Merge pull request #3005 from microsoft/fix/oom-to-v2

fix: bound YAML anchor/alias expansion to prevent OOM (billion laughs) (#3000)

Author: baywet

src/Microsoft.OpenApi.YamlReader/OpenApiYamlReader.cs

@@ -74,6 +74,17 @@ public ReadResult Read(MemoryStream input,
                     Diagnostic = diagnostic,
                 };
             }
+            catch (OpenApiReaderException ex)
+            {
+                var diagnostic = new OpenApiDiagnostic();
+                diagnostic.Errors.Add(new(ex));
+                diagnostic.Format = OpenApiConstants.Yaml;
+                return new()
+                {
+                    Document = null,
+                    Diagnostic = diagnostic,
+                };
+            }
 
             return UpdateFormat(Read(jsonNode, location, settings));
         }

src/Microsoft.OpenApi.YamlReader/PublicAPI.Unshipped.txt

@@ -1 +1,7 @@
 #nullable enable
+const Microsoft.OpenApi.YamlReader.YamlConverter.DefaultMaxDepth = 64 -> uint
+const Microsoft.OpenApi.YamlReader.YamlConverter.DefaultMaxNodeCount = 5000000 -> uint
+static Microsoft.OpenApi.YamlReader.YamlConverter.MaxDepth.get -> uint
+static Microsoft.OpenApi.YamlReader.YamlConverter.MaxDepth.set -> void
+static Microsoft.OpenApi.YamlReader.YamlConverter.MaxNodeCount.get -> uint
+static Microsoft.OpenApi.YamlReader.YamlConverter.MaxNodeCount.set -> void

src/Microsoft.OpenApi.YamlReader/YamlConverter.cs

@@ -14,6 +14,94 @@ namespace Microsoft.OpenApi.YamlReader
     /// </summary>
     public static class YamlConverter
     {
+        /// <summary>
+        /// Default maximum nesting depth allowed when converting a YAML node graph into JSON nodes.
+        /// Mirrors the default System.Text.Json depth limit (64) that already bounds the JSON reader path,
+        /// protecting the recursive conversion from stack exhaustion on deeply nested documents.
+        /// </summary>
+        public const uint DefaultMaxDepth = 64;
+
+        /// <summary>
+        /// Default maximum number of JSON nodes that may be materialized from a single YAML document.
+        /// Guards against YAML anchor/alias expansion ("billion laughs") attacks, where a tiny document
+        /// expands exponentially when its shared node graph is materialized into an independent JSON tree.
+        /// </summary>
+        public const uint DefaultMaxNodeCount = 5_000_000;
+
+        private static uint _maxDepth = DefaultMaxDepth;
+        private static uint _maxNodeCount = DefaultMaxNodeCount;
+
+        /// <summary>
+        /// Gets or sets the maximum nesting depth allowed when converting a YAML node graph into JSON nodes.
+        /// Defaults to <see cref="DefaultMaxDepth"/>. Raise this if legitimate deeply nested documents are
+        /// being rejected, or lower it to fail faster when only shallow documents are expected.
+        /// </summary>
+        /// <exception cref="ArgumentOutOfRangeException">Thrown when set to zero.</exception>
+        public static uint MaxDepth
+        {
+            get => _maxDepth;
+            set
+            {
+                if (value == 0)
+                {
+                    throw new ArgumentOutOfRangeException(nameof(value), "MaxDepth must be greater than zero.");
+                }
+
+                _maxDepth = value;
+            }
+        }
+
+        /// <summary>
+        /// Gets or sets the maximum number of JSON nodes that may be materialized from a single YAML document.
+        /// Defaults to <see cref="DefaultMaxNodeCount"/>, guarding against YAML anchor/alias expansion
+        /// ("billion laughs") attacks. Raise this if legitimate large documents are being rejected, or lower
+        /// it to fail faster when only small documents are expected.
