Compare commits

..

131 Commits

Author SHA1 Message Date
copilot-swe-agent[bot]
4a6e5232ae Add opt-in multi-threading support with SupportsMultiThreadedExtraction flag
- Added IArchive.SupportsMultiThreadedExtraction property to indicate if multi-threading is supported
- Added ReaderOptions.EnableMultiThreadedExtraction option to opt-in to multi-threading
- Updated SeekableZipFilePart, TarFilePart, and SeekableFilePart to check the flag
- Added test to verify multi-threading flag behavior
- Multi-threading is now disabled by default for backward compatibility

Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-18 16:27:59 +00:00
copilot-swe-agent[bot]
3e23a6e5a6 Add multi-threading support for file-based archives - sync test passing
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-18 16:09:54 +00:00
copilot-swe-agent[bot]
e0a43e9727 Initial plan 2026-01-18 15:56:08 +00:00
Adam Hathcock
f5d83c0e33 Merge pull request #1135 from adamhathcock/copilot/consolidate-compile-flags 2026-01-15 18:47:37 +00:00
copilot-swe-agent[bot]
d2cb792d91 Change NET6_0_OR_GREATER to NET8_0_OR_GREATER
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-15 18:31:06 +00:00
copilot-swe-agent[bot]
52fef492a5 Additional simplifications: Remove NETCF, fix NET60 typo, consolidate NETCOREAPP2_1 pattern
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-15 18:09:06 +00:00
copilot-swe-agent[bot]
a5300f3383 Replace NETFRAMEWORK and NETSTANDARD2_0 with LEGACY_DOTNET compile flag
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-15 18:05:14 +00:00
copilot-swe-agent[bot]
cab3e7d498 Initial analysis: Planning compile flags consolidation
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-15 17:55:37 +00:00
copilot-swe-agent[bot]
405dbb30cd Initial plan 2026-01-15 17:50:54 +00:00
Adam Hathcock
3e219fa9ec Merge pull request #1131 from adamhathcock/adam/async-again
More test fixes and some perf changes
2026-01-15 16:20:25 +00:00
Adam Hathcock
ec310c87de merge fixes and fmt 2026-01-15 15:20:52 +00:00
Adam Hathcock
c55a383112 Merge remote-tracking branch 'origin/master' into adam/async-again
# Conflicts:
#	tests/SharpCompress.Test/GZip/GZipReaderAsyncTests.cs
#	tests/SharpCompress.Test/Rar/RarArchiveAsyncTests.cs
#	tests/SharpCompress.Test/SevenZip/SevenZipArchiveAsyncTests.cs
#	tests/SharpCompress.Test/Tar/TarArchiveAsyncTests.cs
#	tests/SharpCompress.Test/Tar/TarReaderAsyncTests.cs
#	tests/SharpCompress.Test/Zip/Zip64AsyncTests.cs
#	tests/SharpCompress.Test/Zip/ZipMemoryArchiveWithCrcAsyncTests.cs
2026-01-15 15:18:05 +00:00
Adam Hathcock
227fec66ad more pooling 2026-01-15 15:16:11 +00:00
Adam Hathcock
38eec23e07 rar byte[] better 2026-01-15 15:16:04 +00:00
Adam Hathcock
437271c6a2 change byte[] to memory using pool 2026-01-15 15:15:57 +00:00
Adam Hathcock
81a2060c75 reduce memory usage on headers 2026-01-15 15:15:50 +00:00
Adam Hathcock
5e90cfd6c5 Merge pull request #1128 from adamhathcock/adam/async-interface
Change interfaces to be consistent for new Async paths (definitely breaks things)
2026-01-15 15:13:33 +00:00
Adam Hathcock
b41296194f more updates to docs 2026-01-15 13:29:57 +00:00
Adam Hathcock
bf7416753a update docs 2026-01-15 13:04:09 +00:00
Adam Hathcock
7fbd751d27 change tests to work 2026-01-15 13:00:05 +00:00
Adam Hathcock
85b28dfe68 more refactoring 2026-01-15 12:20:35 +00:00
Adam Hathcock
779fba5deb finish the open refactor? 2026-01-15 12:06:54 +00:00
Adam Hathcock
2756b1f6f8 more refactor 2026-01-15 11:55:56 +00:00
Adam Hathcock
7b76858ae1 refactoring naming again 2026-01-15 11:41:30 +00:00
Adam Hathcock
84b5b5a717 add more tests 2026-01-14 14:33:20 +00:00
Adam Hathcock
ebfa16f09f more test fixes 2026-01-14 14:12:53 +00:00
Adam Hathcock
c1d240b516 Fix more tests 2026-01-14 14:06:39 +00:00
Adam Hathcock
5c4719f4a9 missing extensions 2026-01-14 13:39:11 +00:00
Adam Hathcock
95d2278d8b fmt 2026-01-14 12:13:29 +00:00
Adam Hathcock
e63ee57ef0 same for writers 2026-01-14 09:29:44 +00:00
Adam Hathcock
775efa1b26 Reader open factories 2026-01-14 09:23:18 +00:00
Adam Hathcock
3677b4b193 add default interfaces to enforce consistency 2026-01-14 08:57:12 +00:00
Adam Hathcock
c32f4b4f2a fix test reference 2026-01-14 08:33:49 +00:00
Adam Hathcock
8d34f88ca6 fix up gitignore 2026-01-13 16:42:54 +00:00
Adam Hathcock
ca4cf25a1f clean up lazy readonly collections and add tests 2026-01-13 16:39:55 +00:00
Adam Hathcock
4fa976b478 remove unused ref 2026-01-13 15:29:02 +00:00
Adam Hathcock
767f3a4985 fix up extensions to more like polyfills 2026-01-13 15:26:45 +00:00
Adam Hathcock
ddc08e068e fix async error 2026-01-13 15:16:38 +00:00
Adam Hathcock
a1a86cdde8 fmt 2026-01-13 14:29:10 +00:00
Adam Hathcock
fc85f1fa2c more tar async fixes 2026-01-13 14:28:45 +00:00
Adam Hathcock
0b8081f320 gzip fixes 2026-01-13 14:24:36 +00:00
Adam Hathcock
0b5371d986 more async fixing 2026-01-13 14:06:14 +00:00
Adam Hathcock
cdca909d84 fmt 2026-01-13 13:58:31 +00:00
Adam Hathcock
ec7d2e357d fix lock? 2026-01-13 13:58:03 +00:00
Adam Hathcock
1c0183ef11 force async tests 2026-01-13 13:56:56 +00:00
Adam Hathcock
9cf2b3129c fixed up async writer 2026-01-13 13:54:15 +00:00
Adam Hathcock
9a4e864f5e fix usage of ArchiveFactory 2026-01-13 13:49:35 +00:00
Adam Hathcock
4df952db1b split out factories for archive 2026-01-12 16:32:26 +00:00
Adam Hathcock
1b4cedfa13 misc fixes 2026-01-12 16:21:20 +00:00
Adam Hathcock
6d6103afd6 update docs 2026-01-12 16:08:16 +00:00
Adam Hathcock
d727d76299 Merge remote-tracking branch 'origin/master' into adam/async-interface 2026-01-12 15:02:19 +00:00
Adam Hathcock
0502ff545e test fixes and fmt 2026-01-12 15:01:29 +00:00
Adam Hathcock
fce4a96718 make Writable interfaces for archive 2026-01-12 14:57:13 +00:00
Adam Hathcock
38203fb950 Fix async reader variable types - Remove double await on ReaderFactory.OpenAsync and use IAsyncReader
- Removed 'await' keyword before ReaderFactory.OpenAsync() calls since the method returns IAsyncReader directly (not Task)
- Changed ZipReader.Open() to ReaderFactory.OpenAsync() in Zip64AsyncTests.ReadForwardOnlyAsync()
- Changed TarReader.Open() to ReaderFactory.OpenAsync() in TarReaderAsyncTests.Tar_BZip2_Entry_Stream_Async()
- Fixed EntryStream disposal from 'await using' to 'using' since EntryStream doesn't implement IAsyncDisposable
- These changes fix compilation errors where async methods were being called on IReader (synchronous) instead of IAsyncReader (asynchronous)
2026-01-12 14:14:46 +00:00
Adam Hathcock
65dba509e0 Merge pull request #1121 from adamhathcock/adam/async
More async for ZipReader and ZipWriter
2026-01-12 14:07:37 +00:00
Adam Hathcock
0615d17b8b fix async interfacing for open 2026-01-12 13:45:21 +00:00
Adam Hathcock
c1f8580d89 Remove unnecessary ValueTask wrappers from async factory methods
Change return types from ValueTask<T> to direct interface types (IAsyncArchive, IAsyncReader, IWriter) for wrapper methods that don't perform async work. This eliminates unnecessary async state machine allocations while maintaining the same public API behavior.

Changes:
- Interface definitions: Updated IArchiveFactory, IMultiArchiveFactory, IReaderFactory, IWriterFactory
- Concrete factories: Updated archive factories (Zip, Tar, Rar, GZip, SevenZip) and reader-only factories (Ace, Arc, Arj)
- Static factory methods: Updated ReaderFactory, ArchiveFactory, WriterFactory to use new signatures
- Archive classes: Updated static OpenAsync methods in ZipArchive, TarArchive, RarArchive, SevenZipArchive, GZipArchive
- Supporting changes: Updated Factory.cs and async polyfills

Performance benefit: Reduced GC pressure by eliminating unnecessary state machine overhead for non-async wrapper methods.
2026-01-12 13:16:44 +00:00
Adam Hathcock
c5a6f900df Merge branch 'adam/async' into adam/async-interface 2026-01-12 13:02:51 +00:00
Adam Hathcock
3807c3ce2a Merge pull request #1127 from adamhathcock/copilot/sub-pr-1121-yet-again
Fix async test method naming in ZipArchiveAsyncTests
2026-01-12 12:10:51 +00:00
Adam Hathcock
d2f328af01 Merge pull request #1126 from adamhathcock/copilot/sub-pr-1121-another-one
Fix typo in TestBase.cs comment
2026-01-12 12:10:11 +00:00
Adam Hathcock
c3ffcf4fe8 Merge pull request #1125 from adamhathcock/copilot/sub-pr-1121-again
[WIP] Update ZipReader and ZipWriter based on review feedback
2026-01-12 12:09:42 +00:00
copilot-swe-agent[bot]
95c409d979 Change File.Create to File.OpenWrite in TarReaderAsyncTests for consistency
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-12 12:09:12 +00:00
Adam Hathcock
dadf9e71bb Merge pull request #1124 from adamhathcock/copilot/sub-pr-1121
[WIP] Address feedback on async implementation for ZipReader and ZipWriter
2026-01-12 12:09:10 +00:00
Adam Hathcock
05ebf22009 Start using the interface to draw distinction between async and sync 2026-01-12 12:08:25 +00:00
copilot-swe-agent[bot]
f4b1780d8a Rename async test methods to use _Async suffix instead of _Sync
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-12 12:08:23 +00:00
copilot-swe-agent[bot]
921cff00a5 Fix async test method naming: rename Sync to Async
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-12 12:08:09 +00:00
copilot-swe-agent[bot]
64a09eb0f8 Fix typo in TestBase.cs comment: 'akways' -> 'always'
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-12 12:06:47 +00:00
copilot-swe-agent[bot]
3a636531e8 Initial plan 2026-01-12 12:05:34 +00:00
copilot-swe-agent[bot]
292da90184 Initial plan 2026-01-12 12:05:22 +00:00
copilot-swe-agent[bot]
90c8ff8650 Initial plan 2026-01-12 12:05:11 +00:00
copilot-swe-agent[bot]
0f37049aad Initial plan 2026-01-12 12:05:04 +00:00
Adam Hathcock
3fb07d129f Use async dispose always 2026-01-12 10:19:01 +00:00
Adam Hathcock
8d0ac5062f Merge pull request #1123 from adamhathcock/adam/add-more-docs
Add more documentation
2026-01-11 12:20:55 +00:00
Adam Hathcock
b2d1505e5c remove section on allocations 2026-01-11 12:20:40 +00:00
Adam Hathcock
a35e65ee42 use ifdefs for creating files? 2026-01-08 16:52:23 +00:00
Adam Hathcock
d1fcf31f7e fmt 2026-01-08 16:31:11 +00:00
Adam Hathcock
17cd934b5b use async methods where we can 2026-01-08 16:24:11 +00:00
Adam Hathcock
ae614cd3fe update references 2026-01-08 16:14:40 +00:00
Adam Hathcock
47037a4b9d docs: Add explicit guideline that agents should never commit to git
Agents should stage files and leave committing to the user. Only create
commits when the user explicitly requests them.
2026-01-08 15:49:10 +00:00
Adam Hathcock
507b1e35d8 docs: Remove broken references to non-existent files
- Remove CONTRIBUTING.md reference from ARCHITECTURE.md
- Remove ERRORS.md reference from API.md
- Remove TROUBLESHOOTING.md reference from ENCODING.md
- Remove TROUBLESHOOTING.md reference from PERFORMANCE.md

All markdown files now reference only existing documentation.
2026-01-08 15:47:57 +00:00
Adam Hathcock
2839e1d33f Update docs/ENCODING.md
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2026-01-08 15:39:47 +00:00
Adam Hathcock
ef0b9d525c merge conflicts 2026-01-08 15:37:55 +00:00
Adam Hathcock
01e6e04a78 Merge branch 'master' into adam/async
# Conflicts:
#	src/SharpCompress/Common/Zip/Headers/LocalEntryHeader.cs
#	src/SharpCompress/Common/Zip/Headers/ZipFileEntry.cs
2026-01-08 15:36:41 +00:00
Adam Hathcock
8c876c70af Add more documentation 2026-01-08 15:34:48 +00:00
Adam Hathcock
a7d6d6493e add version numbers again to get them 2026-01-08 14:22:54 +00:00
Adam Hathcock
b6cc95af73 Merge pull request #1122 from adamhathcock/adam/netstandard-20-readd
Readd netstandard 2.0
2026-01-08 14:03:30 +00:00
Adam Hathcock
bdcc1d32c2 fix scratch dir creation 2026-01-08 14:01:35 +00:00
Adam Hathcock
90d91cc7c2 Merge pull request #1117 from adamhathcock/adam/rework-archive-encoding
Change ArchiveEncoding to interface.
2026-01-08 13:39:30 +00:00
Adam Hathcock
ec83cf588f Readd netstandard 2.0 2026-01-08 13:33:36 +00:00
Adam Hathcock
4f0a2e3c95 disable zip64 tests 2026-01-08 12:55:16 +00:00
Adam Hathcock
3747a27109 Task to ValueTask 2026-01-08 12:35:12 +00:00
Adam Hathcock
b501bac54a better names for new interfaces 2026-01-08 12:02:26 +00:00
Adam Hathcock
7aec98d652 read async interface for reader 2026-01-08 11:28:15 +00:00
Adam Hathcock
406b198e0e can't dispose before returning 2026-01-08 10:24:33 +00:00
Adam Hathcock
8e42296c3a switch Task to ValueTask 2026-01-08 10:22:53 +00:00
Adam Hathcock
60e5220bd0 fmt 2026-01-08 09:41:48 +00:00
Adam Hathcock
0f37cbfd0b archive async path uses new async interface 2026-01-08 09:39:04 +00:00
Adam Hathcock
541fd136d5 IArchiveAsync 2026-01-08 09:14:46 +00:00
Adam Hathcock
60d42ca9c3 fmt 2026-01-07 16:38:48 +00:00
Adam Hathcock
ac0716ddeb write testing 2026-01-07 15:01:04 +00:00
Adam Hathcock
b9792ca491 fix async zip decompression 2026-01-07 14:54:32 +00:00
Adam Hathcock
c3fd42057a Pass more Zip tests 2026-01-07 14:47:20 +00:00
Adam Hathcock
39d85ff4f6 conflicts from merge 2026-01-07 12:18:14 +00:00
Adam Hathcock
fbce3e77ba Merge branch 'master' into adam/async
# Conflicts:
#	src/SharpCompress/Utility.cs
2026-01-07 12:11:19 +00:00
Adam Hathcock
66e9de2685 fixed comment 2026-01-07 11:26:42 +00:00
Adam Hathcock
321520408b fmt 2026-01-07 11:12:02 +00:00
Adam Hathcock
68451bd75f Use explicit enum, add comments 2026-01-07 11:10:15 +00:00
Adam Hathcock
486fdf118b move to own files and refactor UTF8 usage 2026-01-07 10:39:18 +00:00
Adam Hathcock
bd3cda0617 Some restoring of functionality 2026-01-07 10:32:02 +00:00
Adam Hathcock
725503d1ce Change ArchiveEncoding to interface. Simplify class. Question what to do about Forced and complex access 2026-01-07 08:44:12 +00:00
Adam Hathcock
b825e15406 Merge pull request #1100 from adamhathcock/copilot/fix-read-method-implementations
Consolidate stream extension methods and simplify with framework methods
2026-01-05 17:24:54 +00:00
copilot-swe-agent[bot]
9bd86f64c9 Replace manual TransferTo implementation with Stream.CopyTo framework methods
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-03 18:41:02 +00:00
copilot-swe-agent[bot]
77015224f6 Add input validation for ReadBytesAsync count parameter
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-03 17:19:55 +00:00
copilot-swe-agent[bot]
372ecb77d0 Use threshold-based ArrayPool strategy for BinaryReaderExtensions
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-03 17:17:59 +00:00
copilot-swe-agent[bot]
05642cbdc6 Use ArrayPool for temporary buffers in BinaryReaderExtensions
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-03 17:12:43 +00:00
copilot-swe-agent[bot]
1a71c01fd4 Consolidate ReadExact and ReadFully methods into Utility.cs
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-03 15:58:27 +00:00
Adam Hathcock
54640548ed Consolidate reads 2026-01-03 15:47:10 +00:00
copilot-swe-agent[bot]
ea02d31096 Add IsArchiveAsync overloads for Zip and GZip factories
- Added IsArchiveAsync interface method to IFactory
- Implemented async versions of IsZipFile, IsZipMulti, IsGZipFile
- Updated ZipFactory and GZipFactory to override IsArchiveAsync
- Updated ReaderFactory.OpenAsync to use IsArchiveAsync
- Fixed Zip_Reader_Disposal_Test2_Async to use ReaderFactory.OpenAsync
- Fixed TestStream to properly forward ReadAsync calls
- Removed BufferedStream wrapping from AsyncBinaryReader as it uses sync Read
- Added default implementation in Factory base class

Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2026-01-02 18:10:54 +00:00
Adam Hathcock
d04830ba90 Add some OpenAsync 2026-01-02 17:52:18 +00:00
Adam Hathcock
8533b09091 start of implementing zip reading async 2025-12-31 14:53:55 +00:00
Adam Hathcock
44b7955d85 reader tests 2025-12-31 14:43:15 +00:00
Adam Hathcock
038b9f18c6 Merge remote-tracking branch 'origin/master' into copilot/add-buffered-stream-async-read 2025-12-31 14:24:31 +00:00
copilot-swe-agent[bot]
6e0e20ba6e Fix zip64_locator to use ReadUInt32Async instead of ReadUInt16Async
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2025-12-30 11:50:41 +00:00
copilot-swe-agent[bot]
ec31cb9987 Fix Zip headers to support both sync and async reading
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2025-12-30 11:47:31 +00:00
copilot-swe-agent[bot]
39a0b4ce78 Use BufferedStream for async reading in AsyncBinaryReader
Co-authored-by: adamhathcock <527620+adamhathcock@users.noreply.github.com>
2025-12-30 11:23:55 +00:00
Adam Hathcock
af719707bf Merge branch 'adam/async-binary-reader' into copilot/add-buffered-stream-async-read 2025-12-30 11:17:16 +00:00
copilot-swe-agent[bot]
8415a19912 Initial plan 2025-12-30 11:15:28 +00:00
Adam Hathcock
1607d2768e Merge branch 'master' into adam/async-binary-reader 2025-12-30 11:13:14 +00:00
Adam Hathcock
fb76bd82f2 first commit of async reader 2025-11-26 08:09:20 +00:00
Adam Hathcock
3bdaba46a9 fmt 2025-11-25 15:39:43 +00:00
Adam Hathcock
7c3c94ed7f Add ArcReaderAsync tests 2025-11-25 14:44:03 +00:00
263 changed files with 11695 additions and 4066 deletions

View File

@@ -307,7 +307,6 @@ dotnet_diagnostic.CS8602.severity = error
dotnet_diagnostic.CS8604.severity = error
dotnet_diagnostic.CS8618.severity = error
dotnet_diagnostic.CS0618.severity = suggestion
dotnet_diagnostic.CS1998.severity = error
dotnet_diagnostic.CS4014.severity = error
dotnet_diagnostic.CS8600.severity = error
dotnet_diagnostic.CS8603.severity = error
@@ -368,6 +367,9 @@ dotnet_diagnostic.NX0001.severity = error
dotnet_diagnostic.NX0002.severity = silent
dotnet_diagnostic.NX0003.severity = silent
dotnet_diagnostic.VSTHRD110.severity = error
dotnet_diagnostic.VSTHRD107.severity = error
##########################################
# Styles
##########################################

4
.gitignore vendored
View File

@@ -4,8 +4,8 @@ _ReSharper.SharpCompress/
bin/
*.suo
*.user
TestArchives/Scratch/
TestArchives/Scratch2/
tests/TestArchives/Scratch/
tests/TestArchives/Scratch2/
TestResults/
*.nupkg
packages/*/

View File

@@ -14,6 +14,7 @@ SharpCompress is a pure C# compression library supporting multiple archive forma
- Follow the existing code style and patterns in the codebase.
## General Instructions
- **Agents should NEVER commit to git** - Agents should stage files and leave committing to the user. Only create commits when the user explicitly requests them.
- Make only high confidence suggestions when reviewing code changes.
- Write code with good maintainability practices, including comments on why certain design decisions were made.
- Handle edge cases and write clear exception handling.
@@ -110,7 +111,7 @@ SharpCompress supports multiple archive and compression formats:
- **Archive Formats**: Zip, Tar, 7Zip, Rar (read-only)
- **Compression**: DEFLATE, BZip2, LZMA/LZMA2, PPMd, ZStandard (decompress only), Deflate64 (decompress only)
- **Combined Formats**: Tar.GZip, Tar.BZip2, Tar.LZip, Tar.XZ, Tar.ZStandard
- See FORMATS.md for complete format support matrix
- See [docs/FORMATS.md](docs/FORMATS.md) for complete format support matrix
### Stream Handling Rules
- **Disposal**: As of version 0.21, SharpCompress closes wrapped streams by default
@@ -177,5 +178,4 @@ SharpCompress supports multiple archive and compression formats:
2. **Solid archives (Rar, 7Zip)** - Use `ExtractAllEntries()` for best performance, not individual entry extraction
3. **Stream disposal** - Always set `LeaveStreamOpen` explicitly when needed (default is to close)
4. **Tar + non-seekable stream** - Must provide file size or it will throw
5. **Multi-framework differences** - Some features differ between .NET Framework and modern .NET (e.g., Mono.Posix)
6. **Format detection** - Use `ReaderFactory.Open()` for auto-detection, test with actual archive files

View File

@@ -12,5 +12,6 @@
<RunAnalyzersDuringBuild>False</RunAnalyzersDuringBuild>
<ManagePackageVersionsCentrally>true</ManagePackageVersionsCentrally>
<RestorePackagesWithLockFile>true</RestorePackagesWithLockFile>
<CentralPackageTransitivePinningEnabled>true</CentralPackageTransitivePinningEnabled>
</PropertyGroup>
</Project>

View File

@@ -5,6 +5,7 @@
<PackageVersion Include="Glob" Version="1.1.9" />
<PackageVersion Include="JetBrains.Profiler.SelfApi" Version="2.5.15" />
<PackageVersion Include="Microsoft.Bcl.AsyncInterfaces" Version="10.0.0" />
<PackageVersion Include="Microsoft.NET.ILLink.Task" Version="10.0.0" />
<PackageVersion Include="Microsoft.NET.Test.Sdk" Version="18.0.1" />
<PackageVersion Include="Mono.Posix.NETStandard" Version="1.0.0" />
<PackageVersion Include="SimpleExec" Version="13.0.0" />
@@ -13,7 +14,11 @@
<PackageVersion Include="System.Memory" Version="4.6.3" />
<PackageVersion Include="xunit" Version="2.9.3" />
<PackageVersion Include="xunit.runner.visualstudio" Version="3.1.5" />
<GlobalPackageReference Include="Microsoft.SourceLink.GitHub" Version="10.0.102" />
<GlobalPackageReference Include="Microsoft.SourceLink.GitHub" Version="8.0.0" />
<GlobalPackageReference Include="Microsoft.NETFramework.ReferenceAssemblies" Version="1.0.3" />
<GlobalPackageReference
Include="Microsoft.VisualStudio.Threading.Analyzers"
Version="17.14.15"
/>
</ItemGroup>
</Project>

View File

@@ -4,7 +4,7 @@ SharpCompress is a compression library in pure C# for .NET Framework 4.8, .NET 8
The major feature is support for non-seekable streams so large files can be processed on the fly (i.e. download stream).
**NEW:** All I/O operations now support async/await for improved performance and scalability. See the [USAGE.md](USAGE.md#async-examples) for examples.
**NEW:** All I/O operations now support async/await for improved performance and scalability. See the [USAGE.md](docs/USAGE.md#async-examples) for examples.
GitHub Actions Build -
[![SharpCompress](https://github.com/adamhathcock/sharpcompress/actions/workflows/dotnetcore.yml/badge.svg)](https://github.com/adamhathcock/sharpcompress/actions/workflows/dotnetcore.yml)
@@ -14,7 +14,7 @@ GitHub Actions Build -
Post Issues on Github!
Check the [Supported Formats](FORMATS.md) and [Basic Usage.](USAGE.md)
Check the [Supported Formats](docs/FORMATS.md) and [Basic Usage.](docs/USAGE.md)
## Recommended Formats

View File

@@ -18,12 +18,11 @@ Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "Config", "Config", "{CDB425
Directory.Build.props = Directory.Build.props
global.json = global.json
.editorconfig = .editorconfig
.gitignore = .gitignore
Directory.Packages.props = Directory.Packages.props
NuGet.config = NuGet.config
.github\workflows\nuget-release.yml = .github\workflows\nuget-release.yml
USAGE.md = USAGE.md
README.md = README.md
FORMATS.md = FORMATS.md
AGENTS.md = AGENTS.md
EndProjectSection
EndProject

View File

@@ -230,7 +230,7 @@ static async Task<(string version, bool isPrerelease)> GetVersion()
}
else
{
// Not tagged - create prerelease version
// Not tagged - create prerelease version based on next minor version
var allTags = (await GetGitOutput("tag", "--list"))
.Split('\n', StringSplitOptions.RemoveEmptyEntries)
.Where(tag => Regex.IsMatch(tag.Trim(), @"^\d+\.\d+\.\d+$"))
@@ -240,22 +240,8 @@ static async Task<(string version, bool isPrerelease)> GetVersion()
var lastTag = allTags.OrderBy(tag => Version.Parse(tag)).LastOrDefault() ?? "0.0.0";
var lastVersion = Version.Parse(lastTag);
// Determine version increment based on branch
var currentBranch = await GetCurrentBranch();
Version nextVersion;
if (currentBranch == "release")
{
// Release branch: increment patch version
nextVersion = new Version(lastVersion.Major, lastVersion.Minor, lastVersion.Build + 1);
Console.WriteLine($"Building prerelease for release branch (patch increment)");
}
else
{
// Master or other branches: increment minor version
nextVersion = new Version(lastVersion.Major, lastVersion.Minor + 1, 0);
Console.WriteLine($"Building prerelease for {currentBranch} branch (minor increment)");
}
// Increment minor version for next release
var nextVersion = new Version(lastVersion.Major, lastVersion.Minor + 1, 0);
// Use commit count since the last version tag if available; otherwise, fall back to total count
var revListArgs = allTags.Any() ? $"--count {lastTag}..HEAD" : "--count HEAD";
@@ -267,28 +253,6 @@ static async Task<(string version, bool isPrerelease)> GetVersion()
}
}
static async Task<string> GetCurrentBranch()
{
// In GitHub Actions, GITHUB_REF_NAME contains the branch name
var githubRefName = Environment.GetEnvironmentVariable("GITHUB_REF_NAME");
if (!string.IsNullOrEmpty(githubRefName))
{
return githubRefName;
}
// Fallback to git command for local builds
try
{
var (output, _) = await ReadAsync("git", "branch --show-current");
return output.Trim();
}
catch (Exception ex)
{
Console.WriteLine($"Warning: Could not determine current branch: {ex.Message}");
return "unknown";
}
}
static async Task<string> GetGitOutput(string command, string args)
{
try

View File

@@ -25,14 +25,20 @@
},
"Microsoft.SourceLink.GitHub": {
"type": "Direct",
"requested": "[10.0.102, )",
"resolved": "10.0.102",
"contentHash": "Oxq3RCIJSdtpIU4hLqO7XaDe/Ra3HS9Wi8rJl838SAg6Zu1iQjerA0+xXWBgUFYbgknUGCLOU0T+lzMLkvY9Qg==",
"requested": "[8.0.0, )",
"resolved": "8.0.0",
"contentHash": "G5q7OqtwIyGTkeIOAc3u2ZuV/kicQaec5EaRnc0pIeSnh9LUjj+PYQrJYBURvDt7twGl2PKA7nSN0kz1Zw5bnQ==",
"dependencies": {
"Microsoft.Build.Tasks.Git": "10.0.102",
"Microsoft.SourceLink.Common": "10.0.102"
"Microsoft.Build.Tasks.Git": "8.0.0",
"Microsoft.SourceLink.Common": "8.0.0"
}
},
"Microsoft.VisualStudio.Threading.Analyzers": {
"type": "Direct",
"requested": "[17.14.15, )",
"resolved": "17.14.15",
"contentHash": "mXQPJsbuUD2ydq4/ffd8h8tSOFCXec+2xJOVNCvXjuMOq/+5EKHq3D2m2MC2+nUaXeFMSt66VS/J4HdKBixgcw=="
},
"SimpleExec": {
"type": "Direct",
"requested": "[13.0.0, )",
@@ -41,8 +47,8 @@
},
"Microsoft.Build.Tasks.Git": {
"type": "Transitive",
"resolved": "10.0.102",
"contentHash": "0i81LYX31U6UiXz4NOLbvc++u+/mVDmOt+PskrM/MygpDxkv9THKQyRUmavBpLK6iBV0abNWnn+CQgSRz//Pwg=="
"resolved": "8.0.0",
"contentHash": "bZKfSIKJRXLTuSzLudMFte/8CempWjVamNUR5eHJizsy+iuOuO/k2gnh7W0dHJmYY0tBf+gUErfluCv5mySAOQ=="
},
"Microsoft.NETFramework.ReferenceAssemblies.net461": {
"type": "Transitive",
@@ -51,8 +57,8 @@
},
"Microsoft.SourceLink.Common": {
"type": "Transitive",
"resolved": "10.0.102",
"contentHash": "Mk1IMb9q5tahC2NltxYXFkLBtuBvfBoCQ3pIxYQWfzbCE9o1OB9SsHe0hnNGo7lWgTA/ePbFAJLWu6nLL9K17A=="
"resolved": "8.0.0",
"contentHash": "dk9JPxTCIevS75HyEQ0E4OVAFhB2N+V9ShCXf8Q6FkUQZDkgLI12y679Nym1YqsiSysuQskT7Z+6nUf3yab6Vw=="
}
}
}

