[release/7.0] Fix unzipping 4GB+ zip files (#77605)

* add a failing test

* fix

* extend the test to verify more, move it to outerloop as it takes a LOT of time to execute it

* test fewer things, but way faster so the test can be run more frequently

* add a comment

* address code review feedback

* use temp directory for temporary test files

* address code review feedback

Co-authored-by: Adam Sitnik <adam.sitnik@gmail.com>

3 files changed, 83 insertions, 8 deletions

diff --git a/src/libraries/System.IO.Compression/src/System/IO/Compression/ZipBlocks.cs b/src/libraries/System.IO.Compression/src/System/IO/Compression/ZipBlocks.cs
index 44511243b5e..fb186dfcfc9 100644
--- a/src/libraries/System.IO.Compression/src/System/IO/Compression/ZipBlocks.cs
+++ b/src/libraries/System.IO.Compression/src/System/IO/Compression/ZipBlocks.cs
@@ -199,30 +199,55 @@ namespace System.IO.Compression
                 if (extraField.Size < sizeof(long))
                     return true;
 
-                long value64 = reader.ReadInt64();
+                // Advancing the stream (by reading from it) is possible only when:
+                // 1. There is an explicit ask to do that (valid files, corresponding boolean flag(s) set to true).
+                // 2. When the size indicates that all the information is available ("slightly invalid files").
+                bool readAllFields = extraField.Size >= sizeof(long) + sizeof(long) + sizeof(long) + sizeof(int);
+
                 if (readUncompressedSize)
-                    zip64Block._uncompressedSize = value64;
+                {
+                    zip64Block._uncompressedSize = reader.ReadInt64();
+                }
+                else if (readAllFields)
+                {
+                    _ = reader.ReadInt64();
+                }
 
                 if (ms.Position > extraField.Size - sizeof(long))
                     return true;
 
-                value64 = reader.ReadInt64();
                 if (readCompressedSize)
-                    zip64Block._compressedSize = value64;
+                {
+                    zip64Block._compressedSize = reader.ReadInt64();
+                }
+                else if (readAllFields)
+                {
+                    _ = reader.ReadInt64();
+                }
 
                 if (ms.Position > extraField.Size - sizeof(long))
                     return true;
 
-                value64 = reader.ReadInt64();
                 if (readLocalHeaderOffset)
-                    zip64Block._localHeaderOffset = value64;
+                {
+                    zip64Block._localHeaderOffset = reader.ReadInt64();
+                }
+                else if (readAllFields)
+                {
+                    _ = reader.ReadInt64();
+                }
 
                 if (ms.Position > extraField.Size - sizeof(int))
                     return true;
 
-                int value32 = reader.ReadInt32();
                 if (readStartDiskNumber)
-                    zip64Block._startDiskNumber = value32;
+                {
+                    zip64Block._startDiskNumber = reader.ReadInt32();
+                }
+                else if (readAllFields)
+                {
+                    _ = reader.ReadInt32();
+                }
 
