[12074] | 1 | // ZipInputStream.cs
|
---|
| 2 | //
|
---|
| 3 | // ------------------------------------------------------------------
|
---|
| 4 | //
|
---|
| 5 | // Copyright (c) 2009-2010 Dino Chiesa.
|
---|
| 6 | // All rights reserved.
|
---|
| 7 | //
|
---|
| 8 | // This code module is part of DotNetZip, a zipfile class library.
|
---|
| 9 | //
|
---|
| 10 | // ------------------------------------------------------------------
|
---|
| 11 | //
|
---|
| 12 | // This code is licensed under the Microsoft Public License.
|
---|
| 13 | // See the file License.txt for the license details.
|
---|
| 14 | // More info on: http://dotnetzip.codeplex.com
|
---|
| 15 | //
|
---|
| 16 | // ------------------------------------------------------------------
|
---|
| 17 | //
|
---|
| 18 | // last saved (in emacs):
|
---|
| 19 | // Time-stamp: <2011-July-31 14:48:30>
|
---|
| 20 | //
|
---|
| 21 | // ------------------------------------------------------------------
|
---|
| 22 | //
|
---|
| 23 | // This module defines the ZipInputStream class, which is a stream metaphor for
|
---|
| 24 | // reading zip files. This class does not depend on Ionic.Zip.ZipFile, but rather
|
---|
| 25 | // stands alongside it as an alternative "container" for ZipEntry, when reading zips.
|
---|
| 26 | //
|
---|
| 27 | // It adds one interesting method to the normal "stream" interface: GetNextEntry.
|
---|
| 28 | //
|
---|
| 29 | // ------------------------------------------------------------------
|
---|
| 30 | //
|
---|
| 31 |
|
---|
| 32 | using System;
|
---|
| 33 | using System.Threading;
|
---|
| 34 | using System.Collections.Generic;
|
---|
| 35 | using System.IO;
|
---|
| 36 | using Ionic.Zip;
|
---|
| 37 | using OfficeOpenXml.Packaging.Ionic.Zip;
|
---|
| 38 | using OfficeOpenXml.Packaging.Ionic.Crc;
|
---|
| 39 |
|
---|
| 40 | namespace Ionic.Zip
|
---|
| 41 | {
|
---|
| 42 | /// <summary>
|
---|
| 43 | /// Provides a stream metaphor for reading zip files.
|
---|
| 44 | /// </summary>
|
---|
| 45 | ///
|
---|
| 46 | /// <remarks>
|
---|
| 47 | /// <para>
|
---|
| 48 | /// This class provides an alternative programming model for reading zip files to
|
---|
| 49 | /// the one enabled by the <see cref="ZipFile"/> class. Use this when reading zip
|
---|
| 50 | /// files, as an alternative to the <see cref="ZipFile"/> class, when you would
|
---|
| 51 | /// like to use a Stream class to read the file.
|
---|
| 52 | /// </para>
|
---|
| 53 | ///
|
---|
| 54 | /// <para>
|
---|
| 55 | /// Some application designs require a readable stream for input. This stream can
|
---|
| 56 | /// be used to read a zip file, and extract entries.
|
---|
| 57 | /// </para>
|
---|
| 58 | ///
|
---|
| 59 | /// <para>
|
---|
| 60 | /// Both the <c>ZipInputStream</c> class and the <c>ZipFile</c> class can be used
|
---|
| 61 | /// to read and extract zip files. Both of them support many of the common zip
|
---|
| 62 | /// features, including Unicode, different compression levels, and ZIP64. The
|
---|
| 63 | /// programming models differ. For example, when extracting entries via calls to
|
---|
| 64 | /// the <c>GetNextEntry()</c> and <c>Read()</c> methods on the
|
---|
| 65 | /// <c>ZipInputStream</c> class, the caller is responsible for creating the file,
|
---|
| 66 | /// writing the bytes into the file, setting the attributes on the file, and
|
---|
| 67 | /// setting the created, last modified, and last accessed timestamps on the
|
---|
| 68 | /// file. All of these things are done automatically by a call to <see
|
---|
| 69 | /// cref="ZipEntry.Extract()">ZipEntry.Extract()</see>. For this reason, the
|
---|
| 70 | /// <c>ZipInputStream</c> is generally recommended for when your application wants
|
---|
| 71 | /// to extract the data, without storing that data into a file.
|
---|
| 72 | /// </para>
|
---|
| 73 | ///
|
---|
| 74 | /// <para>
|
---|
| 75 | /// Aside from the obvious differences in programming model, there are some
|
---|
| 76 | /// differences in capability between the <c>ZipFile</c> class and the
|
---|
| 77 | /// <c>ZipInputStream</c> class.
|
---|
| 78 | /// </para>
|
---|
| 79 | ///
|
---|
| 80 | /// <list type="bullet">
|
---|
| 81 | /// <item>
|
---|
| 82 | /// <c>ZipFile</c> can be used to create or update zip files, or read and
|
---|
| 83 | /// extract zip files. <c>ZipInputStream</c> can be used only to read and
|
---|
| 84 | /// extract zip files. If you want to use a stream to create zip files, check
|
---|
| 85 | /// out the <see cref="ZipOutputStream"/>.
|
---|
| 86 | /// </item>
|
---|
| 87 | ///
|
---|
| 88 | /// <item>
|
---|
| 89 | /// <c>ZipInputStream</c> cannot read segmented or spanned
|
---|
| 90 | /// zip files.
|
---|
| 91 | /// </item>
|
---|
| 92 | ///
|
---|
| 93 | /// <item>
|
---|
| 94 | /// <c>ZipInputStream</c> will not read Zip file comments.