512
docs/API.md Normal file
View File

@@ -0,0 +1,512 @@
# API Quick Reference
Quick reference for commonly used SharpCompress APIs.
## Factory Methods
### Opening Archives
```csharp
// Auto-detect format
using (var reader = ReaderFactory.OpenReader(stream))
{
// Works with Zip, Tar, GZip, Rar, 7Zip, etc.
}
// Specific format - Archive API
using (var archive = ZipArchive.OpenArchive("file.zip"))
using (var archive = TarArchive.OpenArchive("file.tar"))
using (var archive = RarArchive.OpenArchive("file.rar"))
using (var archive = SevenZipArchive.OpenArchive("file.7z"))
using (var archive = GZipArchive.OpenArchive("file.gz"))
// With options
var options = new ReaderOptions
{
Password = "password",
LeaveStreamOpen = true,
ArchiveEncoding = new ArchiveEncoding { Default = Encoding.GetEncoding(932) }
};
using (var archive = ZipArchive.OpenArchive("encrypted.zip", options))
```
### Creating Archives
```csharp
// Writer Factory
using (var writer = WriterFactory.OpenWriter(stream, ArchiveType.Zip, CompressionType.Deflate))
{
// Write entries
}
// Specific writer
using (var archive = ZipArchive.CreateArchive())
using (var archive = TarArchive.CreateArchive())
using (var archive = GZipArchive.CreateArchive())
// With options
var options = new WriterOptions(CompressionType.Deflate)
{
CompressionLevel = 9,
LeaveStreamOpen = false
};
using (var archive = ZipArchive.CreateArchive())
{
archive.SaveTo("output.zip", options);
}
```
---
## Archive API Methods
### Reading/Extracting
```csharp
using (var archive = ZipArchive.OpenArchive("file.zip"))
{
// Get all entries
IEnumerable<IArchiveEntry> entries = archive.Entries;
// Find specific entry
var entry = archive.Entries.FirstOrDefault(e => e.Key == "file.txt");
// Extract all
archive.WriteToDirectory(@"C:\output", new ExtractionOptions
{
ExtractFullPath = true,
Overwrite = true
});
// Extract single entry
var entry = archive.Entries.First();
entry.WriteToFile(@"C:\output\file.txt");
entry.WriteToFile(@"C:\output\file.txt", new ExtractionOptions { Overwrite = true });
// Get entry stream
using (var stream = entry.OpenEntryStream())
{
stream.CopyTo(outputStream);
}
}
// Async extraction (requires IAsyncArchive)
using (var asyncArchive = await ZipArchive.OpenAsyncArchive("file.zip"))
{
await asyncArchive.WriteToDirectoryAsync(
@"C:\output",
new ExtractionOptions { ExtractFullPath = true, Overwrite = true },
cancellationToken: cancellationToken
);
}
using (var stream = await entry.OpenEntryStreamAsync(cancellationToken))
{
// ...
}
```
### Entry Properties
```csharp
foreach (var entry in archive.Entries)
{
string name = entry.Key; // Entry name/path
long size = entry.Size; // Uncompressed size
long compressedSize = entry.CompressedSize;
bool isDir = entry.IsDirectory;
DateTime? modTime = entry.LastModifiedTime;
CompressionType compression = entry.CompressionType;
}
```
### Creating Archives
```csharp
using (var archive = ZipArchive.CreateArchive())
{
// Add file
archive.AddEntry("file.txt", @"C:\source\file.txt");
// Add multiple files
archive.AddAllFromDirectory(@"C:\source");
archive.AddAllFromDirectory(@"C:\source", "*.txt"); // Pattern
// Save to file
archive.SaveTo("output.zip", CompressionType.Deflate);
// Save to stream
archive.SaveTo(outputStream, new WriterOptions(CompressionType.Deflate)
{
CompressionLevel = 9,
LeaveStreamOpen = true
});
}
```
---
## Reader API Methods
### Forward-Only Reading
```csharp
using (var stream = File.OpenRead("file.zip"))
using (var reader = ReaderFactory.OpenReader(stream))
{
while (reader.MoveToNextEntry())
{
IArchiveEntry entry = reader.Entry;
if (!entry.IsDirectory)
{
// Extract entry
reader.WriteEntryToDirectory(@"C:\output");
reader.WriteEntryToFile(@"C:\output\file.txt");
// Or get stream
using (var entryStream = reader.OpenEntryStream())
{
entryStream.CopyTo(outputStream);
}
}
}
}
// Async variants (use OpenAsyncReader to get IAsyncReader)
using (var stream = File.OpenRead("file.zip"))
using (var reader = await ReaderFactory.OpenAsyncReader(stream))
{
while (await reader.MoveToNextEntryAsync())
{
await reader.WriteEntryToFileAsync(
@"C:\output\" + reader.Entry.Key,
cancellationToken: cancellationToken
);
}
// Async extraction of all entries
await reader.WriteAllToDirectoryAsync(
@"C:\output",
new ExtractionOptions { ExtractFullPath = true, Overwrite = true },
cancellationToken
);
}
```
---
## Writer API Methods
### Creating Archives (Streaming)
```csharp
using (var stream = File.Create("output.zip"))
using (var writer = WriterFactory.OpenWriter(stream, ArchiveType.Zip, CompressionType.Deflate))
{
// Write single file
using (var fileStream = File.OpenRead("source.txt"))
{
writer.Write("entry.txt", fileStream, DateTime.Now);
}
// Write directory
writer.WriteAll("C:\\source", "*", SearchOption.AllDirectories);
writer.WriteAll("C:\\source", "*.txt", SearchOption.TopDirectoryOnly);
// Async variants
using (var fileStream = File.OpenRead("source.txt"))
{
await writer.WriteAsync("entry.txt", fileStream, DateTime.Now, cancellationToken);
}
await writer.WriteAllAsync("C:\\source", "*", SearchOption.AllDirectories, cancellationToken);
}
```
---
## Common Options
### ReaderOptions
```csharp
var options = new ReaderOptions
{
Password = "password", // For encrypted archives
LeaveStreamOpen = true, // Don't close wrapped stream
ArchiveEncoding = new ArchiveEncoding // Custom character encoding
{
Default = Encoding.GetEncoding(932)
}
};
using (var archive = ZipArchive.OpenArchive("file.zip", options))
{
// ...
}
```
### WriterOptions
```csharp
var options = new WriterOptions(CompressionType.Deflate)
{
CompressionLevel = 9, // 0-9 for Deflate
LeaveStreamOpen = true, // Don't close stream
};
archive.SaveTo("output.zip", options);
```
### ExtractionOptions
```csharp
var options = new ExtractionOptions
{
ExtractFullPath = true, // Recreate directory structure
Overwrite = true, // Overwrite existing files
PreserveFileTime = true // Keep original timestamps
};
archive.WriteToDirectory(@"C:\output", options);
```
---
## Compression Types
### Available Compressions
```csharp
// For creating archives
CompressionType.None // No compression (store)
CompressionType.Deflate // DEFLATE (default for ZIP/GZip)
CompressionType.Deflate64 // Deflate64
CompressionType.BZip2 // BZip2
CompressionType.LZMA // LZMA (for 7Zip, LZip, XZ)
CompressionType.PPMd // PPMd (for ZIP)
CompressionType.Rar // RAR compression (read-only)
CompressionType.ZStandard // ZStandard
ArchiveType.Arc
ArchiveType.Arj
ArchiveType.Ace
// For Tar archives with compression
// Use WriterFactory to create compressed tar archives
using (var writer = WriterFactory.OpenWriter(stream, ArchiveType.Tar, CompressionType.GZip)) // Tar.GZip
using (var writer = WriterFactory.OpenWriter(stream, ArchiveType.Tar, CompressionType.BZip2)) // Tar.BZip2
```
### Archive Types
```csharp
ArchiveType.Zip
ArchiveType.Tar
ArchiveType.GZip
ArchiveType.BZip2
ArchiveType.Rar
ArchiveType.SevenZip
ArchiveType.XZ
ArchiveType.ZStandard
```
---
## Patterns & Examples
### Extract with Error Handling
```csharp
try
{
using (var archive = ZipArchive.Open("archive.zip",
new ReaderOptions { Password = "password" }))
{
archive.WriteToDirectory(@"C:\output", new ExtractionOptions
{
ExtractFullPath = true,
Overwrite = true
});
}
}
catch (PasswordRequiredException)
{
Console.WriteLine("Password required");
}
catch (InvalidArchiveException)
{
Console.WriteLine("Archive is invalid");
}
catch (SharpCompressException ex)
{
Console.WriteLine($"Error: {ex.Message}");
}
```
### Extract with Progress
```csharp
var progress = new Progress<ProgressReport>(report =>
{
Console.WriteLine($"Extracting {report.EntryPath}: {report.PercentComplete}%");
});
var options = new ReaderOptions { Progress = progress };
using (var archive = ZipArchive.OpenArchive("archive.zip", options))
{
archive.WriteToDirectory(@"C:\output");
}
```
### Async Extract with Cancellation
```csharp
var cts = new CancellationTokenSource();
cts.CancelAfter(TimeSpan.FromMinutes(5));
try
{
using (var archive = await ZipArchive.OpenAsyncArchive("archive.zip"))
{
await archive.WriteToDirectoryAsync(
@"C:\output",
new ExtractionOptions { ExtractFullPath = true, Overwrite = true },
cancellationToken: cts.Token
);
}
}
catch (OperationCanceledException)
{
Console.WriteLine("Extraction cancelled");
}
```
### Create with Custom Compression
```csharp
using (var archive = ZipArchive.CreateArchive())
{
archive.AddAllFromDirectory(@"D:\source");
// Fastest
archive.SaveTo("fast.zip", new WriterOptions(CompressionType.Deflate)
{
CompressionLevel = 1
});
// Balanced (default)
archive.SaveTo("normal.zip", CompressionType.Deflate);
// Best compression
archive.SaveTo("best.zip", new WriterOptions(CompressionType.Deflate)
{
CompressionLevel = 9
});
}
```
### Stream Processing (No File I/O)
```csharp
using (var outputStream = new MemoryStream())
using (var archive = ZipArchive.CreateArchive())
{
// Add content from memory
using (var contentStream = new MemoryStream(Encoding.UTF8.GetBytes("Hello")))
{
archive.AddEntry("file.txt", contentStream);
}
// Save to memory
archive.SaveTo(outputStream, CompressionType.Deflate);
// Get bytes
byte[] archiveBytes = outputStream.ToArray();
}
```
### Extract Specific Files
```csharp
using (var archive = ZipArchive.OpenArchive("archive.zip"))
{
var filesToExtract = new[] { "file1.txt", "file2.txt" };
foreach (var entry in archive.Entries.Where(e => filesToExtract.Contains(e.Key)))
{
entry.WriteToFile(@"C:\output\" + entry.Key);
}
}
```
### List Archive Contents
```csharp
using (var archive = ZipArchive.OpenArchive("archive.zip"))
{
foreach (var entry in archive.Entries)
{
if (entry.IsDirectory)
Console.WriteLine($"[DIR] {entry.Key}");
else
Console.WriteLine($"[FILE] {entry.Key} ({entry.Size} bytes)");
}
}
```
---
## Common Mistakes
### ✗ Wrong - Stream not disposed
```csharp
var stream = File.OpenRead("archive.zip");
var archive = ZipArchive.OpenArchive(stream);
archive.WriteToDirectory(@"C:\output");
// stream not disposed - leaked resource
```
### ✓ Correct - Using blocks
```csharp
using (var stream = File.OpenRead("archive.zip"))
using (var archive = ZipArchive.OpenArchive(stream))
{
archive.WriteToDirectory(@"C:\output");
}
// Both properly disposed
```
### ✗ Wrong - Mixing API styles
```csharp
// Loading entire archive then iterating
using (var archive = ZipArchive.OpenArchive("large.zip"))
{
var entries = archive.Entries.ToList(); // Loads all in memory
foreach (var e in entries)
{
e.WriteToFile(...); // Then extracts each
}
}
```
### ✓ Correct - Use Reader for large files
```csharp
// Streaming iteration
using (var stream = File.OpenRead("large.zip"))
using (var reader = ReaderFactory.OpenReader(stream))
{
while (reader.MoveToNextEntry())
{
reader.WriteEntryToDirectory(@"C:\output");
}
}
```
---
## Related Documentation
- [USAGE.md](USAGE.md) - Complete code examples
- [FORMATS.md](FORMATS.md) - Supported formats
- [PERFORMANCE.md](PERFORMANCE.md) - API selection guide

659
docs/ARCHITECTURE.md Normal file
View File

@@ -0,0 +1,659 @@
# SharpCompress Architecture Guide
This guide explains the internal architecture and design patterns of SharpCompress for contributors.
## Overview
SharpCompress is organized into three main layers:
```
┌─────────────────────────────────────────┐
│ User-Facing APIs (Top Layer) │
│ Archive, Reader, Writer Factories │
├─────────────────────────────────────────┤
│ Format-Specific Implementations │
│ ZipArchive, TarReader, GZipWriter, │
│ RarArchive, SevenZipArchive, etc. │
├─────────────────────────────────────────┤
│ Compression & Crypto (Bottom Layer) │
│ Deflate, LZMA, BZip2, AES, CRC32 │
└─────────────────────────────────────────┘
```
---
## Directory Structure
### `src/SharpCompress/`
#### `Archives/` - Archive Implementations
Contains `IArchive` implementations for seekable, random-access APIs.
**Key Files:**
- `AbstractArchive.cs` - Base class for all archives
- `IArchive.cs` - Archive interface definition
- `ArchiveFactory.cs` - Factory for opening archives
- Format-specific: `ZipArchive.cs`, `TarArchive.cs`, `RarArchive.cs`, `SevenZipArchive.cs`, `GZipArchive.cs`
**Use Archive API when:**
- Stream is seekable (file, memory)
- Need random access to entries
- Archive fits in memory
- Simplicity is important
#### `Readers/` - Reader Implementations
Contains `IReader` implementations for forward-only, non-seekable APIs.
**Key Files:**
- `AbstractReader.cs` - Base reader class
- `IReader.cs` - Reader interface
- `ReaderFactory.cs` - Auto-detection factory
- `ReaderOptions.cs` - Configuration for readers
- Format-specific: `ZipReader.cs`, `TarReader.cs`, `GZipReader.cs`, `RarReader.cs`, etc.
**Use Reader API when:**
- Stream is non-seekable (network, pipe, compressed)
- Processing large files
- Memory is limited
- Forward-only processing is acceptable
#### `Writers/` - Writer Implementations
Contains `IWriter` implementations for forward-only writing.
**Key Files:**
- `AbstractWriter.cs` - Base writer class
- `IWriter.cs` - Writer interface
- `WriterFactory.cs` - Factory for creating writers
- `WriterOptions.cs` - Configuration for writers
- Format-specific: `ZipWriter.cs`, `TarWriter.cs`, `GZipWriter.cs`
#### `Factories/` - Format Detection
Factory classes for auto-detecting archive format and creating appropriate readers/writers.
**Key Files:**
- `Factory.cs` - Base factory class
- `IFactory.cs` - Factory interface
- Format-specific: `ZipFactory.cs`, `TarFactory.cs`, `RarFactory.cs`, etc.
**How It Works:**
1. `ReaderFactory.OpenReader(stream)` probes stream signatures
2. Identifies format by magic bytes
3. Creates appropriate reader instance
4. Returns generic `IReader` interface
#### `Common/` - Shared Types
Common types, options, and enumerations used across formats.
**Key Files:**
- `IEntry.cs` - Entry interface (file within archive)
- `Entry.cs` - Entry implementation
- `ArchiveType.cs` - Enum for archive formats
- `CompressionType.cs` - Enum for compression methods
- `ArchiveEncoding.cs` - Character encoding configuration
- `ExtractionOptions.cs` - Extraction configuration
- Format-specific headers: `Zip/Headers/`, `Tar/Headers/`, `Rar/Headers/`, etc.
#### `Compressors/` - Compression Algorithms
Low-level compression streams implementing specific algorithms.
**Algorithms:**
- `Deflate/` - DEFLATE compression (Zip default)
- `BZip2/` - BZip2 compression
- `LZMA/` - LZMA compression (7Zip, XZ, LZip)
- `PPMd/` - Prediction by Partial Matching (Zip, 7Zip)
- `ZStandard/` - ZStandard compression (decompression only)
- `Xz/` - XZ format (decompression only)
- `Rar/` - RAR-specific unpacking
- `Arj/`, `Arc/`, `Ace/` - Legacy format decompression
- `Filters/` - BCJ/BCJ2 filters for executable compression
**Each Compressor:**
- Implements a `Stream` subclass
- Provides both compression and decompression
- Some are read-only (decompression only)
#### `Crypto/` - Encryption & Hashing
Cryptographic functions and stream wrappers.
**Key Files:**
- `Crc32Stream.cs` - CRC32 calculation wrapper
- `BlockTransformer.cs` - Block cipher transformations
- AES, PKWare, WinZip encryption implementations
#### `IO/` - Stream Utilities
Stream wrappers and utilities.
**Key Classes:**
- `SharpCompressStream` - Base stream class
- `ProgressReportingStream` - Progress tracking wrapper
- `MarkingBinaryReader` - Binary reader with position marks
- `BufferedSubStream` - Buffered read-only substream
- `ReadOnlySubStream` - Read-only view of parent stream
- `NonDisposingStream` - Prevents wrapped stream disposal
---
## Design Patterns
### 1. Factory Pattern
**Purpose:** Auto-detect format and create appropriate reader/writer.
**Example:**
```csharp
// User calls factory
using (var reader = ReaderFactory.OpenReader(stream)) // Returns IReader
{
while (reader.MoveToNextEntry())
{
// Process entry
}
}
// Behind the scenes:
// 1. Factory.Open() probes stream signatures
// 2. Detects format (Zip, Tar, Rar, etc.)
// 3. Creates appropriate reader (ZipReader, TarReader, etc.)
// 4. Returns as generic IReader interface
```
**Files:**
- `src/SharpCompress/Factories/ReaderFactory.cs`
- `src/SharpCompress/Factories/WriterFactory.cs`
- `src/SharpCompress/Factories/ArchiveFactory.cs`
### 2. Strategy Pattern
**Purpose:** Encapsulate compression algorithms as swappable strategies.
**Example:**
```csharp
// Different compression strategies
CompressionType.Deflate // DEFLATE
CompressionType.BZip2 // BZip2
CompressionType.LZMA // LZMA
CompressionType.PPMd // PPMd
// Writer uses strategy pattern
var archive = ZipArchive.CreateArchive();
archive.SaveTo("output.zip", CompressionType.Deflate); // Use Deflate
archive.SaveTo("output.bz2", CompressionType.BZip2); // Use BZip2
```
**Files:**
- `src/SharpCompress/Compressors/` - Strategy implementations
### 3. Decorator Pattern
**Purpose:** Wrap streams with additional functionality.
**Example:**
```csharp
// Progress reporting decorator
var progressStream = new ProgressReportingStream(baseStream, progressReporter);
progressStream.Read(buffer, 0, buffer.Length); // Reports progress
// Non-disposing decorator
var nonDisposingStream = new NonDisposingStream(baseStream);
using (var compressor = new DeflateStream(nonDisposingStream))
{
// baseStream won't be disposed when compressor is disposed
}
```
**Files:**
- `src/SharpCompress/IO/ProgressReportingStream.cs`
- `src/SharpCompress/IO/NonDisposingStream.cs`
### 4. Template Method Pattern
**Purpose:** Define algorithm skeleton in base class, let subclasses fill details.
**Example:**
```csharp
// AbstractArchive defines common archive operations
public abstract class AbstractArchive : IArchive
{
// Template methods
public virtual void WriteToDirectory(string destinationDirectory, ExtractionOptions options)
{
// Common extraction logic
foreach (var entry in Entries)
{
// Call subclass method
entry.WriteToFile(destinationPath, options);
}
}
// Subclasses override format-specific details
protected abstract Entry CreateEntry(EntryData data);
}
```
**Files:**
- `src/SharpCompress/Archives/AbstractArchive.cs`
- `src/SharpCompress/Readers/AbstractReader.cs`
### 5. Iterator Pattern
**Purpose:** Provide sequential access to entries.
**Example:**
```csharp
// Archive API - provides collection
IEnumerable<IEntry> entries = archive.Entries;
foreach (var entry in entries)
{
// Random access - entries already in memory
}
// Reader API - provides iterator
IReader reader = ReaderFactory.OpenReader(stream);
while (reader.MoveToNextEntry())
{
// Forward-only iteration - one entry at a time
var entry = reader.Entry;
}
```
---
## Key Interfaces
### IArchive - Random Access API
```csharp
public interface IArchive : IDisposable
{
IEnumerable<IEntry> Entries { get; }
void WriteToDirectory(string destinationDirectory,
ExtractionOptions options = null);
IEntry FirstOrDefault(Func<IEntry, bool> predicate);
// ... format-specific methods
}
```
**Implementations:** `ZipArchive`, `TarArchive`, `RarArchive`, `SevenZipArchive`, `GZipArchive`
### IReader - Forward-Only API
```csharp
public interface IReader : IDisposable
{
IEntry Entry { get; }
bool MoveToNextEntry();
void WriteEntryToDirectory(string destinationDirectory,
ExtractionOptions options = null);
Stream OpenEntryStream();
// ... async variants
}
```
**Implementations:** `ZipReader`, `TarReader`, `RarReader`, `GZipReader`, etc.
### IWriter - Writing API
```csharp
public interface IWriter : IDisposable
{
void Write(string entryPath, Stream source,
DateTime? modificationTime = null);
void WriteAll(string sourceDirectory, string searchPattern,
SearchOption searchOption);
// ... async variants
}
```
**Implementations:** `ZipWriter`, `TarWriter`, `GZipWriter`
### IEntry - Archive Entry
```csharp
public interface IEntry
{
string Key { get; }
uint Size { get; }
uint CompressedSize { get; }
bool IsDirectory { get; }
DateTime? LastModifiedTime { get; }
CompressionType CompressionType { get; }
void WriteToFile(string fullPath, ExtractionOptions options = null);
void WriteToStream(Stream destinationStream);
Stream OpenEntryStream();
// ... async variants
}
```
---
## Adding Support for a New Format
### Step 1: Understand the Format
- Research format specification
- Understand compression/encryption used
- Study existing similar formats in codebase
### Step 2: Create Format Structure Classes
**Create:** `src/SharpCompress/Common/NewFormat/`
```csharp
// Headers and data structures
public class NewFormatHeader
{
public uint Magic { get; set; }
public ushort Version { get; set; }
// ... other fields
public static NewFormatHeader Read(BinaryReader reader)
{
// Deserialize from binary
}
}
public class NewFormatEntry
{
public string FileName { get; set; }
public uint CompressedSize { get; set; }
public uint UncompressedSize { get; set; }
// ... other fields
}
```
### Step 3: Create Archive Implementation
**Create:** `src/SharpCompress/Archives/NewFormat/NewFormatArchive.cs`
```csharp
public class NewFormatArchive : AbstractArchive
{
private NewFormatHeader _header;
private List<NewFormatEntry> _entries;
public static NewFormatArchive OpenArchive(Stream stream)
{
var archive = new NewFormatArchive();
archive._header = NewFormatHeader.Read(stream);
archive.LoadEntries(stream);
return archive;
}
public override IEnumerable<IEntry> Entries => _entries.Select(e => new Entry(e));
protected override Stream OpenEntryStream(Entry entry)
{
// Return decompressed stream for entry
}
// ... other abstract method implementations
}
```
### Step 4: Create Reader Implementation
**Create:** `src/SharpCompress/Readers/NewFormat/NewFormatReader.cs`
```csharp
public class NewFormatReader : AbstractReader
{
private NewFormatHeader _header;
private BinaryReader _reader;
public NewFormatReader(Stream stream)
{
_reader = new BinaryReader(stream);
_header = NewFormatHeader.Read(_reader);
}
public override bool MoveToNextEntry()
{
// Read next entry header
if (!_reader.BaseStream.CanRead) return false;
var entryData = NewFormatEntry.Read(_reader);
// ... set this.Entry
return entryData != null;
}
// ... other abstract method implementations
}
```
### Step 5: Create Factory
**Create:** `src/SharpCompress/Factories/NewFormatFactory.cs`
```csharp
public class NewFormatFactory : Factory, IArchiveFactory, IReaderFactory
{
// Archive format magic bytes (signature)
private static readonly byte[] NewFormatSignature = new byte[] { 0x4E, 0x46 }; // "NF"
public static NewFormatFactory Instance { get; } = new();
public IArchive CreateArchive(Stream stream)
=> NewFormatArchive.OpenArchive(stream);
public IReader CreateReader(Stream stream, ReaderOptions options)
=> new NewFormatReader(stream) { Options = options };
public bool Matches(Stream stream, ReadOnlySpan<byte> signature)
=> signature.StartsWith(NewFormatSignature);
}
```
### Step 6: Register Factory
**Update:** `src/SharpCompress/Factories/ArchiveFactory.cs`
```csharp
private static readonly IFactory[] Factories =
{
ZipFactory.Instance,
TarFactory.Instance,
RarFactory.Instance,
SevenZipFactory.Instance,
GZipFactory.Instance,
NewFormatFactory.Instance, // Add here
// ... other factories
};
```
### Step 7: Add Tests
**Create:** `tests/SharpCompress.Test/NewFormat/NewFormatTests.cs`
```csharp
public class NewFormatTests : TestBase
{
[Fact]
public void NewFormat_Extracts_Successfully()
{
var archivePath = Path.Combine(TEST_ARCHIVES_PATH, "archive.newformat");
using (var archive = NewFormatArchive.OpenArchive(archivePath))
{
archive.WriteToDirectory(SCRATCH_FILES_PATH);
// Assert extraction
}
}
[Fact]
public void NewFormat_Reader_Works()
{
var archivePath = Path.Combine(TEST_ARCHIVES_PATH, "archive.newformat");
using (var stream = File.OpenRead(archivePath))
using (var reader = new NewFormatReader(stream))
{
Assert.True(reader.MoveToNextEntry());
Assert.NotNull(reader.Entry);
}
}
}
```
### Step 8: Add Test Archives
Place test files in `tests/TestArchives/Archives/NewFormat/` directory.
### Step 9: Document
Update `docs/FORMATS.md` with format support information.
---
## Compression Algorithm Implementation
### Creating a New Compression Stream
**Example:** Creating `CustomStream` for a custom compression algorithm
```csharp
public class CustomStream : Stream
{
private readonly Stream _baseStream;
private readonly bool _leaveOpen;
public CustomStream(Stream baseStream, bool leaveOpen = false)
{
_baseStream = baseStream;
_leaveOpen = leaveOpen;
}
public override int Read(byte[] buffer, int offset, int count)
{
// Decompress data from _baseStream into buffer
// Return number of decompressed bytes
}
public override void Write(byte[] buffer, int offset, int count)
{
// Compress data from buffer into _baseStream
}
protected override void Dispose(bool disposing)
{
if (disposing && !_leaveOpen)
{
_baseStream?.Dispose();
}
base.Dispose(disposing);
}
}
```
---
## Stream Handling Best Practices
### Disposal Pattern
```csharp
// Correct: Nested using blocks
using (var fileStream = File.OpenRead("archive.zip"))
using (var archive = ZipArchive.OpenArchive(fileStream))
{
archive.WriteToDirectory(@"C:\output");
}
// Both archive and fileStream properly disposed
// Correct: Using with options
var options = new ReaderOptions { LeaveStreamOpen = true };
var stream = File.OpenRead("archive.zip");
using (var archive = ZipArchive.OpenArchive(stream, options))
{
archive.WriteToDirectory(@"C:\output");
}
stream.Dispose(); // Manually dispose if LeaveStreamOpen = true
```
### NonDisposingStream Wrapper
```csharp
// Prevent unwanted stream closure
var baseStream = File.OpenRead("data.bin");
var nonDisposing = new NonDisposingStream(baseStream);
using (var compressor = new DeflateStream(nonDisposing))
{
// Compressor won't close baseStream when disposed
}
// baseStream still usable
baseStream.Position = 0; // Works
baseStream.Dispose(); // Manual disposal
```
---
## Performance Considerations
### Memory Efficiency
1. **Avoid loading entire archive in memory** - Use Reader API for large files
2. **Process entries sequentially** - Especially for solid archives
3. **Use appropriate buffer sizes** - Larger buffers for network I/O
4. **Dispose streams promptly** - Free resources when done
### Algorithm Selection
1. **Archive API** - Fast for small archives with random access
2. **Reader API** - Efficient for large files or streaming
3. **Solid archives** - Sequential extraction much faster
4. **Compression levels** - Trade-off between speed and size
---
## Testing Guidelines
### Test Coverage
1. **Happy path** - Normal extraction works
2. **Edge cases** - Empty archives, single file, many files
3. **Corrupted data** - Handle gracefully
4. **Error cases** - Missing passwords, unsupported compression
5. **Async operations** - Both sync and async code paths
### Test Archives
- Use `tests/TestArchives/` for test data
- Create format-specific subdirectories
- Include encrypted, corrupted, and edge case archives
- Don't recreate existing archives
### Test Patterns
```csharp
[Fact]
public void Archive_Extraction_Works()
{
// Arrange
var testArchive = Path.Combine(TEST_ARCHIVES_PATH, "test.zip");
// Act
using (var archive = ZipArchive.OpenArchive(testArchive))
{
archive.WriteToDirectory(SCRATCH_FILES_PATH);
}
// Assert
Assert.True(File.Exists(Path.Combine(SCRATCH_FILES_PATH, "file.txt")));
}
```
---
## Related Documentation
- [AGENTS.md](../AGENTS.md) - Development guidelines
- [FORMATS.md](FORMATS.md) - Supported formats

610
docs/ENCODING.md Normal file
View File

@@ -0,0 +1,610 @@
# SharpCompress Character Encoding Guide
This guide explains how SharpCompress handles character encoding for archive entries (filenames, comments, etc.).
## Overview
Most archive formats store filenames and metadata as bytes. SharpCompress must convert these bytes to strings using the appropriate character encoding.
**Common Problem:** Archives created on systems with non-UTF8 encodings (especially Japanese, Chinese systems) appear with corrupted filenames when extracted on systems that assume UTF8.
---
## ArchiveEncoding Class
### Basic Usage
```csharp
using SharpCompress.Common;
using SharpCompress.Readers;
// Configure encoding before opening archive
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding(932) // cp932 for Japanese
}
};
using (var archive = ZipArchive.OpenArchive("japanese.zip", options))
{
foreach (var entry in archive.Entries)
{
Console.WriteLine(entry.Key); // Now shows correct characters
}
}
```
### ArchiveEncoding Properties
| Property | Purpose |
|----------|---------|
| `Default` | Default encoding for filenames (fallback) |
| `CustomDecoder` | Custom decoding function for special cases |
### Setting for Different APIs
**Archive API:**
```csharp
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding { Default = Encoding.GetEncoding(932) }
};
using (var archive = ZipArchive.OpenArchive("file.zip", options))
{
// Use archive with correct encoding
}
```
**Reader API:**
```csharp
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding { Default = Encoding.GetEncoding(932) }
};
using (var stream = File.OpenRead("file.zip"))
using (var reader = ReaderFactory.OpenReader(stream, options))
{
while (reader.MoveToNextEntry())
{
// Filenames decoded correctly
}
}
```
---
## Common Encodings
### Asian Encodings
#### cp932 (Japanese)
```csharp
// Windows-31J, Shift-JIS variant used on Japanese Windows
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding(932)
}
};
using (var archive = ZipArchive.OpenArchive("japanese.zip", options))
{
// Correctly decodes Japanese filenames
}
```
**When to use:**
- Archives from Japanese Windows systems
- Files with Japanese characters in names
#### gb2312 (Simplified Chinese)
```csharp
// Simplified Chinese
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding("gb2312")
}
};
```
#### gbk (Extended Simplified Chinese)
```csharp
// Extended Simplified Chinese (more characters than gb2312)
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding("gbk")
}
};
```
#### big5 (Traditional Chinese)
```csharp
// Traditional Chinese (Taiwan, Hong Kong)
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding("big5")
}
};
```
#### euc-jp (Japanese, Unix)
```csharp
// Extended Unix Code for Japanese
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding("eucjp")
}
};
```
#### euc-kr (Korean)
```csharp
// Extended Unix Code for Korean
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding("euc-kr")
}
};
```
### Western European Encodings
#### iso-8859-1 (Latin-1)
```csharp
// Western European (includes accented characters)
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding("iso-8859-1")
}
};
```
**When to use:**
- Archives from French, German, Spanish systems
- Files with accented characters (é, ñ, ü, etc.)
#### cp1252 (Windows-1252)
```csharp
// Windows Western European
// Very similar to iso-8859-1 but with additional printable characters
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding("cp1252")
}
};
```
**When to use:**
- Archives from older Western European Windows systems
- Files with smart quotes and other Windows-specific characters
#### iso-8859-15 (Latin-9)
```csharp
// Western European with Euro symbol support
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding("iso-8859-15")
}
};
```
### Cyrillic Encodings
#### cp1251 (Windows Cyrillic)
```csharp
// Russian, Serbian, Bulgarian, etc.
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding("cp1251")
}
};
```
#### koi8-r (KOI8 Russian)
```csharp
// Russian (Unix standard)
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding("koi8-r")
}
};
```
### UTF Encodings (Modern)
#### UTF-8 (Default)
```csharp
// Modern standard - usually correct for new archives
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.UTF8
}
};
```
#### UTF-16
```csharp
// Unicode - rarely used in archives
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.Unicode
}
};
```
---
## Encoding Auto-Detection
SharpCompress attempts to auto-detect encoding, but this isn't always reliable:
```csharp
// Auto-detection (default)
using (var archive = ZipArchive.OpenArchive("file.zip")) // Uses UTF8 by default
{
// May show corrupted characters if archive uses different encoding
}
// Explicit encoding (more reliable)
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding { Default = Encoding.GetEncoding(932) }
};
using (var archive = ZipArchive.OpenArchive("file.zip", options))
{
// Correct characters displayed
}
```
### When Manual Override is Needed
| Situation | Solution |
|-----------|----------|
| Archive shows corrupted characters | Specify the encoding explicitly |
| Archives from specific region | Use that region's encoding |
| Mixed encodings in archive | Use CustomDecoder |
| Testing with international files | Try different encodings |
---
## Custom Decoder
For complex scenarios where a single encoding isn't sufficient:
### Basic Custom Decoder
```csharp
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
CustomDecoder = (data, offset, length) =>
{
// Custom decoding logic
var bytes = new byte[length];
Array.Copy(data, offset, bytes, 0, length);
// Try UTF8 first
try
{
return Encoding.UTF8.GetString(bytes);
}
catch
{
// Fallback to cp932 if UTF8 fails
return Encoding.GetEncoding(932).GetString(bytes);
}
}
}
};
using (var archive = ZipArchive.OpenArchive("mixed.zip", options))
{
foreach (var entry in archive.Entries)
{
Console.WriteLine(entry.Key); // Uses custom decoder
}
}
```
### Advanced: Detect Encoding by Content
```csharp
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
CustomDecoder = DetectAndDecode
}
};
private static string DetectAndDecode(byte[] data, int offset, int length)
{
var bytes = new byte[length];
Array.Copy(data, offset, bytes, 0, length);
// Try UTF8 (most modern archives)
try
{
var str = Encoding.UTF8.GetString(bytes);
// Verify it decoded correctly (no replacement characters)
if (!str.Contains('\uFFFD'))
return str;
}
catch { }
// Try cp932 (Japanese)
try
{
var str = Encoding.GetEncoding(932).GetString(bytes);
if (!str.Contains('\uFFFD'))
return str;
}
catch { }
// Fallback to iso-8859-1 (always succeeds)
return Encoding.GetEncoding("iso-8859-1").GetString(bytes);
}
```
---
## Code Examples
### Extract Archive with Japanese Filenames
```csharp
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding(932) // cp932
}
};
using (var archive = ZipArchive.OpenArchive("japanese_files.zip", options))
{
archive.WriteToDirectory(@"C:\output", new ExtractionOptions
{
ExtractFullPath = true,
Overwrite = true
});
}
// Files extracted with correct Japanese names
```
### Extract Archive with Western European Filenames
```csharp
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding("iso-8859-1")
}
};
using (var archive = ZipArchive.OpenArchive("french_files.zip", options))
{
archive.WriteToDirectory(@"C:\output");
}
// Accented characters (é, è, ê, etc.) display correctly
```
### Extract Archive with Chinese Filenames
```csharp
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding("gbk") // Simplified Chinese
}
};
using (var archive = ZipArchive.OpenArchive("chinese_files.zip", options))
{
archive.WriteToDirectory(@"C:\output");
}
```
### Extract Archive with Russian Filenames
```csharp
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding("cp1251") // Windows Cyrillic
}
};
using (var archive = ZipArchive.OpenArchive("russian_files.zip", options))
{
archive.WriteToDirectory(@"C:\output");
}
```
### Reader API with Encoding
```csharp
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding(932)
}
};
using (var stream = File.OpenRead("japanese.zip"))
using (var reader = ReaderFactory.OpenReader(stream, options))
{
while (reader.MoveToNextEntry())
{
if (!reader.Entry.IsDirectory)
{
Console.WriteLine(reader.Entry.Key); // Correct characters
reader.WriteEntryToDirectory(@"C:\output");
}
}
}
```
---
## Creating Archives with Correct Encoding
When creating archives, SharpCompress uses UTF8 by default (recommended):
```csharp
// Create with UTF8 (default, recommended)
using (var archive = ZipArchive.CreateArchive())
{
archive.AddAllFromDirectory(@"D:\my_files");
archive.SaveTo("output.zip", CompressionType.Deflate);
// Archives created with UTF8 encoding
}
```
If you need to create archives for systems that expect specific encodings:
```csharp
// Note: SharpCompress Writer API uses UTF8 for encoding
// To create archives with other encodings, consider:
// 1. Let users on those systems create archives
// 2. Use system tools (7-Zip, WinRAR) with desired encoding
// 3. Post-process archives if absolutely necessary
// For now, recommend modern UTF8-based archives
```
---
## Troubleshooting Encoding Issues
### Filenames Show Question Marks (?)
```
✗ Wrong encoding detected
test文件.txt → test???.txt
```
**Solution:** Specify correct encoding explicitly
```csharp
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
Default = Encoding.GetEncoding("gbk") // Try different encodings
}
};
```
### Filenames Show Replacement Character (￿)
```
✗ Invalid bytes for selected encoding
café.txt → caf￿.txt
```
**Solution:**
1. Try a different encoding (see Common Encodings table)
2. Use CustomDecoder with fallback encoding
3. Archive might be corrupted
### Mixed Encodings in Single Archive
```csharp
// Use CustomDecoder to handle mixed encodings
var options = new ReaderOptions
{
ArchiveEncoding = new ArchiveEncoding
{
CustomDecoder = (data, offset, length) =>
{
// Try multiple encodings in priority order
var bytes = new byte[length];
Array.Copy(data, offset, bytes, 0, length);
foreach (var encoding in new[]
{
Encoding.UTF8,
Encoding.GetEncoding(932),
Encoding.GetEncoding("iso-8859-1")
})
{
try
{
var str = encoding.GetString(bytes);
if (!str.Contains('\uFFFD'))
return str;
}
catch { }
}
// Final fallback
return Encoding.GetEncoding("iso-8859-1").GetString(bytes);
}
}
};
```
---
## Encoding Reference Table
| Encoding | Code | Use Case |
|----------|------|----------|
| UTF-8 | (default) | Modern archives, recommended |
| cp932 | 932 | Japanese Windows |
| gb2312 | "gb2312" | Simplified Chinese |
| gbk | "gbk" | Extended Simplified Chinese |
| big5 | "big5" | Traditional Chinese |
| iso-8859-1 | "iso-8859-1" | Western European |
| cp1252 | "cp1252" | Windows Western European |
| cp1251 | "cp1251" | Russian/Cyrillic |
| euc-jp | "euc-jp" | Japanese Unix |
| euc-kr | "euc-kr" | Korean |
---
## Best Practices
1. **Use UTF-8 for new archives** - Most modern systems support it
2. **Ask the archive creator** - When receiving archives with corrupted names
3. **Provide encoding options** - If your app handles user archives
4. **Document your assumption** - Tell users what encoding you're using
5. **Test with international files** - Before releasing production code
---
## Related Documentation
- [USAGE.md](USAGE.md#extract-zip-which-has-non-utf8-encoded-filenamycp932) - Usage examples