+        /// </summary>
+        /// <exception cref="ArgumentOutOfRangeException">Thrown when set to zero.</exception>
+        public static uint MaxNodeCount
+        {
+            get => _maxNodeCount;
+            set
+            {
+                if (value == 0)
+                {
+                    throw new ArgumentOutOfRangeException(nameof(value), "MaxNodeCount must be greater than zero.");
+                }
+
+                _maxNodeCount = value;
+            }
+        }
+
+        /// <summary>
+        /// Tracks and enforces resource limits while converting a YAML node graph into JSON nodes,
+        /// failing fast when a hostile document would otherwise exhaust memory or the stack.
+        /// </summary>
+        private sealed class YamlConversionBudget
+        {
+            private readonly uint _maxDepth;
+            private readonly uint _maxNodeCount;
+            private uint _nodeCount;
+
+            public YamlConversionBudget(uint maxDepth, uint maxNodeCount)
+            {
+                _maxDepth = maxDepth;
+                _maxNodeCount = maxNodeCount;
+            }
+
+            public void EnterNode(uint depth)
+            {
+                if (depth > _maxDepth)
+                {
+                    throw new OpenApiReaderException($"The YAML document exceeds the maximum supported nesting depth of {_maxDepth}.");
+                }
+
+                if (++_nodeCount > _maxNodeCount)
+                {
+                    throw new OpenApiReaderException($"The YAML document expands to more than the maximum supported number of nodes ({_maxNodeCount}). This may indicate a YAML anchor/alias expansion (billion laughs) attack.");
+                }
+            }
+        }
+
         /// <summary>
         /// Converts all of the documents in a YAML stream to <see cref="JsonNode"/>s.
         /// </summary>
@@ -42,10 +130,16 @@ public static JsonNode ToJsonNode(this YamlDocument yaml)
         /// <exception cref="NotSupportedException">Thrown for YAML that is not compatible with JSON.</exception>
         public static JsonNode ToJsonNode(this YamlNode yaml)
         {
+            return yaml.ToJsonNode(new YamlConversionBudget(MaxDepth, MaxNodeCount), 0);
+        }
+
+        private static JsonNode ToJsonNode(this YamlNode yaml, YamlConversionBudget budget, uint depth)
+        {
+            budget.EnterNode(depth);
             return yaml switch
             {
-                YamlMappingNode map => map.ToJsonObject(),
-                YamlSequenceNode seq => seq.ToJsonArray(),
+                YamlMappingNode map => map.ToJsonObject(budget, depth),
+                YamlSequenceNode seq => seq.ToJsonArray(budget, depth),
                 YamlScalarNode scalar => scalar.ToJsonValue(),
                 _ => throw new NotSupportedException("This yaml isn't convertible to JSON")
             };
@@ -78,12 +172,17 @@ public static YamlNode ToYamlNode(this JsonNode json)
         /// <param name="yaml"></param>
         /// <returns></returns>
         public static JsonObject ToJsonObject(this YamlMappingNode yaml)
+        {
+            return yaml.ToJsonObject(new YamlConversionBudget(MaxDepth, MaxNodeCount), 0);
+        }
+
+        private static JsonObject ToJsonObject(this YamlMappingNode yaml, YamlConversionBudget budget, uint depth)
         {
             var node = new JsonObject();
             foreach (var keyValuePair in yaml)
             {
                 var key = ((YamlScalarNode)keyValuePair.Key).Value!;
-                node[key] = keyValuePair.Value.ToJsonNode();
+                node[key] = keyValuePair.Value.ToJsonNode(budget, depth + 1);
             }
 