|
---|
| 95 | /// </item>
|
---|
| 96 | ///
|
---|
| 97 | /// <item>
|
---|
| 98 | /// When reading larger files, <c>ZipInputStream</c> will always underperform
|
---|
| 99 | /// <c>ZipFile</c>. This is because the <c>ZipInputStream</c> does a full scan on the
|
---|
| 100 | /// zip file, while the <c>ZipFile</c> class reads the central directory of the
|
---|
| 101 | /// zip file.
|
---|
| 102 | /// </item>
|
---|
| 103 | ///
|
---|
| 104 | /// </list>
|
---|
| 105 | ///
|
---|
| 106 | /// </remarks>
|
---|
| 107 | internal class ZipInputStream : Stream
|
---|
| 108 | {
|
---|
| 109 | /// <summary>
|
---|
| 110 | /// Create a <c>ZipInputStream</c>, wrapping it around an existing stream.
|
---|
| 111 | /// </summary>
|
---|
| 112 | ///
|
---|
| 113 | /// <remarks>
|
---|
| 114 | ///
|
---|
| 115 | /// <para>
|
---|
| 116 | /// While the <see cref="ZipFile"/> class is generally easier
|
---|
| 117 | /// to use, this class provides an alternative to those
|
---|
| 118 | /// applications that want to read from a zipfile directly,
|
---|
| 119 | /// using a <see cref="System.IO.Stream"/>.
|
---|
| 120 | /// </para>
|
---|
| 121 | ///
|
---|
| 122 | /// <para>
|
---|
| 123 | /// Both the <c>ZipInputStream</c> class and the <c>ZipFile</c> class can be used
|
---|
| 124 | /// to read and extract zip files. Both of them support many of the common zip
|
---|
| 125 | /// features, including Unicode, different compression levels, and ZIP64. The
|
---|
| 126 | /// programming models differ. For example, when extracting entries via calls to
|
---|
| 127 | /// the <c>GetNextEntry()</c> and <c>Read()</c> methods on the
|
---|
| 128 | /// <c>ZipInputStream</c> class, the caller is responsible for creating the file,
|
---|
| 129 | /// writing the bytes into the file, setting the attributes on the file, and
|
---|
| 130 | /// setting the created, last modified, and last accessed timestamps on the
|
---|
| 131 | /// file. All of these things are done automatically by a call to <see
|
---|
| 132 | /// cref="ZipEntry.Extract()">ZipEntry.Extract()</see>. For this reason, the
|
---|
| 133 | /// <c>ZipInputStream</c> is generally recommended for when your application wants
|
---|
| 134 | /// to extract the data, without storing that data into a file.
|
---|
| 135 | /// </para>
|
---|
| 136 | ///
|
---|
| 137 | /// <para>
|
---|
| 138 | /// Aside from the obvious differences in programming model, there are some
|
---|
| 139 | /// differences in capability between the <c>ZipFile</c> class and the
|
---|
| 140 | /// <c>ZipInputStream</c> class.
|
---|
| 141 | /// </para>
|
---|
| 142 | ///
|
---|
| 143 | /// <list type="bullet">
|
---|
| 144 | /// <item>
|
---|
| 145 | /// <c>ZipFile</c> can be used to create or update zip files, or read and extract
|
---|
| 146 | /// zip files. <c>ZipInputStream</c> can be used only to read and extract zip
|
---|
| 147 | /// files. If you want to use a stream to create zip files, check out the <see
|
---|
| 148 | /// cref="ZipOutputStream"/>.
|
---|
| 149 | /// </item>
|
---|
| 150 | ///
|
---|
| 151 | /// <item>
|
---|
| 152 | /// <c>ZipInputStream</c> cannot read segmented or spanned
|
---|
| 153 | /// zip files.
|
---|
| 154 | /// </item>
|
---|
| 155 | ///
|
---|
| 156 | /// <item>
|
---|
| 157 | /// <c>ZipInputStream</c> will not read Zip file comments.
|
---|
| 158 | /// </item>
|
---|
| 159 | ///
|
---|
| 160 | /// <item>
|
---|
| 161 | /// When reading larger files, <c>ZipInputStream</c> will always underperform
|
---|
| 162 | /// <c>ZipFile</c>. This is because the <c>ZipInputStream</c> does a full scan on the
|
---|
| 163 | /// zip file, while the <c>ZipFile</c> class reads the central directory of the
|
---|
| 164 | /// zip file.
|
---|
| 165 | /// </item>
|
---|
| 166 | ///
|
---|
| 167 | /// </list>
|
---|
| 168 | ///
|
---|
| 169 | /// </remarks>
|
---|
| 170 | ///
|
---|
| 171 | /// <param name="stream">
|
---|
| 172 | /// The stream to read. It must be readable. This stream will be closed at
|
---|
| 173 | /// the time the <c>ZipInputStream</c> is closed.
|
---|
| 174 | /// </param>
|
---|
| 175 | ///
|
---|
| 176 | /// <example>
|
---|
| 177 | ///
|
---|
| 178 | /// This example shows how to read a zip file, and extract entries, using the
|
---|
| 179 | /// <c>ZipInputStream</c> class.
