ViewVC Help
View File | Revision Log | Show Annotations | View Changeset | Root Listing
root/gclib/gclib/GBase.h
Revision: 16
Committed: Mon Jul 18 20:56:02 2011 UTC (7 years, 11 months ago) by gpertea
File size: 13918 byte(s)
Log Message:
sync with local source

Line File contents
1 #ifndef G_BASE_DEFINED
2 #define G_BASE_DEFINED
3 #ifndef _POSIX_SOURCE
4 //mostly for MinGW
5 #define _POSIX_SOURCE
6 #endif
7 #ifdef HAVE_CONFIG_H
8 #include "config.h"
9 #endif
10 #include <string.h>
11 #include <stdlib.h>
12 #include <stdio.h>
13 #include <math.h>
14 #include <limits.h>
15 #include <sys/types.h>
16 #include <sys/stat.h>
17 #include <stdint.h>
18
19 #if defined __WIN32__ || defined WIN32 || defined _WIN32 || defined _WIN32_
20 #ifndef __WIN32__
21 #define __WIN32__
22 #endif
23 #include <windows.h>
24 #include <io.h>
25 #define CHPATHSEP '\\'
26 #undef off_t
27 #define off_t int64_t
28 #ifdef _fseeki64
29 #define fseeko(stream, offset, origin) _fseeki64(stream, offset, origin)
30 #else
31 /*
32 #define _DEFINE_WIN32_FSEEKO
33 int fseeko(FILE *stream, off_t offset, int whence);
34 */
35 #define fseeko fseek
36 #endif
37 #ifdef _ftelli64
38 #define ftello(stream) _ftelli64(stream)
39 #else
40 /*
41 #define _DEFINE_WIN32_FTELLO
42 off_t ftello(FILE *stream);
43 */
44 #define ftello ftell
45 #endif
46 #else
47 #define CHPATHSEP '/'
48 #include <unistd.h>
49 #endif
50
51
52 #ifdef DEBUG
53 #undef NDEBUG
54 #endif
55
56 typedef int32_t int32;
57 typedef uint32_t uint32;
58
59 typedef unsigned char uchar;
60 typedef unsigned char byte;
61
62 #ifndef MAXUINT
63 #define MAXUINT ((unsigned int)-1)
64 #endif
65
66 #ifndef MAXINT
67 #define MAXINT INT_MAX
68 #endif
69
70 #ifndef MAX_UINT
71 #define MAX_UINT ((unsigned int)-1)
72 #endif
73
74 #ifndef MAX_INT
75 #define MAX_INT INT_MAX
76 #endif
77
78 typedef int64_t int64;
79 typedef uint64_t uint64;
80
81 /****************************************************************************/
82
83 #ifndef EXIT_FAILURE
84 #define EXIT_FAILURE 1
85 #endif
86
87 #ifndef EXIT_SUCCESS
88 #define EXIT_SUCCESS 0
89 #endif
90
91 /****************************************************************************/
92 #define ERR_ALLOC "Error allocating memory.\n"
93
94 //-------------------
95
96 // Debug helpers
97 #ifndef NDEBUG
98 #define GASSERT(exp) ((exp)?((void)0):(void)GAssert(#exp,__FILE__,__LINE__))
99 #ifdef TRACE
100 #define GTRACE(exp) (GMessage exp)
101 #else
102 #define GTRACE(exp) ((void)0)
103 #endif
104 #else
105 #define GASSERT(exp) ((void)0)
106 #define GTRACE(exp) ((void)0)
107 #endif
108
109 #define GERROR(exp) (GError exp)
110 /********************************** Macros ***********************************/
111 // Abolute value
112 #define GABS(val) (((val)>=0)?(val):-(val))
113
114 // Min and Max
115 #define GMAX(a,b) (((a)>(b))?(a):(b))
116 #define GMIN(a,b) (((a)>(b))?(b):(a))
117
118 // Min of three
119 #define GMIN3(x,y,z) ((x)<(y)?GMIN(x,z):GMIN(y,z))
120
121 // Max of three
122 #define GMAX3(x,y,z) ((x)>(y)?GMAX(x,z):GMAX(y,z))
123
124 // Return minimum and maximum of a, b
125 #define GMINMAX(lo,hi,a,b) ((a)<(b)?((lo)=(a),(hi)=(b)):((lo)=(b),(hi)=(a)))
126
127 // Clamp value x to range [lo..hi]
128 #define GCLAMP(lo,x,hi) ((x)<(lo)?(lo):((x)>(hi)?(hi):(x)))
129
130 typedef void* pointer;
131 typedef unsigned int uint;
132
133 typedef int GCompareProc(const pointer item1, const pointer item2);
134 typedef void GFreeProc(pointer item); //usually just delete,
135 //but may also support structures with embedded dynamic members
136
137 #define GMALLOC(ptr,size) if (!GMalloc((pointer*)(&ptr),size)) \
138 GError(ERR_ALLOC)
139 #define GCALLOC(ptr,size) if (!GCalloc((pointer*)(&ptr),size)) \
140 GError(ERR_ALLOC)