474
docs/PERFORMANCE.md Normal file
View File

@@ -0,0 +1,474 @@
# SharpCompress Performance Guide
This guide helps you optimize SharpCompress for performance in various scenarios.
## API Selection Guide
### Archive API vs Reader API
Choose the right API based on your use case:
| Aspect | Archive API | Reader API |
|--------|------------|-----------|
| **Stream Type** | Seekable only | Non-seekable OK |
| **Memory Usage** | All entries in memory | One entry at a time |
| **Random Access** | ✓ Yes | ✗ No |
| **Best For** | Small-to-medium archives | Large or streaming data |
| **Performance** | Fast for random access | Better for large files |
### Archive API (Fast for Random Access)
```csharp
// Use when:
// - Archive fits in memory
// - You need random access to entries
// - Stream is seekable (file, MemoryStream)
using (var archive = ZipArchive.OpenArchive("archive.zip"))
{
// Random access - all entries available
var specific = archive.Entries.FirstOrDefault(e => e.Key == "file.txt");
if (specific != null)
{
specific.WriteToFile(@"C:\output\file.txt");
}
}
```
**Performance Characteristics:**
- ✓ Instant entry lookup
- ✓ Parallel extraction possible
- ✗ Entire archive in memory
- ✗ Can't process while downloading
### Reader API (Best for Large Files)
```csharp
// Use when:
// - Processing large archives (>100 MB)
// - Streaming from network/pipe
// - Memory is constrained
// - Forward-only processing is acceptable
using (var stream = File.OpenRead("large.zip"))
using (var reader = ReaderFactory.OpenReader(stream))
{
while (reader.MoveToNextEntry())
{
// Process one entry at a time
reader.WriteEntryToDirectory(@"C:\output");
}
}
```
**Performance Characteristics:**
- ✓ Minimal memory footprint
- ✓ Works with non-seekable streams
- ✓ Can process while downloading
- ✗ Forward-only (no random access)
- ✗ Entry lookup requires iteration
---
## Buffer Sizing
### Understanding Buffers
SharpCompress uses internal buffers for reading compressed data. Buffer size affects:
- **Speed:** Larger buffers = fewer I/O operations = faster
- **Memory:** Larger buffers = higher memory usage
### Recommended Buffer Sizes
| Scenario | Size | Notes |
|----------|------|-------|
| Embedded/IoT devices | 4-8 KB | Minimal memory usage |
| Memory-constrained | 16-32 KB | Conservative default |
| Standard use (default) | 64 KB | Recommended default |
| Large file streaming | 256 KB | Better throughput |
| High-speed SSD | 512 KB - 1 MB | Maximum throughput |
### How Buffer Size Affects Performance
```csharp
// SharpCompress manages buffers internally
// You can't directly set buffer size, but you can:
// 1. Use Stream.CopyTo with explicit buffer size
using (var entryStream = reader.OpenEntryStream())
using (var fileStream = File.Create(@"C:\output\file.txt"))
{
// 64 KB buffer (default)
entryStream.CopyTo(fileStream);
// Or specify larger buffer for faster copy
entryStream.CopyTo(fileStream, bufferSize: 262144); // 256 KB
}
// 2. Use custom buffer for writing
using (var entryStream = reader.OpenEntryStream())
using (var fileStream = File.Create(@"C:\output\file.txt"))
{
byte[] buffer = new byte[262144]; // 256 KB
int bytesRead;
while ((bytesRead = entryStream.Read(buffer, 0, buffer.Length)) > 0)
{
fileStream.Write(buffer, 0, bytesRead);
}
}
```
---
## Streaming Large Files
### Non-Seekable Stream Patterns
For processing archives from downloads or pipes:
```csharp
// Download stream (non-seekable)
using (var httpStream = await httpClient.GetStreamAsync(url))
using (var reader = ReaderFactory.OpenReader(httpStream))
{
// Process entries as they arrive
while (reader.MoveToNextEntry())
{
if (!reader.Entry.IsDirectory)
{
reader.WriteEntryToDirectory(@"C:\output");
}
}
}
```
**Performance Tips:**
- Don't try to buffer the entire stream
- Process entries immediately
- Use async APIs for better responsiveness
### Download-Then-Extract vs Streaming
Choose based on your constraints:
| Approach | When to Use |
|----------|------------|
| **Download then extract** | Moderate size, need random access |
| **Stream during download** | Large files, bandwidth limited, memory constrained |
```csharp
// Download then extract (requires disk space)
var archivePath = await DownloadFile(url, @"C:\temp\archive.zip");
using (var archive = ZipArchive.OpenArchive(archivePath))
{
archive.WriteToDirectory(@"C:\output");
}
// Stream during download (on-the-fly extraction)
using (var httpStream = await httpClient.GetStreamAsync(url))
using (var reader = ReaderFactory.OpenReader(httpStream))
{
while (reader.MoveToNextEntry())
{
reader.WriteEntryToDirectory(@"C:\output");
}
}
```
---
## Solid Archive Optimization
### Why Solid Archives Are Slow
Solid archives (Rar, 7Zip) group files together in a single compressed stream:
```
Solid Archive Layout:
[Header] [Compressed Stream] [Footer]
├─ File1 compressed data
├─ File2 compressed data
├─ File3 compressed data
└─ File4 compressed data
```
Extracting File3 requires decompressing File1 and File2 first.
### Sequential vs Random Extraction
**Random Extraction (Slow):**
```csharp
using (var archive = RarArchive.OpenArchive("solid.rar"))
{
foreach (var entry in archive.Entries)
{
entry.WriteToFile(@"C:\output\" + entry.Key); // ✗ Slow!
// Each entry triggers full decompression from start
}
}
```
**Sequential Extraction (Fast):**
```csharp
using (var archive = RarArchive.OpenArchive("solid.rar"))
{
// Method 1: Use WriteToDirectory (recommended)
archive.WriteToDirectory(@"C:\output", new ExtractionOptions
{
ExtractFullPath = true,
Overwrite = true
});
// Method 2: Use ExtractAllEntries
archive.ExtractAllEntries();
// Method 3: Use Reader API (also sequential)
using (var reader = RarReader.Open(File.OpenRead("solid.rar")))
{
while (reader.MoveToNextEntry())
{
reader.WriteEntryToDirectory(@"C:\output");
}
}
}
```
**Performance Impact:**
- Random extraction: O(n²) - very slow for many files
- Sequential extraction: O(n) - 10-100x faster
### Best Practices for Solid Archives
1. **Always extract sequentially** when possible
2. **Use Reader API** for large solid archives
3. **Process entries in order** from the archive
4. **Consider using 7Zip command-line** for scripted extractions
---
## Compression Level Trade-offs
### Deflate/GZip Levels
```csharp
// Level 1 = Fastest, largest size
// Level 6 = Default (balanced)
// Level 9 = Slowest, best compression
// Write with different compression levels
using (var archive = ZipArchive.CreateArchive())
{
archive.AddAllFromDirectory(@"D:\data");
// Fast compression (level 1)
archive.SaveTo("fast.zip", new WriterOptions(CompressionType.Deflate)
{
CompressionLevel = 1
});
// Default compression (level 6)
archive.SaveTo("default.zip", CompressionType.Deflate);
// Best compression (level 9)
archive.SaveTo("best.zip", new WriterOptions(CompressionType.Deflate)
{
CompressionLevel = 9
});
}
```
**Speed vs Size:**
| Level | Speed | Size | Use Case |
|-------|-------|------|----------|
| 1 | 10x | 90% | Network, streaming |
| 6 | 1x | 75% | Default (good balance) |
| 9 | 0.1x | 65% | Archival, static storage |
### BZip2 Block Size
```csharp
// BZip2 block size affects memory and compression
// 100K to 900K (default 900K)
// Smaller block size = lower memory, faster
// Larger block size = better compression, slower
using (var archive = TarArchive.CreateArchive())
{
archive.AddAllFromDirectory(@"D:\data");
// These are preset in WriterOptions via CompressionLevel
archive.SaveTo("archive.tar.bz2", CompressionType.BZip2);
}
```
### LZMA Settings
LZMA compression is very powerful but memory-intensive:
```csharp
// LZMA (7Zip, .tar.lzma):
// - Dictionary size: 16 KB to 1 GB (default 32 MB)
// - Faster preset: smaller dictionary
// - Better compression: larger dictionary
// Preset via CompressionType
using (var archive = TarArchive.CreateArchive())
{
archive.AddAllFromDirectory(@"D:\data");
archive.SaveTo("archive.tar.xz", CompressionType.LZMA); // Default settings
}
```
---
## Async Performance
### When Async Helps
Async is beneficial when:
- **Long I/O operations** (network, slow disks)
- **UI responsiveness** needed (Windows Forms, WPF, Blazor)
- **Server applications** (ASP.NET, multiple concurrent operations)
```csharp
// Async extraction (non-blocking)
using (var archive = ZipArchive.OpenArchive("archive.zip"))
{
await archive.WriteToDirectoryAsync(
@"C:\output",
new ExtractionOptions { ExtractFullPath = true, Overwrite = true },
cancellationToken
);
}
// Thread can handle other work while I/O happens
```
### When Async Doesn't Help
Async doesn't improve performance for:
- **CPU-bound operations** (already fast)
- **Local SSD I/O** (I/O is fast enough)
- **Single-threaded scenarios** (no parallelism benefit)
```csharp
// Sync extraction (simpler, same performance on fast I/O)
using (var archive = ZipArchive.OpenArchive("archive.zip"))
{
archive.WriteToDirectory(
@"C:\output",
new ExtractionOptions { ExtractFullPath = true, Overwrite = true }
);
}
// Simple and fast - no async needed
```
### Cancellation Pattern
```csharp
var cts = new CancellationTokenSource();
// Cancel after 5 minutes
cts.CancelAfter(TimeSpan.FromMinutes(5));
try
{
using (var archive = ZipArchive.OpenArchive("archive.zip"))
{
await archive.WriteToDirectoryAsync(
@"C:\output",
new ExtractionOptions { ExtractFullPath = true, Overwrite = true },
cts.Token
);
}
}
catch (OperationCanceledException)
{
Console.WriteLine("Extraction cancelled");
// Clean up partial extraction if needed
}
```
---
## Practical Performance Tips
### 1. Choose the Right API
| Scenario | API | Why |
|----------|-----|-----|
| Small archives | Archive | Faster random access |
| Large archives | Reader | Lower memory |
| Streaming | Reader | Works on non-seekable streams |
| Download streams | Reader | Async extraction while downloading |
### 2. Batch Operations
```csharp
// ✗ Slow - opens each archive separately
foreach (var file in files)
{
using (var archive = ZipArchive.OpenArchive("archive.zip"))
{
archive.WriteToDirectory(@"C:\output");
}
}
// ✓ Better - process multiple entries at once
using (var archive = ZipArchive.OpenArchive("archive.zip"))
{
archive.WriteToDirectory(@"C:\output");
}
```
### 3. Profile Your Code
```csharp
var sw = Stopwatch.StartNew();
using (var archive = ZipArchive.OpenArchive("large.zip"))
{
archive.WriteToDirectory(@"C:\output");
}
sw.Stop();
Console.WriteLine($"Extraction took {sw.ElapsedMilliseconds}ms");
// Measure memory before/after
var beforeMem = GC.GetTotalMemory(true);
// ... do work ...
var afterMem = GC.GetTotalMemory(true);
Console.WriteLine($"Memory used: {(afterMem - beforeMem) / 1024 / 1024}MB");
```
---
## Troubleshooting Performance
### Extraction is Slow
1. **Check if solid archive** → Use sequential extraction
2. **Check API** → Reader API might be faster for large files
3. **Check compression level** → Higher levels are slower to decompress
4. **Check I/O** → Network drives are much slower than SSD
5. **Check buffer size** → May need larger buffers for network
### High Memory Usage
1. **Use Reader API** instead of Archive API
2. **Process entries immediately** rather than buffering
3. **Reduce compression level** if writing
4. **Check for memory leaks** in your code
### CPU Usage at 100%
1. **Normal for compression** - especially with high compression levels
2. **Consider lower level** for faster processing
3. **Reduce parallelism** if processing multiple archives
4. **Check if awaiting properly** in async code
---
## Related Documentation
- [PERFORMANCE.md](USAGE.md) - Usage examples with performance considerations
- [FORMATS.md](FORMATS.md) - Format-specific performance notes

View File

@@ -1,6 +1,6 @@
# SharpCompress Usage
## Async/Await Support
## Async/Await Support (Beta)
SharpCompress now provides full async/await support for all I/O operations. All `Read`, `Write`, and extraction operations have async equivalents ending in `Async` that accept an optional `CancellationToken`. This enables better performance and scalability for I/O-bound operations.
@@ -13,7 +13,7 @@ SharpCompress now provides full async/await support for all I/O operations. All
See [Async Examples](#async-examples) section below for usage patterns.
## Stream Rules (changed with 0.21)
## Stream Rules
When dealing with Streams, the rule should be that you don't close a stream you didn't create. This, in effect, should mean you should always put a Stream in a using block to dispose it.
@@ -48,7 +48,7 @@ Also, look over the tests for more thorough [examples](https://github.com/adamha
### Create Zip Archive from multiple files
```C#
using(var archive = ZipArchive.Create())
using(var archive = ZipArchive.CreateArchive())
{
archive.AddEntry("file01.txt", "C:\\file01.txt");
archive.AddEntry("file02.txt", "C:\\file02.txt");
@@ -61,7 +61,7 @@ using(var archive = ZipArchive.Create())
### Create Zip Archive from all files in a directory to a file
```C#
using (var archive = ZipArchive.Create())
using (var archive = ZipArchive.CreateArchive())
{
archive.AddAllFromDirectory("D:\\temp");
archive.SaveTo("C:\\temp.zip", CompressionType.Deflate);
@@ -72,7 +72,7 @@ using (var archive = ZipArchive.Create())
```C#
var memoryStream = new MemoryStream();
using (var archive = ZipArchive.Create())
using (var archive = ZipArchive.CreateArchive())
{
archive.AddAllFromDirectory("D:\\temp");
archive.SaveTo(memoryStream, new WriterOptions(CompressionType.Deflate)
@@ -90,7 +90,7 @@ Note: Extracting a solid rar or 7z file needs to be done in sequential order to
`ExtractAllEntries` is primarily intended for solid archives (like solid Rar) or 7Zip archives, where sequential extraction provides the best performance. For general/simple extraction with any supported archive type, use `archive.WriteToDirectory()` instead.
```C#
using (var archive = RarArchive.Open("Test.rar"))
using (var archive = RarArchive.OpenArchive("Test.rar"))
{
// Simple extraction with RarArchive; this WriteToDirectory pattern works for all archive types
archive.WriteToDirectory(@"D:\temp", new ExtractionOptions()
@@ -104,7 +104,7 @@ using (var archive = RarArchive.Open("Test.rar"))
### Iterate over all files from a Rar file using RarArchive
```C#
using (var archive = RarArchive.Open("Test.rar"))
using (var archive = RarArchive.OpenArchive("Test.rar"))
{
foreach (var entry in archive.Entries.Where(entry => !entry.IsDirectory))
{
@@ -126,7 +126,7 @@ var progress = new Progress<ProgressReport>(report =>
Console.WriteLine($"Extracting {report.EntryPath}: {report.PercentComplete}%");
});
using (var archive = RarArchive.Open("archive.rar", new ReaderOptions { Progress = progress })) // Must be solid Rar or 7Zip
using (var archive = RarArchive.OpenArchive("archive.rar", new ReaderOptions { Progress = progress })) // Must be solid Rar or 7Zip
{
archive.WriteToDirectory(@"D:\output", new ExtractionOptions()
{
@@ -140,7 +140,7 @@ using (var archive = RarArchive.Open("archive.rar", new ReaderOptions { Progress
```C#
using (Stream stream = File.OpenRead("Tar.tar.bz2"))
using (var reader = ReaderFactory.Open(stream))
using (var reader = ReaderFactory.OpenReader(stream))
{
while (reader.MoveToNextEntry())
{
@@ -161,7 +161,7 @@ using (var reader = ReaderFactory.Open(stream))
```C#
using (Stream stream = File.OpenRead("Tar.tar.bz2"))
using (var reader = ReaderFactory.Open(stream))
using (var reader = ReaderFactory.OpenReader(stream))
{
while (reader.MoveToNextEntry())
{
@@ -180,7 +180,7 @@ using (var reader = ReaderFactory.Open(stream))
```C#
using (Stream stream = File.OpenWrite("C:\\temp.tgz"))
using (var writer = WriterFactory.Open(stream, ArchiveType.Tar, new WriterOptions(CompressionType.GZip)
using (var writer = WriterFactory.OpenWriter(stream, ArchiveType.Tar, new WriterOptions(CompressionType.GZip)
{
LeaveOpenStream = true
}))
@@ -199,7 +199,7 @@ opts.ArchiveEncoding.CustomDecoder = (data, x, y) =>
{
return encoding.GetString(data);
};
var tr = SharpCompress.Archives.Zip.ZipArchive.Open("test.zip", opts);
var tr = SharpCompress.Archives.Zip.ZipArchive.OpenArchive("test.zip", opts);
foreach(var entry in tr.Entries)
{
Console.WriteLine($"{entry.Key}");
@@ -213,7 +213,7 @@ foreach(var entry in tr.Entries)
**Extract single entry asynchronously:**
```C#
using (Stream stream = File.OpenRead("archive.zip"))
using (var reader = ReaderFactory.Open(stream))
using (var reader = ReaderFactory.OpenReader(stream))
{
while (reader.MoveToNextEntry())
{
@@ -234,7 +234,7 @@ using (var reader = ReaderFactory.Open(stream))
**Extract all entries asynchronously:**
```C#
using (Stream stream = File.OpenRead("archive.tar.gz"))
using (var reader = ReaderFactory.Open(stream))
using (var reader = ReaderFactory.OpenReader(stream))
{
await reader.WriteAllToDirectoryAsync(
@"D:\temp",
@@ -250,7 +250,7 @@ using (var reader = ReaderFactory.Open(stream))
**Open and process entry stream asynchronously:**
```C#
using (var archive = ZipArchive.Open("archive.zip"))
using (var archive = ZipArchive.OpenArchive("archive.zip"))
{
foreach (var entry in archive.Entries.Where(e => !e.IsDirectory))
{
@@ -268,7 +268,7 @@ using (var archive = ZipArchive.Open("archive.zip"))
**Write single file asynchronously:**
```C#
using (Stream archiveStream = File.OpenWrite("output.zip"))
using (var writer = WriterFactory.Open(archiveStream, ArchiveType.Zip, CompressionType.Deflate))
using (var writer = WriterFactory.OpenWriter(archiveStream, ArchiveType.Zip, CompressionType.Deflate))
{
using (Stream fileStream = File.OpenRead("input.txt"))
{
@@ -280,7 +280,7 @@ using (var writer = WriterFactory.Open(archiveStream, ArchiveType.Zip, Compressi
**Write entire directory asynchronously:**
```C#
using (Stream stream = File.OpenWrite("backup.tar.gz"))
using (var writer = WriterFactory.Open(stream, ArchiveType.Tar, new WriterOptions(CompressionType.GZip)))
using (var writer = WriterFactory.OpenWriter(stream, ArchiveType.Tar, new WriterOptions(CompressionType.GZip)))
{
await writer.WriteAllAsync(
@"D:\files",
@@ -299,7 +299,7 @@ var cts = new CancellationTokenSource();
cts.CancelAfter(TimeSpan.FromMinutes(5));
using (Stream stream = File.OpenWrite("archive.zip"))
using (var writer = WriterFactory.Open(stream, ArchiveType.Zip, CompressionType.Deflate))
using (var writer = WriterFactory.OpenWriter(stream, ArchiveType.Zip, CompressionType.Deflate))
{
try
{
@@ -316,7 +316,7 @@ using (var writer = WriterFactory.Open(stream, ArchiveType.Zip, CompressionType.
**Extract from archive asynchronously:**
```C#
using (var archive = ZipArchive.Open("archive.zip"))
using (var archive = ZipArchive.OpenArchive("archive.zip"))
{
// Simple async extraction - works for all archive types
await archive.WriteToDirectoryAsync(

View File

@@ -1,7 +1,7 @@
// Copyright (c) Six Labors.
// Licensed under the Apache License, Version 2.0.
#if !NETSTANDARD2_0 && !NETSTANDARD2_1 && !NETFRAMEWORK
#if !LEGACY_DOTNET
#define SUPPORTS_RUNTIME_INTRINSICS
#define SUPPORTS_HOTPATH
#endif

View File

@@ -1,14 +1,13 @@
using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Threading.Tasks;
using SharpCompress.Common;
using SharpCompress.IO;
using SharpCompress.Readers;
namespace SharpCompress.Archives;
public abstract class AbstractArchive<TEntry, TVolume> : IArchive
public abstract class AbstractArchive<TEntry, TVolume> : IArchive, IAsyncArchive
where TEntry : IArchiveEntry
where TVolume : IVolume
{
@@ -26,6 +25,12 @@ public abstract class AbstractArchive<TEntry, TVolume> : IArchive
_sourceStream = sourceStream;
_lazyVolumes = new LazyReadOnlyCollection<TVolume>(LoadVolumes(_sourceStream));
_lazyEntries = new LazyReadOnlyCollection<TEntry>(LoadEntries(Volumes));
_lazyVolumesAsync = new LazyAsyncReadOnlyCollection<TVolume>(
LoadVolumesAsync(_sourceStream)
);
_lazyEntriesAsync = new LazyAsyncReadOnlyCollection<TEntry>(
LoadEntriesAsync(_lazyVolumesAsync)
);
}
internal AbstractArchive(ArchiveType type)
@@ -34,19 +39,16 @@ public abstract class AbstractArchive<TEntry, TVolume> : IArchive
ReaderOptions = new();
_lazyVolumes = new LazyReadOnlyCollection<TVolume>(Enumerable.Empty<TVolume>());
_lazyEntries = new LazyReadOnlyCollection<TEntry>(Enumerable.Empty<TEntry>());
_lazyVolumesAsync = new LazyAsyncReadOnlyCollection<TVolume>(
AsyncEnumerableEx.Empty<TVolume>()
);
_lazyEntriesAsync = new LazyAsyncReadOnlyCollection<TEntry>(
AsyncEnumerableEx.Empty<TEntry>()
);
}
public ArchiveType Type { get; }
private static Stream CheckStreams(Stream stream)
{
if (!stream.CanSeek || !stream.CanRead)
{
throw new ArchiveException("Archive streams must be Readable and Seekable");
}
return stream;
}
/// <summary>
/// Returns an ReadOnlyCollection of all the RarArchiveEntries across the one or many parts of the RarArchive.
/// </summary>
@@ -66,12 +68,25 @@ public abstract class AbstractArchive<TEntry, TVolume> : IArchive
/// <summary>
/// The total size of the files as uncompressed in the archive.
/// </summary>
public virtual long TotalUncompressSize =>
public virtual long TotalUncompressedSize =>
Entries.Aggregate(0L, (total, cf) => total + cf.Size);
protected abstract IEnumerable<TVolume> LoadVolumes(SourceStream sourceStream);
protected abstract IEnumerable<TEntry> LoadEntries(IEnumerable<TVolume> volumes);
protected virtual IAsyncEnumerable<TVolume> LoadVolumesAsync(SourceStream sourceStream) =>
LoadVolumes(sourceStream).ToAsyncEnumerable();
protected virtual async IAsyncEnumerable<TEntry> LoadEntriesAsync(
IAsyncEnumerable<TVolume> volumes
)
{
foreach (var item in LoadEntries(await volumes.ToListAsync()))
{
yield return item;
}
}
IEnumerable<IArchiveEntry> IArchive.Entries => Entries.Cast<IArchiveEntry>();
IEnumerable<IVolume> IArchive.Volumes => _lazyVolumes.Cast<IVolume>();
@@ -118,6 +133,7 @@ public abstract class AbstractArchive<TEntry, TVolume> : IArchive
}
protected abstract IReader CreateReaderForSolidExtraction();
protected abstract ValueTask<IAsyncReader> CreateReaderForSolidExtractionAsync();
/// <summary>
/// Archive is SOLID (this means the Archive saved bytes by reusing information which helps for archives containing many small files).
@@ -129,6 +145,19 @@ public abstract class AbstractArchive<TEntry, TVolume> : IArchive
/// </summary>
public virtual bool IsEncrypted => false;
/// <summary>
/// Returns whether multi-threaded extraction is supported for this archive.
/// Multi-threading is supported when:
/// 1. The archive is opened from a FileInfo or file path (not a stream)
/// 2. Multi-threading is explicitly enabled in ReaderOptions
/// 3. The archive is not SOLID (SOLID archives should use sequential extraction)
/// </summary>
public virtual bool SupportsMultiThreadedExtraction =>
_sourceStream is not null
&& _sourceStream.IsFileMode
&& ReaderOptions.EnableMultiThreadedExtraction
&& !IsSolid;
/// <summary>
/// The archive can find all the parts of the archive needed to fully extract the archive. This forces the parsing of the entire archive.
/// </summary>
@@ -140,4 +169,85 @@ public abstract class AbstractArchive<TEntry, TVolume> : IArchive
return Entries.All(x => x.IsComplete);
}
}
#region Async Support
private readonly LazyAsyncReadOnlyCollection<TVolume> _lazyVolumesAsync;
private readonly LazyAsyncReadOnlyCollection<TEntry> _lazyEntriesAsync;
public virtual async ValueTask DisposeAsync()
{
if (!_disposed)
{
await foreach (var v in _lazyVolumesAsync)
{
v.Dispose();
}
foreach (var v in _lazyEntriesAsync.GetLoaded().Cast<Entry>())
{
v.Close();
}
_sourceStream?.Dispose();
_disposed = true;
}
}
private async ValueTask EnsureEntriesLoadedAsync()
{
await _lazyEntriesAsync.EnsureFullyLoaded();
await _lazyVolumesAsync.EnsureFullyLoaded();
}
public virtual IAsyncEnumerable<TEntry> EntriesAsync => _lazyEntriesAsync;
private async IAsyncEnumerable<IArchiveEntry> EntriesAsyncCast()
{
await foreach (var entry in EntriesAsync)
{
yield return entry;
}
}
IAsyncEnumerable<IArchiveEntry> IAsyncArchive.EntriesAsync => EntriesAsyncCast();
private async IAsyncEnumerable<IVolume> VolumesAsyncCast()
{
await foreach (var volume in VolumesAsync)
{
yield return volume;
}
}
public IAsyncEnumerable<IVolume> VolumesAsync => VolumesAsyncCast();
public async ValueTask<IAsyncReader> ExtractAllEntriesAsync()
{
if (!IsSolid && Type != ArchiveType.SevenZip)
{
throw new SharpCompressException(
"ExtractAllEntries can only be used on solid archives or 7Zip archives (which require random access)."
);
}
await EnsureEntriesLoadedAsync();
return await CreateReaderForSolidExtractionAsync();
}
public virtual ValueTask<bool> IsSolidAsync() => new(false);
public async ValueTask<bool> IsCompleteAsync()
{
await EnsureEntriesLoadedAsync();
return await EntriesAsync.AllAsync(x => x.IsComplete);
}
public async ValueTask<long> TotalSizeAsync() =>
await EntriesAsync.AggregateAsync(0L, (total, cf) => total + cf.CompressedSize);
public async ValueTask<long> TotalUncompressedSizeAsync() =>
await EntriesAsync.AggregateAsync(0L, (total, cf) => total + cf.Size);
public ValueTask<bool> IsEncryptedAsync() => new(IsEncrypted);
#endregion
}

View File

@@ -12,7 +12,8 @@ namespace SharpCompress.Archives;
public abstract class AbstractWritableArchive<TEntry, TVolume>
: AbstractArchive<TEntry, TVolume>,
IWritableArchive
IWritableArchive,
IWritableAsyncArchive
where TEntry : IArchiveEntry
where TVolume : IVolume
{
@@ -83,12 +84,12 @@ public abstract class AbstractWritableArchive<TEntry, TVolume>
}
}
void IWritableArchive.RemoveEntry(IArchiveEntry entry) => RemoveEntry((TEntry)entry);
void IWritableArchiveCommon.RemoveEntry(IArchiveEntry entry) => RemoveEntry((TEntry)entry);
public TEntry AddEntry(string key, Stream source, long size = 0, DateTime? modified = null) =>
AddEntry(key, source, false, size, modified);
IArchiveEntry IWritableArchive.AddEntry(
IArchiveEntry IWritableArchiveCommon.AddEntry(
string key,
Stream source,
bool closeStream,
@@ -96,7 +97,7 @@ public abstract class AbstractWritableArchive<TEntry, TVolume>
DateTime? modified
) => AddEntry(key, source, closeStream, size, modified);
IArchiveEntry IWritableArchive.AddDirectoryEntry(string key, DateTime? modified) =>
IArchiveEntry IWritableArchiveCommon.AddDirectoryEntry(string key, DateTime? modified) =>
AddDirectoryEntry(key, modified);
public TEntry AddEntry(
@@ -162,7 +163,7 @@ public abstract class AbstractWritableArchive<TEntry, TVolume>
SaveTo(stream, options, OldEntries, newEntries);
}
public async Task SaveToAsync(
public async ValueTask SaveToAsync(
Stream stream,
WriterOptions options,
CancellationToken cancellationToken = default
@@ -208,7 +209,7 @@ public abstract class AbstractWritableArchive<TEntry, TVolume>
IEnumerable<TEntry> newEntries
);
protected abstract Task SaveToAsync(
protected abstract ValueTask SaveToAsync(
Stream stream,
WriterOptions options,
IEnumerable<TEntry> oldEntries,

View File

@@ -2,6 +2,8 @@ using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common;
using SharpCompress.Factories;
using SharpCompress.IO;
@@ -11,20 +13,26 @@ namespace SharpCompress.Archives;
public static class ArchiveFactory
{
/// <summary>
/// Opens an Archive for random access
/// </summary>
/// <param name="stream"></param>
/// <param name="readerOptions"></param>
/// <returns></returns>
public static IArchive Open(Stream stream, ReaderOptions? readerOptions = null)
public static IArchive OpenArchive(Stream stream, ReaderOptions? readerOptions = null)
{
readerOptions ??= new ReaderOptions();
stream = SharpCompressStream.Create(stream, bufferSize: readerOptions.BufferSize);
return FindFactory<IArchiveFactory>(stream).Open(stream, readerOptions);
return FindFactory<IArchiveFactory>(stream).OpenArchive(stream, readerOptions);
}
public static IWritableArchive Create(ArchiveType type)
public static async ValueTask<IAsyncArchive> OpenAsyncArchive(
Stream stream,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
readerOptions ??= new ReaderOptions();
stream = SharpCompressStream.Create(stream, bufferSize: readerOptions.BufferSize);
var factory = await FindFactoryAsync<IArchiveFactory>(stream, cancellationToken);
return factory.OpenAsyncArchive(stream, readerOptions);
}
public static IWritableArchive CreateArchive(ArchiveType type)
{
var factory = Factory
.Factories.OfType<IWriteableArchiveFactory>()
@@ -32,41 +40,51 @@ public static class ArchiveFactory
if (factory != null)
{
return factory.CreateWriteableArchive();
return factory.CreateArchive();
}
throw new NotSupportedException("Cannot create Archives of type: " + type);
}
/// <summary>
/// Constructor expects a filepath to an existing file.
/// </summary>
/// <param name="filePath"></param>
/// <param name="options"></param>
public static IArchive Open(string filePath, ReaderOptions? options = null)
public static IArchive OpenArchive(string filePath, ReaderOptions? options = null)
{
filePath.NotNullOrEmpty(nameof(filePath));
return Open(new FileInfo(filePath), options);
return OpenArchive(new FileInfo(filePath), options);
}
/// <summary>
/// Constructor with a FileInfo object to an existing file.
/// </summary>
/// <param name="fileInfo"></param>
/// <param name="options"></param>
public static IArchive Open(FileInfo fileInfo, ReaderOptions? options = null)
public static ValueTask<IAsyncArchive> OpenAsyncArchive(
string filePath,
ReaderOptions? options = null,
CancellationToken cancellationToken = default
)
{
filePath.NotNullOrEmpty(nameof(filePath));
return OpenAsyncArchive(new FileInfo(filePath), options, cancellationToken);
}
public static IArchive OpenArchive(FileInfo fileInfo, ReaderOptions? options = null)
{
options ??= new ReaderOptions { LeaveStreamOpen = false };
return FindFactory<IArchiveFactory>(fileInfo).Open(fileInfo, options);
return FindFactory<IArchiveFactory>(fileInfo).OpenArchive(fileInfo, options);
}
/// <summary>
/// Constructor with IEnumerable FileInfo objects, multi and split support.
/// </summary>
/// <param name="fileInfos"></param>
/// <param name="options"></param>
public static IArchive Open(IEnumerable<FileInfo> fileInfos, ReaderOptions? options = null)
public static async ValueTask<IAsyncArchive> OpenAsyncArchive(
FileInfo fileInfo,
ReaderOptions? options = null,
CancellationToken cancellationToken = default
)
{
options ??= new ReaderOptions { LeaveStreamOpen = false };
var factory = await FindFactoryAsync<IArchiveFactory>(fileInfo, cancellationToken);
return factory.OpenAsyncArchive(fileInfo, options, cancellationToken);
}
public static IArchive OpenArchive(
IEnumerable<FileInfo> fileInfos,
ReaderOptions? options = null
)
{
fileInfos.NotNull(nameof(fileInfos));
var filesArray = fileInfos.ToArray();
@@ -78,21 +96,42 @@ public static class ArchiveFactory
var fileInfo = filesArray[0];
if (filesArray.Length == 1)
{
return Open(fileInfo, options);
return OpenArchive(fileInfo, options);
}
fileInfo.NotNull(nameof(fileInfo));
options ??= new ReaderOptions { LeaveStreamOpen = false };
return FindFactory<IMultiArchiveFactory>(fileInfo).Open(filesArray, options);
return FindFactory<IMultiArchiveFactory>(fileInfo).OpenArchive(filesArray, options);
}
/// <summary>
/// Constructor with IEnumerable FileInfo objects, multi and split support.
/// </summary>
/// <param name="streams"></param>
/// <param name="options"></param>
public static IArchive Open(IEnumerable<Stream> streams, ReaderOptions? options = null)
public static async ValueTask<IAsyncArchive> OpenAsyncArchive(
IEnumerable<FileInfo> fileInfos,
ReaderOptions? options = null,
CancellationToken cancellationToken = default
)
{
fileInfos.NotNull(nameof(fileInfos));
var filesArray = fileInfos.ToArray();
if (filesArray.Length == 0)
{
throw new InvalidOperationException("No files to open");
}
var fileInfo = filesArray[0];
if (filesArray.Length == 1)
{
return await OpenAsyncArchive(fileInfo, options, cancellationToken);
}
fileInfo.NotNull(nameof(fileInfo));
options ??= new ReaderOptions { LeaveStreamOpen = false };
var factory = await FindFactoryAsync<IMultiArchiveFactory>(fileInfo, cancellationToken);
return factory.OpenAsyncArchive(filesArray, options, cancellationToken);
}
public static IArchive OpenArchive(IEnumerable<Stream> streams, ReaderOptions? options = null)
{
streams.NotNull(nameof(streams));
var streamsArray = streams.ToArray();
@@ -104,25 +143,49 @@ public static class ArchiveFactory
var firstStream = streamsArray[0];
if (streamsArray.Length == 1)
{
return Open(firstStream, options);
return OpenArchive(firstStream, options);
}
firstStream.NotNull(nameof(firstStream));
options ??= new ReaderOptions();
return FindFactory<IMultiArchiveFactory>(firstStream).Open(streamsArray, options);
return FindFactory<IMultiArchiveFactory>(firstStream).OpenArchive(streamsArray, options);
}
public static async ValueTask<IAsyncArchive> OpenAsyncArchive(
IEnumerable<Stream> streams,
ReaderOptions? options = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
streams.NotNull(nameof(streams));
var streamsArray = streams.ToArray();
if (streamsArray.Length == 0)
{
throw new InvalidOperationException("No streams");
}
var firstStream = streamsArray[0];
if (streamsArray.Length == 1)
{
return await OpenAsyncArchive(firstStream, options, cancellationToken);
}
firstStream.NotNull(nameof(firstStream));
options ??= new ReaderOptions();
var factory = FindFactory<IMultiArchiveFactory>(firstStream);
return factory.OpenAsyncArchive(streamsArray, options);
}
/// <summary>
/// Extract to specific directory, retaining filename
/// </summary>
public static void WriteToDirectory(
string sourceArchive,
string destinationDirectory,
ExtractionOptions? options = null
)
{
using var archive = Open(sourceArchive);
using var archive = OpenArchive(sourceArchive);
archive.WriteToDirectory(destinationDirectory, options);
}
@@ -166,14 +229,68 @@ public static class ArchiveFactory
);
}
public static bool IsArchive(string filePath, out ArchiveType? type)
private static async ValueTask<T> FindFactoryAsync<T>(
FileInfo finfo,
CancellationToken cancellationToken
)
where T : IFactory
{
finfo.NotNull(nameof(finfo));
using Stream stream = finfo.OpenRead();
return await FindFactoryAsync<T>(stream, cancellationToken);
}
private static async ValueTask<T> FindFactoryAsync<T>(
Stream stream,
CancellationToken cancellationToken
)
where T : IFactory
{
stream.NotNull(nameof(stream));
if (!stream.CanRead || !stream.CanSeek)
{
throw new ArgumentException("Stream should be readable and seekable");
}
var factories = Factory.Factories.OfType<T>();
var startPosition = stream.Position;
foreach (var factory in factories)
{
stream.Seek(startPosition, SeekOrigin.Begin);
if (await factory.IsArchiveAsync(stream, cancellationToken: cancellationToken))
{
stream.Seek(startPosition, SeekOrigin.Begin);
return factory;
}
}
var extensions = string.Join(", ", factories.Select(item => item.Name));
throw new InvalidOperationException(
$"Cannot determine compressed stream type. Supported Archive Formats: {extensions}"
);
}
public static bool IsArchive(
string filePath,
out ArchiveType? type,
int bufferSize = ReaderOptions.DefaultBufferSize
)
{
filePath.NotNullOrEmpty(nameof(filePath));
using Stream s = File.OpenRead(filePath);
return IsArchive(s, out type);
return IsArchive(s, out type, bufferSize);
}
public static bool IsArchive(Stream stream, out ArchiveType? type)
public static bool IsArchive(
Stream stream,
out ArchiveType? type,
int bufferSize = ReaderOptions.DefaultBufferSize
)
{
type = null;
stream.NotNull(nameof(stream));
@@ -200,22 +317,12 @@ public static class ArchiveFactory
return false;
}
/// <summary>
/// From a passed in archive (zip, rar, 7z, 001), return all parts.
/// </summary>
/// <param name="part1"></param>
/// <returns></returns>
public static IEnumerable<string> GetFileParts(string part1)
{
part1.NotNullOrEmpty(nameof(part1));
return GetFileParts(new FileInfo(part1)).Select(a => a.FullName);
}
/// <summary>
/// From a passed in archive (zip, rar, 7z, 001), return all parts.
/// </summary>
/// <param name="part1"></param>
/// <returns></returns>
public static IEnumerable<FileInfo> GetFileParts(FileInfo part1)
{
part1.NotNull(nameof(part1));
@@ -229,7 +336,7 @@ public static class ArchiveFactory
if (part != null)
{
yield return part;
while ((part = factory.GetFilePart(i++, part1)) != null) //tests split too
while ((part = factory.GetFilePart(i++, part1)) != null)
{
yield return part;
}