|
---|
| 180 | ///
|
---|
| 181 | /// <code lang="C#">
|
---|
| 182 | /// private void Unzip()
|
---|
| 183 | /// {
|
---|
| 184 | /// byte[] buffer= new byte[2048];
|
---|
| 185 | /// int n;
|
---|
| 186 | /// using (var raw = File.Open(inputFileName, FileMode.Open, FileAccess.Read))
|
---|
| 187 | /// {
|
---|
| 188 | /// using (var input= new ZipInputStream(raw))
|
---|
| 189 | /// {
|
---|
| 190 | /// ZipEntry e;
|
---|
| 191 | /// while (( e = input.GetNextEntry()) != null)
|
---|
| 192 | /// {
|
---|
| 193 | /// if (e.IsDirectory) continue;
|
---|
| 194 | /// string outputPath = Path.Combine(extractDir, e.FileName);
|
---|
| 195 | /// using (var output = File.Open(outputPath, FileMode.Create, FileAccess.ReadWrite))
|
---|
| 196 | /// {
|
---|
| 197 | /// while ((n= input.Read(buffer, 0, buffer.Length)) > 0)
|
---|
| 198 | /// {
|
---|
| 199 | /// output.Write(buffer,0,n);
|
---|
| 200 | /// }
|
---|
| 201 | /// }
|
---|
| 202 | /// }
|
---|
| 203 | /// }
|
---|
| 204 | /// }
|
---|
| 205 | /// }
|
---|
| 206 | /// </code>
|
---|
| 207 | ///
|
---|
| 208 | /// <code lang="VB">
|
---|
| 209 | /// Private Sub UnZip()
|
---|
| 210 | /// Dim inputFileName As String = "MyArchive.zip"
|
---|
| 211 | /// Dim extractDir As String = "extract"
|
---|
| 212 | /// Dim buffer As Byte() = New Byte(2048) {}
|
---|
| 213 | /// Using raw As FileStream = File.Open(inputFileName, FileMode.Open, FileAccess.Read)
|
---|
| 214 | /// Using input As ZipInputStream = New ZipInputStream(raw)
|
---|
| 215 | /// Dim e As ZipEntry
|
---|
| 216 | /// Do While (Not e = input.GetNextEntry Is Nothing)
|
---|
| 217 | /// If Not e.IsDirectory Then
|
---|
| 218 | /// Using output As FileStream = File.Open(Path.Combine(extractDir, e.FileName), _
|
---|
| 219 | /// FileMode.Create, FileAccess.ReadWrite)
|
---|
| 220 | /// Dim n As Integer
|
---|
| 221 | /// Do While (n = input.Read(buffer, 0, buffer.Length) > 0)
|
---|
| 222 | /// output.Write(buffer, 0, n)
|
---|
| 223 | /// Loop
|
---|
| 224 | /// End Using
|
---|
| 225 | /// End If
|
---|
| 226 | /// Loop
|
---|
| 227 | /// End Using
|
---|
| 228 | /// End Using
|
---|
| 229 | /// End Sub
|
---|
| 230 | /// </code>
|
---|
| 231 | /// </example>
|
---|
| 232 | public ZipInputStream(Stream stream) : this (stream, false) { }
|
---|
| 233 |
|
---|
| 234 |
|
---|
| 235 |
|
---|
| 236 | /// <summary>
|
---|
| 237 | /// Create a <c>ZipInputStream</c>, given the name of an existing zip file.
|
---|
| 238 | /// </summary>
|
---|
| 239 | ///
|
---|
| 240 | /// <remarks>
|
---|
| 241 | ///
|
---|
| 242 | /// <para>
|
---|
| 243 | /// This constructor opens a <c>FileStream</c> for the given zipfile, and
|
---|
| 244 | /// wraps a <c>ZipInputStream</c> around that. See the documentation for the
|
---|
| 245 | /// <see cref="ZipInputStream(Stream)"/> constructor for full details.
|
---|
| 246 | /// </para>
|
---|
| 247 | ///
|
---|
| 248 | /// <para>
|
---|
| 249 | /// While the <see cref="ZipFile"/> class is generally easier
|
---|
| 250 | /// to use, this class provides an alternative to those
|
---|
| 251 | /// applications that want to read from a zipfile directly,
|
---|
| 252 | /// using a <see cref="System.IO.Stream"/>.
|
---|
| 253 | /// </para>
|
---|
| 254 | ///
|
---|
| 255 | /// </remarks>
|
---|
| 256 | ///
|
---|
| 257 | /// <param name="fileName">
|
---|
| 258 | /// The name of the filesystem file to read.
|
---|
| 259 | /// </param>
|
---|
| 260 | ///
|
---|
| 261 | /// <example>
|
---|
| 262 | ///
|
---|
| 263 | /// This example shows how to read a zip file, and extract entries, using the
|
---|
| 264 | /// <c>ZipInputStream</c> class.
|
---|
| 265 | ///
|
---|
| 266 | /// <code lang="C#">
|
---|
| 267 | /// private void Unzip()
|
---|
| 268 | /// {
|
---|
| 269 | /// byte[] buffer= new byte[2048];
|
---|
| 270 | /// int n;
|
---|
| 271 | /// using (var input= new ZipInputStream(inputFileName))
|
---|
| 272 | /// {
|
---|
| 273 | /// ZipEntry e;
|
---|
| 274 | /// while (( e = input.GetNextEntry()) != null)
|
---|
| 275 | /// {
|
---|
| 276 | /// if (e.IsDirectory) continue;
|
---|
| 277 | /// string outputPath = Path.Combine(extractDir, e.FileName);
|
---|
| 278 | /// using (var output = File.Open(outputPath, FileMode.Create, FileAccess.ReadWrite))
|
---|
| 279 | /// {
|
---|
| 280 | /// while ((n= input.Read(buffer, 0, buffer.Length)) > 0)
|
---|
| 281 | /// {
|
---|
| 282 | /// output.Write(buffer,0,n);
|
---|
| 283 | /// }
|
---|
| 284 | /// }
|
---|
| 285 | /// }
|
---|
| 286 | /// }
|
---|
| 287 | /// }
|
---|
| 288 | /// </code>
|
---|
| 289 | ///
|
---|
| 290 | /// <code lang="VB">
|
---|
| 291 | /// Private Sub UnZip()
|
---|
| 292 | /// Dim inputFileName As String = "MyArchive.zip"
|
---|
| 293 | /// Dim extractDir As String = "extract"
|
---|
| 294 | /// Dim buffer As Byte() = New Byte(2048) {}
|
---|
| 295 | /// Using input As ZipInputStream = New ZipInputStream(inputFileName)
|
---|
| 296 | /// Dim e As ZipEntry
|
---|
| 297 | /// Do While (Not e = input.GetNextEntry Is Nothing)
|
---|
| 298 | /// If Not e.IsDirectory Then
|
---|
| 299 | /// Using output As FileStream = File.Open(Path.Combine(extractDir, e.FileName), _
|
---|
| 300 | /// FileMode.Create, FileAccess.ReadWrite)
|
---|
| 301 | /// Dim n As Integer
|
---|
| 302 | /// Do While (n = input.Read(buffer, 0, buffer.Length) > 0)
|
---|
| 303 | /// output.Write(buffer, 0, n)
|
---|
| 304 | /// Loop
|
---|
| 305 | /// End Using
|
---|
| 306 | /// End If
|
---|
| 307 | /// Loop
|
---|
| 308 | /// End Using
|
---|
| 309 | /// End Sub
|
---|
| 310 | /// </code>
|
---|
| 311 | /// </example>
|
---|
| 312 | public ZipInputStream(String fileName)
|
---|
| 313 | {
|
---|
| 314 | Stream stream = File.Open(fileName, FileMode.Open, FileAccess.Read, FileShare.Read );
|
---|
| 315 | _Init(stream, false, fileName);
|
---|
| 316 | }
|
---|
| 317 |
|
---|
| 318 |
|
---|
| 319 | /// <summary>
|
---|
| 320 | /// Create a <c>ZipInputStream</c>, explicitly specifying whether to
|
---|
| 321 | /// keep the underlying stream open.
