+                    N(SyntaxKind.ColonToken);
+                    N(SyntaxKind.StringLiteralToken, "z");
+                    N(SyntaxKind.EndOfDirectiveToken);
+                }
+            }
+        }
+        EOF();
     }
 
     [Fact]