                 // original values are unsigned, so implies value is too big to fit in signed integer
                 if (zip64Block._uncompressedSize < 0) throw new InvalidDataException(SR.FieldTooBigUncompressedSize);
diff --git a/src/libraries/System.IO.Compression/tests/System.IO.Compression.Tests.csproj b/src/libraries/System.IO.Compression/tests/System.IO.Compression.Tests.csproj
index 229119e3aa3..ef1862eb0bf 100644
--- a/src/libraries/System.IO.Compression/tests/System.IO.Compression.Tests.csproj
+++ b/src/libraries/System.IO.Compression/tests/System.IO.Compression.Tests.csproj
@@ -24,6 +24,7 @@
     <Compile Include="ZipArchive\zip_InvalidParametersAndStrangeFiles.cs" />
     <Compile Include="ZipArchive\zip_ManualAndCompatibilityTests.cs" />
     <Compile Include="ZipArchive\zip_netcoreappTests.cs" />
+    <Compile Include="ZipArchive\zip_LargeFiles.cs" />
     <Compile Include="ZipArchive\zip_ReadTests.cs" />
     <Compile Include="ZipArchive\zip_UpdateTests.cs" />
     <Compile Include="ZipArchive\zip_UpdateTests.Comments.cs" />
@@ -36,6 +37,7 @@
     <Compile Include="$(CommonTestPath)System\IO\Compression\StreamHelpers.cs" Link="Common\System\IO\Compression\StreamHelpers.cs" />
     <Compile Include="$(CommonTestPath)System\IO\TempFile.cs" Link="Common\System\IO\TempFile.cs" />
     <Compile Include="$(CommonTestPath)System\IO\Compression\ZipTestHelper.cs" Link="Common\System\IO\Compression\ZipTestHelper.cs" />
+    <Compile Include="$(CommonTestPath)TestUtilities\System\DisableParallelization.cs" Link="Common\TestUtilities\System\DisableParallelization.cs" />
     <Compile Include="$(CommonPath)System\Threading\Tasks\TaskToApm.cs" Link="Common\System\Threading\Tasks\TaskToApm.cs" />
     <Compile Include="$(CommonTestPath)System\IO\ConnectedStreams.cs" Link="Common\System\IO\ConnectedStreams.cs" />
     <Compile Include="$(CommonPath)System\Net\MultiArrayBuffer.cs" Link="ProductionCode\Common\System\Net\MultiArrayBuffer.cs" />
diff --git a/src/libraries/System.IO.Compression/tests/ZipArchive/zip_LargeFiles.cs b/src/libraries/System.IO.Compression/tests/ZipArchive/zip_LargeFiles.cs
new file mode 100644
index 00000000000..d240a176b2b
--- /dev/null
+++ b/src/libraries/System.IO.Compression/tests/ZipArchive/zip_LargeFiles.cs
@@ -0,0 +1,48 @@
+// Licensed to the .NET Foundation under one or more agreements.
+// The .NET Foundation licenses this file to you under the MIT license.
+
+using System.Collections.Generic;
+using System.Linq;
+using Xunit;
+
+namespace System.IO.Compression.Tests
+{
+    [Collection(nameof(DisableParallelization))]
+    public class zip_LargeFiles : ZipFileTestBase
+    {
+        [ConditionalFact(typeof(PlatformDetection), nameof(PlatformDetection.IsSpeedOptimized), nameof(PlatformDetection.Is64BitProcess))] // don't run it on slower runtimes
+        [OuterLoop("It requires almost 12 GB of free disk space")]
+        public static void UnzipOver4GBZipFile()
+        {
+            byte[] buffer = GC.AllocateUninitializedArray<byte>(1_000_000_000); // 1 GB
+
+            string zipArchivePath = Path.Combine(Path.GetTempPath(), "over4GB.zip");
+            DirectoryInfo tempDir = Directory.CreateDirectory(Path.Combine(Path.GetTempPath(), "over4GB"));
+
+            try
+            {
+                for (byte i = 0; i < 6; i++)
+                {
+                    File.WriteAllBytes(Path.Combine(tempDir.FullName, $"{i}.test"), buffer);
+                }
+
+                ZipFile.CreateFromDirectory(tempDir.FullName, zipArchivePath, CompressionLevel.NoCompression, includeBaseDirectory: false);
+
+                using ZipArchive zipArchive = ZipFile.OpenRead(zipArchivePath);
+                foreach (ZipArchiveEntry entry in zipArchive.Entries)
+                {
+                    using Stream entryStream = entry.Open();
+
+                    Assert.True(entryStream.CanRead);
+                    Assert.Equal(buffer.Length, entryStream.Length);
+                }
+            }
+            finally
+            {
+                File.Delete(zipArchivePath);
+
+                tempDir.Delete(recursive: true);
+            }
+        }
+    }
+}