|
---|
| 322 | /// </summary>
|
---|
| 323 | ///
|
---|
| 324 | /// <remarks>
|
---|
| 325 | /// See the documentation for the <see
|
---|
| 326 | /// cref="ZipInputStream(Stream)">ZipInputStream(Stream)</see>
|
---|
| 327 | /// constructor for a discussion of the class, and an example of how to use the class.
|
---|
| 328 | /// </remarks>
|
---|
| 329 | ///
|
---|
| 330 | /// <param name="stream">
|
---|
| 331 | /// The stream to read from. It must be readable.
|
---|
| 332 | /// </param>
|
---|
| 333 | ///
|
---|
| 334 | /// <param name="leaveOpen">
|
---|
| 335 | /// true if the application would like the stream
|
---|
| 336 | /// to remain open after the <c>ZipInputStream</c> has been closed.
|
---|
| 337 | /// </param>
|
---|
| 338 | public ZipInputStream(Stream stream, bool leaveOpen)
|
---|
| 339 | {
|
---|
| 340 | _Init(stream, leaveOpen, null);
|
---|
| 341 | }
|
---|
| 342 |
|
---|
| 343 | private void _Init(Stream stream, bool leaveOpen, string name)
|
---|
| 344 | {
|
---|
| 345 | _inputStream = stream;
|
---|
| 346 | if (!_inputStream.CanRead)
|
---|
| 347 | throw new ZipException("The stream must be readable.");
|
---|
| 348 | _container= new ZipContainer(this);
|
---|
| 349 | _provisionalAlternateEncoding = System.Text.Encoding.GetEncoding("IBM437");
|
---|
| 350 | _leaveUnderlyingStreamOpen = leaveOpen;
|
---|
| 351 | _findRequired= true;
|
---|
| 352 | _name = name ?? "(stream)";
|
---|
| 353 | }
|
---|
| 354 |
|
---|
| 355 |
|
---|
| 356 | /// <summary>Provides a string representation of the instance.</summary>
|
---|
| 357 | /// <remarks>
|
---|
| 358 | /// <para>
|
---|
| 359 | /// This can be useful for debugging purposes.
|
---|
| 360 | /// </para>
|
---|
| 361 | /// </remarks>
|
---|
| 362 | /// <returns>a string representation of the instance.</returns>
|
---|
| 363 | public override String ToString()
|
---|
| 364 | {
|
---|
| 365 | return String.Format ("ZipInputStream::{0}(leaveOpen({1})))", _name, _leaveUnderlyingStreamOpen);
|
---|
| 366 | }
|
---|
| 367 |
|
---|
| 368 |
|
---|
| 369 | /// <summary>
|
---|
| 370 | /// The text encoding to use when reading entries into the zip archive, for
|
---|
| 371 | /// those entries whose filenames or comments cannot be encoded with the
|
---|
| 372 | /// default (IBM437) encoding.
|
---|
| 373 | /// </summary>
|
---|
| 374 | ///
|
---|
| 375 | /// <remarks>
|
---|
| 376 | /// <para>
|
---|
| 377 | /// In <see href="http://www.pkware.com/documents/casestudies/APPNOTE.TXT">its
|
---|
| 378 | /// zip specification</see>, PKWare describes two options for encoding
|
---|
| 379 | /// filenames and comments: using IBM437 or UTF-8. But, some archiving tools
|
---|
| 380 | /// or libraries do not follow the specification, and instead encode
|
---|
| 381 | /// characters using the system default code page. For example, WinRAR when
|
---|
| 382 | /// run on a machine in Shanghai may encode filenames with the Big-5 Chinese
|
---|
| 383 | /// (950) code page. This behavior is contrary to the Zip specification, but
|
---|
| 384 | /// it occurs anyway.
|
---|
| 385 | /// </para>
|
---|
| 386 | ///
|
---|
| 387 | /// <para>
|
---|
| 388 | /// When using DotNetZip to read zip archives that use something other than
|
---|
| 389 | /// UTF-8 or IBM437, set this property to specify the code page to use when
|
---|
| 390 | /// reading encoded filenames and comments for each <c>ZipEntry</c> in the zip
|
---|
| 391 | /// file.
