2 Copyright (c) 2004-2009, John Hurst
5 Redistribution and use in source and binary forms, with or without
6 modification, are permitted provided that the following conditions
8 1. Redistributions of source code must retain the above copyright
9 notice, this list of conditions and the following disclaimer.
10 2. Redistributions in binary form must reproduce the above copyright
11 notice, this list of conditions and the following disclaimer in the
12 documentation and/or other materials provided with the distribution.
13 3. The name of the author may not be used to endorse or promote products
14 derived from this software without specific prior written permission.
16 THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
17 IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
18 OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
19 IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
20 INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
21 NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
22 DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
23 THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
24 (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
25 THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
28 \version $Id: KM_fileio.h,v 1.17 2009/06/22 05:49:02 jhurst Exp $
29 \brief portable file i/o
37 #include <boost/filesystem.hpp>
38 #include <openssl/md5.h>
46 # include <sys/types.h>
60 Result_t Open(const char *);
61 Result_t GetNext(char *);
64 boost::filesystem::directory_iterator _iterator;
68 typedef __int64 fsize_t;
69 typedef __int64 fpos_t;
70 typedef HANDLE FileHandle;
73 SP_BEGIN = FILE_BEGIN,
74 SP_POS = FILE_CURRENT,
78 typedef off_t fsize_t;
80 typedef int FileHandle;
81 const FileHandle INVALID_HANDLE_VALUE = -1L;
91 #ifndef KM_SMALL_FILES_OK
92 template <bool sizecheck> void compile_time_size_checker();
93 template <> inline void compile_time_size_checker<false>() {}
95 // READ THIS if your compiler is complaining about a previously declared implementation of
96 // compile_time_size_checker(). For example, GCC 4.0.1 looks like this:
98 // error: 'void Kumu::compile_time_size_checker() [with bool sizecheck = false]' previously declared here
100 // This is happening because the equality being tested below is false. The reason for this
101 // will depend on your OS, but on Linux it is probably because you have not used -D_FILE_OFFSET_BITS=64
102 // Adding this magic macro to your CFLAGS will get you going again. If you are on a system that
103 // does not support 64-bit files, you can disable this check by using -DKM_SMALL_FILES_OK. You
104 // will then of course be limited to file sizes < 4GB.
106 template <> inline void compile_time_size_checker<sizeof(Kumu::fsize_t)==sizeof(ui64_t)>() {}
110 const ui32_t Kilobyte = 1024;
111 const ui32_t Megabyte = Kilobyte * Kilobyte;
112 const ui32_t Gigabyte = Megabyte * Kilobyte;
114 const ui32_t MaxFilePath = Kilobyte;
117 //------------------------------------------------------------------------------------------
119 //------------------------------------------------------------------------------------------
122 typedef std::list<std::string> PathCompList_t; // a list of path components
123 typedef std::list<std::string> PathList_t; // a list of paths
126 bool PathExists(const std::string& Path); // true if the path exists in the filesystem
127 bool PathIsFile(const std::string& Path); // true if the path exists in the filesystem and is a file
128 bool PathIsDirectory(const std::string& Path); // true if the path exists in the filesystem and is a directory
129 fsize_t FileSize(const std::string& Path); // returns the size of a regular file, 0 for a directory or device
130 bool PathsAreEquivalent(const std::string& lhs, const std::string& rhs); // true if paths point to the same filesystem entry
132 // Returns free space and total space available for the given path
133 Result_t FreeSpaceForPath(const std::string& path, Kumu::fsize_t& free_space, Kumu::fsize_t& total_space);
135 // split and reassemble paths as lists of path components
136 PathCompList_t& PathToComponents(const std::string& Path, PathCompList_t& CList, char separator = '/'); // removes '//'
137 std::string ComponentsToPath(const PathCompList_t& CList, char separator = '/');
138 std::string ComponentsToAbsolutePath(const PathCompList_t& CList, char separator = '/'); // add separator to the front
139 bool PathHasComponents(const std::string& Path, char separator = '/'); // true if paths starts with separator
141 bool PathIsAbsolute(const std::string& Path, char separator = '/'); // true if path begins with separator
142 std::string PathMakeAbsolute(const std::string& Path, char separator = '/'); // compute position of relative path using getcwd()
143 std::string PathMakeLocal(const std::string& Path, const std::string& Parent); // remove Parent from front of Path, if it exists
144 std::string PathMakeCanonical(const std::string& Path, char separator = '/'); // remove '.' and '..'
147 std::string PathBasename(const std::string& Path, char separator = '/'); // returns right-most path element (list back())
148 std::string PathDirname(const std::string& Path, char separator = '/'); // returns everything but the right-most element
149 std::string PathGetExtension(const std::string& Path); // returns everything in the right-most element following the right-most '.'
