[API Proposal]: Add APIs to `System.IO.Compression` that can open compressed `tar` archives like `tar.gz`

[carlossanlop]

The APIs we added in System.Formats.Tar are meant for handling the tar format only. No compression handling was added there.

Currently, to handle compressed tar.gz files with those APIs, we can use these two methods:

We can extract the whole archive in one call:

using FileStream compressedStream = File.OpenRead("/home/dotnet/SourceDirectory/compressed.tar.gz");
using GZipStream decompressor = new(compressedStream, CompressionMode.Decompress);
TarFile.ExtractToDirectory(source: decompressor, destinationDirectoryName: "/home/dotnet/DestinationDirectory/", overwriteFiles: false);

Or we can iterate through each individual entry and decide what to do with each:

using FileStream compressedStream = File.OpenRead(@"D:\SourceDirectory\compressed.tar.gz");
using GZipStream decompressor = new(compressedStream, CompressionMode.Decompress);
using TarReader reader = new(decompressor, leaveOpen: true);
TarEntry? entry;

while ((entry = GetNextEntry(copyData: true)) != null)
{
 Console.WriteLine($"Entry type: {entry.EntryType}, entry name: {entry.Name}");
}

If we want to add APIs that can handle archives compressed with gzip (or any compression algorithm), we should add them to System.IO.Compression. I am opening this issue to initiate the discussion.

Note: We have an open issue that is trying to standardize the way we use all our compression algorithms: Standardize pattern for exposing advanced configuration for compression streams. It's possible this proposal depends on solving that one first.

Tagging subscribers to this area: @dotnet/area-system-io-compression
See info in area-owners.md if you want to be subscribed.

Issue Details

---

The APIs we added in System.Formats.Tar are meant for handling the tar format only. No compression handling was added there.

Currently, to handle compressed tar.gz files with those APIs, we can use these two methods:

We can extract the whole archive in one call:

using FileStream compressedStream = File.OpenRead("/home/dotnet/SourceDirectory/compressed.tar.gz");
using GZipStream decompressor = new(compressedStream, CompressionMode.Decompress);
TarFile.ExtractToDirectory(source: decompressor, destinationDirectoryName: "/home/dotnet/DestinationDirectory/", overwriteFiles: false);

Or we can iterate through each individual entry and decide what to do with each:

using FileStream compressedStream = File.OpenRead(@"D:\SourceDirectory\compressed.tar.gz");
using GZipStream decompressor = new(compressedStream, CompressionMode.Decompress);
using TarReader reader = new(decompressor, leaveOpen: true);
TarEntry? entry;

while ((entry = GetNextEntry(copyData: true)) != null)
{
 Console.WriteLine($"Entry type: {entry.EntryType}, entry name: {entry.Name}");
}

If we want to add APIs that can handle archives compressed with gzip (or any compression algorithm), we should add them to System.IO.Compression. I am opening this issue to initiate the discussion.

Note: We have an open issue that is trying to standardize the way we use all our compression algorithms: Standardize pattern for exposing advanced configuration for compression streams. It's possible this proposal depends on solving that one first.

Author:	carlossanlop
Assignees:	-
Labels:	area-System.IO.Compression
Milestone:	-

[iSazonov]

Example of code to extract an archive:

 TarEntry? entry = null;

 using (var sourceStream = File.Open(Source, FileMode.Open, FileAccess.ReadWrite, FileShare.ReadWrite))
 {
 using var reader = new TarReader(sourceStream);
 try
 {
 entry = reader.GetNextEntry();
 }
 catch
 {
 // It is not tar archive. Go to next format detection.
 }
 }

 if (entry is not null)
 {
 TarFile.ExtractToDirectory(Source, ToDirectory, Otherwrite);
 return;
 }

 using (var sourceStream = File.Open(Source, FileMode.Open, FileAccess.Read, FileShare.Read))
 {
 using var decompressorStream = new GZipStream(sourceStream, CompressionMode.Decompress);
 using var reader = new TarReader(decompressorStream);
 try
 {
 entry = reader.GetNextEntry();
 }
 catch
 {
 // It is not tar archive. Go to next format detection.
 }
 }

 if (entry is not null)
 {
 using var sourceStream = File.Open(Source, FileMode.Open, FileAccess.Read, FileShare.Read);
 using var decompressorStream = new GZipStream(sourceStream, CompressionMode.Decompress);
 TarFile.ExtractToDirectory(decompressorStream, ToDirectory, Otherwrite);

 return;
 }

 ExtractZipToDirectory(this, Source, ToDirectory, Encoding, Otherwrite);

Notice, we have to put Zip detection on last place - after .gz