View File

@@ -1,12 +1,14 @@
using System;
using System.Collections.Generic;
using System.IO;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common;
using SharpCompress.Readers;
namespace SharpCompress.Archives;
class AutoArchiveFactory : IArchiveFactory
internal class AutoArchiveFactory : IArchiveFactory
{
public string Name => nameof(AutoArchiveFactory);
@@ -14,14 +16,37 @@ class AutoArchiveFactory : IArchiveFactory
public IEnumerable<string> GetSupportedExtensions() => throw new NotSupportedException();
public bool IsArchive(Stream stream, string? password = null) =>
throw new NotSupportedException();
public bool IsArchive(
Stream stream,
string? password = null,
int bufferSize = ReaderOptions.DefaultBufferSize
) => throw new NotSupportedException();
public ValueTask<bool> IsArchiveAsync(
Stream stream,
string? password = null,
int bufferSize = ReaderOptions.DefaultBufferSize,
CancellationToken cancellationToken = default
) => throw new NotSupportedException();
public FileInfo? GetFilePart(int index, FileInfo part1) => throw new NotSupportedException();
public IArchive Open(Stream stream, ReaderOptions? readerOptions = null) =>
ArchiveFactory.Open(stream, readerOptions);
public IArchive OpenArchive(Stream stream, ReaderOptions? readerOptions = null) =>
ArchiveFactory.OpenArchive(stream, readerOptions);
public IArchive Open(FileInfo fileInfo, ReaderOptions? readerOptions = null) =>
ArchiveFactory.Open(fileInfo, readerOptions);
public IAsyncArchive OpenAsyncArchive(Stream stream, ReaderOptions? readerOptions = null) =>
(IAsyncArchive)OpenArchive(stream, readerOptions);
public IArchive OpenArchive(FileInfo fileInfo, ReaderOptions? readerOptions = null) =>
ArchiveFactory.OpenArchive(fileInfo, readerOptions);
public IAsyncArchive OpenAsyncArchive(
FileInfo fileInfo,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IAsyncArchive)OpenArchive(fileInfo, readerOptions);
}
}

View File

@@ -0,0 +1,197 @@
using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common;
using SharpCompress.Common.GZip;
using SharpCompress.IO;
using SharpCompress.Readers;
using SharpCompress.Readers.GZip;
using SharpCompress.Writers;
using SharpCompress.Writers.GZip;
namespace SharpCompress.Archives.GZip;
public partial class GZipArchive
#if NET8_0_OR_GREATER
: IWritableArchiveOpenable,
IMultiArchiveOpenable<IWritableArchive, IWritableAsyncArchive>
#endif
{
public static IWritableAsyncArchive OpenAsyncArchive(
string path,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
path.NotNullOrEmpty(nameof(path));
return (IWritableAsyncArchive)OpenArchive(
new FileInfo(path),
readerOptions ?? new ReaderOptions()
);
}
public static IWritableArchive OpenArchive(string filePath, ReaderOptions? readerOptions = null)
{
filePath.NotNullOrEmpty(nameof(filePath));
return OpenArchive(new FileInfo(filePath), readerOptions ?? new ReaderOptions());
}
public static IWritableArchive OpenArchive(
FileInfo fileInfo,
ReaderOptions? readerOptions = null
)
{
fileInfo.NotNull(nameof(fileInfo));
return new GZipArchive(
new SourceStream(
fileInfo,
i => ArchiveVolumeFactory.GetFilePart(i, fileInfo),
readerOptions ?? new ReaderOptions()
)
);
}
public static IWritableArchive OpenArchive(
IEnumerable<FileInfo> fileInfos,
ReaderOptions? readerOptions = null
)
{
fileInfos.NotNull(nameof(fileInfos));
var files = fileInfos.ToArray();
return new GZipArchive(
new SourceStream(
files[0],
i => i < files.Length ? files[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
public static IWritableArchive OpenArchive(
IEnumerable<Stream> streams,
ReaderOptions? readerOptions = null
)
{
streams.NotNull(nameof(streams));
var strms = streams.ToArray();
return new GZipArchive(
new SourceStream(
strms[0],
i => i < strms.Length ? strms[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
public static IWritableArchive OpenArchive(Stream stream, ReaderOptions? readerOptions = null)
{
stream.NotNull(nameof(stream));
if (stream is not { CanSeek: true })
{
throw new ArgumentException("Stream must be seekable", nameof(stream));
}
return new GZipArchive(
new SourceStream(stream, _ => null, readerOptions ?? new ReaderOptions())
);
}
public static IWritableAsyncArchive OpenAsyncArchive(
Stream stream,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IWritableAsyncArchive)OpenArchive(stream, readerOptions);
}
public static IWritableAsyncArchive OpenAsyncArchive(
FileInfo fileInfo,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IWritableAsyncArchive)OpenArchive(fileInfo, readerOptions);
}
public static IWritableAsyncArchive OpenAsyncArchive(
IReadOnlyList<Stream> streams,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IWritableAsyncArchive)OpenArchive(streams, readerOptions);
}
public static IWritableAsyncArchive OpenAsyncArchive(
IReadOnlyList<FileInfo> fileInfos,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IWritableAsyncArchive)OpenArchive(fileInfos, readerOptions);
}
public static IWritableArchive CreateArchive() => new GZipArchive();
public static IWritableAsyncArchive CreateAsyncArchive() => new GZipArchive();
public static bool IsGZipFile(string filePath) => IsGZipFile(new FileInfo(filePath));
public static bool IsGZipFile(FileInfo fileInfo)
{
if (!fileInfo.Exists)
{
return false;
}
using Stream stream = fileInfo.OpenRead();
return IsGZipFile(stream);
}
public static bool IsGZipFile(Stream stream)
{
Span<byte> header = stackalloc byte[10];
if (!stream.ReadFully(header))
{
return false;
}
if (header[0] != 0x1F || header[1] != 0x8B || header[2] != 8)
{
return false;
}
return true;
}
public static async ValueTask<bool> IsGZipFileAsync(
Stream stream,
CancellationToken cancellationToken = default
)
{
byte[] header = new byte[10];
if (!await stream.ReadFullyAsync(header, cancellationToken).ConfigureAwait(false))
{
return false;
}
if (header[0] != 0x1F || header[1] != 0x8B || header[2] != 8)
{
return false;
}
return true;
}
}

View File

@@ -14,122 +14,20 @@ using SharpCompress.Writers.GZip;
namespace SharpCompress.Archives.GZip;
public class GZipArchive : AbstractWritableArchive<GZipArchiveEntry, GZipVolume>
public partial class GZipArchive : AbstractWritableArchive<GZipArchiveEntry, GZipVolume>
{
/// <summary>
/// Constructor expects a filepath to an existing file.
/// </summary>
/// <param name="filePath"></param>
/// <param name="readerOptions"></param>
public static GZipArchive Open(string filePath, ReaderOptions? readerOptions = null)
{
filePath.NotNullOrEmpty(nameof(filePath));
return Open(new FileInfo(filePath), readerOptions ?? new ReaderOptions());
}
/// <summary>
/// Constructor with a FileInfo object to an existing file.
/// </summary>
/// <param name="fileInfo"></param>
/// <param name="readerOptions"></param>
public static GZipArchive Open(FileInfo fileInfo, ReaderOptions? readerOptions = null)
{
fileInfo.NotNull(nameof(fileInfo));
return new GZipArchive(
new SourceStream(
fileInfo,
i => ArchiveVolumeFactory.GetFilePart(i, fileInfo),
readerOptions ?? new ReaderOptions()
)
);
}
/// <summary>
/// Constructor with all file parts passed in
/// </summary>
/// <param name="fileInfos"></param>
/// <param name="readerOptions"></param>
public static GZipArchive Open(
IEnumerable<FileInfo> fileInfos,
ReaderOptions? readerOptions = null
)
{
fileInfos.NotNull(nameof(fileInfos));
var files = fileInfos.ToArray();
return new GZipArchive(
new SourceStream(
files[0],
i => i < files.Length ? files[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
/// <summary>
/// Constructor with all stream parts passed in
/// </summary>
/// <param name="streams"></param>
/// <param name="readerOptions"></param>
public static GZipArchive Open(IEnumerable<Stream> streams, ReaderOptions? readerOptions = null)
{
streams.NotNull(nameof(streams));
var strms = streams.ToArray();
return new GZipArchive(
new SourceStream(
strms[0],
i => i < strms.Length ? strms[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
/// <summary>
/// Takes a seekable Stream as a source
/// </summary>
/// <param name="stream"></param>
/// <param name="readerOptions"></param>
public static GZipArchive Open(Stream stream, ReaderOptions? readerOptions = null)
{
stream.NotNull(nameof(stream));
if (stream is not { CanSeek: true })
{
throw new ArgumentException("Stream must be seekable", nameof(stream));
}
return new GZipArchive(
new SourceStream(stream, _ => null, readerOptions ?? new ReaderOptions())
);
}
public static GZipArchive Create() => new();
/// <summary>
/// Constructor with a SourceStream able to handle FileInfo and Streams.
/// </summary>
/// <param name="sourceStream"></param>
private GZipArchive(SourceStream sourceStream)
: base(ArchiveType.GZip, sourceStream) { }
internal GZipArchive()
: base(ArchiveType.GZip) { }
protected override IEnumerable<GZipVolume> LoadVolumes(SourceStream sourceStream)
{
sourceStream.LoadAllParts();
return sourceStream.Streams.Select(a => new GZipVolume(a, ReaderOptions, 0));
}
public static bool IsGZipFile(string filePath) => IsGZipFile(new FileInfo(filePath));
public static bool IsGZipFile(FileInfo fileInfo)
{
if (!fileInfo.Exists)
{
return false;
}
using Stream stream = fileInfo.OpenRead();
return IsGZipFile(stream);
}
public void SaveTo(string filePath) => SaveTo(new FileInfo(filePath));
public void SaveTo(FileInfo fileInfo)
@@ -138,38 +36,19 @@ public class GZipArchive : AbstractWritableArchive<GZipArchiveEntry, GZipVolume>
SaveTo(stream, new WriterOptions(CompressionType.GZip));
}
public Task SaveToAsync(string filePath, CancellationToken cancellationToken = default) =>
public ValueTask SaveToAsync(string filePath, CancellationToken cancellationToken = default) =>
SaveToAsync(new FileInfo(filePath), cancellationToken);
public async Task SaveToAsync(FileInfo fileInfo, CancellationToken cancellationToken = default)
public async ValueTask SaveToAsync(
FileInfo fileInfo,
CancellationToken cancellationToken = default
)
{
using var stream = fileInfo.Open(FileMode.Create, FileAccess.Write);
await SaveToAsync(stream, new WriterOptions(CompressionType.GZip), cancellationToken)
.ConfigureAwait(false);
}
public static bool IsGZipFile(Stream stream)
{
// read the header on the first read
Span<byte> header = stackalloc byte[10];
// workitem 8501: handle edge case (decompress empty stream)
if (!stream.ReadFully(header))
{
return false;
}
if (header[0] != 0x1F || header[1] != 0x8B || header[2] != 8)
{
return false;
}
return true;
}
internal GZipArchive()
: base(ArchiveType.GZip) { }
protected override GZipArchiveEntry CreateEntryInternal(
string filePath,
Stream source,
@@ -213,7 +92,7 @@ public class GZipArchive : AbstractWritableArchive<GZipArchiveEntry, GZipVolume>
}
}
protected override async Task SaveToAsync(
protected override async ValueTask SaveToAsync(
Stream stream,
WriterOptions options,
IEnumerable<GZipArchiveEntry> oldEntries,
@@ -240,7 +119,18 @@ public class GZipArchive : AbstractWritableArchive<GZipArchiveEntry, GZipVolume>
var stream = volumes.Single().Stream;
yield return new GZipArchiveEntry(
this,
new GZipFilePart(stream, ReaderOptions.ArchiveEncoding)
GZipFilePart.Create(stream, ReaderOptions.ArchiveEncoding)
);
}
protected override async IAsyncEnumerable<GZipArchiveEntry> LoadEntriesAsync(
IAsyncEnumerable<GZipVolume> volumes
)
{
var stream = (await volumes.SingleAsync()).Stream;
yield return new GZipArchiveEntry(
this,
await GZipFilePart.CreateAsync(stream, ReaderOptions.ArchiveEncoding)
);
}
@@ -248,6 +138,13 @@ public class GZipArchive : AbstractWritableArchive<GZipArchiveEntry, GZipVolume>
{
var stream = Volumes.Single().Stream;
stream.Position = 0;
return GZipReader.Open(stream);
return GZipReader.OpenReader(stream);
}
protected override ValueTask<IAsyncReader> CreateReaderForSolidExtractionAsync()
{
var stream = Volumes.Single().Stream;
stream.Position = 0;
return new((IAsyncReader)GZipReader.OpenReader(stream));
}
}

View File

@@ -23,10 +23,10 @@ public class GZipArchiveEntry : GZipEntry, IArchiveEntry
return Parts.Single().GetCompressedStream().NotNull();
}
public virtual Task<Stream> OpenEntryStreamAsync(CancellationToken cancellationToken = default)
public ValueTask<Stream> OpenEntryStreamAsync(CancellationToken cancellationToken = default)
{
// GZip synchronous implementation is fast enough, just wrap it
return Task.FromResult(OpenEntryStream());
return new(OpenEntryStream());
}
#region IArchiveEntry Members

View File

@@ -38,5 +38,18 @@ public interface IArchive : IDisposable
/// <summary>
/// The total size of the files as uncompressed in the archive.
/// </summary>
long TotalUncompressSize { get; }
long TotalUncompressedSize { get; }
/// <summary>
/// Returns whether the archive is encrypted.
/// </summary>
bool IsEncrypted { get; }
/// <summary>
/// Returns whether multi-threaded extraction is supported for this archive.
/// Multi-threading is supported when the archive is opened from a FileInfo or file path
/// (not a stream) and the format supports random access (e.g., Zip, Tar, Rar).
/// SOLID archives (some Rar, all 7Zip) should use sequential extraction for best performance.
/// </summary>
bool SupportsMultiThreadedExtraction { get; }
}

View File

@@ -17,7 +17,7 @@ public interface IArchiveEntry : IEntry
/// Opens the current entry as a stream that will decompress as it is read asynchronously.
/// Read the entire stream or use SkipEntry on EntryStream.
/// </summary>
Task<Stream> OpenEntryStreamAsync(CancellationToken cancellationToken = default);
ValueTask<Stream> OpenEntryStreamAsync(CancellationToken cancellationToken = default);
/// <summary>
/// The archive can find all the parts of the archive needed to extract this entry.

View File

@@ -9,6 +9,8 @@ namespace SharpCompress.Archives;
public static class IArchiveEntryExtensions
{
private const int BufferSize = 81920;
/// <param name="archiveEntry">The archive entry to extract.</param>
extension(IArchiveEntry archiveEntry)
{
@@ -26,7 +28,7 @@ public static class IArchiveEntryExtensions
using var entryStream = archiveEntry.OpenEntryStream();
var sourceStream = WrapWithProgress(entryStream, archiveEntry, progress);
sourceStream.CopyTo(streamToWriteTo, Constants.BufferSize);
sourceStream.CopyTo(streamToWriteTo, BufferSize);
}
/// <summary>
@@ -35,7 +37,7 @@ public static class IArchiveEntryExtensions
/// <param name="streamToWriteTo">The stream to write the entry content to.</param>
/// <param name="cancellationToken">Cancellation token.</param>
/// <param name="progress">Optional progress reporter for tracking extraction progress.</param>
public async Task WriteToAsync(
public async ValueTask WriteToAsync(
Stream streamToWriteTo,
IProgress<ProgressReport>? progress = null,
CancellationToken cancellationToken = default
@@ -49,7 +51,7 @@ public static class IArchiveEntryExtensions
using var entryStream = await archiveEntry.OpenEntryStreamAsync(cancellationToken);
var sourceStream = WrapWithProgress(entryStream, archiveEntry, progress);
await sourceStream
.CopyToAsync(streamToWriteTo, Constants.BufferSize, cancellationToken)
.CopyToAsync(streamToWriteTo, BufferSize, cancellationToken)
.ConfigureAwait(false);
}
}
@@ -108,18 +110,20 @@ public static class IArchiveEntryExtensions
/// <summary>
/// Extract to specific directory asynchronously, retaining filename
/// </summary>
public Task WriteToDirectoryAsync(
public async ValueTask WriteToDirectoryAsync(
string destinationDirectory,
ExtractionOptions? options = null,
CancellationToken cancellationToken = default
) =>
ExtractionMethods.WriteEntryToDirectoryAsync(
entry,
destinationDirectory,
options,
entry.WriteToFileAsync,
cancellationToken
);
await ExtractionMethods
.WriteEntryToDirectoryAsync(
entry,
destinationDirectory,
options,
entry.WriteToFileAsync,
cancellationToken
)
.ConfigureAwait(false);
/// <summary>
/// Extract to specific file
@@ -139,21 +143,23 @@ public static class IArchiveEntryExtensions
/// <summary>
/// Extract to specific file asynchronously
/// </summary>
public Task WriteToFileAsync(
public async ValueTask WriteToFileAsync(
string destinationFileName,
ExtractionOptions? options = null,
CancellationToken cancellationToken = default
) =>
ExtractionMethods.WriteEntryToFileAsync(
entry,
destinationFileName,
options,
async (x, fm, ct) =>
{
using var fs = File.Open(destinationFileName, fm);
await entry.WriteToAsync(fs, null, ct).ConfigureAwait(false);
},
cancellationToken
);
await ExtractionMethods
.WriteEntryToFileAsync(
entry,
destinationFileName,
options,
async (x, fm, ct) =>
{
using var fs = File.Open(destinationFileName, fm);
await entry.WriteToAsync(fs, null, ct).ConfigureAwait(false);
},
cancellationToken
)
.ConfigureAwait(false);
}
}

View File

@@ -1,8 +1,6 @@
using System;
using System.Collections.Generic;
using System.IO;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common;
using SharpCompress.Readers;
@@ -10,7 +8,6 @@ namespace SharpCompress.Archives;
public static class IArchiveExtensions
{
/// <param name="archive">The archive to extract.</param>
extension(IArchive archive)
{
/// <summary>
@@ -25,7 +22,6 @@ public static class IArchiveExtensions
IProgress<ProgressReport>? progress = null
)
{
// For solid archives (Rar, 7Zip), use the optimized reader-based approach
if (archive.IsSolid || archive.Type == ArchiveType.SevenZip)
{
using var reader = archive.ExtractAllEntries();
@@ -33,7 +29,6 @@ public static class IArchiveExtensions
}
else
{
// For non-solid archives, extract entries directly
archive.WriteToDirectoryInternal(destinationDirectory, options, progress);
}
}
@@ -44,14 +39,10 @@ public static class IArchiveExtensions
IProgress<ProgressReport>? progress
)
{
// Prepare for progress reporting
var totalBytes = archive.TotalUncompressSize;
var totalBytes = archive.TotalUncompressedSize;
var bytesRead = 0L;
// Tracking for created directories.
var seenDirectories = new HashSet<string>();
// Extract
foreach (var entry in archive.Entries)
{
if (entry.IsDirectory)
@@ -70,94 +61,8 @@ public static class IArchiveExtensions
continue;
}
// Use the entry's WriteToDirectory method which respects ExtractionOptions
entry.WriteToDirectory(destinationDirectory, options);
// Update progress
bytesRead += entry.Size;
progress?.Report(
new ProgressReport(entry.Key ?? string.Empty, bytesRead, totalBytes)
);
}
}
/// <summary>
/// Extract to specific directory asynchronously with progress reporting and cancellation support
/// </summary>
/// <param name="destinationDirectory">The folder to extract into.</param>
/// <param name="options">Extraction options.</param>
/// <param name="progress">Optional progress reporter for tracking extraction progress.</param>
/// <param name="cancellationToken">Optional cancellation token.</param>
public async Task WriteToDirectoryAsync(
string destinationDirectory,
ExtractionOptions? options = null,
IProgress<ProgressReport>? progress = null,
CancellationToken cancellationToken = default
)
{
// For solid archives (Rar, 7Zip), use the optimized reader-based approach
if (archive.IsSolid || archive.Type == ArchiveType.SevenZip)
{
using var reader = archive.ExtractAllEntries();
await reader.WriteAllToDirectoryAsync(
destinationDirectory,
options,
cancellationToken
);
}
else
{
// For non-solid archives, extract entries directly
await archive.WriteToDirectoryAsyncInternal(
destinationDirectory,
options,
progress,
cancellationToken
);
}
}
private async Task WriteToDirectoryAsyncInternal(
string destinationDirectory,
ExtractionOptions? options,
IProgress<ProgressReport>? progress,
CancellationToken cancellationToken
)
{
// Prepare for progress reporting
var totalBytes = archive.TotalUncompressSize;
var bytesRead = 0L;
// Tracking for created directories.
var seenDirectories = new HashSet<string>();
// Extract
foreach (var entry in archive.Entries)
{
cancellationToken.ThrowIfCancellationRequested();
if (entry.IsDirectory)
{
var dirPath = Path.Combine(
destinationDirectory,
entry.Key.NotNull("Entry Key is null")
);
if (
Path.GetDirectoryName(dirPath + "/") is { } parentDirectory
&& seenDirectories.Add(dirPath)
)
{
Directory.CreateDirectory(parentDirectory);
}
continue;
}
// Use the entry's WriteToDirectoryAsync method which respects ExtractionOptions
await entry
.WriteToDirectoryAsync(destinationDirectory, options, cancellationToken)
.ConfigureAwait(false);
// Update progress
bytesRead += entry.Size;
progress?.Report(
new ProgressReport(entry.Key ?? string.Empty, bytesRead, totalBytes)

View File

@@ -1,4 +1,5 @@
using System.IO;
using System.Threading;
using SharpCompress.Factories;
using SharpCompress.Readers;
@@ -24,12 +25,31 @@ public interface IArchiveFactory : IFactory
/// </summary>
/// <param name="stream">An open, readable and seekable stream.</param>
/// <param name="readerOptions">reading options.</param>
IArchive Open(Stream stream, ReaderOptions? readerOptions = null);
IArchive OpenArchive(Stream stream, ReaderOptions? readerOptions = null);
/// <summary>
/// Opens an Archive for random access asynchronously.
/// </summary>
/// <param name="stream">An open, readable and seekable stream.</param>
/// <param name="readerOptions">reading options.</param>
IAsyncArchive OpenAsyncArchive(Stream stream, ReaderOptions? readerOptions = null);
/// <summary>
/// Constructor with a FileInfo object to an existing file.
/// </summary>
/// <param name="fileInfo">the file to open.</param>
/// <param name="readerOptions">reading options.</param>
IArchive Open(FileInfo fileInfo, ReaderOptions? readerOptions = null);
IArchive OpenArchive(FileInfo fileInfo, ReaderOptions? readerOptions = null);
/// <summary>
/// Opens an Archive from a FileInfo object asynchronously.
/// </summary>
/// <param name="fileInfo">the file to open.</param>
/// <param name="readerOptions">reading options.</param>
/// <param name="cancellationToken">Cancellation token.</param>
IAsyncArchive OpenAsyncArchive(
FileInfo fileInfo,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
);
}

View File

@@ -0,0 +1,40 @@
#if NET8_0_OR_GREATER
using System.IO;
using System.Threading;
using SharpCompress.Readers;
namespace SharpCompress.Archives;
public interface IArchiveOpenable<TSync, TASync>
where TSync : IArchive
where TASync : IAsyncArchive
{
public static abstract TSync OpenArchive(string filePath, ReaderOptions? readerOptions = null);
public static abstract TSync OpenArchive(
FileInfo fileInfo,
ReaderOptions? readerOptions = null
);
public static abstract TSync OpenArchive(Stream stream, ReaderOptions? readerOptions = null);
public static abstract TASync OpenAsyncArchive(
string path,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
);
public static abstract TASync OpenAsyncArchive(
Stream stream,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
);
public static abstract TASync OpenAsyncArchive(
FileInfo fileInfo,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
);
}
#endif

View File

@@ -0,0 +1,48 @@
using System;
using System.Collections.Generic;
using System.Threading.Tasks;
using SharpCompress.Common;
using SharpCompress.Readers;
namespace SharpCompress.Archives;
public interface IAsyncArchive : IAsyncDisposable
{
IAsyncEnumerable<IArchiveEntry> EntriesAsync { get; }
IAsyncEnumerable<IVolume> VolumesAsync { get; }
ArchiveType Type { get; }
/// <summary>
/// Use this method to extract all entries in an archive in order.
/// This is primarily for SOLID Rar Archives or 7Zip Archives as they need to be
/// extracted sequentially for the best performance.
/// </summary>
ValueTask<IAsyncReader> ExtractAllEntriesAsync();
/// <summary>
/// Archive is SOLID (this means the Archive saved bytes by reusing information which helps for archives containing many small files).
/// Rar Archives can be SOLID while all 7Zip archives are considered SOLID.
/// </summary>
ValueTask<bool> IsSolidAsync();
/// <summary>
/// This checks to see if all the known entries have IsComplete = true
/// </summary>
ValueTask<bool> IsCompleteAsync();
/// <summary>
/// The total size of the files compressed in the archive.
/// </summary>
ValueTask<long> TotalSizeAsync();
/// <summary>
/// The total size of the files as uncompressed in the archive.
/// </summary>
ValueTask<long> TotalUncompressedSizeAsync();
/// <summary>
/// Returns whether the archive is encrypted.
/// </summary>
ValueTask<bool> IsEncryptedAsync();
}

View File

@@ -0,0 +1,92 @@
using System;
using System.Collections.Generic;
using System.IO;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common;
using SharpCompress.Readers;
namespace SharpCompress.Archives;
public static class IAsyncArchiveExtensions
{
extension(IAsyncArchive archive)
{
/// <summary>
/// Extract to specific directory asynchronously with progress reporting and cancellation support
/// </summary>
/// <param name="archive">The archive to extract.</param>
/// <param name="destinationDirectory">The folder to extract into.</param>
/// <param name="options">Extraction options.</param>
/// <param name="progress">Optional progress reporter for tracking extraction progress.</param>
/// <param name="cancellationToken">Optional cancellation token.</param>
public async Task WriteToDirectoryAsync(
string destinationDirectory,
ExtractionOptions? options = null,
IProgress<ProgressReport>? progress = null,
CancellationToken cancellationToken = default
)
{
if (await archive.IsSolidAsync() || archive.Type == ArchiveType.SevenZip)
{
await using var reader = await archive.ExtractAllEntriesAsync();
await reader.WriteAllToDirectoryAsync(
destinationDirectory,
options,
cancellationToken
);
}
else
{
await archive.WriteToDirectoryAsyncInternal(
destinationDirectory,
options,
progress,
cancellationToken
);
}
}
private async Task WriteToDirectoryAsyncInternal(
string destinationDirectory,
ExtractionOptions? options,
IProgress<ProgressReport>? progress,
CancellationToken cancellationToken
)
{
var totalBytes = await archive.TotalUncompressedSizeAsync();
var bytesRead = 0L;
var seenDirectories = new HashSet<string>();
await foreach (var entry in archive.EntriesAsync.WithCancellation(cancellationToken))
{
cancellationToken.ThrowIfCancellationRequested();
if (entry.IsDirectory)
{
var dirPath = Path.Combine(
destinationDirectory,
entry.Key.NotNull("Entry Key is null")
);
if (
Path.GetDirectoryName(dirPath + "/") is { } parentDirectory
&& seenDirectories.Add(dirPath)
)
{
Directory.CreateDirectory(parentDirectory);
}
continue;
}
await entry
.WriteToDirectoryAsync(destinationDirectory, options, cancellationToken)
.ConfigureAwait(false);
bytesRead += entry.Size;
progress?.Report(
new ProgressReport(entry.Key ?? string.Empty, bytesRead, totalBytes)
);
}
}
}
}

View File

@@ -1,5 +1,6 @@
using System.Collections.Generic;
using System.IO;
using System.Threading;
using SharpCompress.Factories;
using SharpCompress.Readers;
@@ -25,12 +26,34 @@ public interface IMultiArchiveFactory : IFactory
/// </summary>
/// <param name="streams"></param>
/// <param name="readerOptions">reading options.</param>
IArchive Open(IReadOnlyList<Stream> streams, ReaderOptions? readerOptions = null);
IArchive OpenArchive(IReadOnlyList<Stream> streams, ReaderOptions? readerOptions = null);
/// <summary>
/// Opens a multi-part archive from streams asynchronously.
/// </summary>
/// <param name="streams"></param>
/// <param name="readerOptions">reading options.</param>
IAsyncArchive OpenAsyncArchive(
IReadOnlyList<Stream> streams,
ReaderOptions? readerOptions = null
);
/// <summary>
/// Constructor with IEnumerable Stream objects, multi and split support.
/// </summary>
/// <param name="fileInfos"></param>
/// <param name="readerOptions">reading options.</param>
IArchive Open(IReadOnlyList<FileInfo> fileInfos, ReaderOptions? readerOptions = null);
IArchive OpenArchive(IReadOnlyList<FileInfo> fileInfos, ReaderOptions? readerOptions = null);
/// <summary>
/// Opens a multi-part archive from files asynchronously.
/// </summary>
/// <param name="fileInfos"></param>
/// <param name="readerOptions">reading options.</param>
/// <param name="cancellationToken">Cancellation token.</param>
IAsyncArchive OpenAsyncArchive(
IReadOnlyList<FileInfo> fileInfos,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
);
}

View File

@@ -0,0 +1,35 @@
#if NET8_0_OR_GREATER
using System.Collections.Generic;
using System.IO;
using System.Threading;
using SharpCompress.Readers;
namespace SharpCompress.Archives;
public interface IMultiArchiveOpenable<TSync, TASync>
where TSync : IArchive
where TASync : IAsyncArchive
{
public static abstract TSync OpenArchive(
IEnumerable<FileInfo> fileInfos,
ReaderOptions? readerOptions = null
);
public static abstract TSync OpenArchive(
IEnumerable<Stream> streams,
ReaderOptions? readerOptions = null
);
public static abstract TASync OpenAsyncArchive(
IReadOnlyList<Stream> streams,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
);
public static abstract TASync OpenAsyncArchive(
IReadOnlyList<FileInfo> fileInfos,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
);
}
#endif

View File

@@ -6,8 +6,17 @@ using SharpCompress.Writers;
namespace SharpCompress.Archives;
public interface IWritableArchive : IArchive
public interface IWritableArchiveCommon
{
/// <summary>
/// Use this to pause entry rebuilding when adding large collections of entries. Dispose when complete. A using statement is recommended.
/// </summary>
/// <returns>IDisposeable to resume entry rebuilding</returns>
IDisposable PauseEntryRebuilding();
/// <summary>
/// Removes the specified entry from the archive.
/// </summary>
void RemoveEntry(IArchiveEntry entry);
IArchiveEntry AddEntry(
@@ -19,18 +28,24 @@ public interface IWritableArchive : IArchive
);
IArchiveEntry AddDirectoryEntry(string key, DateTime? modified = null);
}
public interface IWritableArchive : IArchive, IWritableArchiveCommon
{
/// <summary>
/// Saves the archive to the specified stream using the given writer options.
/// </summary>
void SaveTo(Stream stream, WriterOptions options);
}
Task SaveToAsync(
public interface IWritableAsyncArchive : IAsyncArchive, IWritableArchiveCommon
{
/// <summary>
/// Asynchronously saves the archive to the specified stream using the given writer options.
/// </summary>
ValueTask SaveToAsync(
Stream stream,
WriterOptions options,
CancellationToken cancellationToken = default
);
/// <summary>
/// Use this to pause entry rebuilding when adding large collections of entries. Dispose when complete. A using statement is recommended.
/// </summary>
/// <returns>IDisposeable to resume entry rebuilding</returns>
IDisposable PauseEntryRebuilding();
}

View File

@@ -0,0 +1,59 @@
using System;
using System.IO;
namespace SharpCompress.Archives;
public static class IWritableArchiveCommonExtensions
{
extension(IWritableArchiveCommon writableArchive)
{
public void AddAllFromDirectory(
string filePath,
string searchPattern = "*.*",
SearchOption searchOption = SearchOption.AllDirectories
)
{
using (writableArchive.PauseEntryRebuilding())
{
foreach (
var path in Directory.EnumerateFiles(filePath, searchPattern, searchOption)
)
{
var fileInfo = new FileInfo(path);
writableArchive.AddEntry(
path.Substring(filePath.Length),
fileInfo.OpenRead(),
true,
fileInfo.Length,
fileInfo.LastWriteTime
);
}
}
}
public IArchiveEntry AddEntry(string key, string file) =>
writableArchive.AddEntry(key, new FileInfo(file));
public IArchiveEntry AddEntry(
string key,
Stream source,
long size = 0,
DateTime? modified = null
) => writableArchive.AddEntry(key, source, false, size, modified);
public IArchiveEntry AddEntry(string key, FileInfo fileInfo)
{
if (!fileInfo.Exists)
{
throw new ArgumentException("FileInfo does not exist.");
}
return writableArchive.AddEntry(
key,
fileInfo.OpenRead(),
true,
fileInfo.Length,
fileInfo.LastWriteTime
);
}
}
}