|
---|
| 392 | /// </para>
|
---|
| 393 | ///
|
---|
| 394 | /// <para>
|
---|
| 395 | /// This property is "provisional". When the entry in the zip archive is not
|
---|
| 396 | /// explicitly marked as using UTF-8, then IBM437 is used to decode filenames
|
---|
| 397 | /// and comments. If a loss of data would result from using IBM436 -
|
---|
| 398 | /// specifically when encoding and decoding is not reflexive - the codepage
|
---|
| 399 | /// specified here is used. It is possible, therefore, to have a given entry
|
---|
| 400 | /// with a <c>Comment</c> encoded in IBM437 and a <c>FileName</c> encoded with
|
---|
| 401 | /// the specified "provisional" codepage.
|
---|
| 402 | /// </para>
|
---|
| 403 | ///
|
---|
| 404 | /// <para>
|
---|
| 405 | /// When a zip file uses an arbitrary, non-UTF8 code page for encoding, there
|
---|
| 406 | /// is no standard way for the reader application - whether DotNetZip, WinZip,
|
---|
| 407 | /// WinRar, or something else - to know which codepage has been used for the
|
---|
| 408 | /// entries. Readers of zip files are not able to inspect the zip file and
|
---|
| 409 | /// determine the codepage that was used for the entries contained within it.
|
---|
| 410 | /// It is left to the application or user to determine the necessary codepage
|
---|
| 411 | /// when reading zip files encoded this way. If you use an incorrect codepage
|
---|
| 412 | /// when reading a zipfile, you will get entries with filenames that are
|
---|
| 413 | /// incorrect, and the incorrect filenames may even contain characters that
|
---|
| 414 | /// are not legal for use within filenames in Windows. Extracting entries with
|
---|
| 415 | /// illegal characters in the filenames will lead to exceptions. It's too bad,
|
---|
| 416 | /// but this is just the way things are with code pages in zip files. Caveat
|
---|
| 417 | /// Emptor.
|
---|
| 418 | /// </para>
|
---|
| 419 | ///
|
---|
| 420 | /// </remarks>
|
---|
| 421 | public System.Text.Encoding ProvisionalAlternateEncoding
|
---|
| 422 | {
|
---|
| 423 | get
|
---|
| 424 | {
|
---|
| 425 | return _provisionalAlternateEncoding;
|
---|
| 426 | }
|
---|
| 427 | set
|
---|
| 428 | {
|
---|
| 429 | _provisionalAlternateEncoding = value;
|
---|
| 430 | }
|
---|
| 431 | }
|
---|
| 432 |
|
---|
| 433 |
|
---|
| 434 | /// <summary>
|
---|
| 435 | /// Size of the work buffer to use for the ZLIB codec during decompression.
|
---|
| 436 | /// </summary>
|
---|
| 437 | ///
|
---|
| 438 | /// <remarks>
|
---|
| 439 | /// Setting this affects the performance and memory efficiency of compression
|
---|
| 440 | /// and decompression. For larger files, setting this to a larger size may
|
---|
| 441 | /// improve performance, but the exact numbers vary depending on available
|
---|
| 442 | /// memory, and a bunch of other variables. I don't have good firm
|
---|
| 443 | /// recommendations on how to set it. You'll have to test it yourself. Or
|
---|
| 444 | /// just leave it alone and accept the default.
|
---|
| 445 | /// </remarks>
|
---|
| 446 | public int CodecBufferSize
|
---|
| 447 | {
|
---|
| 448 | get;
|
---|
| 449 | set;
|
---|
| 450 | }
|
---|
| 451 |
|
---|
| 452 |
|
---|
| 453 | /// <summary>
|
---|
| 454 | /// Sets the password to be used on the <c>ZipInputStream</c> instance.
|
---|
| 455 | /// </summary>
|
---|
| 456 | ///
|
---|
| 457 | /// <remarks>
|
---|
| 458 | ///
|
---|
| 459 | /// <para>
|
---|
| 460 | /// When reading a zip archive, this password is used to read and decrypt the
|
---|
| 461 | /// entries that are encrypted within the zip file. When entries within a zip
|
---|
| 462 | /// file use different passwords, set the appropriate password for the entry
|
---|
| 463 | /// before the first call to <c>Read()</c> for each entry.
|
---|
| 464 | /// </para>
|
---|
| 465 | ///
|
---|
| 466 | /// <para>
|
---|
| 467 | /// When reading an entry that is not encrypted, the value of this property is
|
---|
| 468 | /// ignored.
|
---|
| 469 | /// </para>
|
---|
| 470 | ///
|
---|
| 471 | /// </remarks>
|
---|
| 472 | ///
|
---|
| 473 | /// <example>
|
---|
| 474 | ///
|
---|
| 475 | /// This example uses the ZipInputStream to read and extract entries from a
|
---|
| 476 | /// zip file, using a potentially different password for each entry.
|
---|
| 477 | ///
|
---|
| 478 | /// <code lang="C#">
|
---|
| 479 | /// byte[] buffer= new byte[2048];
|
---|
| 480 | /// int n;
|
---|
| 481 | /// using (var raw = File.Open(_inputFileName, FileMode.Open, FileAccess.Read ))
|
---|
| 482 | /// {
|
---|
| 483 | /// using (var input= new ZipInputStream(raw))
|
---|
| 484 | /// {
|
---|
| 485 | /// ZipEntry e;
|
---|
| 486 | /// while (( e = input.GetNextEntry()) != null)
|
---|
| 487 | /// {
|
---|
| 488 | /// input.Password = PasswordForEntry(e.FileName);
|
---|
| 489 | /// if (e.IsDirectory) continue;
|
---|
| 490 | /// string outputPath = Path.Combine(_extractDir, e.FileName);
|
---|
| 491 | /// using (var output = File.Open(outputPath, FileMode.Create, FileAccess.ReadWrite))
|
---|
| 492 | /// {
|
---|
| 493 | /// while ((n= input.Read(buffer,0,buffer.Length)) > 0)
|
---|
| 494 | /// {
|
---|
| 495 | /// output.Write(buffer,0,n);
|
---|
| 496 | /// }
|
---|
| 497 | /// }
|
---|
| 498 | /// }
|
---|
| 499 | /// }
|
---|
| 500 | /// }
|
---|
| 501 | ///
|
---|
| 502 | /// </code>
|
---|
| 503 | /// </example>
|
---|
| 504 | public String Password
|
---|
| 505 | {
|
---|
| 506 | set
|
---|
| 507 | {
|
---|
| 508 | if (_closed)
|
---|
| 509 | {
|
---|
| 510 | _exceptionPending = true;
|
---|
| 511 | throw new System.InvalidOperationException("The stream has been closed.");
|
---|
| 512 | }
|
---|
| 513 | _Password = value;
|
---|
| 514 | }
|
---|
| 515 | }
|
---|
| 516 |
|
---|
| 517 |
|
---|
| 518 | private void SetupStream()
|
---|
| 519 | {
|
---|
| 520 | // Seek to the correct posn in the file, and open a
|
---|
| 521 | // stream that can be read.