141 #define GREALLOC(ptr,size) if (!GRealloc((pointer*)(&ptr),size)) \
142 GError(ERR_ALLOC)
143 #define GFREE(ptr) GFree((pointer*)(&ptr))
144
145 inline char* strMin(char *arg1, char *arg2) {
146 return (strcmp(arg1, arg2) < 0)? arg1 : arg2;
147 }
148
149 inline char* strMax(char *arg1, char *arg2) {
150 return (strcmp(arg2, arg1) < 0)? arg1 : arg2;
151 }
152
153 inline int iround(double x) {
154 return (int)floor(x + 0.5);
155 }
156
157 /****************************************************************************/
158
159 inline int Gintcmp(int a, int b) {
160 //return (a>b)? 1 : ((a==b)?0:-1);
161 return a-b;
162 }
163
164 int Gstrcmp(char* a, char* b);
165 //same as strcmp but doesn't crash on NULL pointers
166
167 int Gstricmp(const char* a, const char* b);
168
169 inline void swap(int &arg1, int &arg2){
170 //arg1 ^= arg2;
171 //arg2 ^= arg1;
172 //arg1 ^= arg2;
173 register int swp=arg1;
174 arg1=arg2; arg2=swp;
175 }
176
177 inline void swap(char* &arg1, char* &arg2){ //swap pointers!
178 register char* swp=arg1;
179 arg1=arg2; arg2=swp;
180 }
181
182 inline void swap(uint &arg1, uint &arg2) {
183 register uint swp=arg1;
184 arg1=arg2; arg2=swp;
185 }
186
187 inline void swap(short &arg1, short &arg2) {
188 register short swp=arg1;
189 arg1=arg2; arg2=swp;
190 }
191
192 inline void swap(unsigned short &arg1, unsigned short &arg2) {
193 register unsigned short swp=arg1;
194 arg1=arg2; arg2=swp;
195 }
196
197 inline void swap(long &arg1, long &arg2) {
198 register long swp=arg1;
199 arg1=arg2; arg2=swp;
200 }
201
202 inline void swap(unsigned long &arg1, unsigned long &arg2) {
203 register unsigned long swp=arg1;
204 arg1=arg2; arg2=swp;
205 }
206
207
208 inline void swap(char &arg1, char &arg2) {
209 register char swp=arg1;
210 arg1=arg2; arg2=swp;
211 }
212
213 inline void swap(unsigned char &arg1, unsigned char &arg2) {
214 register unsigned char swp=arg1;
215 arg1=arg2; arg2=swp;
216 }
217
218 /**************** Memory management ***************************/
219
220 bool GMalloc(pointer* ptr, unsigned long size); // Allocate memory
221 bool GCalloc(pointer* ptr, unsigned long size); // Allocate and initialize memory
222 bool GRealloc(pointer* ptr,unsigned long size); // Resize memory
223 void GFree(pointer* ptr); // Free memory, resets ptr to NULL
224
225
226 int saprintf(char **retp, const char *fmt, ...);
227
228 void GError(const char* format,...); // Error routine (aborts program)
229 void GMessage(const char* format,...);// Log message to stderr
230 // Assert failed routine:- usually not called directly but through GASSERT
231 void GAssert(const char* expression, const char* filename, unsigned int lineno);
232
233 // ****************** string manipulation *************************
234 char *Gstrdup(const char* str);
235 //duplicate a string by allocating a copy for it and returning it
236 char* Gstrdup(const char* sfrom, const char* sto);
237 //same as GStrdup, but with an early termination (e.g. on delimiter)
238
239 char* Gsubstr(const char* str, char* from, char* to=NULL);
240 //extracts a substring, allocating it, including boundaries (from/to)
241
242 int strsplit(char* str, char** fields, int maxfields, const char* delim);
243 int strsplit(char* str, char** fields, int maxfields, const char delim);
244 int strsplit(char* str, char** fields, int maxfields); //splits by tab or space
245
246 char* replaceStr(char* &str, char* newvalue);
247
248 //conversion: to Lower/Upper case
249 // creating a new string:
250 char* upCase(const char* str);
251 char* loCase(const char* str);
252 // changing string in place:
253 char* strlower(char * str);
254 char* strupper(char * str);
255
256 //strstr but for memory zones: scans a memory region
257 //for a substring:
258 void* Gmemscan(void *mem, unsigned int len,