View File

@@ -1,106 +1,20 @@
using System;
using System.IO;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common;
using SharpCompress.Writers;
namespace SharpCompress.Archives;
public static class IWritableArchiveExtensions
{
public static void AddEntry(
this IWritableArchive writableArchive,
string entryPath,
string filePath
)
extension(IWritableArchive writableArchive)
{
var fileInfo = new FileInfo(filePath);
if (!fileInfo.Exists)
public void SaveTo(string filePath, WriterOptions? options = null) =>
writableArchive.SaveTo(new FileInfo(filePath), options ?? new(CompressionType.Deflate));
public void SaveTo(FileInfo fileInfo, WriterOptions? options = null)
{
throw new FileNotFoundException("Could not AddEntry: " + filePath);
using var stream = fileInfo.Open(FileMode.Create, FileAccess.Write);
writableArchive.SaveTo(stream, options ?? new(CompressionType.Deflate));
}
writableArchive.AddEntry(
entryPath,
new FileInfo(filePath).OpenRead(),
true,
fileInfo.Length,
fileInfo.LastWriteTime
);
}
public static void SaveTo(
this IWritableArchive writableArchive,
string filePath,
WriterOptions options
) => writableArchive.SaveTo(new FileInfo(filePath), options);
public static void SaveTo(
this IWritableArchive writableArchive,
FileInfo fileInfo,
WriterOptions options
)
{
using var stream = fileInfo.Open(FileMode.Create, FileAccess.Write);
writableArchive.SaveTo(stream, options);
}
public static Task SaveToAsync(
this IWritableArchive writableArchive,
string filePath,
WriterOptions options,
CancellationToken cancellationToken = default
) => writableArchive.SaveToAsync(new FileInfo(filePath), options, cancellationToken);
public static async Task SaveToAsync(
this IWritableArchive writableArchive,
FileInfo fileInfo,
WriterOptions options,
CancellationToken cancellationToken = default
)
{
using var stream = fileInfo.Open(FileMode.Create, FileAccess.Write);
await writableArchive.SaveToAsync(stream, options, cancellationToken).ConfigureAwait(false);
}
public static void AddAllFromDirectory(
this IWritableArchive writableArchive,
string filePath,
string searchPattern = "*.*",
SearchOption searchOption = SearchOption.AllDirectories
)
{
using (writableArchive.PauseEntryRebuilding())
{
foreach (var path in Directory.EnumerateFiles(filePath, searchPattern, searchOption))
{
var fileInfo = new FileInfo(path);
writableArchive.AddEntry(
path.Substring(filePath.Length),
fileInfo.OpenRead(),
true,
fileInfo.Length,
fileInfo.LastWriteTime
);
}
}
}
public static IArchiveEntry AddEntry(
this IWritableArchive writableArchive,
string key,
FileInfo fileInfo
)
{
if (!fileInfo.Exists)
{
throw new ArgumentException("FileInfo does not exist.");
}
return writableArchive.AddEntry(
key,
fileInfo.OpenRead(),
true,
fileInfo.Length,
fileInfo.LastWriteTime
);
}
}

View File

@@ -0,0 +1,10 @@
#if NET8_0_OR_GREATER
namespace SharpCompress.Archives;
public interface IWritableArchiveOpenable
: IArchiveOpenable<IWritableArchive, IWritableAsyncArchive>
{
public static abstract IWritableArchive CreateArchive();
public static abstract IWritableAsyncArchive CreateAsyncArchive();
}
#endif

View File

@@ -0,0 +1,36 @@
using System.IO;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common;
using SharpCompress.Writers;
namespace SharpCompress.Archives;
public static class IWritableAsyncArchiveExtensions
{
extension(IWritableAsyncArchive writableArchive)
{
public ValueTask SaveToAsync(
string filePath,
WriterOptions? options = null,
CancellationToken cancellationToken = default
) =>
writableArchive.SaveToAsync(
new FileInfo(filePath),
options ?? new(CompressionType.Deflate),
cancellationToken
);
public async ValueTask SaveToAsync(
FileInfo fileInfo,
WriterOptions? options = null,
CancellationToken cancellationToken = default
)
{
using var stream = fileInfo.Open(FileMode.Create, FileAccess.Write);
await writableArchive
.SaveToAsync(stream, options ?? new(CompressionType.Deflate), cancellationToken)
.ConfigureAwait(false);
}
}
}

View File

@@ -16,5 +16,5 @@ public interface IWriteableArchiveFactory : Factories.IFactory
/// Creates a new, empty archive, ready to be written.
/// </summary>
/// <returns></returns>
IWritableArchive CreateWriteableArchive();
IWritableArchive CreateArchive();
}

View File

@@ -1,18 +1,36 @@
using System.Linq;
using System.Linq;
using System.Threading.Tasks;
using SharpCompress.Common.Rar;
namespace SharpCompress.Archives.Rar;
public static class RarArchiveExtensions
{
/// <summary>
/// RarArchive is the first volume of a multi-part archive. If MultipartVolume is true and IsFirstVolume is false then the first volume file must be missing.
/// </summary>
public static bool IsFirstVolume(this RarArchive archive) =>
archive.Volumes.First().IsFirstVolume;
extension(IRarArchive archive)
{
/// <summary>
/// RarArchive is the first volume of a multi-part archive. If MultipartVolume is true and IsFirstVolume is false then the first volume file must be missing.
/// </summary>
public bool IsFirstVolume() => archive.Volumes.Cast<RarVolume>().First().IsFirstVolume;
/// <summary>
/// RarArchive is part of a multi-part archive.
/// </summary>
public static bool IsMultipartVolume(this RarArchive archive) =>
archive.Volumes.First().IsMultiVolume;
/// <summary>
/// RarArchive is part of a multi-part archive.
/// </summary>
public bool IsMultipartVolume() => archive.Volumes.Cast<RarVolume>().First().IsMultiVolume;
}
extension(IRarAsyncArchive archive)
{
/// <summary>
/// RarArchive is the first volume of a multi-part archive. If MultipartVolume is true and IsFirstVolume is false then the first volume file must be missing.
/// </summary>
public async ValueTask<bool> IsFirstVolumeAsync() =>
(await archive.VolumesAsync.CastAsync<RarVolume>().FirstAsync()).IsFirstVolume;
/// <summary>
/// RarArchive is part of a multi-part archive.
/// </summary>
public async ValueTask<bool> IsMultipartVolumeAsync() =>
(await archive.VolumesAsync.CastAsync<RarVolume>().FirstAsync()).IsMultiVolume;
}
}

View File

@@ -0,0 +1,166 @@
using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Threading;
using SharpCompress.Common;
using SharpCompress.Common.Rar;
using SharpCompress.Common.Rar.Headers;
using SharpCompress.Compressors.Rar;
using SharpCompress.IO;
using SharpCompress.Readers;
using SharpCompress.Readers.Rar;
namespace SharpCompress.Archives.Rar;
public partial class RarArchive
#if NET8_0_OR_GREATER
: IArchiveOpenable<IRarArchive, IRarAsyncArchive>,
IMultiArchiveOpenable<IRarArchive, IRarAsyncArchive>
#endif
{
public static IRarAsyncArchive OpenAsyncArchive(
string path,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
path.NotNullOrEmpty(nameof(path));
return (IRarAsyncArchive)OpenArchive(new FileInfo(path), readerOptions);
}
public static IRarArchive OpenArchive(string filePath, ReaderOptions? options = null)
{
filePath.NotNullOrEmpty(nameof(filePath));
var fileInfo = new FileInfo(filePath);
return new RarArchive(
new SourceStream(
fileInfo,
i => RarArchiveVolumeFactory.GetFilePart(i, fileInfo),
options ?? new ReaderOptions()
)
);
}
public static IRarArchive OpenArchive(FileInfo fileInfo, ReaderOptions? options = null)
{
fileInfo.NotNull(nameof(fileInfo));
return new RarArchive(
new SourceStream(
fileInfo,
i => RarArchiveVolumeFactory.GetFilePart(i, fileInfo),
options ?? new ReaderOptions()
)
);
}
public static IRarArchive OpenArchive(Stream stream, ReaderOptions? options = null)
{
stream.NotNull(nameof(stream));
if (stream is not { CanSeek: true })
{
throw new ArgumentException("Stream must be seekable", nameof(stream));
}
return new RarArchive(new SourceStream(stream, _ => null, options ?? new ReaderOptions()));
}
public static IRarArchive OpenArchive(
IEnumerable<FileInfo> fileInfos,
ReaderOptions? readerOptions = null
)
{
fileInfos.NotNull(nameof(fileInfos));
var files = fileInfos.ToArray();
return new RarArchive(
new SourceStream(
files[0],
i => i < files.Length ? files[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
public static IRarArchive OpenArchive(
IEnumerable<Stream> streams,
ReaderOptions? readerOptions = null
)
{
streams.NotNull(nameof(streams));
var strms = streams.ToArray();
return new RarArchive(
new SourceStream(
strms[0],
i => i < strms.Length ? strms[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
public static IRarAsyncArchive OpenAsyncArchive(
Stream stream,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IRarAsyncArchive)OpenArchive(stream, readerOptions);
}
public static IRarAsyncArchive OpenAsyncArchive(
FileInfo fileInfo,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IRarAsyncArchive)OpenArchive(fileInfo, readerOptions);
}
public static IRarAsyncArchive OpenAsyncArchive(
IReadOnlyList<Stream> streams,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IRarAsyncArchive)OpenArchive(streams, readerOptions);
}
public static IRarAsyncArchive OpenAsyncArchive(
IReadOnlyList<FileInfo> fileInfos,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IRarAsyncArchive)OpenArchive(fileInfos, readerOptions);
}
public static bool IsRarFile(string filePath) => IsRarFile(new FileInfo(filePath));
public static bool IsRarFile(FileInfo fileInfo)
{
if (!fileInfo.Exists)
{
return false;
}
using Stream stream = fileInfo.OpenRead();
return IsRarFile(stream);
}
public static bool IsRarFile(Stream stream, ReaderOptions? options = null)
{
try
{
MarkHeader.Read(stream, true, false);
return true;
}
catch
{
return false;
}
}
}

View File

@@ -2,6 +2,8 @@ using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common;
using SharpCompress.Common.Rar;
using SharpCompress.Common.Rar.Headers;
@@ -12,17 +14,23 @@ using SharpCompress.Readers.Rar;
namespace SharpCompress.Archives.Rar;
public class RarArchive : AbstractArchive<RarArchiveEntry, RarVolume>
public interface IRarArchiveCommon
{
int MinVersion { get; }
int MaxVersion { get; }
}
public interface IRarArchive : IArchive, IRarArchiveCommon { }
public interface IRarAsyncArchive : IAsyncArchive, IRarArchiveCommon { }
public partial class RarArchive : AbstractArchive<RarArchiveEntry, RarVolume>, IRarArchive
{
private bool _disposed;
internal Lazy<IRarUnpack> UnpackV2017 { get; } =
new(() => new Compressors.Rar.UnpackV2017.Unpack());
internal Lazy<IRarUnpack> UnpackV1 { get; } = new(() => new Compressors.Rar.UnpackV1.Unpack());
/// <summary>
/// Constructor with a SourceStream able to handle FileInfo and Streams.
/// </summary>
/// <param name="sourceStream"></param>
private RarArchive(SourceStream sourceStream)
: base(ArchiveType.Rar, sourceStream) { }
@@ -40,15 +48,29 @@ public class RarArchive : AbstractArchive<RarArchiveEntry, RarVolume>
}
}
public override async ValueTask DisposeAsync()
{
if (!_disposed)
{
if (UnpackV1.IsValueCreated && UnpackV1.Value is IDisposable unpackV1)
{
unpackV1.Dispose();
}
_disposed = true;
await base.DisposeAsync();
}
}
protected override IEnumerable<RarArchiveEntry> LoadEntries(IEnumerable<RarVolume> volumes) =>
RarArchiveEntryFactory.GetEntries(this, volumes, ReaderOptions);
protected override IEnumerable<RarVolume> LoadVolumes(SourceStream sourceStream)
{
sourceStream.LoadAllParts(); //request all streams
sourceStream.LoadAllParts();
var streams = sourceStream.Streams.ToArray();
var i = 0;
if (streams.Length > 1 && IsRarFile(streams[1], ReaderOptions)) //test part 2 - true = multipart not split
if (streams.Length > 1 && IsRarFile(streams[1], ReaderOptions))
{
sourceStream.IsVolumes = true;
streams[1].Position = 0;
@@ -61,11 +83,16 @@ public class RarArchive : AbstractArchive<RarArchiveEntry, RarVolume>
));
}
//split mode or single file
return new StreamRarArchiveVolume(sourceStream, ReaderOptions, i++).AsEnumerable();
}
protected override IReader CreateReaderForSolidExtraction()
protected override IReader CreateReaderForSolidExtraction() =>
CreateReaderForSolidExtractionInternal();
protected override ValueTask<IAsyncReader> CreateReaderForSolidExtractionAsync() =>
new(CreateReaderForSolidExtractionInternal());
private RarReader CreateReaderForSolidExtractionInternal()
{
if (this.IsMultipartVolume())
{
@@ -74,12 +101,12 @@ public class RarArchive : AbstractArchive<RarArchiveEntry, RarVolume>
volume.Stream.Position = 0;
return volume.Stream;
});
return RarReader.Open(streams, ReaderOptions);
return (RarReader)RarReader.OpenReader(streams, ReaderOptions);
}
var stream = Volumes.First().Stream;
stream.Position = 0;
return RarReader.Open(stream, ReaderOptions);
return (RarReader)RarReader.OpenReader(stream, ReaderOptions);
}
public override bool IsSolid => Volumes.First().IsSolidArchive;
@@ -88,123 +115,4 @@ public class RarArchive : AbstractArchive<RarArchiveEntry, RarVolume>
public virtual int MinVersion => Volumes.First().MinVersion;
public virtual int MaxVersion => Volumes.First().MaxVersion;
#region Creation
/// <summary>
/// Constructor with a FileInfo object to an existing file.
/// </summary>
/// <param name="filePath"></param>
/// <param name="options"></param>
public static RarArchive Open(string filePath, ReaderOptions? options = null)
{
filePath.NotNullOrEmpty(nameof(filePath));
var fileInfo = new FileInfo(filePath);
return new RarArchive(
new SourceStream(
fileInfo,
i => RarArchiveVolumeFactory.GetFilePart(i, fileInfo),
options ?? new ReaderOptions()
)
);
}
/// <summary>
/// Constructor with a FileInfo object to an existing file.
/// </summary>
/// <param name="fileInfo"></param>
/// <param name="options"></param>
public static RarArchive Open(FileInfo fileInfo, ReaderOptions? options = null)
{
fileInfo.NotNull(nameof(fileInfo));
return new RarArchive(
new SourceStream(
fileInfo,
i => RarArchiveVolumeFactory.GetFilePart(i, fileInfo),
options ?? new ReaderOptions()
)
);
}
/// <summary>
/// Takes a seekable Stream as a source
/// </summary>
/// <param name="stream"></param>
/// <param name="options"></param>
public static RarArchive Open(Stream stream, ReaderOptions? options = null)
{
stream.NotNull(nameof(stream));
if (stream is not { CanSeek: true })
{
throw new ArgumentException("Stream must be seekable", nameof(stream));
}
return new RarArchive(new SourceStream(stream, _ => null, options ?? new ReaderOptions()));
}
/// <summary>
/// Constructor with all file parts passed in
/// </summary>
/// <param name="fileInfos"></param>
/// <param name="readerOptions"></param>
public static RarArchive Open(
IEnumerable<FileInfo> fileInfos,
ReaderOptions? readerOptions = null
)
{
fileInfos.NotNull(nameof(fileInfos));
var files = fileInfos.ToArray();
return new RarArchive(
new SourceStream(
files[0],
i => i < files.Length ? files[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
/// <summary>
/// Constructor with all stream parts passed in
/// </summary>
/// <param name="streams"></param>
/// <param name="readerOptions"></param>
public static RarArchive Open(IEnumerable<Stream> streams, ReaderOptions? readerOptions = null)
{
streams.NotNull(nameof(streams));
var strms = streams.ToArray();
return new RarArchive(
new SourceStream(
strms[0],
i => i < strms.Length ? strms[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
public static bool IsRarFile(string filePath) => IsRarFile(new FileInfo(filePath));
public static bool IsRarFile(FileInfo fileInfo)
{
if (!fileInfo.Exists)
{
return false;
}
using Stream stream = fileInfo.OpenRead();
return IsRarFile(stream);
}
public static bool IsRarFile(Stream stream, ReaderOptions? options = null)
{
try
{
MarkHeader.Read(stream, true, false);
return true;
}
catch
{
return false;
}
}
#endregion
}

View File

@@ -92,7 +92,9 @@ public class RarArchiveEntry : RarEntry, IArchiveEntry
return stream;
}
public async Task<Stream> OpenEntryStreamAsync(CancellationToken cancellationToken = default)
public async ValueTask<Stream> OpenEntryStreamAsync(
CancellationToken cancellationToken = default
)
{
RarStream stream;
if (IsRarV3)

View File

@@ -1,6 +1,7 @@
using System.IO;
using SharpCompress.Common.Rar;
using SharpCompress.Common.Rar.Headers;
using SharpCompress.IO;
namespace SharpCompress.Archives.Rar;
@@ -24,6 +25,76 @@ internal class SeekableFilePart : RarFilePart
internal override Stream GetCompressedStream()
{
Stream streamToUse;
// If the stream is a SourceStream in file mode with multi-threading enabled,
// create an independent stream to support concurrent extraction
if (
_stream is SourceStream sourceStream
&& sourceStream.IsFileMode
&& sourceStream.ReaderOptions.EnableMultiThreadedExtraction
)
{
var independentStream = sourceStream.CreateIndependentStream(0);
if (independentStream is not null)
{
streamToUse = independentStream;
streamToUse.Position = FileHeader.DataStartPosition;
if (FileHeader.R4Salt != null)
{
var cryptKey = new CryptKey3(_password!);
return new RarCryptoWrapper(streamToUse, FileHeader.R4Salt, cryptKey);
}
if (FileHeader.Rar5CryptoInfo != null)
{
var cryptKey = new CryptKey5(_password!, FileHeader.Rar5CryptoInfo);
return new RarCryptoWrapper(
streamToUse,
FileHeader.Rar5CryptoInfo.Salt,
cryptKey
);
}
return streamToUse;
}
}
// Check if the stream wraps a FileStream
Stream? underlyingStream = _stream;
if (_stream is IStreamStack streamStack)
{
underlyingStream = streamStack.BaseStream();
}
if (underlyingStream is FileStream fileStream)
{
// Create a new independent stream from the file
streamToUse = new FileStream(
fileStream.Name,
FileMode.Open,
FileAccess.Read,
FileShare.Read
);
streamToUse.Position = FileHeader.DataStartPosition;
if (FileHeader.R4Salt != null)
{
var cryptKey = new CryptKey3(_password!);
return new RarCryptoWrapper(streamToUse, FileHeader.R4Salt, cryptKey);
}
if (FileHeader.Rar5CryptoInfo != null)
{
var cryptKey = new CryptKey5(_password!, FileHeader.Rar5CryptoInfo);
return new RarCryptoWrapper(streamToUse, FileHeader.Rar5CryptoInfo.Salt, cryptKey);
}
return streamToUse;
}
// Fall back to existing behavior for stream-based sources
_stream.Position = FileHeader.DataStartPosition;
if (FileHeader.R4Salt != null)

View File

@@ -0,0 +1,169 @@
using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common;
using SharpCompress.Common.SevenZip;
using SharpCompress.Compressors.LZMA.Utilites;
using SharpCompress.IO;
using SharpCompress.Readers;
namespace SharpCompress.Archives.SevenZip;
public partial class SevenZipArchive
#if NET8_0_OR_GREATER
: IArchiveOpenable<IArchive, IAsyncArchive>,
IMultiArchiveOpenable<IArchive, IAsyncArchive>
#endif
{
public static IAsyncArchive OpenAsyncArchive(
string path,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
path.NotNullOrEmpty("path");
return (IAsyncArchive)OpenArchive(new FileInfo(path), readerOptions ?? new ReaderOptions());
}
public static IArchive OpenArchive(string filePath, ReaderOptions? readerOptions = null)
{
filePath.NotNullOrEmpty("filePath");
return OpenArchive(new FileInfo(filePath), readerOptions ?? new ReaderOptions());
}
public static IArchive OpenArchive(FileInfo fileInfo, ReaderOptions? readerOptions = null)
{
fileInfo.NotNull("fileInfo");
return new SevenZipArchive(
new SourceStream(
fileInfo,
i => ArchiveVolumeFactory.GetFilePart(i, fileInfo),
readerOptions ?? new ReaderOptions()
)
);
}
public static IArchive OpenArchive(
IEnumerable<FileInfo> fileInfos,
ReaderOptions? readerOptions = null
)
{
fileInfos.NotNull(nameof(fileInfos));
var files = fileInfos.ToArray();
return new SevenZipArchive(
new SourceStream(
files[0],
i => i < files.Length ? files[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
public static IArchive OpenArchive(
IEnumerable<Stream> streams,
ReaderOptions? readerOptions = null
)
{
streams.NotNull(nameof(streams));
var strms = streams.ToArray();
return new SevenZipArchive(
new SourceStream(
strms[0],
i => i < strms.Length ? strms[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
public static IArchive OpenArchive(Stream stream, ReaderOptions? readerOptions = null)
{
stream.NotNull("stream");
if (stream is not { CanSeek: true })
{
throw new ArgumentException("Stream must be seekable", nameof(stream));
}
return new SevenZipArchive(
new SourceStream(stream, _ => null, readerOptions ?? new ReaderOptions())
);
}
public static IAsyncArchive OpenAsyncArchive(
Stream stream,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IAsyncArchive)OpenArchive(stream, readerOptions);
}
public static IAsyncArchive OpenAsyncArchive(
FileInfo fileInfo,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IAsyncArchive)OpenArchive(fileInfo, readerOptions);
}
public static IAsyncArchive OpenAsyncArchive(
IReadOnlyList<Stream> streams,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IAsyncArchive)OpenArchive(streams, readerOptions);
}
public static IAsyncArchive OpenAsyncArchive(
IReadOnlyList<FileInfo> fileInfos,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IAsyncArchive)OpenArchive(fileInfos, readerOptions);
}
public static bool IsSevenZipFile(string filePath) => IsSevenZipFile(new FileInfo(filePath));
public static bool IsSevenZipFile(FileInfo fileInfo)
{
if (!fileInfo.Exists)
{
return false;
}
using Stream stream = fileInfo.OpenRead();
return IsSevenZipFile(stream);
}
public static bool IsSevenZipFile(Stream stream)
{
try
{
return SignatureMatch(stream);
}
catch
{
return false;
}
}
private static ReadOnlySpan<byte> Signature =>
new byte[] { (byte)'7', (byte)'z', 0xBC, 0xAF, 0x27, 0x1C };
private static bool SignatureMatch(Stream stream)
{
var reader = new BinaryReader(stream);
ReadOnlySpan<byte> signatureBytes = reader.ReadBytes(6);
return signatureBytes.SequenceEqual(Signature);
}
}

View File

@@ -12,127 +12,22 @@ using SharpCompress.Readers;
namespace SharpCompress.Archives.SevenZip;
public class SevenZipArchive : AbstractArchive<SevenZipArchiveEntry, SevenZipVolume>
public partial class SevenZipArchive : AbstractArchive<SevenZipArchiveEntry, SevenZipVolume>
{
private ArchiveDatabase? _database;
/// <summary>
/// Constructor expects a filepath to an existing file.
/// </summary>
/// <param name="filePath"></param>
/// <param name="readerOptions"></param>
public static SevenZipArchive Open(string filePath, ReaderOptions? readerOptions = null)
{
filePath.NotNullOrEmpty("filePath");
return Open(new FileInfo(filePath), readerOptions ?? new ReaderOptions());
}
/// <summary>
/// Constructor with a FileInfo object to an existing file.
/// </summary>
/// <param name="fileInfo"></param>
/// <param name="readerOptions"></param>
public static SevenZipArchive Open(FileInfo fileInfo, ReaderOptions? readerOptions = null)
{
fileInfo.NotNull("fileInfo");
return new SevenZipArchive(
new SourceStream(
fileInfo,
i => ArchiveVolumeFactory.GetFilePart(i, fileInfo),
readerOptions ?? new ReaderOptions()
)
);
}
/// <summary>
/// Constructor with all file parts passed in
/// </summary>
/// <param name="fileInfos"></param>
/// <param name="readerOptions"></param>
public static SevenZipArchive Open(
IEnumerable<FileInfo> fileInfos,
ReaderOptions? readerOptions = null
)
{
fileInfos.NotNull(nameof(fileInfos));
var files = fileInfos.ToArray();
return new SevenZipArchive(
new SourceStream(
files[0],
i => i < files.Length ? files[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
/// <summary>
/// Constructor with all stream parts passed in
/// </summary>
/// <param name="streams"></param>
/// <param name="readerOptions"></param>
public static SevenZipArchive Open(
IEnumerable<Stream> streams,
ReaderOptions? readerOptions = null
)
{
streams.NotNull(nameof(streams));
var strms = streams.ToArray();
return new SevenZipArchive(
new SourceStream(
strms[0],
i => i < strms.Length ? strms[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
/// <summary>
/// Takes a seekable Stream as a source
/// </summary>
/// <param name="stream"></param>
/// <param name="readerOptions"></param>
public static SevenZipArchive Open(Stream stream, ReaderOptions? readerOptions = null)
{
stream.NotNull("stream");
if (stream is not { CanSeek: true })
{
throw new ArgumentException("Stream must be seekable", nameof(stream));
}
return new SevenZipArchive(
new SourceStream(stream, _ => null, readerOptions ?? new ReaderOptions())
);
}
/// <summary>
/// Constructor with a SourceStream able to handle FileInfo and Streams.
/// </summary>
/// <param name="sourceStream"></param>
private SevenZipArchive(SourceStream sourceStream)
: base(ArchiveType.SevenZip, sourceStream) { }
protected override IEnumerable<SevenZipVolume> LoadVolumes(SourceStream sourceStream)
{
sourceStream.NotNull("SourceStream is null").LoadAllParts(); //request all streams
return new SevenZipVolume(sourceStream, ReaderOptions, 0).AsEnumerable(); //simple single volume or split, multivolume not supported
}
public static bool IsSevenZipFile(string filePath) => IsSevenZipFile(new FileInfo(filePath));
public static bool IsSevenZipFile(FileInfo fileInfo)
{
if (!fileInfo.Exists)
{
return false;
}
using Stream stream = fileInfo.OpenRead();
return IsSevenZipFile(stream);
}
internal SevenZipArchive()
: base(ArchiveType.SevenZip) { }
protected override IEnumerable<SevenZipVolume> LoadVolumes(SourceStream sourceStream)
{
sourceStream.NotNull("SourceStream is null").LoadAllParts();
return new SevenZipVolume(sourceStream, ReaderOptions, 0).AsEnumerable();
}
protected override IEnumerable<SevenZipArchiveEntry> LoadEntries(
IEnumerable<SevenZipVolume> volumes
)
@@ -158,7 +53,7 @@ public class SevenZipArchive : AbstractArchive<SevenZipArchiveEntry, SevenZipVol
foreach (var entry in group)
{
entry.IsSolid = isSolid;
isSolid = true; //mark others in this group as solid - same as rar behaviour.
isSolid = true;
}
}
@@ -176,31 +71,12 @@ public class SevenZipArchive : AbstractArchive<SevenZipArchiveEntry, SevenZipVol
}
}
public static bool IsSevenZipFile(Stream stream)
{
try
{
return SignatureMatch(stream);
}
catch
{
return false;
}
}
private static ReadOnlySpan<byte> Signature =>
new byte[] { (byte)'7', (byte)'z', 0xBC, 0xAF, 0x27, 0x1C };
private static bool SignatureMatch(Stream stream)
{
var reader = new BinaryReader(stream);
ReadOnlySpan<byte> signatureBytes = reader.ReadBytes(6);
return signatureBytes.SequenceEqual(Signature);
}
protected override IReader CreateReaderForSolidExtraction() =>
new SevenZipReader(ReaderOptions, this);
protected override ValueTask<IAsyncReader> CreateReaderForSolidExtractionAsync() =>
new(new SevenZipReader(ReaderOptions, this));
public override bool IsSolid =>
Entries
.Where(x => !x.IsDirectory)
@@ -212,31 +88,10 @@ public class SevenZipArchive : AbstractArchive<SevenZipArchiveEntry, SevenZipVol
public override long TotalSize =>
_database?._packSizes.Aggregate(0L, (total, packSize) => total + packSize) ?? 0;
internal sealed class SevenZipReader : AbstractReader<SevenZipEntry, SevenZipVolume>
private sealed class SevenZipReader : AbstractReader<SevenZipEntry, SevenZipVolume>
{
private readonly SevenZipArchive _archive;
private SevenZipEntry? _currentEntry;
private Stream? _currentFolderStream;
private CFolder? _currentFolder;
/// <summary>
/// Enables internal diagnostics for tests.
/// When disabled (default), diagnostics properties return null to avoid exposing internal state.
/// </summary>
internal bool DiagnosticsEnabled { get; set; }
/// <summary>
/// Current folder instance used to decide whether the solid folder stream should be reused.
/// Only available when <see cref="DiagnosticsEnabled"/> is true.
/// </summary>
internal object? DiagnosticsCurrentFolder => DiagnosticsEnabled ? _currentFolder : null;
/// <summary>
/// Current shared folder stream instance.
/// Only available when <see cref="DiagnosticsEnabled"/> is true.
/// </summary>
internal Stream? DiagnosticsCurrentFolderStream =>
DiagnosticsEnabled ? _currentFolderStream : null;
internal SevenZipReader(ReaderOptions readerOptions, SevenZipArchive archive)
: base(readerOptions, ArchiveType.SevenZip) => this._archive = archive;
@@ -252,10 +107,6 @@ public class SevenZipArchive : AbstractArchive<SevenZipArchiveEntry, SevenZipVol
_currentEntry = dir;
yield return dir;
}
// For solid archives (entries in the same folder share a compressed stream),
// we must iterate entries sequentially and maintain the folder stream state
// across entries in the same folder to avoid recreating the decompression
// stream for each file, which breaks contiguous streaming.
foreach (var entry in entries.Where(x => !x.IsDirectory))
{
_currentEntry = entry;
@@ -270,60 +121,10 @@ public class SevenZipArchive : AbstractArchive<SevenZipArchiveEntry, SevenZipVol
{
return CreateEntryStream(Stream.Null);
}
var filePart = (SevenZipFilePart)entry.FilePart;
if (!filePart.Header.HasStream)
{
// Entries with no underlying stream (e.g., empty files or anti-items)
// should return an empty stream, matching previous behavior.
return CreateEntryStream(Stream.Null);
}
var folder = filePart.Folder;
// Check if we're starting a new folder - dispose old folder stream if needed
if (folder != _currentFolder)
{
_currentFolderStream?.Dispose();
_currentFolderStream = null;
_currentFolder = folder;
}
// Create the folder stream once per folder
if (_currentFolderStream is null)
{
_currentFolderStream = _archive._database!.GetFolderStream(
_archive.Volumes.Single().Stream,
folder!,
_archive._database.PasswordProvider
);
}
// Wrap with SyncOnlyStream to work around LZMA async bugs
// Return a ReadOnlySubStream that reads from the shared folder stream
return CreateEntryStream(
new SyncOnlyStream(
new ReadOnlySubStream(_currentFolderStream, entry.Size, leaveOpen: true)
)
);
}
public override void Dispose()
{
_currentFolderStream?.Dispose();
_currentFolderStream = null;
base.Dispose();
return CreateEntryStream(new SyncOnlyStream(entry.FilePart.GetCompressedStream()));
}
}
/// <summary>
/// WORKAROUND: Forces async operations to use synchronous equivalents.
/// This is necessary because the LZMA decoder has bugs in its async implementation
/// that cause state corruption (IndexOutOfRangeException, DataErrorException).
///
/// The proper fix would be to repair the LZMA decoder's async methods
/// (LzmaStream.ReadAsync, Decoder.CodeAsync, OutWindow async operations),
/// but that requires deep changes to the decoder state machine.
/// </summary>
private sealed class SyncOnlyStream : Stream
{
private readonly Stream _baseStream;
@@ -353,7 +154,6 @@ public class SevenZipArchive : AbstractArchive<SevenZipArchiveEntry, SevenZipVol
public override void Write(byte[] buffer, int offset, int count) =>
_baseStream.Write(buffer, offset, count);
// Force async operations to use sync equivalents to avoid LZMA decoder bugs
public override Task<int> ReadAsync(
byte[] buffer,
int offset,
@@ -384,7 +184,7 @@ public class SevenZipArchive : AbstractArchive<SevenZipArchiveEntry, SevenZipVol
return Task.CompletedTask;
}
#if !NETFRAMEWORK && !NETSTANDARD2_0
#if !LEGACY_DOTNET
public override ValueTask<int> ReadAsync(
Memory<byte> buffer,
CancellationToken cancellationToken = default

View File

@@ -12,8 +12,8 @@ public class SevenZipArchiveEntry : SevenZipEntry, IArchiveEntry
public Stream OpenEntryStream() => FilePart.GetCompressedStream();
public Task<Stream> OpenEntryStreamAsync(CancellationToken cancellationToken = default) =>
Task.FromResult(OpenEntryStream());
public ValueTask<Stream> OpenEntryStreamAsync(CancellationToken cancellationToken = default) =>
new(OpenEntryStream());
public IArchive Archive { get; }

View File

@@ -0,0 +1,171 @@
using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common;
using SharpCompress.Common.Tar;
using SharpCompress.Common.Tar.Headers;
using SharpCompress.IO;
using SharpCompress.Readers;
using SharpCompress.Writers;
using SharpCompress.Writers.Tar;
namespace SharpCompress.Archives.Tar;
public partial class TarArchive
#if NET8_0_OR_GREATER
: IWritableArchiveOpenable,
IMultiArchiveOpenable<IWritableArchive, IWritableAsyncArchive>
#endif
{
public static IWritableArchive OpenArchive(string filePath, ReaderOptions? readerOptions = null)
{
filePath.NotNullOrEmpty(nameof(filePath));
return OpenArchive(new FileInfo(filePath), readerOptions ?? new ReaderOptions());
}
public static IWritableArchive OpenArchive(
FileInfo fileInfo,
ReaderOptions? readerOptions = null
)
{
fileInfo.NotNull(nameof(fileInfo));
return new TarArchive(
new SourceStream(
fileInfo,
i => ArchiveVolumeFactory.GetFilePart(i, fileInfo),
readerOptions ?? new ReaderOptions()
)
);
}
public static IWritableArchive OpenArchive(
IEnumerable<FileInfo> fileInfos,
ReaderOptions? readerOptions = null
)
{
fileInfos.NotNull(nameof(fileInfos));
var files = fileInfos.ToArray();
return new TarArchive(
new SourceStream(
files[0],
i => i < files.Length ? files[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
public static IWritableArchive OpenArchive(
IEnumerable<Stream> streams,
ReaderOptions? readerOptions = null
)
{
streams.NotNull(nameof(streams));
var strms = streams.ToArray();
return new TarArchive(
new SourceStream(
strms[0],
i => i < strms.Length ? strms[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
public static IWritableArchive OpenArchive(Stream stream, ReaderOptions? readerOptions = null)
{
stream.NotNull(nameof(stream));
if (stream is not { CanSeek: true })
{
throw new ArgumentException("Stream must be seekable", nameof(stream));
}
return new TarArchive(
new SourceStream(stream, i => null, readerOptions ?? new ReaderOptions())
);
}
public static IWritableAsyncArchive OpenAsyncArchive(
Stream stream,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IWritableAsyncArchive)OpenArchive(stream, readerOptions);
}
public static IWritableAsyncArchive OpenAsyncArchive(
string path,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IWritableAsyncArchive)OpenArchive(new FileInfo(path), readerOptions);
}
public static IWritableAsyncArchive OpenAsyncArchive(
FileInfo fileInfo,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IWritableAsyncArchive)OpenArchive(fileInfo, readerOptions);
}
public static IWritableAsyncArchive OpenAsyncArchive(
IReadOnlyList<Stream> streams,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IWritableAsyncArchive)OpenArchive(streams, readerOptions);
}
public static IWritableAsyncArchive OpenAsyncArchive(
IReadOnlyList<FileInfo> fileInfos,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IWritableAsyncArchive)OpenArchive(fileInfos, readerOptions);
}
public static bool IsTarFile(string filePath) => IsTarFile(new FileInfo(filePath));
public static bool IsTarFile(FileInfo fileInfo)
{
if (!fileInfo.Exists)
{
return false;
}
using Stream stream = fileInfo.OpenRead();
return IsTarFile(stream);
}
public static bool IsTarFile(Stream stream)
{
try
{
var tarHeader = new TarHeader(new ArchiveEncoding());
var readSucceeded = tarHeader.Read(new BinaryReader(stream));
var isEmptyArchive =
tarHeader.Name?.Length == 0
&& tarHeader.Size == 0
&& Enum.IsDefined(typeof(EntryType), tarHeader.EntryType);
return readSucceeded || isEmptyArchive;
}
catch { }
return false;
}
public static IWritableArchive CreateArchive() => new TarArchive();
public static IWritableAsyncArchive CreateAsyncArchive() => new TarArchive();
}