             return node;
@@ -103,11 +202,16 @@ private static YamlMappingNode ToYamlMapping(this JsonObject obj)
         /// <param name="yaml"></param>
         /// <returns></returns>
         public static JsonArray ToJsonArray(this YamlSequenceNode yaml)
+        {
+            return yaml.ToJsonArray(new YamlConversionBudget(MaxDepth, MaxNodeCount), 0);
+        }
+
+        private static JsonArray ToJsonArray(this YamlSequenceNode yaml, YamlConversionBudget budget, uint depth)
         {
             var node = new JsonArray();
             foreach (var value in yaml)
             {
-                node.Add(value.ToJsonNode());
+                node.Add(value.ToJsonNode(budget, depth + 1));
             }
 
             return node;

test/Microsoft.OpenApi.Readers.Tests/OpenApiYamlReaderTests.cs

@@ -0,0 +1,137 @@
+using System;
+using System.IO;
+using System.Text;
+using System.Threading;
+using System.Threading.Tasks;
+using Microsoft.OpenApi.Reader;
+using Microsoft.OpenApi.YamlReader;
+using Xunit;
+
+namespace Microsoft.OpenApi.Readers.Tests;
+
+public class OpenApiYamlReaderTests
+{
+    private static readonly Uri DocumentLocation = new("https://contoso.test/openapi.yaml");
+
+    [Fact]
+    public async Task ReadAsyncParsesDocumentsFromNonMemoryStreams()
+    {
+        var reader = new OpenApiYamlReader();
+        await using var stream = new NonMemoryStream(CreateStream(
+            """
+            openapi: 3.0.1
+            info:
+              title: Sample API
+              version: 1.0.0
+            paths: {}
+            """));
+
+        var result = await reader.ReadAsync(stream, DocumentLocation, SettingsFixture.ReaderSettings, CancellationToken.None);
+
+        Assert.NotNull(result.Document);
+        Assert.Equal("Sample API", result.Document.Info.Title);
+        Assert.Equal(OpenApiConstants.Yaml, result.Diagnostic.Format);
+    }
+
+    [Fact]
+    public void ReadThrowsWhenYamlDoesNotContainADocument()
+    {
+        var reader = new OpenApiYamlReader();
+        using var stream = CreateStream(string.Empty);
+
+        var exception = Assert.Throws<InvalidOperationException>(() => reader.Read(stream, DocumentLocation, SettingsFixture.ReaderSettings));
+
+        Assert.Equal("No documents found in the YAML stream.", exception.Message);
+    }
+
+    [Fact]
+    public void ReadFragmentParsesSchemaFragments()
+    {
+        var reader = new OpenApiYamlReader();
+        using var stream = CreateStream(
+            """
+            type: string
+            description: A reusable schema
+            """);
+
+        var schema = reader.ReadFragment<OpenApiSchema>(
+            stream,
+            OpenApiSpecVersion.OpenApi3_0,
+            new OpenApiDocument(),
+            out var diagnostic);
+
+        Assert.NotNull(schema);
+        Assert.Empty(diagnostic.Errors);
+        Assert.Equal(JsonSchemaType.String, schema.Type);
+        Assert.Equal("A reusable schema", schema.Description);
+    }
+
+    [Fact]
+    public void ReadThrowsWhenSettingsIsNull()
+    {
+        var reader = new OpenApiYamlReader();
+        using var stream = CreateStream("openapi: 3.0.1");
+
+        Assert.Throws<ArgumentNullException>(() => reader.Read(stream, DocumentLocation, null!));
+    }
+
+    [Fact]
+    public void ReadReturnsDiagnosticErrorForExponentialAliasExpansion()
+    {
+        // A "billion laughs" YAML bomb must surface as a diagnostic error with no document,
+        // rather than throwing or exhausting memory.
+        var reader = new OpenApiYamlReader();
+        using var stream = CreateStream(
+            """
+            a: &a ["x","x","x","x","x","x","x","x","x"]
+            b: &b [*a,*a,*a,*a,*a,*a,*a,*a,*a]
+            c: &c [*b,*b,*b,*b,*b,*b,*b,*b,*b]
+            d: &d [*c,*c,*c,*c,*c,*c,*c,*c,*c]
+            e: &e [*d,*d,*d,*d,*d,*d,*d,*d,*d]
+            f: &f [*e,*e,*e,*e,*e,*e,*e,*e,*e]
+            g: &g [*f,*f,*f,*f,*f,*f,*f,*f,*f]
+            h: &h [*g,*g,*g,*g,*g,*g,*g,*g,*g]
+            i: &i [*h,*h,*h,*h,*h,*h,*h,*h,*h]
+            """);
+
+        var result = reader.Read(stream, DocumentLocation, SettingsFixture.ReaderSettings);
+
+        Assert.Null(result.Document);
+        Assert.NotEmpty(result.Diagnostic.Errors);
+        Assert.Equal(OpenApiConstants.Yaml, result.Diagnostic.Format);
+    }
+
+    private static MemoryStream CreateStream(string yaml)
+    {
+        return new MemoryStream(Encoding.UTF8.GetBytes(yaml));
+    }
+
+    private sealed class NonMemoryStream(Stream innerStream) : Stream
+    {
+        public override bool CanRead => innerStream.CanRead;
+        public override bool CanSeek => innerStream.CanSeek;
+        public override bool CanWrite => innerStream.CanWrite;
+        public override long Length => innerStream.Length;
+        public override long Position
+        {
+            get => innerStream.Position;
+            set => innerStream.Position = value;
+        }
+
+        public override void Flush() => innerStream.Flush();
+        public override int Read(byte[] buffer, int offset, int count) => innerStream.Read(buffer, offset, count);
+        public override long Seek(long offset, SeekOrigin origin) => innerStream.Seek(offset, origin);
+        public override void SetLength(long value) => innerStream.SetLength(value);
+        public override void Write(byte[] buffer, int offset, int count) => innerStream.Write(buffer, offset, count);
+        public override ValueTask DisposeAsync() => innerStream.DisposeAsync();
+        protected override void Dispose(bool disposing)
+        {
+            if (disposing)
+            {
+                innerStream.Dispose();
+            }
+
+            base.Dispose(disposing);
+        }
+    }
+}