zip_reader.h revision 5d1f7b1de12d16ceb2c938c56701a3e8bfa558f7
1// Copyright (c) 2011 The Chromium Authors. All rights reserved. 2// Use of this source code is governed by a BSD-style license that can be 3// found in the LICENSE file. 4#ifndef THIRD_PARTY_ZLIB_GOOGLE_ZIP_READER_H_ 5#define THIRD_PARTY_ZLIB_GOOGLE_ZIP_READER_H_ 6 7#include <string> 8 9#include "base/basictypes.h" 10#include "base/callback.h" 11#include "base/file_util.h" 12#include "base/files/file_path.h" 13#include "base/memory/scoped_ptr.h" 14#include "base/memory/weak_ptr.h" 15#include "base/platform_file.h" 16#include "base/time/time.h" 17 18#if defined(USE_SYSTEM_MINIZIP) 19#include <minizip/unzip.h> 20#else 21#include "third_party/zlib/contrib/minizip/unzip.h" 22#endif 23 24namespace zip { 25 26// This class is used for reading zip files. A typical use case of this 27// class is to scan entries in a zip file and extract them. The code will 28// look like: 29// 30// ZipReader reader; 31// reader.Open(zip_file_path); 32// while (reader.HasMore()) { 33// reader.OpenCurrentEntryInZip(); 34// reader.ExtractCurrentEntryToDirectory(output_directory_path); 35// reader.AdvanceToNextEntry(); 36// } 37// 38// For simplicty, error checking is omitted in the example code above. The 39// production code should check return values from all of these functions. 40// 41// This calls can also be used for random access of contents in a zip file 42// using LocateAndOpenEntry(). 43// 44class ZipReader { 45 public: 46 // A callback that is called when the operation is successful. 47 typedef base::Closure SuccessCallback; 48 // A callback that is called when the operation fails. 49 typedef base::Closure FailureCallback; 50 // A callback that is called periodically during the operation with the number 51 // of bytes that have been processed so far. 52 typedef base::Callback<void(int64)> ProgressCallback; 53 54 // This class represents information of an entry (file or directory) in 55 // a zip file. 56 class EntryInfo { 57 public: 58 EntryInfo(const std::string& filename_in_zip, 59 const unz_file_info& raw_file_info); 60 61 // Returns the file path. The path is usually relative like 62 // "foo/bar.txt", but if it's absolute, is_unsafe() returns true. 63 const base::FilePath& file_path() const { return file_path_; } 64 65 // Returns the size of the original file (i.e. after uncompressed). 66 // Returns 0 if the entry is a directory. 67 int64 original_size() const { return original_size_; } 68 69 // Returns the last modified time. If the time stored in the zip file was 70 // not valid, the unix epoch will be returned. 71 // 72 // The time stored in the zip archive uses the MS-DOS date and time format. 73 // http://msdn.microsoft.com/en-us/library/ms724247(v=vs.85).aspx 74 // As such the following limitations apply: 75 // * only years from 1980 to 2107 can be represented. 76 // * the time stamp has a 2 second resolution. 77 // * there's no timezone information, so the time is interpreted as local. 78 base::Time last_modified() const { return last_modified_; } 79 80 // Returns true if the entry is a directory. 81 bool is_directory() const { return is_directory_; } 82 83 // Returns true if the entry is unsafe, like having ".." or invalid 84 // UTF-8 characters in its file name, or the file path is absolute. 85 bool is_unsafe() const { return is_unsafe_; } 86 87 private: 88 const base::FilePath file_path_; 89 int64 original_size_; 90 base::Time last_modified_; 91 bool is_directory_; 92 bool is_unsafe_; 93 DISALLOW_COPY_AND_ASSIGN(EntryInfo); 94 }; 95 96 ZipReader(); 97 ~ZipReader(); 98 99 // Opens the zip file specified by |zip_file_path|. Returns true on 100 // success. 101 bool Open(const base::FilePath& zip_file_path); 102 103 // Opens the zip file referred to by the platform file |zip_fd|. 104 // Returns true on success. 105 bool OpenFromPlatformFile(base::PlatformFile zip_fd); 106 107 // Opens the zip data stored in |data|. This class uses a weak reference to 108 // the given sring while extracting files, i.e. the caller should keep the 109 // string until it finishes extracting files. 