Fix async decompression of .7z files by implementing Memory<byte> ReadAsync overload

[Copilot]

Async extraction of .7z files with LZMA/LZMA2 compression threw DataErrorException when using CopyToAsync(), while synchronous CopyTo() worked correctly.

Root Cause

In .NET 6+, ReadExactlyAsync calls ReadAsync(Memory<byte>, CancellationToken). BufferedSubStream only implemented the legacy byte[] overload, causing the base Stream class to fall back to synchronous reads. This corrupted cache state when LZMA's RangeCoder mixed sync ReadByte() calls with async operations.

Changes

• BufferedSubStream: Added ReadAsync(Memory<byte>, CancellationToken) and RefillCacheAsync() for true async I/O
• Tests: Added async test coverage for LZMA, LZMA2, Solid, BZip2, and PPMd archives

Example

// This now works correctly with async operations
using var archive = ArchiveFactory.Open(archivePath);
foreach (var entry in archive.Entries.Where(e => !e.IsDirectory))
{
    using var stream = await entry.OpenEntryStreamAsync(cancellationToken);
    await stream.CopyToAsync(outputStream, cancellationToken);  // Previously threw DataErrorException
}

The fix ensures async operations remain async throughout the decompression pipeline, preventing sync-over-async patterns.

Original prompt

---

This section details on the original issue you should resolve

<issue_title>decompressing big .7z file throws error</issue_title>
<issue_description>lib version 0.42.1
under .net 10

code:

public class SharpCompressExtractor : IArchiveExtractor
{
    public async Task<IReadOnlyCollection<FileInfo>> ExtractAsync(
        string archivePath,
        string destinationDirectory,
        CancellationToken token)
    {
        if (!File.Exists(archivePath))
        {
            throw new FileNotFoundException($"Nie znaleziono archiwum: {archivePath}");
        }

        var extractedFiles = new List<FileInfo>();

        using var archive = ArchiveFactory.Open(archivePath);

        foreach (var entry in archive.Entries)
        {
            if (entry.IsDirectory)
            {
                continue;
            }

            token.ThrowIfCancellationRequested();

            var targetPath = Path.Combine(destinationDirectory, entry.Key);

            var targetDir = Path.GetDirectoryName(targetPath);

            if (!string.IsNullOrEmpty(targetDir) && !Directory.Exists(targetDir))
            {
                Directory.CreateDirectory(targetDir);
            }

            using var sourceStream = await entry.OpenEntryStreamAsync(token);

            var fileOptions = new FileStreamOptions
            {
                Mode = FileMode.Create,
                Access = FileAccess.Write,
                Share = FileShare.None,
                BufferSize = 1_024 * 1_024,
                Options = FileOptions.Asynchronous
            };

            await using var targetStream = new FileStream(targetPath, fileOptions);

            await sourceStream.CopyToAsync(targetStream, token);

            var fileInfo = new FileInfo(targetPath);

            extractedFiles.Add(fileInfo);
        }

        return extractedFiles;
    }
}

usage

            try
            {
                var files = await fileExtractor.ExtractAsync(
    "C:\\whitelist\\20251223.7z",
    "c:/whitelist/out",
    CancellationToken.None);
            }
            catch (Exception ex)
            {
                
            }

Data Error

at SharpCompress.Compressors.LZMA.LzmaStream.d__52.MoveNext()
at SharpCompress.Compressors.LZMA.LzmaStream.d__56.MoveNext()
at System.Runtime.CompilerServices.ConfiguredValueTaskAwaitable`1.ConfiguredValueTaskAwaiter.GetResult()
at SharpCompress.IO.ReadOnlySubStream.d__24.MoveNext()
at System.IO.Stream.<g__Core|30_0>d.MoveNext()
at SharpCompressExtractor.d__0.MoveNext() in SharpCompressExtractor.cs

file : probably any from https://www.podatki.gov.pl/narzedzia/white-list/

but for sure exist for this url: https://plikplaski.mf.gov.pl/pliki//20251223.7z

this file perfectly uncompress under win11 using native extractor or 7zip manager
</issue_description>

<agent_instructions>compare the synchronous code path to the new asynchronous code path to reproduce the issue</agent_instructions>

Comments on the Issue (you are @copilot in this section)

@adamhathcock Sounds like the fix I did for https://github.com//pull/1081

I can validate next week though.</comment_new>
<comment_new>@adamhathcock
You're right....testing this myself the async path is broken. However, the sync path works.

If you change CopyToAsync to CopytTo then it works for me

gonna look at a real fix though</comment_new>

• Fixes decompressing big .7z file throws error #1086

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[Copilot]

Pull request overview

This PR fixes a critical bug where async extraction of .7z files with LZMA/LZMA2 compression threw DataErrorException, while synchronous extraction worked correctly. The root cause was that BufferedSubStream only implemented the legacy byte[] ReadAsync overload, causing .NET 6+ to fall back to synchronous reads when ReadExactlyAsync called the Memory<byte> overload. This corrupted cache state when LZMA's RangeCoder mixed synchronous ReadByte() calls with async operations.

The fix implements true async I/O by adding ReadAsync(Memory<byte>, CancellationToken) overloads throughout the decompression pipeline:

• Added ReadAsync(Memory<byte>) and RefillCacheAsync() to BufferedSubStream for async cache management
• Added ReadAsync(Memory<byte>) to LzmaStream with proper async decompression handling
• Enhanced LzOutWindow with Memory support and converted async helpers from Task to ValueTask
• Added comprehensive async test coverage for LZMA, LZMA2, Solid, BZip2, and PPMd archives

Reviewed changes

Copilot reviewed 6 out of 6 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
tests/SharpCompress.Test/SevenZip/SevenZipArchiveAsyncTests.cs	New test file adding async extraction tests for all 7z compression types (LZMA, LZMA2, Solid, BZip2, PPMd)
src/SharpCompress/IO/BufferedSubStream.cs	Added Memory ReadAsync overload and RefillCacheAsync for true async buffered reads
src/SharpCompress/IO/ReadOnlySubStream.cs	Minor refactoring of ReadAsync Memory implementation
src/SharpCompress/Compressors/LZMA/LzmaStream.cs	Added Memory ReadAsync overload; changed DecodeChunkHeaderAsync return type to ValueTask
src/SharpCompress/Compressors/LZMA/LzmaDecoder.cs	Changed CodeAsync return type from Task to ValueTask for better performance
src/SharpCompress/Compressors/LZMA/LZ/LzOutWindow.cs	Added Read(Memory) overload and converted async methods to ValueTask; optimized async pattern

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.