150 std::string PathSetExtension(const std::string& Path, const std::string& Extension); // empty extension removes '.' as well
152 std::string PathJoin(const std::string& Path1, const std::string& Path2, char separator = '/');
153 std::string PathJoin(const std::string& Path1, const std::string& Path2, const std::string& Path3, char separator = '/');
154 std::string PathJoin(const std::string& Path1, const std::string& Path2,
155 const std::string& Path3, const std::string& Path4, char separator = '/');
158 //------------------------------------------------------------------------------------------
160 //------------------------------------------------------------------------------------------
162 // An interface for a path matching function, used by FindInPath() and FindInPaths() below
167 virtual ~IPathMatch() {}
168 virtual bool Match(const std::string& s) const = 0;
171 // matches any pathname
172 class PathMatchAny : public IPathMatch
175 virtual ~PathMatchAny() {}
176 inline bool Match(const std::string&) const { return true; }
180 // matches pathnames using a regular expression
181 class PathMatchRegex : public IPathMatch
185 const PathMatchRegex& operator=(const PathMatchRegex&);
188 PathMatchRegex(const std::string& Pattern);
189 PathMatchRegex(const PathMatchRegex&);
190 virtual ~PathMatchRegex();
191 bool Match(const std::string& s) const;
194 // matches pathnames using a Bourne shell glob expression
195 class PathMatchGlob : public IPathMatch
199 const PathMatchGlob& operator=(const PathMatchGlob&);
202 PathMatchGlob(const std::string& Pattern);
203 PathMatchGlob(const PathMatchGlob&);
204 virtual ~PathMatchGlob();
205 bool Match(const std::string& s) const;
207 #endif /* !KM_WIN32 */
209 // Search all paths in SearchPaths for filenames matching Pattern (no directories are returned).
210 // Put results in FoundPaths. Returns after first find if one_shot is true.
211 PathList_t& FindInPath(const IPathMatch& Pattern, const std::string& SearchDir,
212 PathList_t& FoundPaths, bool one_shot = false, char separator = '/');
214 PathList_t& FindInPaths(const IPathMatch& Pattern, const PathList_t& SearchPaths,
215 PathList_t& FoundPaths, bool one_shot = false, char separator = '/');
217 //------------------------------------------------------------------------------------------
218 // Directory Manipulation
219 //------------------------------------------------------------------------------------------
221 // Create a directory, creates intermediate directories as necessary
222 Result_t CreateDirectoriesInPath(const std::string& Path);
224 // Delete a file (fails if the path points to a directory)
225 Result_t DeleteFile(const std::string& filename);
227 // Recursively remove a file or directory
228 Result_t DeletePath(const std::string& pathname);
230 //------------------------------------------------------------------------------------------
232 //------------------------------------------------------------------------------------------
234 // Instant IO for strings
236 // Reads an entire file into a string.
237 Result_t ReadFileIntoString(const char* filename, std::string& outString, ui32_t max_size = 8 * Megabyte);
239 // Writes a string to a file, overwrites the existing file if present.
240 Result_t WriteStringIntoFile(const char* filename, const std::string& inString);
242 // Instant IO for archivable objects
244 // Unarchives a file into an object
245 Result_t ReadFileIntoObject(const std::string& Filename, IArchive& Object, ui32_t max_size = 8 * Kumu::Megabyte);
247 // Archives an object into a file
248 Result_t WriteObjectIntoFile(const IArchive& Object, const std::string& Filename);
250 // Instant IO for memory buffers
252 // Unarchives a file into a buffer
253 Result_t ReadFileIntoBuffer(const std::string& Filename, Kumu::ByteString& Buffer,
254 ui32_t max_size = 8 * Kumu::Megabyte);
256 // Archives a buffer into a file
257 Result_t WriteBufferIntoFile(const Kumu::ByteString& Buffer, const std::string& Filename);
260 //------------------------------------------------------------------------------------------
262 //------------------------------------------------------------------------------------------
267 KM_NO_COPY_CONSTRUCT(FileReader);
270 std::string m_Filename;
274 FileReader() : m_Handle(INVALID_HANDLE_VALUE) {}
275 virtual ~FileReader() { Close(); }
277 Result_t OpenRead(const char*) const; // open the file for reading
278 Result_t Close() const; // close the file
279 fsize_t Size() const; // returns the file's current size
280 Result_t Seek(Kumu::fpos_t = 0, SeekPos_t = SP_BEGIN) const; // move the file pointer
281 Result_t Tell(Kumu::fpos_t* pos) const; // report the file pointer's location
282 Result_t Read(byte_t*, ui32_t, ui32_t* = 0) const; // read a buffer of data
284 inline Kumu::fpos_t Tell() const // report the file pointer's location
286 Kumu::fpos_t tmp_pos;
291 inline bool IsOpen() { // returns true if the file is open
292 return (m_Handle != INVALID_HANDLE_VALUE);
297 class FileWriter : public FileReader
300 mem_ptr<h__iovec> m_IOVec;
301 KM_NO_COPY_CONSTRUCT(FileWriter);
303 MD5_CTX m_MD5Context;
307 virtual ~FileWriter();
309 Result_t OpenWrite(const char*); // open a new file, overwrites existing
310 Result_t OpenModify(const char*); // open a file for read/write
312 // this part of the interface takes advantage of the iovec structure on
313 // platforms that support it. For each call to Writev(const byte_t*, ui32_t, ui32_t*),
314 // the given buffer is added to an internal iovec struct. All items on the list
315 // are written to disk by a call to Writev();
316 Result_t Writev(const byte_t*, ui32_t); // queue buffer for "gather" write
317 Result_t Writev(ui32_t* = 0); // write all queued buffers
319 // if you call this while there are unwritten items on the iovec list,
320 // the iovec list will be written to disk before the given buffer,as though
321 // you had called Writev() first.
322 Result_t Write(const byte_t*, ui32_t, ui32_t* = 0); // write buffer to disk
325 void MaybeHash(void const *, int);
326 std::string StopHashing();
329 Result_t CreateDirectoriesInPath(const std::string& Path);
330 Result_t FreeSpaceForPath(const std::string& path, Kumu::fsize_t& free_space, Kumu::fsize_t& total_space);
331 Result_t DeleteFile(const std::string& filename);
332 Result_t DeletePath(const std::string& pathname);
337 #endif // _KM_FILEIO_H_