259 void *part, unsigned int partlen);
260
261 // test if a char is in a string:
262 bool chrInStr(char c, const char* str);
263
264 char* rstrchr(char* str, char ch);
265 /* returns a pointer to the rightmost
266 occurence of ch in str - like rindex for platforms missing it*/
267
268 char* strchrs(const char* s, const char* chrs);
269 //strchr but with a set of chars instead of only one
270
271 char* rstrfind(const char* str, const char *substr); /* like rindex() but for strings
272 or like the right side version of strstr()
273 */
274 //reverse character string or
275 char* reverseChars(char* str, int slen=0);
276
277 char* rstrstr(const char* rstart, const char *lend, const char* substr);
278 /*the reversed, rightside equivalent of strstr: starts searching
279 from right end (rstart), going back to left end (lend) and returns
280 a pointer to the last (right) matching character in str */
281
282 char* strifind(const char* str, const char* substr);
283 // the case insensitive version of strstr -- finding a string within a strin
284
285
286 //Determines if a string begins with a given prefix
287 //(returns false when any of the params is NULL,
288 // but true when prefix is '' (empty string)!)
289 bool startsWith(const char* s, const char* prefix);
290
291 bool endsWith(const char* s, const char* suffix);
292 //Note: returns true if suffix is empty string, but false if it's NULL
293
294
295 // ELF hash function for strings
296 int strhash(const char* str);
297
298
299
300 //---- generic base GSeg : genomic segment (interval) --
301 // coordinates are considered 1-based (so 0 is invalid)
302 class GSeg {
303 public:
304 uint start; //start<end always!
305 uint end;
306 GSeg(uint s=0,uint e=0) {
307 if (s>e) { start=e;end=s; }
308 else { start=s;end=e; }
309 }
310 //check for overlap with other segment
311 uint len() { return end-start+1; }
312 bool overlap(GSeg* d) {
313 //return start<d->start ? (d->start<=end) : (start<=d->end);
314 return (start<=d->end && end>=d->start);
315 }
316
317 bool overlap(GSeg& d) {
318 //return start<d.start ? (d.start<=end) : (start<=d.end);
319 return (start<=d.end && end>=d.start);
320 }
321
322 bool overlap(GSeg& d, int fuzz) {
323 //return start<d.start ? (d.start<=end+fuzz) : (start<=d.end+fuzz);
324 return (start<=d.end+fuzz && end+fuzz>=d.start);
325 }
326
327 bool overlap(uint s, uint e) {
328 if (s>e) { swap(s,e); }
329 //return start<s ? (s<=end) : (start<=e);
330 return (start<=e && end>=s);
331 }
332
333 //return the length of overlap between two segments
334 int overlapLen(GSeg* r) {
335 if (start<r->start) {
336 if (r->start>end) return 0;
337 return (r->end>end) ? end-r->start+1 : r->end-r->start+1;
338 }
339 else { //r->start<=start
340 if (start>r->end) return 0;
341 return (r->end<end)? r->end-start+1 : end-start+1;
342 }
343 }
344 int overlapLen(uint rstart, uint rend) {
345 if (rstart>rend) { swap(rstart,rend); }
346 if (start<rstart) {
347 if (rstart>end) return 0;
348 return (rend>end) ? end-rstart+1 : rend-rstart+1;
349 }
350 else { //rstart<=start
351 if (start>rend) return 0;
352 return (rend<end)? rend-start+1 : end-start+1;
353 }
354 }
355
356 //fuzzy coordinate matching:
357 bool coordMatch(GSeg* s, uint fuzz=0) {
358 if (fuzz==0) return (start==s->start && end==s->end);
359 uint sd = (start>s->start) ? start-s->start : s->start-start;
360 uint ed = (end>s->end) ? end-s->end : s->end-end;
361 return (sd<=fuzz && ed<=fuzz);
362 }
363 //comparison operators required for sorting
364 bool operator==(GSeg& d){
365 return (start==d.start && end==d.end);
366 }
367 bool operator>(GSeg& d){
368 return (start==d.start)?(end>d.end):(start>d.start);
369 }
370 bool operator<(GSeg& d){
371 return (start==d.start)?(end<d.end):(start<d.start);