|
---|
| 522 | _crcStream= _currentEntry.InternalOpenReader(_Password);
|
---|
| 523 | _LeftToRead = _crcStream.Length;
|
---|
| 524 | _needSetup = false;
|
---|
| 525 | }
|
---|
| 526 |
|
---|
| 527 |
|
---|
| 528 |
|
---|
| 529 | internal Stream ReadStream
|
---|
| 530 | {
|
---|
| 531 | get
|
---|
| 532 | {
|
---|
| 533 | return _inputStream;
|
---|
| 534 | }
|
---|
| 535 | }
|
---|
| 536 |
|
---|
| 537 |
|
---|
| 538 | /// <summary>
|
---|
| 539 | /// Read the data from the stream into the buffer.
|
---|
| 540 | /// </summary>
|
---|
| 541 | ///
|
---|
| 542 | /// <remarks>
|
---|
| 543 | /// <para>
|
---|
| 544 | /// The data for the zipentry will be decrypted and uncompressed, as
|
---|
| 545 | /// necessary, before being copied into the buffer.
|
---|
| 546 | /// </para>
|
---|
| 547 | ///
|
---|
| 548 | /// <para>
|
---|
| 549 | /// You must set the <see cref="Password"/> property before calling
|
---|
| 550 | /// <c>Read()</c> the first time for an encrypted entry. To determine if an
|
---|
| 551 | /// entry is encrypted and requires a password, check the <see
|
---|
| 552 | /// cref="ZipEntry.Encryption">ZipEntry.Encryption</see> property.
|
---|
| 553 | /// </para>
|
---|
| 554 | /// </remarks>
|
---|
| 555 | ///
|
---|
| 556 | /// <param name="buffer">The buffer to hold the data read from the stream.</param>
|
---|
| 557 | /// <param name="offset">the offset within the buffer to copy the first byte read.</param>
|
---|
| 558 | /// <param name="count">the number of bytes to read.</param>
|
---|
| 559 | /// <returns>the number of bytes read, after decryption and decompression.</returns>
|
---|
| 560 | public override int Read(byte[] buffer, int offset, int count)
|
---|
| 561 | {
|
---|
| 562 | if (_closed)
|
---|
| 563 | {
|
---|
| 564 | _exceptionPending = true;
|
---|
| 565 | throw new System.InvalidOperationException("The stream has been closed.");
|
---|
| 566 | }
|
---|
| 567 |
|
---|
| 568 | if (_needSetup)
|
---|
| 569 | SetupStream();
|
---|
| 570 |
|
---|
| 571 | if (_LeftToRead == 0) return 0;
|
---|
| 572 |
|
---|
| 573 | int len = (_LeftToRead > count) ? count : (int)_LeftToRead;
|
---|
| 574 | int n = _crcStream.Read(buffer, offset, len);
|
---|
| 575 |
|
---|
| 576 | _LeftToRead -= n;
|
---|
| 577 |
|
---|
| 578 | if (_LeftToRead == 0)
|
---|
| 579 | {
|
---|
| 580 | int CrcResult = _crcStream.Crc;
|
---|
| 581 | _currentEntry.VerifyCrcAfterExtract(CrcResult);
|
---|
| 582 | _inputStream.Seek(_endOfEntry, SeekOrigin.Begin);
|
---|
| 583 | // workitem 10178
|
---|
| 584 | SharedUtilities.Workaround_Ladybug318918(_inputStream);
|
---|
| 585 | }
|
---|
| 586 |
|
---|
| 587 | return n;
|
---|
| 588 | }
|
---|
| 589 |
|
---|
| 590 |
|
---|
| 591 |
|
---|
| 592 | /// <summary>
|
---|
| 593 | /// Read the next entry from the zip file.
|
---|
| 594 | /// </summary>
|
---|
| 595 | ///
|
---|
| 596 | /// <remarks>
|
---|
| 597 | /// <para>
|
---|
| 598 | /// Call this method just before calling <see cref="Read(byte[], int, int)"/>,
|
---|
| 599 | /// to position the pointer in the zip file to the next entry that can be
|
---|
| 600 | /// read. Subsequent calls to <c>Read()</c>, will decrypt and decompress the
|
---|
| 601 | /// data in the zip file, until <c>Read()</c> returns 0.
|
---|
| 602 | /// </para>
|
---|
| 603 | ///
|
---|
| 604 | /// <para>
|
---|
| 605 | /// Each time you call <c>GetNextEntry()</c>, the pointer in the wrapped
|
---|
| 606 | /// stream is moved to the next entry in the zip file. If you call <see
|
---|
| 607 | /// cref="Seek(long, SeekOrigin)"/>, and thus re-position the pointer within
|
---|
| 608 | /// the file, you will need to call <c>GetNextEntry()</c> again, to insure
|
---|
| 609 | /// that the file pointer is positioned at the beginning of a zip entry.
