Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 8 additions & 2 deletions RecursiveExtractor.Tests/ExtractorTests/ExpectedNumFilesTests.cs
Original file line number Diff line number Diff line change
Expand Up @@ -37,7 +37,9 @@ public static IEnumerable<object[]> ArchiveData
new object[] { "TestData.vhdx",3 },
new object[] { "TestData.wim",3 },
new object[] { "EmptyFile.txt", 1 },
new object[] { "TestDataArchivesNested.Zip", 54 }
new object[] { "TestDataArchivesNested.Zip", 54 },
new object[] { "UdfTest.iso", 3 },
new object[] { "UdfTestWithMultiSystem.iso", 3 }
};
}
}
Expand Down Expand Up @@ -170,7 +172,11 @@ public void ExtractArchive(string fileName, int expectedNumFiles)
var extractor = new Extractor();
var path = Path.Combine(Directory.GetCurrentDirectory(), "TestData", "TestDataArchives", fileName);
var results = extractor.Extract(path, GetExtractorOptions()).ToList();
Assert.AreEqual(expectedNumFiles, results.Count());
foreach (var result in results)
{
Assert.AreNotEqual(FileEntryStatus.FailedArchive, result.EntryStatus);
}
Assert.AreEqual(expectedNumFiles, results.Count);
}

[TestMethod]
Expand Down
2 changes: 1 addition & 1 deletion RecursiveExtractor.Tests/ExtractorTests/MiniMagicTests.cs
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,7 @@ public class MiniMagicTests : BaseExtractorTestClass
[DataRow("sysvbanner_1.0-17fakesync1_amd64.deb", ArchiveFileType.DEB)]
[DataRow("TestData.a", ArchiveFileType.AR)]
[DataRow("TestData.iso", ArchiveFileType.ISO_9660)]
// [DataRow("TestData.vhd", ArchiveFileType.VHD)]
[DataRow("UdfTest.iso", ArchiveFileType.UDF)]
[DataRow("TestData.vhdx", ArchiveFileType.VHDX)]
[DataRow("TestData.wim", ArchiveFileType.WIM)]
[DataRow("Empty.vmdk", ArchiveFileType.VMDK)]
Expand Down
6 changes: 6 additions & 0 deletions RecursiveExtractor.Tests/RecursiveExtractor.Tests.csproj
Original file line number Diff line number Diff line change
Expand Up @@ -267,5 +267,11 @@
<None Update="TestData\TestData\Lorem.txt">
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
</None>
<None Update="TestData\TestDataArchives\UdfTest.iso">
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
</None>
<None Update="TestData\TestDataArchives\UdfTestWithMultiSystem.iso">
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
</None>
</ItemGroup>
</Project>
11 changes: 2 additions & 9 deletions RecursiveExtractor.Tests/SanitizePathTests.cs
Original file line number Diff line number Diff line change
@@ -1,18 +1,11 @@
// Copyright (c) Microsoft Corporation. Licensed under the MIT License.

using Microsoft.CST.RecursiveExtractor;
using Microsoft.VisualStudio.TestTools.UnitTesting;
using NLog;
using NLog.Config;
using NLog.Targets;
using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Runtime.InteropServices;
using System.Text.RegularExpressions;
using System.Threading.Tasks;

namespace Microsoft.CST.RecursiveExtractor.Tests
namespace RecursiveExtractor.Tests
{
[TestClass]
public class SanitizePathTests
Expand Down
Binary file not shown.
Binary file not shown.
3 changes: 2 additions & 1 deletion RecursiveExtractor/Extractor.cs
Original file line number Diff line number Diff line change
Expand Up @@ -51,6 +51,7 @@ public void SetDefaultExtractors()
SetExtractor(ArchiveFileType.AR, new GnuArExtractor(this));
SetExtractor(ArchiveFileType.GZIP, new GzipExtractor(this));
SetExtractor(ArchiveFileType.ISO_9660, new IsoExtractor(this));
SetExtractor(ArchiveFileType.UDF, new UdfExtractor(this));
SetExtractor(ArchiveFileType.RAR, new RarExtractor(this));
SetExtractor(ArchiveFileType.RAR5, new RarExtractor(this));
SetExtractor(ArchiveFileType.P7ZIP, new SevenZipExtractor(this));
Expand Down Expand Up @@ -663,4 +664,4 @@ public IEnumerable<FileEntry> Extract(FileEntry fileEntry, ExtractorOptions? opt
}
}
}
}
}
157 changes: 157 additions & 0 deletions RecursiveExtractor/Extractors/UdfExtractor.cs
Original file line number Diff line number Diff line change
@@ -0,0 +1,157 @@
using DiscUtils.Udf;
using System;
using System.Collections.Concurrent;
using System.Collections.Generic;
using System.IO;
using System.Linq;

namespace Microsoft.CST.RecursiveExtractor.Extractors
{
/// <summary>
/// The UDF disc image extractor implementation.
/// </summary>
public class UdfExtractor : AsyncExtractorInterface
{
/// <summary>
/// The constructor takes the Extractor context for recursion.
/// </summary>
/// <param name="context">The Extractor context.</param>
public UdfExtractor(Extractor context)
{
Context = context;
}
private readonly NLog.Logger Logger = NLog.LogManager.GetCurrentClassLogger();