110 bool OpenFromString(const std::string& data); 111 112 // Closes the currently opened zip file. This function is called in the 113 // destructor of the class, so you usually don't need to call this. 114 void Close(); 115 116 // Returns true if there is at least one entry to read. This function is 117 // used to scan entries with AdvanceToNextEntry(), like: 118 // 119 // while (reader.HasMore()) { 120 // // Do something with the current file here. 121 // reader.AdvanceToNextEntry(); 122 // } 123 bool HasMore(); 124 125 // Advances the next entry. Returns true on success. 126 bool AdvanceToNextEntry(); 127 128 // Opens the current entry in the zip file. On success, returns true and 129 // updates the the current entry state (i.e. current_entry_info() is 130 // updated). This function should be called before operations over the 131 // current entry like ExtractCurrentEntryToFile(). 132 // 133 // Note that there is no CloseCurrentEntryInZip(). The the current entry 134 // state is reset automatically as needed. 135 bool OpenCurrentEntryInZip(); 136 137 // Locates an entry in the zip file and opens it. Returns true on 138 // success. This function internally calls OpenCurrentEntryInZip() on 139 // success. On failure, current_entry_info() becomes NULL. 140 bool LocateAndOpenEntry(const base::FilePath& path_in_zip); 141 142 // Extracts the current entry to the given output file path. If the 143 // current file is a directory, just creates a directory 144 // instead. Returns true on success. OpenCurrentEntryInZip() must be 145 // called beforehand. 146 // 147 // This function preserves the timestamp of the original entry. If that 148 // timestamp is not valid, the timestamp will be set to the current time. 149 bool ExtractCurrentEntryToFilePath(const base::FilePath& output_file_path); 150 151 // Asynchronously extracts the current entry to the given output file path. 152 // If the current entry is a directory it just creates the directory 153 // synchronously instead. OpenCurrentEntryInZip() must be called beforehand. 154 // success_callback will be called on success and failure_callback will be 155 // called on failure. progress_callback will be called at least once. 156 // Callbacks will be posted to the current MessageLoop in-order. 157 void ExtractCurrentEntryToFilePathAsync( 158 const base::FilePath& output_file_path, 159 const SuccessCallback& success_callback, 160 const FailureCallback& failure_callback, 161 const ProgressCallback& progress_callback); 162 163 // Extracts the current entry to the given output directory path using 164 // ExtractCurrentEntryToFilePath(). Sub directories are created as needed 165 // based on the file path of the current entry. For example, if the file 166 // path in zip is "foo/bar.txt", and the output directory is "output", 167 // "output/foo/bar.txt" will be created. 168 // 169 // Returns true on success. OpenCurrentEntryInZip() must be called 170 // beforehand. 171 // 172 // This function preserves the timestamp of the original entry. If that 173 // timestamp is not valid, the timestamp will be set to the current time. 174 bool ExtractCurrentEntryIntoDirectory( 175 const base::FilePath& output_directory_path); 176 177#if defined(OS_POSIX) 178 // Extracts the current entry by writing directly to a file descriptor. 179 // Does not close the file descriptor. Returns true on success. 180 bool ExtractCurrentEntryToFd(int fd); 181#endif 182 183 // Returns the current entry info. Returns NULL if the current entry is 184 // not yet opened. OpenCurrentEntryInZip() must be called beforehand. 185 EntryInfo* current_entry_info() const { 186 return current_entry_info_.get(); 187 } 188 189 // Returns the number of entries in the zip file. 190 // Open() must be called beforehand. 191 int num_entries() const { return num_entries_; } 192 193 private: 194 // Common code used both in Open and OpenFromFd. 195 bool OpenInternal(); 196 197 // Resets the internal state. 198 void Reset(); 199 200 // Extracts a chunk of the file to the target. Will post a task for the next 201 // chunk and success/failure/progress callbacks as necessary. 202 void ExtractChunk(base::PlatformFile target_file, 203 const SuccessCallback& success_callback, 204 const FailureCallback& failure_callback, 205 const ProgressCallback& progress_callback, 206 const int64 offset); 207 208 unzFile zip_file_; 209 int num_entries_; 210 bool reached_end_; 211 scoped_ptr<EntryInfo> current_entry_info_; 212 213 base::WeakPtrFactory<ZipReader> weak_ptr_factory_; 214 215 DISALLOW_COPY_AND_ASSIGN(ZipReader); 216}; 217 218} // namespace zip 219 220#endif // THIRD_PARTY_ZLIB_GOOGLE_ZIP_READER_H_ 221