372 }
373 };
374
375
376
377 //--------------------------------------------------------
378 // ************** simple line reading class for text files
379
380 //GLineReader -- text line reading/buffering class
381 class GLineReader {
382 bool closeFile;
383 int len;
384 int allocated;
385 char* buf;
386 bool isEOF;
387 FILE* file;
388 off_t filepos; //current position
389 bool pushed; //pushed back
390 int lcount; //line counter (read lines)
391 public:
392 char* chars() { return buf; }
393 char* line() { return buf; }
394 int readcount() { return lcount; } //number of lines read
395 void setFile(FILE* stream) { file=stream; }
396 int length() { return len; }
397 int size() { return len; } //same as size();
398 bool isEof() {return isEOF; }
399 bool eof() { return isEOF; }
400 off_t getfpos() { return filepos; }
401 off_t getFpos() { return filepos; }
402 char* nextLine() { return getLine(); }
403 char* getLine() { if (pushed) { pushed=false; return buf; }
404 else return getLine(file); }
405 char* getLine(FILE* stream) {
406 if (pushed) { pushed=false; return buf; }
407 else return getLine(stream, filepos); }
408 char* getLine(FILE* stream, off_t& f_pos); //read a line from a stream and update
409 // the given file position
410 void pushBack() { if (lcount>0) pushed=true; } // "undo" the last getLine request
411 // so the next call will in fact return the same line
412 GLineReader(const char* fname) {
413 FILE* f=fopen(fname, "rb");
414 if (f==NULL) GError("Error opening file '%s'!\n",fname);
415 closeFile=true;
416 init(f);
417 }
418 GLineReader(FILE* stream=NULL, off_t fpos=0) {
419 closeFile=false;
420 init(stream,fpos);
421 }
422 void init(FILE* stream, off_t fpos=0) {
423 len=0;
424 isEOF=false;
425 allocated=1024;
426 GMALLOC(buf,allocated);
427 lcount=0;
428 buf[0]=0;
429 file=stream;
430 filepos=fpos;
431 pushed=false;
432 }
433 ~GLineReader() {
434 GFREE(buf);
435 if (closeFile) fclose(file);
436 }
437 };
438
439
440 /* extended fgets() - to read one full line from a file and
441 update the file position correctly !
442 buf will be reallocated as necessary, to fit the whole line
443 */
444 char* fgetline(char* & buf, int& buflen, FILE* stream, off_t* f_pos=NULL, int* linelen=NULL);
445
446
447 //print int/values nicely formatted in 3-digit groups
448 char* commaprint(uint64 n);
449
450 /*********************** File management functions *********************/
451
452 // removes the last part (file or directory name) of a full path
453 // WARNING: this is a destructive operation for the given string!
454 void delFileName(char* filepath);
455
456 // returns a pointer to the last file or directory name in a full path
457 const char* getFileName(const char* filepath);
458 // returns a pointer to the file "extension" part in a filename
459 const char* getFileExt(const char* filepath);
460
461
462 int fileExists(const char* fname);
463 //returns 0 if file entry doesn't exist
464 // 1 if it's a directory
465 // 2 if it's a regular file
466 // 3 otherwise (?)
467
468 int64 fileSize(const char* fpath);
469
470 //write a formatted fasta record, fasta formatted
471 void writeFasta(FILE *fw, const char* seqid, const char* descr,
472 const char* seq, int linelen=60, int seqlen=0);
473
474 //parses the next number found in a string at the current position
475 //until a non-digit (and not a '.', 'e','E','-','+') is encountered;
476 //updates the char* pointer to be after the last digit parsed
477 bool parseNumber(char* &p, double& v);
478 bool parseDouble(char* &p, double& v); //just an alias for parseNumber
479
480 bool parseInt(char* &p, int& i);
481 bool parseUInt(char* &p, uint& i);
482 bool parseHex(char* &p, uint& i);
483
484 #endif /* G_BASE_DEFINED */