|
---|
| 610 | /// </para>
|
---|
| 611 | ///
|
---|
| 612 | /// <para>
|
---|
| 613 | /// This method returns the <c>ZipEntry</c>. Using a stream approach, you will
|
---|
| 614 | /// read the raw bytes for an entry in a zip file via calls to <c>Read()</c>.
|
---|
| 615 | /// Alternatively, you can extract an entry into a file, or a stream, by
|
---|
| 616 | /// calling <see cref="ZipEntry.Extract()"/>, or one of its siblings.
|
---|
| 617 | /// </para>
|
---|
| 618 | ///
|
---|
| 619 | /// </remarks>
|
---|
| 620 | ///
|
---|
| 621 | /// <returns>
|
---|
| 622 | /// The <c>ZipEntry</c> read. Returns null (or Nothing in VB) if there are no more
|
---|
| 623 | /// entries in the zip file.
|
---|
| 624 | /// </returns>
|
---|
| 625 | ///
|
---|
| 626 | public ZipEntry GetNextEntry()
|
---|
| 627 | {
|
---|
| 628 | if (_findRequired)
|
---|
| 629 | {
|
---|
| 630 | // find the next signature
|
---|
| 631 | long d = SharedUtilities.FindSignature(_inputStream, ZipConstants.ZipEntrySignature);
|
---|
| 632 | if (d == -1) return null;
|
---|
| 633 | // back up 4 bytes: ReadEntry assumes the file pointer is positioned before the entry signature
|
---|
| 634 | _inputStream.Seek(-4, SeekOrigin.Current);
|
---|
| 635 | // workitem 10178
|
---|
| 636 | SharedUtilities.Workaround_Ladybug318918(_inputStream);
|
---|
| 637 | }
|
---|
| 638 | // workitem 10923
|
---|
| 639 | else if (_firstEntry)
|
---|
| 640 | {
|
---|
| 641 | // we've already read one entry.
|
---|
| 642 | // Seek to the end of it.
|
---|
| 643 | _inputStream.Seek(_endOfEntry, SeekOrigin.Begin);
|
---|
| 644 | SharedUtilities.Workaround_Ladybug318918(_inputStream);
|
---|
| 645 | }
|
---|
| 646 |
|
---|
| 647 | _currentEntry = ZipEntry.ReadEntry(_container, !_firstEntry);
|
---|
| 648 | // ReadEntry leaves the file position after all the entry
|
---|
| 649 | // data and the optional bit-3 data descriptpr. This is
|
---|
| 650 | // where the next entry would normally start.
|
---|
| 651 | _endOfEntry = _inputStream.Position;
|
---|
| 652 | _firstEntry = true;
|
---|
| 653 | _needSetup = true;
|
---|
| 654 | _findRequired= false;
|
---|
| 655 | return _currentEntry;
|
---|
| 656 | }
|
---|
| 657 |
|
---|
| 658 |
|
---|
| 659 | /// <summary>
|
---|
| 660 | /// Dispose the stream.
|
---|
| 661 | /// </summary>
|
---|
| 662 | ///
|
---|
| 663 | /// <remarks>
|
---|
| 664 | /// <para>
|
---|
| 665 | /// This method disposes the ZipInputStream. It may also close the
|
---|
| 666 | /// underlying stream, depending on which constructor was used.
|
---|
| 667 | /// </para>
|
---|
| 668 | ///
|
---|
| 669 | /// <para>
|
---|
| 670 | /// Typically the application will call <c>Dispose()</c> implicitly, via
|
---|
| 671 | /// a <c>using</c> statement in C#, or a <c>Using</c> statement in VB.
|
---|
| 672 | /// </para>
|
---|
| 673 | ///
|
---|
| 674 | /// <para>
|
---|
| 675 | /// Application code won't call this code directly. This method may
|
---|
| 676 | /// be invoked in two distinct scenarios. If disposing == true, the
|
---|
| 677 | /// method has been called directly or indirectly by a user's code,
|
---|
| 678 | /// for example via the public Dispose() method. In this case, both
|
---|
| 679 | /// managed and unmanaged resources can be referenced and disposed.
|
---|
| 680 | /// If disposing == false, the method has been called by the runtime
|
---|
| 681 | /// from inside the object finalizer and this method should not
|
---|
| 682 | /// reference other objects; in that case only unmanaged resources
|
---|
| 683 | /// must be referenced or disposed.
|
---|
| 684 | /// </para>
|
---|
| 685 | /// </remarks>
|
---|
| 686 | ///
|
---|
| 687 | /// <param name="disposing">
|
---|
| 688 | /// true if the Dispose method was invoked by user code.
|
---|
| 689 | /// </param>
|
---|
| 690 | protected override void Dispose(bool disposing)
|
---|
| 691 | {
|
---|
| 692 | if (_closed) return;
|
---|
| 693 |
|
---|
| 694 | if (disposing) // not called from finalizer
|
---|
| 695 | {
|
---|
| 696 | // When ZipInputStream is used within a using clause, and an
|
---|
| 697 | // exception is thrown, Close() is invoked. But we don't want to
|
---|
| 698 | // try to write anything in that case. Eventually the exception
|
---|
| 699 | // will be propagated to the application.
|
---|
| 700 | if (_exceptionPending) return;
|
---|
| 701 |
|
---|
| 702 | if (!_leaveUnderlyingStreamOpen)
|
---|
| 703 | {
|
---|
| 704 | #if NETCF
|
---|
| 705 | _inputStream.Close();
|
---|
| 706 | #else
|
---|
| 707 | _inputStream.Dispose();
|
---|
| 708 | #endif
|
---|
| 709 | }
|
---|
| 710 | }
|
---|
| 711 | _closed= true;
|
---|
| 712 | }
|
---|
| 713 |
|
---|
| 714 |
|
---|
| 715 | /// <summary>
|
---|
| 716 | /// Always returns true.