View File

@@ -15,132 +15,14 @@ using SharpCompress.Writers.Tar;
namespace SharpCompress.Archives.Tar;
public class TarArchive : AbstractWritableArchive<TarArchiveEntry, TarVolume>
public partial class TarArchive : AbstractWritableArchive<TarArchiveEntry, TarVolume>
{
/// <summary>
/// Constructor expects a filepath to an existing file.
/// </summary>
/// <param name="filePath"></param>
/// <param name="readerOptions"></param>
public static TarArchive Open(string filePath, ReaderOptions? readerOptions = null)
{
filePath.NotNullOrEmpty(nameof(filePath));
return Open(new FileInfo(filePath), readerOptions ?? new ReaderOptions());
}
/// <summary>
/// Constructor with a FileInfo object to an existing file.
/// </summary>
/// <param name="fileInfo"></param>
/// <param name="readerOptions"></param>
public static TarArchive Open(FileInfo fileInfo, ReaderOptions? readerOptions = null)
{
fileInfo.NotNull(nameof(fileInfo));
return new TarArchive(
new SourceStream(
fileInfo,
i => ArchiveVolumeFactory.GetFilePart(i, fileInfo),
readerOptions ?? new ReaderOptions()
)
);
}
/// <summary>
/// Constructor with all file parts passed in
/// </summary>
/// <param name="fileInfos"></param>
/// <param name="readerOptions"></param>
public static TarArchive Open(
IEnumerable<FileInfo> fileInfos,
ReaderOptions? readerOptions = null
)
{
fileInfos.NotNull(nameof(fileInfos));
var files = fileInfos.ToArray();
return new TarArchive(
new SourceStream(
files[0],
i => i < files.Length ? files[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
/// <summary>
/// Constructor with all stream parts passed in
/// </summary>
/// <param name="streams"></param>
/// <param name="readerOptions"></param>
public static TarArchive Open(IEnumerable<Stream> streams, ReaderOptions? readerOptions = null)
{
streams.NotNull(nameof(streams));
var strms = streams.ToArray();
return new TarArchive(
new SourceStream(
strms[0],
i => i < strms.Length ? strms[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
/// <summary>
/// Takes a seekable Stream as a source
/// </summary>
/// <param name="stream"></param>
/// <param name="readerOptions"></param>
public static TarArchive Open(Stream stream, ReaderOptions? readerOptions = null)
{
stream.NotNull(nameof(stream));
if (stream is not { CanSeek: true })
{
throw new ArgumentException("Stream must be seekable", nameof(stream));
}
return new TarArchive(
new SourceStream(stream, i => null, readerOptions ?? new ReaderOptions())
);
}
public static bool IsTarFile(string filePath) => IsTarFile(new FileInfo(filePath));
public static bool IsTarFile(FileInfo fileInfo)
{
if (!fileInfo.Exists)
{
return false;
}
using Stream stream = fileInfo.OpenRead();
return IsTarFile(stream);
}
public static bool IsTarFile(Stream stream)
{
try
{
var tarHeader = new TarHeader(new ArchiveEncoding());
var readSucceeded = tarHeader.Read(new BinaryReader(stream));
var isEmptyArchive =
tarHeader.Name?.Length == 0
&& tarHeader.Size == 0
&& Enum.IsDefined(typeof(EntryType), tarHeader.EntryType);
return readSucceeded || isEmptyArchive;
}
catch { }
return false;
}
protected override IEnumerable<TarVolume> LoadVolumes(SourceStream sourceStream)
{
sourceStream.NotNull("SourceStream is null").LoadAllParts(); //request all streams
return new TarVolume(sourceStream, ReaderOptions, 1).AsEnumerable(); //simple single volume or split, multivolume not supported
sourceStream.NotNull("SourceStream is null").LoadAllParts();
return new TarVolume(sourceStream, ReaderOptions, 1).AsEnumerable();
}
/// <summary>
/// Constructor with a SourceStream able to handle FileInfo and Streams.
/// </summary>
/// <param name="sourceStream"></param>
private TarArchive(SourceStream sourceStream)
: base(ArchiveType.Tar, sourceStream) { }
@@ -180,7 +62,7 @@ public class TarArchive : AbstractWritableArchive<TarArchiveEntry, TarVolume>
using (var entryStream = entry.OpenEntryStream())
{
using var memoryStream = new MemoryStream();
entryStream.CopyTo(memoryStream, Constants.BufferSize);
entryStream.CopyTo(memoryStream);
memoryStream.Position = 0;
var bytes = memoryStream.ToArray();
@@ -205,8 +87,6 @@ public class TarArchive : AbstractWritableArchive<TarArchiveEntry, TarVolume>
}
}
public static TarArchive Create() => new();
protected override TarArchiveEntry CreateEntryInternal(
string filePath,
Stream source,
@@ -259,7 +139,7 @@ public class TarArchive : AbstractWritableArchive<TarArchiveEntry, TarVolume>
}
}
protected override async Task SaveToAsync(
protected override async ValueTask SaveToAsync(
Stream stream,
WriterOptions options,
IEnumerable<TarArchiveEntry> oldEntries,
@@ -300,6 +180,13 @@ public class TarArchive : AbstractWritableArchive<TarArchiveEntry, TarVolume>
{
var stream = Volumes.Single().Stream;
stream.Position = 0;
return TarReader.Open(stream);
return TarReader.OpenReader(stream);
}
protected override ValueTask<IAsyncReader> CreateReaderForSolidExtractionAsync()
{
var stream = Volumes.Single().Stream;
stream.Position = 0;
return new((IAsyncReader)TarReader.OpenReader(stream));
}
}

View File

@@ -14,9 +14,8 @@ public class TarArchiveEntry : TarEntry, IArchiveEntry
public virtual Stream OpenEntryStream() => Parts.Single().GetCompressedStream().NotNull();
public virtual Task<Stream> OpenEntryStreamAsync(
CancellationToken cancellationToken = default
) => Task.FromResult(OpenEntryStream());
public ValueTask<Stream> OpenEntryStreamAsync(CancellationToken cancellationToken = default) =>
new(OpenEntryStream());
#region IArchiveEntry Members

View File

@@ -0,0 +1,324 @@
using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common;
using SharpCompress.Common.Zip;
using SharpCompress.Common.Zip.Headers;
using SharpCompress.IO;
using SharpCompress.Readers;
namespace SharpCompress.Archives.Zip;
public partial class ZipArchive
#if NET8_0_OR_GREATER
: IWritableArchiveOpenable,
IMultiArchiveOpenable<IWritableArchive, IWritableAsyncArchive>
#endif
{
public static IWritableArchive OpenArchive(string filePath, ReaderOptions? readerOptions = null)
{
filePath.NotNullOrEmpty(nameof(filePath));
return OpenArchive(new FileInfo(filePath), readerOptions ?? new ReaderOptions());
}
public static IWritableArchive OpenArchive(
FileInfo fileInfo,
ReaderOptions? readerOptions = null
)
{
fileInfo.NotNull(nameof(fileInfo));
return new ZipArchive(
new SourceStream(
fileInfo,
i => ZipArchiveVolumeFactory.GetFilePart(i, fileInfo),
readerOptions ?? new ReaderOptions()
)
);
}
public static IWritableArchive OpenArchive(
IEnumerable<FileInfo> fileInfos,
ReaderOptions? readerOptions = null
)
{
fileInfos.NotNull(nameof(fileInfos));
var files = fileInfos.ToArray();
return new ZipArchive(
new SourceStream(
files[0],
i => i < files.Length ? files[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
public static IWritableArchive OpenArchive(
IEnumerable<Stream> streams,
ReaderOptions? readerOptions = null
)
{
streams.NotNull(nameof(streams));
var strms = streams.ToArray();
return new ZipArchive(
new SourceStream(
strms[0],
i => i < strms.Length ? strms[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
public static IWritableArchive OpenArchive(Stream stream, ReaderOptions? readerOptions = null)
{
stream.NotNull(nameof(stream));
if (stream is not { CanSeek: true })
{
throw new ArgumentException("Stream must be seekable", nameof(stream));
}
return new ZipArchive(
new SourceStream(stream, i => null, readerOptions ?? new ReaderOptions())
);
}
public static IWritableAsyncArchive OpenAsyncArchive(
string path,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IWritableAsyncArchive)OpenArchive(path, readerOptions);
}
public static IWritableAsyncArchive OpenAsyncArchive(
Stream stream,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IWritableAsyncArchive)OpenArchive(stream, readerOptions);
}
public static IWritableAsyncArchive OpenAsyncArchive(
FileInfo fileInfo,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IWritableAsyncArchive)OpenArchive(fileInfo, readerOptions);
}
public static IWritableAsyncArchive OpenAsyncArchive(
IReadOnlyList<Stream> streams,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IWritableAsyncArchive)OpenArchive(streams, readerOptions);
}
public static IWritableAsyncArchive OpenAsyncArchive(
IReadOnlyList<FileInfo> fileInfos,
ReaderOptions? readerOptions = null,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
return (IWritableAsyncArchive)OpenArchive(fileInfos, readerOptions);
}
public static bool IsZipFile(
string filePath,
string? password = null,
int bufferSize = ReaderOptions.DefaultBufferSize
) => IsZipFile(new FileInfo(filePath), password, bufferSize);
public static bool IsZipFile(
FileInfo fileInfo,
string? password = null,
int bufferSize = ReaderOptions.DefaultBufferSize
)
{
if (!fileInfo.Exists)
{
return false;
}
using Stream stream = fileInfo.OpenRead();
return IsZipFile(stream, password, bufferSize);
}
public static bool IsZipFile(
Stream stream,
string? password = null,
int bufferSize = ReaderOptions.DefaultBufferSize
)
{
var headerFactory = new StreamingZipHeaderFactory(password, new ArchiveEncoding(), null);
try
{
if (stream is not SharpCompressStream)
{
stream = new SharpCompressStream(stream, bufferSize: bufferSize);
}
var header = headerFactory
.ReadStreamHeader(stream)
.FirstOrDefault(x => x.ZipHeaderType != ZipHeaderType.Split);
if (header is null)
{
return false;
}
return Enum.IsDefined(typeof(ZipHeaderType), header.ZipHeaderType);
}
catch (CryptographicException)
{
return true;
}
catch
{
return false;
}
}
public static bool IsZipMulti(
Stream stream,
string? password = null,
int bufferSize = ReaderOptions.DefaultBufferSize
)
{
var headerFactory = new StreamingZipHeaderFactory(password, new ArchiveEncoding(), null);
try
{
if (stream is not SharpCompressStream)
{
stream = new SharpCompressStream(stream, bufferSize: bufferSize);
}
var header = headerFactory
.ReadStreamHeader(stream)
.FirstOrDefault(x => x.ZipHeaderType != ZipHeaderType.Split);
if (header is null)
{
if (stream.CanSeek)
{
var z = new SeekableZipHeaderFactory(password, new ArchiveEncoding());
var x = z.ReadSeekableHeader(stream, useSync: true).FirstOrDefault();
return x?.ZipHeaderType == ZipHeaderType.DirectoryEntry;
}
else
{
return false;
}
}
return Enum.IsDefined(typeof(ZipHeaderType), header.ZipHeaderType);
}
catch (CryptographicException)
{
return true;
}
catch
{
return false;
}
}
public static async ValueTask<bool> IsZipFileAsync(
Stream stream,
string? password = null,
int bufferSize = ReaderOptions.DefaultBufferSize,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
var headerFactory = new StreamingZipHeaderFactory(password, new ArchiveEncoding(), null);
try
{
if (stream is not SharpCompressStream)
{
stream = new SharpCompressStream(stream, bufferSize: bufferSize);
}
var header = await headerFactory
.ReadStreamHeaderAsync(stream)
.Where(x => x.ZipHeaderType != ZipHeaderType.Split)
.FirstOrDefaultAsync(cancellationToken);
if (header is null)
{
return false;
}
return Enum.IsDefined(typeof(ZipHeaderType), header.ZipHeaderType);
}
catch (CryptographicException)
{
return true;
}
catch
{
return false;
}
}
public static IWritableArchive CreateArchive() => new ZipArchive();
public static IWritableAsyncArchive CreateAsyncArchive() => new ZipArchive();
public static async ValueTask<bool> IsZipMultiAsync(
Stream stream,
string? password = null,
int bufferSize = ReaderOptions.DefaultBufferSize,
CancellationToken cancellationToken = default
)
{
cancellationToken.ThrowIfCancellationRequested();
var headerFactory = new StreamingZipHeaderFactory(password, new ArchiveEncoding(), null);
try
{
if (stream is not SharpCompressStream)
{
stream = new SharpCompressStream(stream, bufferSize: bufferSize);
}
var header = headerFactory
.ReadStreamHeader(stream)
.FirstOrDefault(x => x.ZipHeaderType != ZipHeaderType.Split);
if (header is null)
{
if (stream.CanSeek)
{
var z = new SeekableZipHeaderFactory(password, new ArchiveEncoding());
ZipHeader? x = null;
await foreach (
var h in z.ReadSeekableHeaderAsync(stream)
.WithCancellation(cancellationToken)
)
{
x = h;
break;
}
return x?.ZipHeaderType == ZipHeaderType.DirectoryEntry;
}
else
{
return false;
}
}
return Enum.IsDefined(typeof(ZipHeaderType), header.ZipHeaderType);
}
catch (CryptographicException)
{
return true;
}
catch
{
return false;
}
}
}

View File

@@ -16,21 +16,12 @@ using SharpCompress.Writers.Zip;
namespace SharpCompress.Archives.Zip;
public class ZipArchive : AbstractWritableArchive<ZipArchiveEntry, ZipVolume>
public partial class ZipArchive : AbstractWritableArchive<ZipArchiveEntry, ZipVolume>
{
private readonly SeekableZipHeaderFactory? headerFactory;
/// <summary>
/// Gets or sets the compression level applied to files added to the archive,
/// if the compression method is set to deflate
/// </summary>
public CompressionLevel DeflateCompressionLevel { get; set; }
/// <summary>
/// Constructor with a SourceStream able to handle FileInfo and Streams.
/// </summary>
/// <param name="sourceStream"></param>
/// <param name="options"></param>
internal ZipArchive(SourceStream sourceStream)
: base(ArchiveType.Zip, sourceStream) =>
headerFactory = new SeekableZipHeaderFactory(
@@ -38,208 +29,42 @@ public class ZipArchive : AbstractWritableArchive<ZipArchiveEntry, ZipVolume>
sourceStream.ReaderOptions.ArchiveEncoding
);
/// <summary>
/// Constructor expects a filepath to an existing file.
/// </summary>
/// <param name="filePath"></param>
/// <param name="readerOptions"></param>
public static ZipArchive Open(string filePath, ReaderOptions? readerOptions = null)
{
filePath.NotNullOrEmpty(nameof(filePath));
return Open(new FileInfo(filePath), readerOptions ?? new ReaderOptions());
}
/// <summary>
/// Constructor with a FileInfo object to an existing file.
/// </summary>
/// <param name="fileInfo"></param>
/// <param name="readerOptions"></param>
public static ZipArchive Open(FileInfo fileInfo, ReaderOptions? readerOptions = null)
{
fileInfo.NotNull(nameof(fileInfo));
return new ZipArchive(
new SourceStream(
fileInfo,
i => ZipArchiveVolumeFactory.GetFilePart(i, fileInfo),
readerOptions ?? new ReaderOptions()
)
);
}
/// <summary>
/// Constructor with all file parts passed in
/// </summary>
/// <param name="fileInfos"></param>
/// <param name="readerOptions"></param>
public static ZipArchive Open(
IEnumerable<FileInfo> fileInfos,
ReaderOptions? readerOptions = null
)
{
fileInfos.NotNull(nameof(fileInfos));
var files = fileInfos.ToArray();
return new ZipArchive(
new SourceStream(
files[0],
i => i < files.Length ? files[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
/// <summary>
/// Constructor with all stream parts passed in
/// </summary>
/// <param name="streams"></param>
/// <param name="readerOptions"></param>
public static ZipArchive Open(IEnumerable<Stream> streams, ReaderOptions? readerOptions = null)
{
streams.NotNull(nameof(streams));
var strms = streams.ToArray();
return new ZipArchive(
new SourceStream(
strms[0],
i => i < strms.Length ? strms[i] : null,
readerOptions ?? new ReaderOptions()
)
);
}
/// <summary>
/// Takes a seekable Stream as a source
/// </summary>
/// <param name="stream"></param>
/// <param name="readerOptions"></param>
public static ZipArchive Open(Stream stream, ReaderOptions? readerOptions = null)
{
stream.NotNull(nameof(stream));
if (stream is not { CanSeek: true })
{
throw new ArgumentException("Stream must be seekable", nameof(stream));
}
return new ZipArchive(
new SourceStream(stream, i => null, readerOptions ?? new ReaderOptions())
);
}
public static bool IsZipFile(string filePath, string? password = null) =>
IsZipFile(new FileInfo(filePath), password);
public static bool IsZipFile(FileInfo fileInfo, string? password = null)
{
if (!fileInfo.Exists)
{
return false;
}
using Stream stream = fileInfo.OpenRead();
return IsZipFile(stream, password);
}
public static bool IsZipFile(Stream stream, string? password = null)
{
var headerFactory = new StreamingZipHeaderFactory(password, new ArchiveEncoding(), null);
try
{
if (stream is not SharpCompressStream)
{
stream = new SharpCompressStream(stream, bufferSize: Constants.BufferSize);
}
var header = headerFactory
.ReadStreamHeader(stream)
.FirstOrDefault(x => x.ZipHeaderType != ZipHeaderType.Split);
if (header is null)
{
return false;
}
return Enum.IsDefined(typeof(ZipHeaderType), header.ZipHeaderType);
}
catch (CryptographicException)
{
return true;
}
catch
{
return false;
}
}
public static bool IsZipMulti(Stream stream, string? password = null)
{
var headerFactory = new StreamingZipHeaderFactory(password, new ArchiveEncoding(), null);
try
{
if (stream is not SharpCompressStream)
{
stream = new SharpCompressStream(stream, bufferSize: Constants.BufferSize);
}
var header = headerFactory
.ReadStreamHeader(stream)
.FirstOrDefault(x => x.ZipHeaderType != ZipHeaderType.Split);
if (header is null)
{
if (stream.CanSeek) //could be multipart. Test for central directory - might not be z64 safe
{
var z = new SeekableZipHeaderFactory(password, new ArchiveEncoding());
var x = z.ReadSeekableHeader(stream).FirstOrDefault();
return x?.ZipHeaderType == ZipHeaderType.DirectoryEntry;
}
else
{
return false;
}
}
return Enum.IsDefined(typeof(ZipHeaderType), header.ZipHeaderType);
}
catch (CryptographicException)
{
return true;
}
catch
{
return false;
}
}
internal ZipArchive()
: base(ArchiveType.Zip) { }
protected override IEnumerable<ZipVolume> LoadVolumes(SourceStream stream)
{
stream.LoadAllParts(); //request all streams
stream.LoadAllParts();
stream.Position = 0;
var streams = stream.Streams.ToList();
var idx = 0;
if (streams.Count() > 1) //test part 2 - true = multipart not split
if (streams.Count() > 1)
{
streams[1].Position += 4; //skip the POST_DATA_DESCRIPTOR to prevent an exception
var isZip = IsZipFile(streams[1], ReaderOptions.Password);
streams[1].Position += 4;
var isZip = IsZipFile(streams[1], ReaderOptions.Password, ReaderOptions.BufferSize);
streams[1].Position -= 4;
if (isZip)
{
stream.IsVolumes = true;
var tmp = streams[0]; //arcs as zip, z01 ... swap the zip the end
var tmp = streams[0];
streams.RemoveAt(0);
streams.Add(tmp);
//streams[0].Position = 4; //skip the POST_DATA_DESCRIPTOR to prevent an exception
return streams.Select(a => new ZipVolume(a, ReaderOptions, idx++));
}
}
//split mode or single file
return new ZipVolume(stream, ReaderOptions, idx++).AsEnumerable();
}
internal ZipArchive()
: base(ArchiveType.Zip) { }
protected override IEnumerable<ZipArchiveEntry> LoadEntries(IEnumerable<ZipVolume> volumes)
{
var vols = volumes.ToArray();
foreach (var h in headerFactory.NotNull().ReadSeekableHeader(vols.Last().Stream))
foreach (
var h in headerFactory.NotNull().ReadSeekableHeader(vols.Last().Stream, useSync: true)
)
{
if (h != null)
{
@@ -283,6 +108,59 @@ public class ZipArchive : AbstractWritableArchive<ZipArchiveEntry, ZipVolume>
}
}
protected override async IAsyncEnumerable<ZipArchiveEntry> LoadEntriesAsync(
IAsyncEnumerable<ZipVolume> volumes
)
{
var vols = await volumes.ToListAsync();
var volsArray = vols.ToArray();
await foreach (
var h in headerFactory.NotNull().ReadSeekableHeaderAsync(volsArray.Last().Stream)
)
{
if (h != null)
{
switch (h.ZipHeaderType)
{
case ZipHeaderType.DirectoryEntry:
{
var deh = (DirectoryEntryHeader)h;
Stream s;
if (
deh.RelativeOffsetOfEntryHeader + deh.CompressedSize
> volsArray[deh.DiskNumberStart].Stream.Length
)
{
var v = volsArray.Skip(deh.DiskNumberStart).ToArray();
s = new SourceStream(
v[0].Stream,
i => i < v.Length ? v[i].Stream : null,
new ReaderOptions() { LeaveStreamOpen = true }
);
}
else
{
s = volsArray[deh.DiskNumberStart].Stream;
}
yield return new ZipArchiveEntry(
this,
new SeekableZipFilePart(headerFactory.NotNull(), deh, s)
);
}
break;
case ZipHeaderType.DirectoryEnd:
{
var bytes = ((DirectoryEndHeader)h).Comment ?? Array.Empty<byte>();
volsArray.Last().Comment = ReaderOptions.ArchiveEncoding.Decode(bytes);
yield break;
}
}
}
}
}
public void SaveTo(Stream stream) => SaveTo(stream, new WriterOptions(CompressionType.Deflate));
protected override void SaveTo(
@@ -314,7 +192,7 @@ public class ZipArchive : AbstractWritableArchive<ZipArchiveEntry, ZipVolume>
}
}
protected override async Task SaveToAsync(
protected override async ValueTask SaveToAsync(
Stream stream,
WriterOptions options,
IEnumerable<ZipArchiveEntry> oldEntries,
@@ -362,12 +240,17 @@ public class ZipArchive : AbstractWritableArchive<ZipArchiveEntry, ZipVolume>
DateTime? modified
) => new ZipWritableArchiveEntry(this, directoryPath, modified);
public static ZipArchive Create() => new();
protected override IReader CreateReaderForSolidExtraction()
{
var stream = Volumes.Single().Stream;
((IStreamStack)stream).StackSeek(0);
return ZipReader.Open(stream, ReaderOptions, Entries);
return ZipReader.OpenReader(stream, ReaderOptions, Entries);
}
protected override ValueTask<IAsyncReader> CreateReaderForSolidExtractionAsync()
{
var stream = Volumes.Single().Stream;
stream.Position = 0;
return new((IAsyncReader)ZipReader.OpenReader(stream));
}
}

View File

@@ -13,9 +13,17 @@ public class ZipArchiveEntry : ZipEntry, IArchiveEntry
public virtual Stream OpenEntryStream() => Parts.Single().GetCompressedStream().NotNull();
public virtual Task<Stream> OpenEntryStreamAsync(
public async ValueTask<Stream> OpenEntryStreamAsync(
CancellationToken cancellationToken = default
) => Task.FromResult(OpenEntryStream());
)
{
var part = Parts.Single();
if (part is SeekableZipFilePart seekablePart)
{
return (await seekablePart.GetCompressedStreamAsync(cancellationToken)).NotNull();
}
return OpenEntryStream();
}
#region IArchiveEntry Members

View File

@@ -46,7 +46,7 @@ namespace SharpCompress.Common.Ace.Headers
}
}
public AceFileHeader(ArchiveEncoding archiveEncoding)
public AceFileHeader(IArchiveEncoding archiveEncoding)
: base(archiveEncoding, AceHeaderType.FILE) { }
/// <summary>

View File

@@ -31,13 +31,13 @@ namespace SharpCompress.Common.Ace.Headers
(byte)'*',
];
public AceHeader(ArchiveEncoding archiveEncoding, AceHeaderType type)
public AceHeader(IArchiveEncoding archiveEncoding, AceHeaderType type)
{
AceHeaderType = type;
ArchiveEncoding = archiveEncoding;
}
public ArchiveEncoding ArchiveEncoding { get; }
public IArchiveEncoding ArchiveEncoding { get; }
public AceHeaderType AceHeaderType { get; }
public ushort HeaderFlags { get; set; }

View File

@@ -22,7 +22,7 @@ namespace SharpCompress.Common.Ace.Headers
public List<byte> Comment { get; set; } = new();
public byte AceVersion { get; private set; }
public AceMainHeader(ArchiveEncoding archiveEncoding)
public AceMainHeader(IArchiveEncoding archiveEncoding)
: base(archiveEncoding, AceHeaderType.MAIN) { }
/// <summary>

View File

@@ -7,7 +7,7 @@ namespace SharpCompress.Common.Arc
{
public class ArcEntryHeader
{
public ArchiveEncoding ArchiveEncoding { get; }
public IArchiveEncoding ArchiveEncoding { get; }
public CompressionType CompressionMethod { get; private set; }
public string? Name { get; private set; }
public long CompressedSize { get; private set; }
@@ -16,7 +16,7 @@ namespace SharpCompress.Common.Arc
public long OriginalSize { get; private set; }
public long DataStartPosition { get; private set; }
public ArcEntryHeader(ArchiveEncoding archiveEncoding)
public ArcEntryHeader(IArchiveEncoding archiveEncoding)
{
this.ArchiveEncoding = archiveEncoding;
}

View File

@@ -3,55 +3,11 @@ using System.Text;
namespace SharpCompress.Common;
public class ArchiveEncoding
public class ArchiveEncoding : IArchiveEncoding
{
/// <summary>
/// Default encoding to use when archive format doesn't specify one.
/// </summary>
public Encoding? Default { get; set; }
/// <summary>
/// ArchiveEncoding used by encryption schemes which don't comply with RFC 2898.
/// </summary>
public Encoding? Password { get; set; }
/// <summary>
/// Set this encoding when you want to force it for all encoding operations.
/// </summary>
public Encoding Default { get; set; } = Encoding.Default;
public Encoding Password { get; set; } = Encoding.Default;
public Encoding UTF8 { get; set; } = Encoding.UTF8;
public Encoding? Forced { get; set; }
/// <summary>
/// Set this when you want to use a custom method for all decoding operations.
/// </summary>
/// <returns>string Func(bytes, index, length)</returns>
public Func<byte[], int, int, string>? CustomDecoder { get; set; }
public ArchiveEncoding()
: this(Encoding.Default, Encoding.Default) { }
public ArchiveEncoding(Encoding def, Encoding password)
{
Default = def;
Password = password;
}
#if !NETFRAMEWORK
static ArchiveEncoding() => Encoding.RegisterProvider(CodePagesEncodingProvider.Instance);
#endif
public string Decode(byte[] bytes) => Decode(bytes, 0, bytes.Length);
public string Decode(byte[] bytes, int start, int length) =>
GetDecoder().Invoke(bytes, start, length);
public string DecodeUTF8(byte[] bytes) => Encoding.UTF8.GetString(bytes, 0, bytes.Length);
public byte[] Encode(string str) => GetEncoding().GetBytes(str);
public Encoding GetEncoding() => Forced ?? Default ?? Encoding.UTF8;
public Encoding GetPasswordEncoding() => Password ?? Encoding.UTF8;
public Func<byte[], int, int, string> GetDecoder() =>
CustomDecoder ?? ((bytes, index, count) => GetEncoding().GetString(bytes, index, count));
public Func<byte[], int, int, EncodingType, string>? CustomDecoder { get; set; }
}

View File

@@ -0,0 +1,87 @@
using System;
using System.Text;
namespace SharpCompress.Common;
/// <summary>
/// Specifies the type of encoding to use.
/// </summary>
public enum EncodingType
{
/// <summary>
/// Uses the default encoding.
/// </summary>
Default,
/// <summary>
/// Uses UTF-8 encoding.
/// </summary>
UTF8,
}
/// <summary>
/// Provides extension methods for archive encoding.
/// </summary>
public static class ArchiveEncodingExtensions
{
#if !NETFRAMEWORK
/// <summary>
/// Registers the code pages encoding provider.
/// </summary>
static ArchiveEncodingExtensions() =>
Encoding.RegisterProvider(CodePagesEncodingProvider.Instance);
#endif
extension(IArchiveEncoding encoding)
{
/// <summary>
/// Gets the encoding based on the archive encoding settings.
/// </summary>
/// <param name="useUtf8">Whether to use UTF-8.</param>
/// <returns>The encoding.</returns>
public Encoding GetEncoding(bool useUtf8 = false) =>
encoding.Forced ?? (useUtf8 ? encoding.UTF8 : encoding.Default);
/// <summary>
/// Gets the decoder function for the archive encoding.
/// </summary>
/// <returns>The decoder function.</returns>
public Func<byte[], int, int, EncodingType, string> GetDecoder() =>
encoding.CustomDecoder
?? (
(bytes, index, count, type) =>
encoding.GetEncoding(type == EncodingType.UTF8).GetString(bytes, index, count)
);
/// <summary>
/// Encodes a string using the default encoding.
/// </summary>
/// <param name="str">The string to encode.</param>
/// <returns>The encoded bytes.</returns>
public byte[] Encode(string str) => encoding.Default.GetBytes(str);
/// <summary>
/// Decodes bytes using the specified encoding type.
/// </summary>
/// <param name="bytes">The bytes to decode.</param>
/// <param name="type">The encoding type.</param>
/// <returns>The decoded string.</returns>
public string Decode(byte[] bytes, EncodingType type = EncodingType.Default) =>
encoding.Decode(bytes, 0, bytes.Length, type);
/// <summary>
/// Decodes a portion of bytes using the specified encoding type.
/// </summary>
/// <param name="bytes">The bytes to decode.</param>
/// <param name="start">The start index.</param>
/// <param name="length">The length.</param>
/// <param name="type">The encoding type.</param>
/// <returns>The decoded string.</returns>
public string Decode(
byte[] bytes,
int start,
int length,
EncodingType type = EncodingType.Default
) => encoding.GetDecoder()(bytes, start, length, type);
}
}

View File

@@ -0,0 +1,103 @@
using System;
using System.Buffers.Binary;
using System.IO;
using System.Threading;
using System.Threading.Tasks;
namespace SharpCompress.Common
{
public sealed class AsyncBinaryReader : IDisposable
{
private readonly Stream _stream;
private readonly Stream _originalStream;
private readonly bool _leaveOpen;
private readonly byte[] _buffer = new byte[8];
private bool _disposed;
public AsyncBinaryReader(Stream stream, bool leaveOpen = false, int bufferSize = 4096)
{
_originalStream = stream ?? throw new ArgumentNullException(nameof(stream));
_leaveOpen = leaveOpen;
// Use the stream directly without wrapping in BufferedStream
// BufferedStream uses synchronous Read internally which doesn't work with async-only streams
// SharpCompress uses SharpCompressStream for buffering which supports true async reads
_stream = stream;
}
public Stream BaseStream => _stream;
public async ValueTask<byte> ReadByteAsync(CancellationToken ct = default)
{
await _stream.ReadExactAsync(_buffer, 0, 1, ct).ConfigureAwait(false);
return _buffer[0];
}
public async ValueTask<ushort> ReadUInt16Async(CancellationToken ct = default)
{
await _stream.ReadExactAsync(_buffer, 0, 2, ct).ConfigureAwait(false);
return BinaryPrimitives.ReadUInt16LittleEndian(_buffer);
}
public async ValueTask<uint> ReadUInt32Async(CancellationToken ct = default)
{
await _stream.ReadExactAsync(_buffer, 0, 4, ct).ConfigureAwait(false);
return BinaryPrimitives.ReadUInt32LittleEndian(_buffer);
}
public async ValueTask<ulong> ReadUInt64Async(CancellationToken ct = default)
{
await _stream.ReadExactAsync(_buffer, 0, 8, ct).ConfigureAwait(false);
return BinaryPrimitives.ReadUInt64LittleEndian(_buffer);
}
public async ValueTask ReadBytesAsync(
byte[] bytes,
int offset,
int count,
CancellationToken ct = default
)
{
await _stream.ReadExactAsync(bytes, offset, count, ct).ConfigureAwait(false);
}
public async ValueTask SkipAsync(int count, CancellationToken ct = default)
{
await _stream.SkipAsync(count, ct).ConfigureAwait(false);
}
public void Dispose()
{
if (_disposed)
{
return;
}
_disposed = true;
// Dispose the original stream if we own it
if (!_leaveOpen)
{
_originalStream.Dispose();
}
}
#if NET8_0_OR_GREATER
public async ValueTask DisposeAsync()
{
if (_disposed)
{
return;
}
_disposed = true;
// Dispose the original stream if we own it
if (!_leaveOpen)
{
await _originalStream.DisposeAsync().ConfigureAwait(false);
}
}
#endif
}
}

View File

@@ -1,10 +0,0 @@
namespace SharpCompress.Common;
public static class Constants
{
/// <summary>
/// The default buffer size for stream operations, matching .NET's Stream.CopyTo default of 81920 bytes.
/// This can be modified globally at runtime.
/// </summary>
public static int BufferSize { get; set; } = 81920;
}

View File

@@ -56,7 +56,7 @@ public class EntryStream : Stream, IStreamStack
/// <summary>
/// Asynchronously skip the rest of the entry stream.
/// </summary>
public async Task SkipEntryAsync(CancellationToken cancellationToken = default)
public async ValueTask SkipEntryAsync(CancellationToken cancellationToken = default)
{
await this.SkipAsync(cancellationToken).ConfigureAwait(false);
_completed = true;
@@ -93,7 +93,7 @@ public class EntryStream : Stream, IStreamStack
_stream.Dispose();
}
#if !NETFRAMEWORK && !NETSTANDARD2_0
#if !LEGACY_DOTNET
public override async ValueTask DisposeAsync()
{
if (_isDisposed)
@@ -171,7 +171,7 @@ public class EntryStream : Stream, IStreamStack
return read;
}
#if !NETFRAMEWORK && !NETSTANDARD2_0
#if !LEGACY_DOTNET
public override async ValueTask<int> ReadAsync(
Memory<byte> buffer,
CancellationToken cancellationToken = default

View File

@@ -124,11 +124,11 @@ internal static class ExtractionMethods
}
}
public static async Task WriteEntryToDirectoryAsync(
public static async ValueTask WriteEntryToDirectoryAsync(
IEntry entry,
string destinationDirectory,
ExtractionOptions? options,
Func<string, ExtractionOptions?, CancellationToken, Task> writeAsync,
Func<string, ExtractionOptions?, CancellationToken, ValueTask> writeAsync,
CancellationToken cancellationToken = default
)
{
@@ -197,11 +197,11 @@ internal static class ExtractionMethods
}
}
public static async Task WriteEntryToFileAsync(
public static async ValueTask WriteEntryToFileAsync(
IEntry entry,
string destinationFileName,
ExtractionOptions? options,
Func<string, FileMode, CancellationToken, Task> openAndWriteAsync,
Func<string, FileMode, CancellationToken, ValueTask> openAndWriteAsync,
CancellationToken cancellationToken = default
)
{

View File

@@ -1,12 +1,14 @@
using System.IO;
using System.Threading;
using System.Threading.Tasks;
namespace SharpCompress.Common;
public abstract class FilePart
{
protected FilePart(ArchiveEncoding archiveEncoding) => ArchiveEncoding = archiveEncoding;
protected FilePart(IArchiveEncoding archiveEncoding) => ArchiveEncoding = archiveEncoding;
internal ArchiveEncoding ArchiveEncoding { get; }
internal IArchiveEncoding ArchiveEncoding { get; }
internal abstract string? FilePartName { get; }
public int Index { get; set; }
@@ -14,4 +16,8 @@ public abstract class FilePart
internal abstract Stream? GetCompressedStream();
internal abstract Stream? GetRawStream();
internal bool Skipped { get; set; }
internal virtual ValueTask<Stream?> GetCompressedStreamAsync(
CancellationToken cancellationToken = default
) => new(GetCompressedStream());
}