internal Extractor Context { get; }

/// <summary>
/// Extracts an UDF file
/// </summary>
///<inheritdoc />
public async IAsyncEnumerable<FileEntry> ExtractAsync(FileEntry fileEntry, ExtractorOptions options, ResourceGovernor governor, bool topLevel = true)
{
DiscUtils.DiscFileInfo[]? entries = null;
var failed = false;
try
{
using var cd = new UdfReader(fileEntry.Content);
entries = cd.Root.GetFiles("*.*", SearchOption.AllDirectories).ToArray();
}
catch (Exception e)
{
Logger.Debug("Failed to open UDF {0}. ({1}:{2})", fileEntry.FullPath, e.GetType(), e.Message);
failed = true;
}
if (failed)
{
if (options.ExtractSelfOnFail)
{
fileEntry.EntryStatus = FileEntryStatus.FailedArchive;
yield return fileEntry;
}
}
else if (entries != null)
{
foreach (var file in entries)
{
var fileInfo = file;
governor.CheckResourceGovernor(fileInfo.Length);
Stream? stream = null;
try
{
stream = fileInfo.OpenRead();
}
catch (Exception e)
{
Logger.Debug("Failed to extract {0} from UDF {1}. ({2}:{3})", fileInfo.FullName, fileEntry.FullPath, e.GetType(), e.Message);
}
if (stream != null)
{
var name = fileInfo.FullName.Replace('/', Path.DirectorySeparatorChar);
var newFileEntry = await FileEntry.FromStreamAsync(name, stream, fileEntry, fileInfo.CreationTime, fileInfo.LastWriteTime, fileInfo.LastAccessTime, memoryStreamCutoff: options.MemoryStreamCutoff).ConfigureAwait(false);
if (options.Recurse || topLevel)
{
await foreach (var entry in Context.ExtractAsync(newFileEntry, options, governor, false))
{
yield return entry;
}
}
else
{
yield return newFileEntry;
}
}
}
}
}

/// <summary>
/// Extracts an UDF file
/// </summary>
///<inheritdoc />
public IEnumerable<FileEntry> Extract(FileEntry fileEntry, ExtractorOptions options, ResourceGovernor governor, bool topLevel = true)
{
DiscUtils.DiscFileInfo[]? entries = null;
var failed = false;
try
{
using var cd = new UdfReader(fileEntry.Content);
entries = cd.Root.GetFiles("*.*", SearchOption.AllDirectories).ToArray();
}
catch(Exception e)
{
Logger.Debug("Failed to open UDF {0}. ({1}:{2})", fileEntry.FullPath, e.GetType(), e.Message);
failed = true;
}
if (failed)
{
if (options.ExtractSelfOnFail)
{
fileEntry.EntryStatus = FileEntryStatus.FailedArchive;
yield return fileEntry;
}
}
else if (entries != null)
{
foreach (var file in entries)
{
var fileInfo = file;
governor.CheckResourceGovernor(fileInfo.Length);
Stream? stream = null;
try
{
stream = fileInfo.OpenRead();
}
catch (Exception e)
{
Logger.Debug("Failed to extract {0} from UDF {1}. ({2}:{3})", fileInfo.FullName, fileEntry.FullPath, e.GetType(), e.Message);
}
if (stream != null)
{
var name = fileInfo.FullName.Replace('/', Path.DirectorySeparatorChar);
var newFileEntry = new FileEntry(name, stream, fileEntry, createTime: file.CreationTime, modifyTime: file.LastWriteTime, accessTime: file.LastAccessTime, memoryStreamCutoff: options.MemoryStreamCutoff);
if (options.Recurse || topLevel)
{
foreach (var entry in Context.Extract(newFileEntry, options, governor, false))
{
yield return entry;
}
}
else
{
yield return newFileEntry;
}
}
}
}
else
{
if (options.ExtractSelfOnFail)
{
fileEntry.EntryStatus = FileEntryStatus.FailedArchive;
yield return fileEntry;
}
}
}
}
}
10 changes: 9 additions & 1 deletion RecursiveExtractor/MiniMagic.cs
Original file line number Diff line number Diff line change
Expand Up @@ -61,6 +61,10 @@ public enum ArchiveFileType
/// </summary>
ISO_9660,
/// <summary>
/// An UDF disc image. <see cref="Extractors.UdfExtractor"/>
/// </summary>
UDF,
/// <summary>
/// A VHDX disc image. <see cref="Extractors.VhdxExtractor"/>
/// </summary>
VHDX,
Expand Down Expand Up @@ -228,6 +232,10 @@ public static ArchiveFileType DetectFileType(Stream fileStream)
{
return ArchiveFileType.ISO_9660;
}
if (buffer[0] == 'B' && buffer[1] == 'E' && buffer[2] == 'A' && buffer[3] == '0' && buffer[4] == '1')
{
return ArchiveFileType.UDF;
}
}

//https://www.microsoft.com/en-us/download/details.aspx?id=23850 - 'Hard Disk Footer Format'
Expand Down Expand Up @@ -265,4 +273,4 @@ public static ArchiveFileType DetectFileType(Stream fileStream)
/// <returns>The ArchiveFileType detected</returns>
public static ArchiveFileType DetectFileType(FileEntry fileEntry) => DetectFileType(fileEntry?.Content ?? new MemoryStream());
}
}
}
1 change: 1 addition & 0 deletions RecursiveExtractor/RecursiveExtractor.csproj
Original file line number Diff line number Diff line change
Expand Up @@ -32,6 +32,7 @@
<PackageReference Include="DiscUtils.HfsPlus" Version="0.16.13" />
<PackageReference Include="DiscUtils.Iso9660" Version="0.16.13" />
<PackageReference Include="DiscUtils.Ntfs" Version="0.16.13" />
<PackageReference Include="DiscUtils.Udf" Version="0.16.13" />
<PackageReference Include="DiscUtils.Vhd" Version="0.16.13" />
<PackageReference Include="DiscUtils.Vhdx" Version="0.16.13" />
<PackageReference Include="DiscUtils.Vmdk" Version="0.16.13" />
Expand Down