|
---|
| 717 | /// </summary>
|
---|
| 718 | public override bool CanRead { get { return true; }}
|
---|
| 719 |
|
---|
| 720 | /// <summary>
|
---|
| 721 | /// Returns the value of <c>CanSeek</c> for the underlying (wrapped) stream.
|
---|
| 722 | /// </summary>
|
---|
| 723 | public override bool CanSeek { get { return _inputStream.CanSeek; } }
|
---|
| 724 |
|
---|
| 725 | /// <summary>
|
---|
| 726 | /// Always returns false.
|
---|
| 727 | /// </summary>
|
---|
| 728 | public override bool CanWrite { get { return false; } }
|
---|
| 729 |
|
---|
| 730 | /// <summary>
|
---|
| 731 | /// Returns the length of the underlying stream.
|
---|
| 732 | /// </summary>
|
---|
| 733 | public override long Length { get { return _inputStream.Length; }}
|
---|
| 734 |
|
---|
| 735 | /// <summary>
|
---|
| 736 | /// Gets or sets the position of the underlying stream.
|
---|
| 737 | /// </summary>
|
---|
| 738 | /// <remarks>
|
---|
| 739 | /// Setting the position is equivalent to calling <c>Seek(value, SeekOrigin.Begin)</c>.
|
---|
| 740 | /// </remarks>
|
---|
| 741 | public override long Position
|
---|
| 742 | {
|
---|
| 743 | get { return _inputStream.Position;}
|
---|
| 744 | set { Seek(value, SeekOrigin.Begin); }
|
---|
| 745 | }
|
---|
| 746 |
|
---|
| 747 | /// <summary>
|
---|
| 748 | /// This is a no-op.
|
---|
| 749 | /// </summary>
|
---|
| 750 | public override void Flush()
|
---|
| 751 | {
|
---|
| 752 | throw new NotSupportedException("Flush");
|
---|
| 753 | }
|
---|
| 754 |
|
---|
| 755 |
|
---|
| 756 | /// <summary>
|
---|
| 757 | /// This method always throws a NotSupportedException.
|
---|
| 758 | /// </summary>
|
---|
| 759 | /// <param name="buffer">ignored</param>
|
---|
| 760 | /// <param name="offset">ignored</param>
|
---|
| 761 | /// <param name="count">ignored</param>
|
---|
| 762 | public override void Write(byte[] buffer, int offset, int count)
|
---|
| 763 | {
|
---|
| 764 | throw new NotSupportedException("Write");
|
---|
| 765 | }
|
---|
| 766 |
|
---|
| 767 |
|
---|
| 768 | /// <summary>
|
---|
| 769 | /// This method seeks in the underlying stream.
|
---|
| 770 | /// </summary>
|
---|
| 771 | ///
|
---|
| 772 | /// <remarks>
|
---|
| 773 | /// <para>
|
---|
| 774 | /// Call this method if you want to seek around within the zip file for random access.
|
---|
| 775 | /// </para>
|
---|
| 776 | ///
|
---|
| 777 | /// <para>
|
---|
| 778 | /// Applications can intermix calls to <c>Seek()</c> with calls to <see
|
---|
| 779 | /// cref="GetNextEntry()"/>. After a call to <c>Seek()</c>,
|
---|
| 780 | /// <c>GetNextEntry()</c> will get the next <c>ZipEntry</c> that falls after
|
---|
| 781 | /// the current position in the input stream. You're on your own for finding
|
---|
| 782 | /// out just where to seek in the stream, to get to the various entries.
|
---|
| 783 | /// </para>
|
---|
| 784 | ///
|
---|
| 785 | /// </remarks>
|
---|
| 786 | ///
|
---|
| 787 | /// <param name="offset">the offset point to seek to</param>
|
---|
| 788 | /// <param name="origin">the reference point from which to seek</param>
|
---|
| 789 | /// <returns>The new position</returns>
|
---|
| 790 | public override long Seek(long offset, SeekOrigin origin)
|
---|
| 791 | {
|
---|
| 792 | _findRequired= true;
|
---|
| 793 | var x = _inputStream.Seek(offset, origin);
|
---|
| 794 | // workitem 10178
|
---|
| 795 | SharedUtilities.Workaround_Ladybug318918(_inputStream);
|
---|
| 796 | return x;
|
---|
| 797 | }
|
---|
| 798 |
|
---|
| 799 | /// <summary>
|
---|
| 800 | /// This method always throws a NotSupportedException.
|
---|
| 801 | /// </summary>
|
---|
| 802 | /// <param name="value">ignored</param>
|
---|
| 803 | public override void SetLength(long value)
|
---|
| 804 | {
|
---|
| 805 | throw new NotSupportedException();
|
---|
| 806 | }
|
---|
| 807 |
|
---|
| 808 |
|
---|
| 809 | private Stream _inputStream;
|
---|
| 810 | private System.Text.Encoding _provisionalAlternateEncoding;
|
---|
| 811 | private ZipEntry _currentEntry;
|
---|
| 812 | private bool _firstEntry;
|
---|
| 813 | private bool _needSetup;
|
---|
| 814 | private ZipContainer _container;
|
---|
| 815 | private CrcCalculatorStream _crcStream;
|
---|
| 816 | private Int64 _LeftToRead;
|
---|
| 817 | internal String _Password;
|
---|
| 818 | private Int64 _endOfEntry;
|
---|
| 819 | private string _name;
|
---|
| 820 |
|
---|
| 821 | private bool _leaveUnderlyingStreamOpen;
|
---|
| 822 | private bool _closed;
|
---|
| 823 | private bool _findRequired;
|
---|
| 824 | private bool _exceptionPending;
|
---|
| 825 | }
|
---|
| 826 |
|
---|
| 827 |
|
---|
| 828 |
|
---|
| 829 | } |
---|