View File

@@ -40,6 +40,6 @@ public class GZipEntry : Entry
internal static IEnumerable<GZipEntry> GetEntries(Stream stream, OptionsBase options)
{
yield return new GZipEntry(new GZipFilePart(stream, options.ArchiveEncoding));
yield return new GZipEntry(GZipFilePart.Create(stream, options.ArchiveEncoding));
}
}

View File

@@ -2,6 +2,8 @@ using System;
using System.Buffers.Binary;
using System.Collections.Generic;
using System.IO;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common.Tar.Headers;
using SharpCompress.Compressors;
using SharpCompress.Compressors.Deflate;
@@ -13,28 +15,58 @@ internal sealed class GZipFilePart : FilePart
private string? _name;
private readonly Stream _stream;
internal GZipFilePart(Stream stream, ArchiveEncoding archiveEncoding)
: base(archiveEncoding)
internal static GZipFilePart Create(Stream stream, IArchiveEncoding archiveEncoding)
{
_stream = stream;
ReadAndValidateGzipHeader();
var part = new GZipFilePart(stream, archiveEncoding);
part.ReadAndValidateGzipHeader();
if (stream.CanSeek)
{
var position = stream.Position;
stream.Position = stream.Length - 8;
ReadTrailer();
part.ReadTrailer();
stream.Position = position;
EntryStartPosition = position;
part.EntryStartPosition = position;
}
else
{
// For non-seekable streams, we can't read the trailer or track position.
// Set to 0 since the stream will be read sequentially from its current position.
EntryStartPosition = 0;
part.EntryStartPosition = 0;
}
return part;
}
internal long EntryStartPosition { get; }
internal static async ValueTask<GZipFilePart> CreateAsync(
Stream stream,
IArchiveEncoding archiveEncoding,
CancellationToken cancellationToken = default
)
{
var part = new GZipFilePart(stream, archiveEncoding);
await part.ReadAndValidateGzipHeaderAsync(cancellationToken);
if (stream.CanSeek)
{
var position = stream.Position;
stream.Position = stream.Length - 8;
await part.ReadTrailerAsync(cancellationToken);
stream.Position = position;
part.EntryStartPosition = position;
}
else
{
// For non-seekable streams, we can't read the trailer or track position.
// Set to 0 since the stream will be read sequentially from its current position.
part.EntryStartPosition = 0;
}
return part;
}
private GZipFilePart(Stream stream, IArchiveEncoding archiveEncoding)
: base(archiveEncoding) => _stream = stream;
internal long EntryStartPosition { get; private set; }
internal DateTime? DateModified { get; private set; }
internal uint? Crc { get; private set; }
@@ -51,12 +83,22 @@ internal sealed class GZipFilePart : FilePart
{
// Read and potentially verify the GZIP trailer: CRC32 and size mod 2^32
Span<byte> trailer = stackalloc byte[8];
var n = _stream.Read(trailer);
_stream.ReadFully(trailer);
Crc = BinaryPrimitives.ReadUInt32LittleEndian(trailer);
UncompressedSize = BinaryPrimitives.ReadUInt32LittleEndian(trailer.Slice(4));
}
private async ValueTask ReadTrailerAsync(CancellationToken cancellationToken = default)
{
// Read and potentially verify the GZIP trailer: CRC32 and size mod 2^32
var trailer = new byte[8];
_ = await _stream.ReadFullyAsync(trailer, 0, 8, cancellationToken);
Crc = BinaryPrimitives.ReadUInt32LittleEndian(trailer);
UncompressedSize = BinaryPrimitives.ReadUInt32LittleEndian(trailer.AsSpan().Slice(4));
}
private void ReadAndValidateGzipHeader()
{
// read the header on the first read
@@ -109,6 +151,61 @@ internal sealed class GZipFilePart : FilePart
}
}
private async ValueTask ReadAndValidateGzipHeaderAsync(
CancellationToken cancellationToken = default
)
{
// read the header on the first read
var header = new byte[10];
var n = await _stream.ReadAsync(header, 0, 10, cancellationToken);
// workitem 8501: handle edge case (decompress empty stream)
if (n == 0)
{
return;
}
if (n != 10)
{
throw new ZlibException("Not a valid GZIP stream.");
}
if (header[0] != 0x1F || header[1] != 0x8B || header[2] != 8)
{
throw new ZlibException("Bad GZIP header.");
}
var timet = BinaryPrimitives.ReadInt32LittleEndian(header.AsSpan().Slice(4));
DateModified = TarHeader.EPOCH.AddSeconds(timet);
if ((header[3] & 0x04) == 0x04)
{
// read and discard extra field
var lengthField = new byte[2];
_ = await _stream.ReadAsync(lengthField, 0, 2, cancellationToken);
var extraLength = (short)(lengthField[0] + (lengthField[1] * 256));
var extra = new byte[extraLength];
if (!await _stream.ReadFullyAsync(extra, cancellationToken))
{
throw new ZlibException("Unexpected end-of-file reading GZIP header.");
}
}
if ((header[3] & 0x08) == 0x08)
{
_name = await ReadZeroTerminatedStringAsync(_stream, cancellationToken);
}
if ((header[3] & 0x10) == 0x010)
{
await ReadZeroTerminatedStringAsync(_stream, cancellationToken);
}
if ((header[3] & 0x02) == 0x02)
{
var buf = new byte[1];
_ = await _stream.ReadAsync(buf, 0, 1, cancellationToken); // CRC16, ignore
}
}
private string ReadZeroTerminatedString(Stream stream)
{
Span<byte> buf1 = stackalloc byte[1];
@@ -134,4 +231,33 @@ internal sealed class GZipFilePart : FilePart
var buffer = list.ToArray();
return ArchiveEncoding.Decode(buffer);
}
private async ValueTask<string> ReadZeroTerminatedStringAsync(
Stream stream,
CancellationToken cancellationToken = default
)
{
var buf1 = new byte[1];
var list = new List<byte>();
var done = false;
do
{
// workitem 7740
var n = await stream.ReadAsync(buf1, 0, 1, cancellationToken);
if (n != 1)
{
throw new ZlibException("Unexpected EOF reading GZIP header.");
}
if (buf1[0] == 0)
{
done = true;
}
else
{
list.Add(buf1[0]);
}
} while (!done);
var buffer = list.ToArray();
return ArchiveEncoding.Decode(buffer);
}
}

View File

@@ -0,0 +1,36 @@
using System;
using System.Text;
namespace SharpCompress.Common;
/// <summary>
/// Defines the encoding settings for archives.
/// </summary>
public interface IArchiveEncoding
{
/// <summary>
/// Default encoding to use when archive format doesn't specify one. Required and defaults to Encoding.Default.
/// </summary>
public Encoding Default { get; set; }
/// <summary>
/// ArchiveEncoding used by encryption schemes which don't comply with RFC 2898. Required and defaults to Encoding.Default.
/// </summary>
public Encoding Password { get; set; }
/// <summary>
/// Default encoding to use when archive format specifies UTF-8 encoding. Required and defaults to Encoding.UTF8.
/// </summary>
public Encoding UTF8 { get; set; }
/// <summary>
/// Set this encoding when you want to force it for all encoding operations.
/// </summary>
public Encoding? Forced { get; set; }
/// <summary>
/// Set this when you want to use a custom method for all decoding operations.
/// </summary>
/// <returns>string Func(bytes, index, length, EncodingType)</returns>
public Func<byte[], int, int, EncodingType, string>? CustomDecoder { get; set; }
}

View File

@@ -7,5 +7,5 @@ public class OptionsBase
/// </summary>
public bool LeaveStreamOpen { get; set; } = true;
public ArchiveEncoding ArchiveEncoding { get; set; } = new();
public IArchiveEncoding ArchiveEncoding { get; set; } = new ArchiveEncoding();
}

View File

@@ -13,7 +13,7 @@ internal class RarHeader : IRarHeader
internal static RarHeader? TryReadBase(
RarCrcBinaryReader reader,
bool isRar5,
ArchiveEncoding archiveEncoding
IArchiveEncoding archiveEncoding
)
{
try
@@ -26,7 +26,7 @@ internal class RarHeader : IRarHeader
}
}
private RarHeader(RarCrcBinaryReader reader, bool isRar5, ArchiveEncoding archiveEncoding)
private RarHeader(RarCrcBinaryReader reader, bool isRar5, IArchiveEncoding archiveEncoding)
{
_headerType = HeaderType.Null;
_isRar5 = isRar5;
@@ -115,7 +115,7 @@ internal class RarHeader : IRarHeader
protected int HeaderSize { get; }
internal ArchiveEncoding ArchiveEncoding { get; }
internal IArchiveEncoding ArchiveEncoding { get; }
/// <summary>
/// Extra header size.

View File

@@ -15,7 +15,7 @@ internal class SevenZipFilePart : FilePart
ArchiveDatabase database,
int index,
CFileItem fileEntry,
ArchiveEncoding archiveEncoding
IArchiveEncoding archiveEncoding
)
: base(archiveEncoding)
{

View File

@@ -11,7 +11,7 @@ internal sealed class TarHeader
internal static readonly DateTime EPOCH = new(1970, 1, 1, 0, 0, 0, DateTimeKind.Utc);
public TarHeader(
ArchiveEncoding archiveEncoding,
IArchiveEncoding archiveEncoding,
TarHeaderWriteFormat writeFormat = TarHeaderWriteFormat.GNU_TAR_LONG_LINK
)
{
@@ -30,7 +30,7 @@ internal sealed class TarHeader
internal DateTime LastModifiedTime { get; set; }
internal EntryType EntryType { get; set; }
internal Stream? PackedStream { get; set; }
internal ArchiveEncoding ArchiveEncoding { get; }
internal IArchiveEncoding ArchiveEncoding { get; }
internal const int BLOCK_SIZE = 512;

View File

@@ -54,7 +54,7 @@ public class TarEntry : Entry
StreamingMode mode,
Stream stream,
CompressionType compressionType,
ArchiveEncoding archiveEncoding
IArchiveEncoding archiveEncoding
)
{
foreach (var header in TarHeaderFactory.ReadHeader(mode, stream, archiveEncoding))

View File

@@ -1,5 +1,6 @@
using System.IO;
using SharpCompress.Common.Tar.Headers;
using SharpCompress.IO;
namespace SharpCompress.Common.Tar;
@@ -20,8 +21,45 @@ internal sealed class TarFilePart : FilePart
internal override Stream GetCompressedStream()
{
if (_seekableStream != null)
if (_seekableStream is not null)
{
// If the seekable stream is a SourceStream in file mode with multi-threading enabled,
// create an independent stream to support concurrent extraction
if (
_seekableStream is SourceStream sourceStream
&& sourceStream.IsFileMode
&& sourceStream.ReaderOptions.EnableMultiThreadedExtraction
)
{
var independentStream = sourceStream.CreateIndependentStream(0);
if (independentStream is not null)
{
independentStream.Position = Header.DataStartPosition ?? 0;
return new TarReadOnlySubStream(independentStream, Header.Size);
}
}
// Check if the seekable stream wraps a FileStream
Stream? underlyingStream = _seekableStream;
if (_seekableStream is IStreamStack streamStack)
{
underlyingStream = streamStack.BaseStream();
}
if (underlyingStream is FileStream fileStream)
{
// Create a new independent stream from the file
var independentStream = new FileStream(
fileStream.Name,
FileMode.Open,
FileAccess.Read,
FileShare.Read
);
independentStream.Position = Header.DataStartPosition ?? 0;
return new TarReadOnlySubStream(independentStream, Header.Size);
}
// Fall back to existing behavior for stream-based sources
_seekableStream.Position = Header.DataStartPosition ?? 0;
return new TarReadOnlySubStream(_seekableStream, Header.Size);
}

View File

@@ -10,7 +10,7 @@ internal static class TarHeaderFactory
internal static IEnumerable<TarHeader?> ReadHeader(
StreamingMode mode,
Stream stream,
ArchiveEncoding archiveEncoding
IArchiveEncoding archiveEncoding
)
{
while (true)

View File

@@ -66,7 +66,7 @@ internal class TarReadOnlySubStream : SharpCompressStream, IStreamStack
base.Dispose(disposing);
}
#if !NETFRAMEWORK && !NETSTANDARD2_0
#if !LEGACY_DOTNET
public override async System.Threading.Tasks.ValueTask DisposeAsync()
{
if (_isDisposed)
@@ -170,7 +170,7 @@ internal class TarReadOnlySubStream : SharpCompressStream, IStreamStack
return read;
}
#if !NETFRAMEWORK && !NETSTANDARD2_0
#if !LEGACY_DOTNET
public override async System.Threading.Tasks.ValueTask<int> ReadAsync(
System.Memory<byte> buffer,
System.Threading.CancellationToken cancellationToken = default

View File

@@ -1,4 +1,5 @@
using System.IO;
using System.Threading.Tasks;
namespace SharpCompress.Common.Zip.Headers;
@@ -19,6 +20,19 @@ internal class DirectoryEndHeader : ZipHeader
Comment = reader.ReadBytes(CommentLength);
}
internal override async ValueTask Read(AsyncBinaryReader reader)
{
VolumeNumber = await reader.ReadUInt16Async();
FirstVolumeWithDirectory = await reader.ReadUInt16Async();
TotalNumberOfEntriesInDisk = await reader.ReadUInt16Async();
TotalNumberOfEntries = await reader.ReadUInt16Async();
DirectorySize = await reader.ReadUInt32Async();
DirectoryStartOffsetRelativeToDisk = await reader.ReadUInt32Async();
CommentLength = await reader.ReadUInt16Async();
Comment = new byte[CommentLength];
await reader.ReadBytesAsync(Comment, 0, CommentLength);
}
public ushort VolumeNumber { get; private set; }
public ushort FirstVolumeWithDirectory { get; private set; }

View File

@@ -1,11 +1,12 @@
using System.IO;
using System.Linq;
using System.Threading.Tasks;
namespace SharpCompress.Common.Zip.Headers;
internal class DirectoryEntryHeader : ZipFileEntry
{
public DirectoryEntryHeader(ArchiveEncoding archiveEncoding)
public DirectoryEntryHeader(IArchiveEncoding archiveEncoding)
: base(ZipHeaderType.DirectoryEntry, archiveEncoding) { }
internal override void Read(BinaryReader reader)
@@ -31,7 +32,39 @@ internal class DirectoryEntryHeader : ZipFileEntry
var extra = reader.ReadBytes(extraLength);
var comment = reader.ReadBytes(commentLength);
// According to .ZIP File Format Specification
ProcessReadData(name, extra, comment);
}
internal override async ValueTask Read(AsyncBinaryReader reader)
{
Version = await reader.ReadUInt16Async();
VersionNeededToExtract = await reader.ReadUInt16Async();
Flags = (HeaderFlags)await reader.ReadUInt16Async();
CompressionMethod = (ZipCompressionMethod)await reader.ReadUInt16Async();
OriginalLastModifiedTime = LastModifiedTime = await reader.ReadUInt16Async();
OriginalLastModifiedDate = LastModifiedDate = await reader.ReadUInt16Async();
Crc = await reader.ReadUInt32Async();
CompressedSize = await reader.ReadUInt32Async();
UncompressedSize = await reader.ReadUInt32Async();
var nameLength = await reader.ReadUInt16Async();
var extraLength = await reader.ReadUInt16Async();
var commentLength = await reader.ReadUInt16Async();
DiskNumberStart = await reader.ReadUInt16Async();
InternalFileAttributes = await reader.ReadUInt16Async();
ExternalFileAttributes = await reader.ReadUInt32Async();
RelativeOffsetOfEntryHeader = await reader.ReadUInt32Async();
var name = new byte[nameLength];
var extra = new byte[extraLength];
var comment = new byte[commentLength];
await reader.ReadBytesAsync(name, 0, nameLength);
await reader.ReadBytesAsync(extra, 0, extraLength);
await reader.ReadBytesAsync(comment, 0, commentLength);
ProcessReadData(name, extra, comment);
}
private void ProcessReadData(byte[] name, byte[] extra, byte[] comment)
{
//
// For example: https://pkware.cachefly.net/webdocs/casestudies/APPNOTE.TXT
//
@@ -41,8 +74,8 @@ internal class DirectoryEntryHeader : ZipFileEntry
if (Flags.HasFlag(HeaderFlags.Efs))
{
Name = ArchiveEncoding.DecodeUTF8(name);
Comment = ArchiveEncoding.DecodeUTF8(comment);
Name = ArchiveEncoding.Decode(name, EncodingType.UTF8);
Comment = ArchiveEncoding.Decode(comment, EncodingType.UTF8);
}
else
{

View File

@@ -1,4 +1,5 @@
using System.IO;
using System.Threading.Tasks;
namespace SharpCompress.Common.Zip.Headers;
@@ -8,4 +9,6 @@ internal class IgnoreHeader : ZipHeader
: base(type) { }
internal override void Read(BinaryReader reader) { }
internal override ValueTask Read(AsyncBinaryReader reader) => default;
}

View File

@@ -1,13 +1,12 @@
using System.IO;
using System.Linq;
using System.Threading.Tasks;
namespace SharpCompress.Common.Zip.Headers;
internal class LocalEntryHeader : ZipFileEntry
internal class LocalEntryHeader(IArchiveEncoding archiveEncoding)
: ZipFileEntry(ZipHeaderType.LocalEntry, archiveEncoding)
{
public LocalEntryHeader(ArchiveEncoding archiveEncoding)
: base(ZipHeaderType.LocalEntry, archiveEncoding) { }
internal override void Read(BinaryReader reader)
{
Version = reader.ReadUInt16();
@@ -23,7 +22,31 @@ internal class LocalEntryHeader : ZipFileEntry
var name = reader.ReadBytes(nameLength);
var extra = reader.ReadBytes(extraLength);
// According to .ZIP File Format Specification
ProcessReadData(name, extra);
}
internal override async ValueTask Read(AsyncBinaryReader reader)
{
Version = await reader.ReadUInt16Async();
Flags = (HeaderFlags)await reader.ReadUInt16Async();
CompressionMethod = (ZipCompressionMethod)await reader.ReadUInt16Async();
OriginalLastModifiedTime = LastModifiedTime = await reader.ReadUInt16Async();
OriginalLastModifiedDate = LastModifiedDate = await reader.ReadUInt16Async();
Crc = await reader.ReadUInt32Async();
CompressedSize = await reader.ReadUInt32Async();
UncompressedSize = await reader.ReadUInt32Async();
var nameLength = await reader.ReadUInt16Async();
var extraLength = await reader.ReadUInt16Async();
var name = new byte[nameLength];
var extra = new byte[extraLength];
await reader.ReadBytesAsync(name, 0, nameLength);
await reader.ReadBytesAsync(extra, 0, extraLength);
ProcessReadData(name, extra);
}
private void ProcessReadData(byte[] name, byte[] extra)
{
//
// For example: https://pkware.cachefly.net/webdocs/casestudies/APPNOTE.TXT
//
@@ -33,7 +56,7 @@ internal class LocalEntryHeader : ZipFileEntry
if (Flags.HasFlag(HeaderFlags.Efs))
{
Name = ArchiveEncoding.DecodeUTF8(name);
Name = ArchiveEncoding.Decode(name, EncodingType.UTF8);
}
else
{

View File

@@ -1,5 +1,6 @@
using System;
using System.IO;
using System.Threading.Tasks;
namespace SharpCompress.Common.Zip.Headers;
@@ -9,4 +10,7 @@ internal class SplitHeader : ZipHeader
: base(ZipHeaderType.Split) { }
internal override void Read(BinaryReader reader) => throw new NotImplementedException();
internal override ValueTask Read(AsyncBinaryReader reader) =>
throw new NotImplementedException();
}

View File

@@ -1,4 +1,5 @@
using System.IO;
using System.Threading.Tasks;
namespace SharpCompress.Common.Zip.Headers;
@@ -26,6 +27,24 @@ internal class Zip64DirectoryEndHeader : ZipHeader
);
}
internal override async ValueTask Read(AsyncBinaryReader reader)
{
SizeOfDirectoryEndRecord = (long)await reader.ReadUInt64Async();
VersionMadeBy = await reader.ReadUInt16Async();
VersionNeededToExtract = await reader.ReadUInt16Async();
VolumeNumber = await reader.ReadUInt32Async();
FirstVolumeWithDirectory = await reader.ReadUInt32Async();
TotalNumberOfEntriesInDisk = (long)await reader.ReadUInt64Async();
TotalNumberOfEntries = (long)await reader.ReadUInt64Async();
DirectorySize = (long)await reader.ReadUInt64Async();
DirectoryStartOffsetRelativeToDisk = (long)await reader.ReadUInt64Async();
var size = (int)(
SizeOfDirectoryEndRecord - SIZE_OF_FIXED_HEADER_DATA_EXCEPT_SIGNATURE_AND_SIZE_FIELDS
);
DataSector = new byte[size];
await reader.ReadBytesAsync(DataSector, 0, size);
}
private const int SIZE_OF_FIXED_HEADER_DATA_EXCEPT_SIGNATURE_AND_SIZE_FIELDS = 44;
public long SizeOfDirectoryEndRecord { get; private set; }

View File

@@ -1,12 +1,10 @@
using System.IO;
using System.Threading.Tasks;
namespace SharpCompress.Common.Zip.Headers;
internal class Zip64DirectoryEndLocatorHeader : ZipHeader
internal class Zip64DirectoryEndLocatorHeader() : ZipHeader(ZipHeaderType.Zip64DirectoryEndLocator)
{
public Zip64DirectoryEndLocatorHeader()
: base(ZipHeaderType.Zip64DirectoryEndLocator) { }
internal override void Read(BinaryReader reader)
{
FirstVolumeWithDirectory = reader.ReadUInt32();
@@ -14,6 +12,13 @@ internal class Zip64DirectoryEndLocatorHeader : ZipHeader
TotalNumberOfVolumes = reader.ReadUInt32();
}
internal override async ValueTask Read(AsyncBinaryReader reader)
{
FirstVolumeWithDirectory = await reader.ReadUInt32Async();
RelativeOffsetOfTheEndOfDirectoryRecord = (long)await reader.ReadUInt64Async();
TotalNumberOfVolumes = await reader.ReadUInt32Async();
}
public uint FirstVolumeWithDirectory { get; private set; }
public long RelativeOffsetOfTheEndOfDirectoryRecord { get; private set; }

View File

@@ -2,18 +2,14 @@ using System;
using System.Buffers.Binary;
using System.Collections.Generic;
using System.IO;
using System.Threading;
using System.Threading.Tasks;
namespace SharpCompress.Common.Zip.Headers;
internal abstract class ZipFileEntry : ZipHeader
internal abstract class ZipFileEntry(ZipHeaderType type, IArchiveEncoding archiveEncoding)
: ZipHeader(type)
{
protected ZipFileEntry(ZipHeaderType type, ArchiveEncoding archiveEncoding)
: base(type)
{
Extra = new List<ExtraData>();
ArchiveEncoding = archiveEncoding;
}
internal bool IsDirectory
{
get
@@ -30,7 +26,7 @@ internal abstract class ZipFileEntry : ZipHeader
internal Stream? PackedStream { get; set; }
internal ArchiveEncoding ArchiveEncoding { get; }
internal IArchiveEncoding ArchiveEncoding { get; } = archiveEncoding;
internal string? Name { get; set; }
@@ -44,7 +40,7 @@ internal abstract class ZipFileEntry : ZipHeader
internal long UncompressedSize { get; set; }
internal List<ExtraData> Extra { get; set; }
internal List<ExtraData> Extra { get; set; } = new();
public string? Password { get; set; }
@@ -63,6 +59,24 @@ internal abstract class ZipFileEntry : ZipHeader
return encryptionData;
}
internal async ValueTask<PkwareTraditionalEncryptionData> ComposeEncryptionDataAsync(
Stream archiveStream,
CancellationToken cancellationToken = default
)
{
if (archiveStream is null)
{
throw new ArgumentNullException(nameof(archiveStream));
}
var buffer = new byte[12];
await archiveStream.ReadFullyAsync(buffer, 0, 12, cancellationToken).ConfigureAwait(false);
var encryptionData = PkwareTraditionalEncryptionData.ForRead(Password!, this, buffer);
return encryptionData;
}
internal WinzipAesEncryptionData? WinzipAesEncryptionData { get; set; }
/// <summary>

View File

@@ -1,18 +1,14 @@
using System.IO;
using System.Threading.Tasks;
namespace SharpCompress.Common.Zip.Headers;
internal abstract class ZipHeader
internal abstract class ZipHeader(ZipHeaderType type)
{
protected ZipHeader(ZipHeaderType type)
{
ZipHeaderType = type;
HasData = true;
}
internal ZipHeaderType ZipHeaderType { get; }
internal ZipHeaderType ZipHeaderType { get; } = type;
internal abstract void Read(BinaryReader reader);
internal abstract ValueTask Read(AsyncBinaryReader reader);
internal bool HasData { get; set; }
internal bool HasData { get; set; } = true;
}

View File

@@ -8,9 +8,9 @@ internal class PkwareTraditionalEncryptionData
{
private static readonly CRC32 CRC32 = new();
private readonly uint[] _keys = { 0x12345678, 0x23456789, 0x34567890 };
private readonly ArchiveEncoding _archiveEncoding;
private readonly IArchiveEncoding _archiveEncoding;
private PkwareTraditionalEncryptionData(string password, ArchiveEncoding archiveEncoding)
private PkwareTraditionalEncryptionData(string password, IArchiveEncoding archiveEncoding)
{
_archiveEncoding = archiveEncoding;
Initialize(password);
@@ -103,7 +103,7 @@ internal class PkwareTraditionalEncryptionData
internal byte[] StringToByteArray(string value)
{
var a = _archiveEncoding.GetPasswordEncoding().GetBytes(value);
var a = _archiveEncoding.Password.GetBytes(value);
return a;
}

View File

@@ -1,12 +1,17 @@
using System.IO;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common.Zip.Headers;
using SharpCompress.IO;
namespace SharpCompress.Common.Zip;
internal class SeekableZipFilePart : ZipFilePart
{
private bool _isLocalHeaderLoaded;
private volatile bool _isLocalHeaderLoaded;
private readonly SeekableZipHeaderFactory _headerFactory;
private readonly object _headerLock = new();
private readonly SemaphoreSlim _asyncHeaderSemaphore = new(1, 1);
internal SeekableZipFilePart(
SeekableZipHeaderFactory headerFactory,
@@ -19,19 +24,191 @@ internal class SeekableZipFilePart : ZipFilePart
{
if (!_isLocalHeaderLoaded)
{
LoadLocalHeader();
_isLocalHeaderLoaded = true;
lock (_headerLock)
{
if (!_isLocalHeaderLoaded)
{
LoadLocalHeader();
_isLocalHeaderLoaded = true;
}
}
}
return base.GetCompressedStream();
}
private void LoadLocalHeader() =>
Header = _headerFactory.GetLocalHeader(BaseStream, (DirectoryEntryHeader)Header);
internal override async ValueTask<Stream?> GetCompressedStreamAsync(
CancellationToken cancellationToken = default
)
{
if (!_isLocalHeaderLoaded)
{
await _asyncHeaderSemaphore.WaitAsync(cancellationToken);
try
{
if (!_isLocalHeaderLoaded)
{
await LoadLocalHeaderAsync(cancellationToken);
_isLocalHeaderLoaded = true;
}
}
finally
{
_asyncHeaderSemaphore.Release();
}
}
return await base.GetCompressedStreamAsync(cancellationToken);
}
private void LoadLocalHeader()
{
// Use an independent stream for loading the header if multi-threading is enabled
Stream streamToUse = BaseStream;
bool disposeStream = false;
if (
BaseStream is SourceStream sourceStream
&& sourceStream.IsFileMode
&& sourceStream.ReaderOptions.EnableMultiThreadedExtraction
)
{
var independentStream = sourceStream.CreateIndependentStream(0);
if (independentStream is not null)
{
streamToUse = independentStream;
disposeStream = true;
}
}
else
{
// Check if BaseStream wraps a FileStream
Stream? underlyingStream = BaseStream;
if (BaseStream is IStreamStack streamStack)
{
underlyingStream = streamStack.BaseStream();
}
if (underlyingStream is FileStream fileStream)
{
streamToUse = new FileStream(
fileStream.Name,
FileMode.Open,
FileAccess.Read,
FileShare.Read
);
disposeStream = true;
}
}
try
{
Header = _headerFactory.GetLocalHeader(streamToUse, (DirectoryEntryHeader)Header);
}
finally
{
if (disposeStream)
{
streamToUse.Dispose();
}
}
}
private async ValueTask LoadLocalHeaderAsync(CancellationToken cancellationToken = default)
{
// Use an independent stream for loading the header if multi-threading is enabled
Stream streamToUse = BaseStream;
bool disposeStream = false;
if (
BaseStream is SourceStream sourceStream
&& sourceStream.IsFileMode
&& sourceStream.ReaderOptions.EnableMultiThreadedExtraction
)
{
var independentStream = sourceStream.CreateIndependentStream(0);
if (independentStream is not null)
{
streamToUse = independentStream;
disposeStream = true;
}
}
else
{
// Check if BaseStream wraps a FileStream
Stream? underlyingStream = BaseStream;
if (BaseStream is IStreamStack streamStack)
{
underlyingStream = streamStack.BaseStream();
}
if (underlyingStream is FileStream fileStream)
{
streamToUse = new FileStream(
fileStream.Name,
FileMode.Open,
FileAccess.Read,
FileShare.Read
);
disposeStream = true;
}
}
try
{
Header = await _headerFactory.GetLocalHeaderAsync(
streamToUse,
(DirectoryEntryHeader)Header
);
}
finally
{
if (disposeStream)
{
streamToUse.Dispose();
}
}
}
protected override Stream CreateBaseStream()
{
BaseStream.Position = Header.DataStartPosition.NotNull();
// If BaseStream is a SourceStream in file mode with multi-threading enabled,
// create an independent stream to support concurrent extraction
if (
BaseStream is SourceStream sourceStream
&& sourceStream.IsFileMode
&& sourceStream.ReaderOptions.EnableMultiThreadedExtraction
)
{
// Create a new independent stream for this entry
var independentStream = sourceStream.CreateIndependentStream(0);
if (independentStream is not null)
{
independentStream.Position = Header.DataStartPosition.NotNull();
return independentStream;
}
}
// Check if BaseStream wraps a FileStream (for multi-volume archives)
Stream? underlyingStream = BaseStream;
if (BaseStream is IStreamStack streamStack)
{
underlyingStream = streamStack.BaseStream();
}
if (underlyingStream is FileStream fileStream)
{
// Create a new independent stream from the file
var independentStream = new FileStream(
fileStream.Name,
FileMode.Open,
FileAccess.Read,
FileShare.Read
);
independentStream.Position = Header.DataStartPosition.NotNull();
return independentStream;
}
// Fall back to existing behavior for stream-based sources
BaseStream.Position = Header.DataStartPosition.NotNull();
return BaseStream;
}
}

View File

@@ -1,6 +1,8 @@
using System;
using System.Buffers;
using System.Collections.Generic;
using System.IO;
using System.Threading.Tasks;
using SharpCompress.Common.Zip.Headers;
using SharpCompress.IO;
@@ -15,10 +17,79 @@ internal sealed class SeekableZipHeaderFactory : ZipHeaderFactory
private const int MAX_SEARCH_LENGTH_FOR_EOCD = 65557;
private bool _zip64;
internal SeekableZipHeaderFactory(string? password, ArchiveEncoding archiveEncoding)
private static readonly byte[] needle = { 0x06, 0x05, 0x4b, 0x50 };
internal SeekableZipHeaderFactory(string? password, IArchiveEncoding archiveEncoding)
: base(StreamingMode.Seekable, password, archiveEncoding) { }
internal IEnumerable<ZipHeader> ReadSeekableHeader(Stream stream)
internal async IAsyncEnumerable<ZipHeader> ReadSeekableHeaderAsync(Stream stream)
{
var reader = new AsyncBinaryReader(stream);
await SeekBackToHeaderAsync(stream, reader);
var eocd_location = stream.Position;
var entry = new DirectoryEndHeader();
await entry.Read(reader);
if (entry.IsZip64)
{
_zip64 = true;
// ZIP64_END_OF_CENTRAL_DIRECTORY_LOCATOR should be before the EOCD
stream.Seek(eocd_location - ZIP64_EOCD_LENGTH - 4, SeekOrigin.Begin);
uint zip64_locator = await reader.ReadUInt32Async();
if (zip64_locator != ZIP64_END_OF_CENTRAL_DIRECTORY_LOCATOR)
{
throw new ArchiveException("Failed to locate the Zip64 Directory Locator");
}
var zip64Locator = new Zip64DirectoryEndLocatorHeader();
await zip64Locator.Read(reader);
stream.Seek(zip64Locator.RelativeOffsetOfTheEndOfDirectoryRecord, SeekOrigin.Begin);
var zip64Signature = await reader.ReadUInt32Async();
if (zip64Signature != ZIP64_END_OF_CENTRAL_DIRECTORY)
{
throw new ArchiveException("Failed to locate the Zip64 Header");
}
var zip64Entry = new Zip64DirectoryEndHeader();
await zip64Entry.Read(reader);
stream.Seek(zip64Entry.DirectoryStartOffsetRelativeToDisk, SeekOrigin.Begin);
}
else
{
stream.Seek(entry.DirectoryStartOffsetRelativeToDisk, SeekOrigin.Begin);
}
var position = stream.Position;
while (true)
{
stream.Position = position;
var signature = await reader.ReadUInt32Async();
var nextHeader = await ReadHeader(signature, reader, _zip64);
position = stream.Position;
if (nextHeader is null)
{
yield break;
}
if (nextHeader is DirectoryEntryHeader entryHeader)
{
//entry could be zero bytes so we need to know that.
entryHeader.HasData = entryHeader.CompressedSize != 0;
yield return entryHeader;
}
else if (nextHeader is DirectoryEndHeader endHeader)
{
yield return endHeader;
}
}
}
internal IEnumerable<ZipHeader> ReadSeekableHeader(Stream stream, bool useSync)
{
var reader = new BinaryReader(stream);
@@ -85,7 +156,7 @@ internal sealed class SeekableZipHeaderFactory : ZipHeaderFactory
}
}
private static bool IsMatch(byte[] haystack, int position, byte[] needle)
private static bool IsMatch(Span<byte> haystack, int position, byte[] needle)
{
for (var i = 0; i < needle.Length; i++)
{
@@ -98,6 +169,51 @@ internal sealed class SeekableZipHeaderFactory : ZipHeaderFactory
return true;
}
private static async ValueTask SeekBackToHeaderAsync(Stream stream, AsyncBinaryReader reader)
{
// Minimum EOCD length
if (stream.Length < MINIMUM_EOCD_LENGTH)
{
throw new ArchiveException(
"Could not find Zip file Directory at the end of the file. File may be corrupted."
);
}
var len =
stream.Length < MAX_SEARCH_LENGTH_FOR_EOCD
? (int)stream.Length
: MAX_SEARCH_LENGTH_FOR_EOCD;
stream.Seek(-len, SeekOrigin.End);
var seek = ArrayPool<byte>.Shared.Rent(len);
try
{
await reader.ReadBytesAsync(seek, 0, len, default);
var memory = new Memory<byte>(seek, 0, len);
var span = memory.Span;
span.Reverse();
// don't exclude the minimum eocd region, otherwise you fail to locate the header in empty zip files
var max_search_area = len; // - MINIMUM_EOCD_LENGTH;
for (var pos_from_end = 0; pos_from_end < max_search_area; ++pos_from_end)
{
if (IsMatch(span, pos_from_end, needle))
{
stream.Seek(-pos_from_end, SeekOrigin.End);
return;
}
}
throw new ArchiveException("Failed to locate the Zip Header");
}
finally
{
ArrayPool<byte>.Shared.Return(seek);
}
}
private static void SeekBackToHeader(Stream stream, BinaryReader reader)
{
// Minimum EOCD length
@@ -112,9 +228,6 @@ internal sealed class SeekableZipHeaderFactory : ZipHeaderFactory
stream.Length < MAX_SEARCH_LENGTH_FOR_EOCD
? (int)stream.Length
: MAX_SEARCH_LENGTH_FOR_EOCD;
// We search for marker in reverse to find the first occurance
byte[] needle = { 0x06, 0x05, 0x4b, 0x50 };
stream.Seek(-len, SeekOrigin.End);
var seek = reader.ReadBytes(len);
@@ -163,4 +276,31 @@ internal sealed class SeekableZipHeaderFactory : ZipHeaderFactory
}
return localEntryHeader;
}
internal async ValueTask<LocalEntryHeader> GetLocalHeaderAsync(
Stream stream,
DirectoryEntryHeader directoryEntryHeader
)
{
stream.Seek(directoryEntryHeader.RelativeOffsetOfEntryHeader, SeekOrigin.Begin);
var reader = new AsyncBinaryReader(stream);
var signature = await reader.ReadUInt32Async();
if (await ReadHeader(signature, reader, _zip64) is not LocalEntryHeader localEntryHeader)
{
throw new InvalidOperationException();
}
// populate fields only known from the DirectoryEntryHeader
localEntryHeader.HasData = directoryEntryHeader.HasData;
localEntryHeader.ExternalFileAttributes = directoryEntryHeader.ExternalFileAttributes;
localEntryHeader.Comment = directoryEntryHeader.Comment;
if (FlagUtility.HasFlag(localEntryHeader.Flags, HeaderFlags.UsePostDataDescriptor))
{
localEntryHeader.Crc = directoryEntryHeader.Crc;
localEntryHeader.CompressedSize = directoryEntryHeader.CompressedSize;
localEntryHeader.UncompressedSize = directoryEntryHeader.UncompressedSize;
}
return localEntryHeader;
}
}

