3 Copyright (c) 2008 Broad Institute / Massachusetts Institute of Technology
4 2011, 2012 Attractive Chaos <attractor@live.co.uk>
6 Permission is hereby granted, free of charge, to any person obtaining a copy
7 of this software and associated documentation files (the "Software"), to deal
8 in the Software without restriction, including without limitation the rights
9 to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
10 copies of the Software, and to permit persons to whom the Software is
11 furnished to do so, subject to the following conditions:
13 The above copyright notice and this permission notice shall be included in
14 all copies or substantial portions of the Software.
16 THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
17 IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
18 FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
19 AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
20 LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
21 OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
25 /* The BGZF library was originally written by Bob Handsaker from the Broad
26 * Institute. It was later improved by the SAMtools developers. */
34 #include <sys/types.h>
36 #define BGZF_BLOCK_SIZE 0xff00 // make sure compressBound(BGZF_BLOCK_SIZE) < BGZF_MAX_BLOCK_SIZE
37 #define BGZF_MAX_BLOCK_SIZE 0x10000
39 #define BGZF_ERR_ZLIB 1
40 #define BGZF_ERR_HEADER 2
42 #define BGZF_ERR_MISUSE 8
45 int errcode:16, is_write:2, compress_level:14;
47 int block_length, block_offset;
48 int64_t block_address;
49 void *uncompressed_block, *compressed_block;
50 void *cache; // a pointer to a hash table
51 void *fp; // actual file handler; FILE* on writing; FILE* or knetFile* on reading
52 void *mt; // only used for multi-threading
56 #define KSTRING_T kstring_t
57 typedef struct __kstring_t {
72 * Open an existing file descriptor for reading or writing.
74 * @param fd file descriptor
75 * @param mode mode matching /[rwu0-9]+/: 'r' for reading, 'w' for writing and a digit specifies
76 * the zlib compression level; if both 'r' and 'w' are present, 'w' is ignored.
77 * @return BGZF file handler; 0 on error
79 BGZF* bgzf_dopen(int fd, const char *mode);
81 #define bgzf_fdopen(fd, mode) bgzf_dopen((fd), (mode)) // for backward compatibility
84 * Open the specified file for reading or writing.
86 BGZF* bgzf_open(const char* path, const char *mode);
89 * Close the BGZF and free all associated resources.
91 * @param fp BGZF file handler
92 * @return 0 on success and -1 on error
94 int bgzf_close(BGZF *fp);
97 * Read up to _length_ bytes from the file storing into _data_.
99 * @param fp BGZF file handler
100 * @param data data array to read into
101 * @param length size of data to read
102 * @return number of bytes actually read; 0 on end-of-file and -1 on error
104 ssize_t bgzf_read(BGZF *fp, void *data, ssize_t length);
107 * Write _length_ bytes from _data_ to the file.
109 * @param fp BGZF file handler
110 * @param data data array to write
111 * @param length size of data to write
112 * @return number of bytes actually written; -1 on error
114 ssize_t bgzf_write(BGZF *fp, const void *data, ssize_t length);
117 * Write the data in the buffer to the file.
119 int bgzf_flush(BGZF *fp);
122 * Return a virtual file pointer to the current location in the file.
123 * No interpetation of the value should be made, other than a subsequent
124 * call to bgzf_seek can be used to position the file at the same point.
125 * Return value is non-negative on success.
127 #define bgzf_tell(fp) ((fp->block_address << 16) | (fp->block_offset & 0xFFFF))
130 * Set the file to read from the location specified by _pos_.
132 * @param fp BGZF file handler
133 * @param pos virtual file offset returned by bgzf_tell()
134 * @param whence must be SEEK_SET
135 * @return 0 on success and -1 on error
137 int64_t bgzf_seek(BGZF *fp, int64_t pos, int whence);
140 * Check if the BGZF end-of-file (EOF) marker is present
142 * @param fp BGZF file handler opened for reading
143 * @return 1 if EOF is present; 0 if not or on I/O error
145 int bgzf_check_EOF(BGZF *fp);
148 * Check if a file is in the BGZF format
150 * @param fn file name
151 * @return 1 if _fn_ is BGZF; 0 if not or on I/O error
153 int bgzf_is_bgzf(const char *fn);
155 /*********************
156 * Advanced routines *
157 *********************/
160 * Set the cache size. Only effective when compiled with -DBGZF_CACHE.
162 * @param fp BGZF file handler
163 * @param size size of cache in bytes; 0 to disable caching (default)
165 void bgzf_set_cache_size(BGZF *fp, int size);
168 * Flush the file if the remaining buffer size is smaller than _size_
170 int bgzf_flush_try(BGZF *fp, ssize_t size);
173 * Read one byte from a BGZF file. It is faster than bgzf_read()
174 * @param fp BGZF file handler
175 * @return byte read; -1 on end-of-file or error
177 int bgzf_getc(BGZF *fp);
180 * Read one line from a BGZF file. It is faster than bgzf_getc()
182 * @param fp BGZF file handler
183 * @param delim delimitor
184 * @param str string to write to; must be initialized
185 * @return length of the string; 0 on end-of-file; negative on error
187 int bgzf_getline(BGZF *fp, int delim, kstring_t *str);
190 * Read the next BGZF block.
192 int bgzf_read_block(BGZF *fp);
195 * Enable multi-threading (only effective on writing)
197 * @param fp BGZF file handler; must be opened for writing
198 * @param n_threads #threads used for writing
199 * @param n_sub_blks #blocks processed by each thread; a value 64-256 is recommended
201 int bgzf_mt(BGZF *fp, int n_threads, int n_sub_blks);