View File

@@ -1,4 +1,6 @@
using System.IO;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common.Zip.Headers;
using SharpCompress.Compressors.Deflate;
using SharpCompress.IO;
@@ -31,6 +33,28 @@ internal sealed class StreamingZipFilePart : ZipFilePart
return _decompressionStream;
}
internal override async ValueTask<Stream?> GetCompressedStreamAsync(
CancellationToken cancellationToken = default
)
{
if (!Header.HasData)
{
return Stream.Null;
}
_decompressionStream = await CreateDecompressionStreamAsync(
await GetCryptoStreamAsync(CreateBaseStream(), cancellationToken)
.ConfigureAwait(false),
Header.CompressionMethod,
cancellationToken
)
.ConfigureAwait(false);
if (LeaveStreamOpen)
{
return SharpCompressStream.Create(_decompressionStream, leaveOpen: true);
}
return _decompressionStream;
}
internal BinaryReader FixStreamedFileLocation(ref SharpCompressStream rewindableStream)
{
if (Header.IsDirectory)

View File

@@ -2,6 +2,9 @@ using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common;
using SharpCompress.Common.Zip.Headers;
using SharpCompress.IO;
@@ -13,7 +16,7 @@ internal class StreamingZipHeaderFactory : ZipHeaderFactory
internal StreamingZipHeaderFactory(
string? password,
ArchiveEncoding archiveEncoding,
IArchiveEncoding archiveEncoding,
IEnumerable<ZipEntry>? entries
)
: base(StreamingMode.Streaming, password, archiveEncoding) => _entries = entries;
@@ -200,4 +203,331 @@ internal class StreamingZipHeaderFactory : ZipHeaderFactory
yield return header;
}
}
/// <summary>
/// Reads ZIP headers asynchronously for streams that do not support synchronous reads.
/// </summary>
internal IAsyncEnumerable<ZipHeader> ReadStreamHeaderAsync(Stream stream) =>
new StreamHeaderAsyncEnumerable(this, stream);
/// <summary>
/// Invokes the shared async header parsing logic on the base factory.
/// </summary>
private ValueTask<ZipHeader?> ReadHeaderAsyncInternal(
uint headerBytes,
AsyncBinaryReader reader
) => ReadHeader(headerBytes, reader);
/// <summary>
/// Exposes the last parsed local entry header to the async enumerator so it can handle streaming data descriptors.
/// </summary>
private LocalEntryHeader? LastEntryHeader
{
get => _lastEntryHeader;
set => _lastEntryHeader = value;
}
/// <summary>
/// Produces an async enumerator for streaming ZIP headers.
/// </summary>
private sealed class StreamHeaderAsyncEnumerable : IAsyncEnumerable<ZipHeader>
{
private readonly StreamingZipHeaderFactory _headerFactory;
private readonly Stream _stream;
public StreamHeaderAsyncEnumerable(StreamingZipHeaderFactory headerFactory, Stream stream)
{
_headerFactory = headerFactory;
_stream = stream;
}
public IAsyncEnumerator<ZipHeader> GetAsyncEnumerator(
CancellationToken cancellationToken = default
) => new StreamHeaderAsyncEnumerator(_headerFactory, _stream, cancellationToken);
}
/// <summary>
/// Async implementation of <see cref="ReadStreamHeader"/> using <see cref="AsyncBinaryReader"/> to avoid sync reads.
/// </summary>
private sealed class StreamHeaderAsyncEnumerator : IAsyncEnumerator<ZipHeader>, IDisposable
{
private readonly StreamingZipHeaderFactory _headerFactory;
private readonly SharpCompressStream _rewindableStream;
private readonly AsyncBinaryReader _reader;
private readonly CancellationToken _cancellationToken;
private bool _completed;
public StreamHeaderAsyncEnumerator(
StreamingZipHeaderFactory headerFactory,
Stream stream,
CancellationToken cancellationToken
)
{
_headerFactory = headerFactory;
_rewindableStream = EnsureSharpCompressStream(stream);
_reader = new AsyncBinaryReader(_rewindableStream, leaveOpen: true);
_cancellationToken = cancellationToken;
}
private ZipHeader? _current;
public ZipHeader Current =>
_current ?? throw new InvalidOperationException("No current header is available.");
/// <summary>
/// Advances to the next ZIP header in the stream, honoring streaming data descriptors where applicable.
/// </summary>
public async ValueTask<bool> MoveNextAsync()
{
if (_completed)
{
return false;
}
while (true)
{
_cancellationToken.ThrowIfCancellationRequested();
uint headerBytes;
var lastEntryHeader = _headerFactory.LastEntryHeader;
if (
lastEntryHeader != null
&& FlagUtility.HasFlag(lastEntryHeader.Flags, HeaderFlags.UsePostDataDescriptor)
)
{
if (lastEntryHeader.Part is null)
{
continue;
}
var pos = _rewindableStream.CanSeek ? (long?)_rewindableStream.Position : null;
var crc = await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false);
if (crc == POST_DATA_DESCRIPTOR)
{
crc = await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false);
}
lastEntryHeader.Crc = crc;
//attempt 32bit read
ulong compressedSize = await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false);
ulong uncompressedSize = await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false);
headerBytes = await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false);
//check for zip64 sentinel or unexpected header
bool isSentinel =
compressedSize == 0xFFFFFFFF || uncompressedSize == 0xFFFFFFFF;
bool isHeader = headerBytes == 0x04034b50 || headerBytes == 0x02014b50;
if (!isHeader && !isSentinel)
{
//reshuffle into 64-bit values
compressedSize = (uncompressedSize << 32) | compressedSize;
uncompressedSize =
((ulong)headerBytes << 32)
| await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false);
headerBytes = await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false);
}
else if (isSentinel)
{
//standards-compliant zip64 descriptor
compressedSize = await _reader
.ReadUInt64Async(_cancellationToken)
.ConfigureAwait(false);
uncompressedSize = await _reader
.ReadUInt64Async(_cancellationToken)
.ConfigureAwait(false);
}
lastEntryHeader.CompressedSize = (long)compressedSize;
lastEntryHeader.UncompressedSize = (long)uncompressedSize;
if (pos.HasValue)
{
lastEntryHeader.DataStartPosition = pos - lastEntryHeader.CompressedSize;
}
}
else if (lastEntryHeader != null && lastEntryHeader.IsZip64)
{
if (lastEntryHeader.Part is null)
{
continue;
}
var pos = _rewindableStream.CanSeek ? (long?)_rewindableStream.Position : null;
headerBytes = await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false);
_ = await _reader.ReadUInt16Async(_cancellationToken).ConfigureAwait(false); // version
_ = await _reader.ReadUInt16Async(_cancellationToken).ConfigureAwait(false); // flags
_ = await _reader.ReadUInt16Async(_cancellationToken).ConfigureAwait(false); // compressionMethod
_ = await _reader.ReadUInt16Async(_cancellationToken).ConfigureAwait(false); // lastModifiedDate
_ = await _reader.ReadUInt16Async(_cancellationToken).ConfigureAwait(false); // lastModifiedTime
var crc = await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false);
if (crc == POST_DATA_DESCRIPTOR)
{
crc = await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false);
}
lastEntryHeader.Crc = crc;
// The DataDescriptor can be either 64bit or 32bit
var compressedSize = await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false);
var uncompressedSize = await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false);
// Check if we have header or 64bit DataDescriptor
var testHeader = !(headerBytes == 0x04034b50 || headerBytes == 0x02014b50);
var test64Bit = ((long)uncompressedSize << 32) | compressedSize;
if (test64Bit == lastEntryHeader.CompressedSize && testHeader)
{
lastEntryHeader.UncompressedSize =
(
(long)
await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false) << 32
) | headerBytes;
headerBytes = await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false);
}
else
{
lastEntryHeader.UncompressedSize = uncompressedSize;
}
if (pos.HasValue)
{
lastEntryHeader.DataStartPosition = pos - lastEntryHeader.CompressedSize;
// 4 = First 4 bytes of the entry header (i.e. 50 4B 03 04)
_rewindableStream.Position = pos.Value + 4;
}
}
else
{
headerBytes = await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false);
}
_headerFactory.LastEntryHeader = null;
var header = await _headerFactory
.ReadHeaderAsyncInternal(headerBytes, _reader)
.ConfigureAwait(false);
if (header is null)
{
_completed = true;
return false;
}
//entry could be zero bytes so we need to know that.
if (header.ZipHeaderType == ZipHeaderType.LocalEntry)
{
var localHeader = (LocalEntryHeader)header;
var directoryHeader = _headerFactory._entries?.FirstOrDefault(entry =>
entry.Key == localHeader.Name
&& localHeader.CompressedSize == 0
&& localHeader.UncompressedSize == 0
&& localHeader.Crc == 0
&& localHeader.IsDirectory == false
);
if (directoryHeader != null)
{
localHeader.UncompressedSize = directoryHeader.Size;
localHeader.CompressedSize = directoryHeader.CompressedSize;
localHeader.Crc = (uint)directoryHeader.Crc;
}
// If we have CompressedSize, there is data to be read
if (localHeader.CompressedSize > 0)
{
header.HasData = true;
} // Check if zip is streaming ( Length is 0 and is declared in PostDataDescriptor )
else if (localHeader.Flags.HasFlag(HeaderFlags.UsePostDataDescriptor))
{
var nextHeaderBytes = await _reader
.ReadUInt32Async(_cancellationToken)
.ConfigureAwait(false);
((IStreamStack)_rewindableStream).Rewind(sizeof(uint));
// Check if next data is PostDataDescriptor, streamed file with 0 length
header.HasData = !IsHeader(nextHeaderBytes);
}
else // We are not streaming and compressed size is 0, we have no data
{
header.HasData = false;
}
}
_current = header;
return true;
}
}
public ValueTask DisposeAsync()
{
Dispose();
return default;
}
/// <summary>
/// Disposes the underlying reader (without closing the archive stream).
/// </summary>
public void Dispose()
{
_reader.Dispose();
}
/// <summary>
/// Ensures the stream is a <see cref="SharpCompressStream"/> so header parsing can use rewind/buffer helpers.
/// </summary>
private static SharpCompressStream EnsureSharpCompressStream(Stream stream)
{
if (stream is SharpCompressStream sharpCompressStream)
{
return sharpCompressStream;
}
// Ensure the stream is already a SharpCompressStream so the buffer/size is set.
// The original code wrapped this with RewindableStream; use SharpCompressStream so we can get the buffer size.
if (stream is SourceStream src)
{
return new SharpCompressStream(
stream,
src.ReaderOptions.LeaveStreamOpen,
bufferSize: src.ReaderOptions.BufferSize
);
}
throw new ArgumentException("Stream must be a SharpCompressStream", nameof(stream));
}
}
}

View File

@@ -20,7 +20,7 @@ internal class WinzipAesEncryptionData
{
_keySize = keySize;
#if NETFRAMEWORK || NETSTANDARD2_0
#if LEGACY_DOTNET
var rfc2898 = new Rfc2898DeriveBytes(password, salt, RFC2898_ITERATIONS);
KeyBytes = rfc2898.GetBytes(KeySizeInBytes);
IvBytes = rfc2898.GetBytes(KeySizeInBytes);

View File

@@ -2,6 +2,8 @@ using System;
using System.Buffers.Binary;
using System.IO;
using System.Linq;
using System.Threading;
using System.Threading.Tasks;
using SharpCompress.Common.Zip.Headers;
using SharpCompress.Compressors;
using SharpCompress.Compressors.BZip2;
@@ -264,4 +266,244 @@ internal abstract class ZipFilePart : FilePart
}
return plainStream;
}
internal override async ValueTask<Stream?> GetCompressedStreamAsync(
CancellationToken cancellationToken = default
)
{
if (!Header.HasData)
{
return Stream.Null;
}
var decompressionStream = await CreateDecompressionStreamAsync(
await GetCryptoStreamAsync(CreateBaseStream(), cancellationToken)
.ConfigureAwait(false),
Header.CompressionMethod,
cancellationToken
)
.ConfigureAwait(false);
if (LeaveStreamOpen)
{
return SharpCompressStream.Create(decompressionStream, leaveOpen: true);
}
return decompressionStream;
}
protected async Task<Stream> GetCryptoStreamAsync(
Stream plainStream,
CancellationToken cancellationToken = default
)
{
var isFileEncrypted = FlagUtility.HasFlag(Header.Flags, HeaderFlags.Encrypted);
if (Header.CompressedSize == 0 && isFileEncrypted)
{
throw new NotSupportedException("Cannot encrypt file with unknown size at start.");
}
if (
(
Header.CompressedSize == 0
&& FlagUtility.HasFlag(Header.Flags, HeaderFlags.UsePostDataDescriptor)
) || Header.IsZip64
)
{
plainStream = SharpCompressStream.Create(plainStream, leaveOpen: true); //make sure AES doesn't close
}
else
{
plainStream = new ReadOnlySubStream(plainStream, Header.CompressedSize); //make sure AES doesn't close
}
if (isFileEncrypted)
{
switch (Header.CompressionMethod)
{
case ZipCompressionMethod.None:
case ZipCompressionMethod.Shrink:
case ZipCompressionMethod.Reduce1:
case ZipCompressionMethod.Reduce2:
case ZipCompressionMethod.Reduce3:
case ZipCompressionMethod.Reduce4:
case ZipCompressionMethod.Deflate:
case ZipCompressionMethod.Deflate64:
case ZipCompressionMethod.BZip2:
case ZipCompressionMethod.LZMA:
case ZipCompressionMethod.PPMd:
{
return new PkwareTraditionalCryptoStream(
plainStream,
await Header
.ComposeEncryptionDataAsync(plainStream, cancellationToken)
.ConfigureAwait(false),
CryptoMode.Decrypt
);
}
case ZipCompressionMethod.WinzipAes:
{
if (Header.WinzipAesEncryptionData != null)
{
return new WinzipAesCryptoStream(
plainStream,
Header.WinzipAesEncryptionData,
Header.CompressedSize - 10
);
}
return plainStream;
}
default:
{
throw new InvalidOperationException("Header.CompressionMethod is invalid");
}
}
}
return plainStream;
}
protected async Task<Stream> CreateDecompressionStreamAsync(
Stream stream,
ZipCompressionMethod method,
CancellationToken cancellationToken = default
)
{
switch (method)
{
case ZipCompressionMethod.None:
{
if (Header.CompressedSize is 0)
{
return new DataDescriptorStream(stream);
}
return stream;
}
case ZipCompressionMethod.Shrink:
{
return new ShrinkStream(
stream,
CompressionMode.Decompress,
Header.CompressedSize,
Header.UncompressedSize
);
}
case ZipCompressionMethod.Reduce1:
{
return new ReduceStream(stream, Header.CompressedSize, Header.UncompressedSize, 1);
}
case ZipCompressionMethod.Reduce2:
{
return new ReduceStream(stream, Header.CompressedSize, Header.UncompressedSize, 2);
}
case ZipCompressionMethod.Reduce3:
{
return new ReduceStream(stream, Header.CompressedSize, Header.UncompressedSize, 3);
}
case ZipCompressionMethod.Reduce4:
{
return new ReduceStream(stream, Header.CompressedSize, Header.UncompressedSize, 4);
}
case ZipCompressionMethod.Explode:
{
return new ExplodeStream(
stream,
Header.CompressedSize,
Header.UncompressedSize,
Header.Flags
);
}
case ZipCompressionMethod.Deflate:
{
return new DeflateStream(stream, CompressionMode.Decompress);
}
case ZipCompressionMethod.Deflate64:
{
return new Deflate64Stream(stream, CompressionMode.Decompress);
}
case ZipCompressionMethod.BZip2:
{
return new BZip2Stream(stream, CompressionMode.Decompress, false);
}
case ZipCompressionMethod.LZMA:
{
if (FlagUtility.HasFlag(Header.Flags, HeaderFlags.Encrypted))
{
throw new NotSupportedException("LZMA with pkware encryption.");
}
var buffer = new byte[4];
await stream.ReadFullyAsync(buffer, 0, 4, cancellationToken).ConfigureAwait(false);
var version = BinaryPrimitives.ReadUInt16LittleEndian(buffer.AsSpan(0, 2));
var propsSize = BinaryPrimitives.ReadUInt16LittleEndian(buffer.AsSpan(2, 2));
var props = new byte[propsSize];
await stream
.ReadFullyAsync(props, 0, propsSize, cancellationToken)
.ConfigureAwait(false);
return new LzmaStream(
props,
stream,
Header.CompressedSize > 0 ? Header.CompressedSize - 4 - props.Length : -1,
FlagUtility.HasFlag(Header.Flags, HeaderFlags.Bit1)
? -1
: Header.UncompressedSize
);
}
case ZipCompressionMethod.Xz:
{
return new XZStream(stream);
}
case ZipCompressionMethod.ZStandard:
{
return new DecompressionStream(stream);
}
case ZipCompressionMethod.PPMd:
{
var props = new byte[2];
await stream.ReadFullyAsync(props, 0, 2, cancellationToken).ConfigureAwait(false);
return new PpmdStream(new PpmdProperties(props), stream, false);
}
case ZipCompressionMethod.WinzipAes:
{
var data = Header.Extra.SingleOrDefault(x => x.Type == ExtraDataType.WinZipAes);
if (data is null)
{
throw new InvalidFormatException("No Winzip AES extra data found.");
}
if (data.Length != 7)
{
throw new InvalidFormatException("Winzip data length is not 7.");
}
var compressedMethod = BinaryPrimitives.ReadUInt16LittleEndian(data.DataBytes);
if (compressedMethod != 0x01 && compressedMethod != 0x02)
{
throw new InvalidFormatException(
"Unexpected vendor version number for WinZip AES metadata"
);
}
var vendorId = BinaryPrimitives.ReadUInt16LittleEndian(data.DataBytes.AsSpan(2));
if (vendorId != 0x4541)
{
throw new InvalidFormatException(
"Unexpected vendor ID for WinZip AES metadata"
);
}
return await CreateDecompressionStreamAsync(
stream,
(ZipCompressionMethod)
BinaryPrimitives.ReadUInt16LittleEndian(data.DataBytes.AsSpan(5)),
cancellationToken
);
}
default:
{
throw new NotSupportedException("CompressionMethod: " + Header.CompressionMethod);
}
}
}
}

View File

@@ -1,6 +1,8 @@
using System;
using System.IO;
using System.Linq;
using System.Threading.Tasks;
using SharpCompress;
using SharpCompress.Common.Zip.Headers;
using SharpCompress.IO;
@@ -21,12 +23,12 @@ internal class ZipHeaderFactory
protected LocalEntryHeader? _lastEntryHeader;
private readonly string? _password;
private readonly StreamingMode _mode;
private readonly ArchiveEncoding _archiveEncoding;
private readonly IArchiveEncoding _archiveEncoding;
protected ZipHeaderFactory(
StreamingMode mode,
string? password,
ArchiveEncoding archiveEncoding
IArchiveEncoding archiveEncoding
)
{
_mode = mode;
@@ -34,6 +36,82 @@ internal class ZipHeaderFactory
_archiveEncoding = archiveEncoding;
}
protected async ValueTask<ZipHeader?> ReadHeader(
uint headerBytes,
AsyncBinaryReader reader,
bool zip64 = false
)
{
switch (headerBytes)
{
case ENTRY_HEADER_BYTES:
{
var entryHeader = new LocalEntryHeader(_archiveEncoding);
await entryHeader.Read(reader);
await LoadHeaderAsync(entryHeader, reader.BaseStream).ConfigureAwait(false);
_lastEntryHeader = entryHeader;
return entryHeader;
}
case DIRECTORY_START_HEADER_BYTES:
{
var entry = new DirectoryEntryHeader(_archiveEncoding);
await entry.Read(reader);
return entry;
}
case POST_DATA_DESCRIPTOR:
{
if (
_lastEntryHeader != null
&& FlagUtility.HasFlag(
_lastEntryHeader.NotNull().Flags,
HeaderFlags.UsePostDataDescriptor
)
)
{
_lastEntryHeader.Crc = await reader.ReadUInt32Async();
_lastEntryHeader.CompressedSize = zip64
? (long)await reader.ReadUInt64Async()
: await reader.ReadUInt32Async();
_lastEntryHeader.UncompressedSize = zip64
? (long)await reader.ReadUInt64Async()
: await reader.ReadUInt32Async();
}
else
{
await reader.SkipAsync(zip64 ? 20 : 12);
}
return null;
}
case DIGITAL_SIGNATURE:
return null;
case DIRECTORY_END_HEADER_BYTES:
{
var entry = new DirectoryEndHeader();
await entry.Read(reader);
return entry;
}
case SPLIT_ARCHIVE_HEADER_BYTES:
{
return new SplitHeader();
}
case ZIP64_END_OF_CENTRAL_DIRECTORY:
{
var entry = new Zip64DirectoryEndHeader();
await entry.Read(reader);
return entry;
}
case ZIP64_END_OF_CENTRAL_DIRECTORY_LOCATOR:
{
var entry = new Zip64DirectoryEndLocatorHeader();
await entry.Read(reader);
return entry;
}
default:
return null;
}
}
protected ZipHeader? ReadHeader(uint headerBytes, BinaryReader reader, bool zip64 = false)
{
switch (headerBytes)
@@ -205,4 +283,82 @@ internal class ZipHeaderFactory
//}
}
/// <summary>
/// Loads encryption metadata and stream positioning for a header using async reads where needed.
/// </summary>
private async ValueTask LoadHeaderAsync(ZipFileEntry entryHeader, Stream stream)
{
if (FlagUtility.HasFlag(entryHeader.Flags, HeaderFlags.Encrypted))
{
if (
!entryHeader.IsDirectory
&& entryHeader.CompressedSize == 0
&& FlagUtility.HasFlag(entryHeader.Flags, HeaderFlags.UsePostDataDescriptor)
)
{
throw new NotSupportedException(
"SharpCompress cannot currently read non-seekable Zip Streams with encrypted data that has been written in a non-seekable manner."
);
}
if (_password is null)
{
throw new CryptographicException("No password supplied for encrypted zip.");
}
entryHeader.Password = _password;
if (entryHeader.CompressionMethod == ZipCompressionMethod.WinzipAes)
{
var data = entryHeader.Extra.SingleOrDefault(x =>
x.Type == ExtraDataType.WinZipAes
);
if (data != null)
{
var keySize = (WinzipAesKeySize)data.DataBytes[4];
var salt = new byte[WinzipAesEncryptionData.KeyLengthInBytes(keySize) / 2];
var passwordVerifyValue = new byte[2];
await stream.ReadExactAsync(salt, 0, salt.Length).ConfigureAwait(false);
await stream.ReadExactAsync(passwordVerifyValue, 0, 2).ConfigureAwait(false);
entryHeader.WinzipAesEncryptionData = new WinzipAesEncryptionData(
keySize,
salt,
passwordVerifyValue,
_password
);
entryHeader.CompressedSize -= (uint)(salt.Length + 2);
}
}
}
if (entryHeader.IsDirectory)
{
return;
}
switch (_mode)
{
case StreamingMode.Seekable:
{
entryHeader.DataStartPosition = stream.Position;
stream.Position += entryHeader.CompressedSize;
break;
}
case StreamingMode.Streaming:
{
entryHeader.PackedStream = stream;
break;
}
default:
{
throw new InvalidFormatException("Invalid StreamingMode");
}
}
}
}

View File

@@ -104,7 +104,7 @@ public static class ADCBase
/// <param name="bufferSize">Max size for decompressed data</param>
/// <param name="cancellationToken">Cancellation token</param>
/// <returns>Result containing bytes read and decompressed data</returns>
public static async Task<AdcDecompressResult> DecompressAsync(
public static async ValueTask<AdcDecompressResult> DecompressAsync(
byte[] input,
int bufferSize = 262144,
CancellationToken cancellationToken = default
@@ -117,7 +117,7 @@ public static class ADCBase
/// <param name="bufferSize">Max size for decompressed data</param>
/// <param name="cancellationToken">Cancellation token</param>
/// <returns>Result containing bytes read and decompressed data</returns>
public static async Task<AdcDecompressResult> DecompressAsync(
public static async ValueTask<AdcDecompressResult> DecompressAsync(
Stream input,
int bufferSize = 262144,
CancellationToken cancellationToken = default

View File

@@ -30,7 +30,6 @@ public sealed class BZip2Stream : Stream, IStreamStack
private readonly Stream stream;
private bool isDisposed;
private readonly bool leaveOpen;
/// <summary>
/// Create a BZip2Stream
@@ -38,30 +37,19 @@ public sealed class BZip2Stream : Stream, IStreamStack
/// <param name="stream">The stream to read from</param>
/// <param name="compressionMode">Compression Mode</param>
/// <param name="decompressConcatenated">Decompress Concatenated</param>
/// <param name="leaveOpen">Leave the stream open after disposing</param>
public BZip2Stream(
Stream stream,
CompressionMode compressionMode,
bool decompressConcatenated,
bool leaveOpen = false
)
public BZip2Stream(Stream stream, CompressionMode compressionMode, bool decompressConcatenated)
{
#if DEBUG_STREAMS
this.DebugConstruct(typeof(BZip2Stream));
#endif
this.leaveOpen = leaveOpen;
Mode = compressionMode;
if (Mode == CompressionMode.Compress)
{
this.stream = new CBZip2OutputStream(stream, 9, leaveOpen);
this.stream = new CBZip2OutputStream(stream);
}
else
{
this.stream = new CBZip2InputStream(
stream,
decompressConcatenated,
leaveOpen: leaveOpen
);
this.stream = new CBZip2InputStream(stream, decompressConcatenated);
}
}
@@ -110,7 +98,7 @@ public sealed class BZip2Stream : Stream, IStreamStack
public override void SetLength(long value) => stream.SetLength(value);
#if !NETFRAMEWORK && !NETSTANDARD2_0
#if !LEGACY_DOTNET
public override int Read(Span<byte> buffer) => stream.Read(buffer);

View File

@@ -168,7 +168,6 @@ internal class CBZip2InputStream : Stream, IStreamStack
private int computedBlockCRC,
computedCombinedCRC;
private readonly bool decompressConcatenated;
private readonly bool leaveOpen;
private int i2,
count,
@@ -182,10 +181,9 @@ internal class CBZip2InputStream : Stream, IStreamStack
private char z;
private bool isDisposed;
public CBZip2InputStream(Stream zStream, bool decompressConcatenated, bool leaveOpen = false)
public CBZip2InputStream(Stream zStream, bool decompressConcatenated)
{
this.decompressConcatenated = decompressConcatenated;
this.leaveOpen = leaveOpen;
ll8 = null;
tt = null;
BsSetStream(zStream);
@@ -209,10 +207,7 @@ internal class CBZip2InputStream : Stream, IStreamStack
this.DebugDispose(typeof(CBZip2InputStream));
#endif
base.Dispose(disposing);
if (!leaveOpen)
{
bsStream?.Dispose();
}
bsStream?.Dispose();
}
internal static int[][] InitIntArray(int n1, int n2)
@@ -403,10 +398,7 @@ internal class CBZip2InputStream : Stream, IStreamStack
private void BsFinishedWithStream()
{
if (!leaveOpen)
{
bsStream?.Dispose();
}
bsStream?.Dispose();
bsStream = null;
}

View File

@@ -341,14 +341,12 @@ internal sealed class CBZip2OutputStream : Stream, IStreamStack
private int currentChar = -1;
private int runLength;
private readonly bool leaveOpen;
public CBZip2OutputStream(Stream inStream, bool leaveOpen = false)
: this(inStream, 9, leaveOpen) { }
public CBZip2OutputStream(Stream inStream)
: this(inStream, 9) { }
public CBZip2OutputStream(Stream inStream, int inBlockSize, bool leaveOpen = false)
public CBZip2OutputStream(Stream inStream, int inBlockSize)
{
this.leaveOpen = leaveOpen;
block = null;
quadrant = null;
zptr = null;
@@ -483,10 +481,7 @@ internal sealed class CBZip2OutputStream : Stream, IStreamStack
this.DebugDispose(typeof(CBZip2OutputStream));
#endif
Dispose();
if (!leaveOpen)
{
bsStream?.Dispose();
}
bsStream?.Dispose();
bsStream = null;
}
}

View File

@@ -299,7 +299,7 @@ public class DeflateStream : Stream, IStreamStack
await _baseStream.FlushAsync(cancellationToken).ConfigureAwait(false);
}
#if !NETFRAMEWORK && !NETSTANDARD2_0
#if !LEGACY_DOTNET
public override async ValueTask DisposeAsync()
{
if (_disposed)
@@ -370,7 +370,7 @@ public class DeflateStream : Stream, IStreamStack
.ConfigureAwait(false);
}
#if !NETFRAMEWORK && !NETSTANDARD2_0
#if !LEGACY_DOTNET
public override async ValueTask<int> ReadAsync(
Memory<byte> buffer,
CancellationToken cancellationToken = default
@@ -461,7 +461,7 @@ public class DeflateStream : Stream, IStreamStack
.ConfigureAwait(false);
}
#if !NETFRAMEWORK && !NETSTANDARD2_0
#if !LEGACY_DOTNET
public override async ValueTask WriteAsync(
ReadOnlyMemory<byte> buffer,
CancellationToken cancellationToken = default

View File

@@ -345,7 +345,7 @@ public class GZipStream : Stream, IStreamStack
return n;
}
#if !NETFRAMEWORK && !NETSTANDARD2_0
#if !LEGACY_DOTNET
public override async ValueTask<int> ReadAsync(
Memory<byte> buffer,
CancellationToken cancellationToken = default
@@ -454,7 +454,7 @@ public class GZipStream : Stream, IStreamStack
await BaseStream.WriteAsync(buffer, offset, count, cancellationToken).ConfigureAwait(false);
}
#if !NETFRAMEWORK && !NETSTANDARD2_0
#if !LEGACY_DOTNET
public override async ValueTask WriteAsync(
ReadOnlyMemory<byte> buffer,
CancellationToken cancellationToken = default

View File

@@ -64,6 +64,7 @@
// -----------------------------------------------------------------------
using System;
using System.Buffers;
using SharpCompress.Algorithms;
namespace SharpCompress.Compressors.Deflate;
@@ -116,14 +117,14 @@ internal sealed class InflateBlocks
internal int readAt; // window read pointer
internal int table; // table lengths (14 bits)
internal int[] tb = new int[1]; // bit length decoding tree
internal byte[] window; // sliding window
internal IMemoryOwner<byte> window; // sliding window
internal int writeAt; // window write pointer
internal InflateBlocks(ZlibCodec codec, object checkfn, int w)
{
_codec = codec;
hufts = new int[MANY * 3];
window = new byte[w];
window = MemoryPool<byte>.Shared.Rent(w);
end = w;
this.checkfn = checkfn;
mode = InflateBlockMode.TYPE;
@@ -340,7 +341,7 @@ internal sealed class InflateBlocks
{
t = m;
}
Array.Copy(_codec.InputBuffer, p, window, q, t);
_codec.InputBuffer.AsSpan(p, t).CopyTo(window.Memory.Span.Slice(q));
p += t;
n -= t;
q += t;
@@ -715,13 +716,14 @@ internal sealed class InflateBlocks
internal void Free()
{
Reset();
window?.Dispose();
window = null;
hufts = null;
}
internal void SetDictionary(byte[] d, int start, int n)
{
Array.Copy(d, start, window, 0, n);
d.AsSpan(start, n).CopyTo(window.Memory.Span.Slice(0, n));
readAt = writeAt = n;
}
@@ -774,11 +776,16 @@ internal sealed class InflateBlocks
// update check information
if (checkfn != null)
{
_codec._adler32 = check = Adler32.Calculate(check, window.AsSpan(readAt, nBytes));
_codec._adler32 = check = Adler32.Calculate(
check,
window.Memory.Span.Slice(readAt, nBytes)
);
}
// copy as far as end of window
Array.Copy(window, readAt, _codec.OutputBuffer, _codec.NextOut, nBytes);
window
.Memory.Span.Slice(readAt, nBytes)
.CopyTo(_codec.OutputBuffer.AsSpan(_codec.NextOut));
_codec.NextOut += nBytes;
readAt += nBytes;
@@ -1213,7 +1220,7 @@ internal sealed class InflateCodes
}
}
blocks.window[q++] = blocks.window[f++];
blocks.window.Memory.Span[q++] = blocks.window.Memory.Span[f++];
m--;
if (f == blocks.end)
@@ -1259,7 +1266,7 @@ internal sealed class InflateCodes
}
r = ZlibConstants.Z_OK;
blocks.window[q++] = (byte)lit;
blocks.window.Memory.Span[q++] = (byte)lit;
m--;
mode = START;
@@ -1396,7 +1403,7 @@ internal sealed class InflateCodes
b >>= (tp[tp_index_t_3 + 1]);
k -= (tp[tp_index_t_3 + 1]);
s.window[q++] = (byte)tp[tp_index_t_3 + 2];
s.window.Memory.Span[q++] = (byte)tp[tp_index_t_3 + 2];
m--;
continue;
}
@@ -1461,13 +1468,14 @@ internal sealed class InflateCodes
r = q - d;
if (q - r > 0 && 2 > (q - r))
{
s.window[q++] = s.window[r++]; // minimum count is three,
s.window[q++] = s.window[r++]; // so unroll loop a little
s.window.Memory.Span[q++] = s.window.Memory.Span[r++]; // minimum count is three,
s.window.Memory.Span[q++] = s.window.Memory.Span[r++]; // so unroll loop a little
c -= 2;
}
else
{
Array.Copy(s.window, r, s.window, q, 2);
s.window.Memory.Span.Slice(r, 2)
.CopyTo(s.window.Memory.Span.Slice(q));
q += 2;
r += 2;
c -= 2;
@@ -1490,12 +1498,13 @@ internal sealed class InflateCodes
{
do
{
s.window[q++] = s.window[r++];
s.window.Memory.Span[q++] = s.window.Memory.Span[r++];
} while (--e != 0);
}
else
{
Array.Copy(s.window, r, s.window, q, e);
s.window.Memory.Span.Slice(r, e)
.CopyTo(s.window.Memory.Span.Slice(q));
q += e;
r += e;
e = 0;
@@ -1509,12 +1518,13 @@ internal sealed class InflateCodes
{
do
{
s.window[q++] = s.window[r++];
s.window.Memory.Span[q++] = s.window.Memory.Span[r++];
} while (--c != 0);
}
else
{
Array.Copy(s.window, r, s.window, q, c);
s.window.Memory.Span.Slice(r, c)
.CopyTo(s.window.Memory.Span.Slice(q));
q += c;
r += c;
c = 0;
@@ -1560,7 +1570,7 @@ internal sealed class InflateCodes
{
b >>= (tp[tp_index_t_3 + 1]);
k -= (tp[tp_index_t_3 + 1]);
s.window[q++] = (byte)tp[tp_index_t_3 + 2];
s.window.Memory.Span[q++] = (byte)tp[tp_index_t_3 + 2];
m--;
break;
}

Some files were not shown because too many files have changed in this diff Show More