Annotation of embedaddon/pcre/pcregrep.c, revision 1.1.1.5
1.1 misho 1: /*************************************************
2: * pcregrep program *
3: *************************************************/
4:
5: /* This is a grep program that uses the PCRE regular expression library to do
1.1.1.5 ! misho 6: its pattern matching. On Unix-like, Windows, and native z/OS systems it can
! 7: recurse into directories, and in z/OS it can handle PDS files.
1.1 misho 8:
1.1.1.5 ! misho 9: Note that for native z/OS, in addition to defining the NATIVE_ZOS macro, an
! 10: additional header is required. That header is not included in the main PCRE
! 11: distribution because other apparatus is needed to compile pcregrep for z/OS.
! 12: The header can be found in the special z/OS distribution, which is available
! 13: from www.zaconsultants.net or from www.cbttape.org.
! 14:
! 15: Copyright (c) 1997-2013 University of Cambridge
1.1 misho 16:
17: -----------------------------------------------------------------------------
18: Redistribution and use in source and binary forms, with or without
19: modification, are permitted provided that the following conditions are met:
20:
21: * Redistributions of source code must retain the above copyright notice,
22: this list of conditions and the following disclaimer.
23:
24: * Redistributions in binary form must reproduce the above copyright
25: notice, this list of conditions and the following disclaimer in the
26: documentation and/or other materials provided with the distribution.
27:
28: * Neither the name of the University of Cambridge nor the names of its
29: contributors may be used to endorse or promote products derived from
30: this software without specific prior written permission.
31:
32: THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
33: AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
34: IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
35: ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
36: LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
37: CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
38: SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
39: INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
40: CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
41: ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
42: POSSIBILITY OF SUCH DAMAGE.
43: -----------------------------------------------------------------------------
44: */
45:
46: #ifdef HAVE_CONFIG_H
47: #include "config.h"
48: #endif
49:
50: #include <ctype.h>
51: #include <locale.h>
52: #include <stdio.h>
53: #include <string.h>
54: #include <stdlib.h>
55: #include <errno.h>
56:
57: #include <sys/types.h>
58: #include <sys/stat.h>
59:
60: #ifdef HAVE_UNISTD_H
61: #include <unistd.h>
62: #endif
63:
64: #ifdef SUPPORT_LIBZ
65: #include <zlib.h>
66: #endif
67:
68: #ifdef SUPPORT_LIBBZ2
69: #include <bzlib.h>
70: #endif
71:
72: #include "pcre.h"
73:
74: #define FALSE 0
75: #define TRUE 1
76:
77: typedef int BOOL;
78:
79: #define OFFSET_SIZE 99
80:
81: #if BUFSIZ > 8192
1.1.1.4 misho 82: #define MAXPATLEN BUFSIZ
1.1 misho 83: #else
1.1.1.4 misho 84: #define MAXPATLEN 8192
1.1 misho 85: #endif
86:
1.1.1.4 misho 87: #define PATBUFSIZE (MAXPATLEN + 10) /* Allows for prefix+suffix */
88:
1.1 misho 89: /* Values for the "filenames" variable, which specifies options for file name
90: output. The order is important; it is assumed that a file name is wanted for
91: all values greater than FN_DEFAULT. */
92:
93: enum { FN_NONE, FN_DEFAULT, FN_MATCH_ONLY, FN_NOMATCH_ONLY, FN_FORCE };
94:
95: /* File reading styles */
96:
97: enum { FR_PLAIN, FR_LIBZ, FR_LIBBZ2 };
98:
99: /* Actions for the -d and -D options */
100:
101: enum { dee_READ, dee_SKIP, dee_RECURSE };
102: enum { DEE_READ, DEE_SKIP };
103:
104: /* Actions for special processing options (flag bits) */
105:
106: #define PO_WORD_MATCH 0x0001
107: #define PO_LINE_MATCH 0x0002
108: #define PO_FIXED_STRINGS 0x0004
109:
110: /* Line ending types */
111:
112: enum { EL_LF, EL_CR, EL_CRLF, EL_ANY, EL_ANYCRLF };
113:
1.1.1.3 misho 114: /* Binary file options */
115:
116: enum { BIN_BINARY, BIN_NOMATCH, BIN_TEXT };
117:
1.1 misho 118: /* In newer versions of gcc, with FORTIFY_SOURCE set (the default in some
119: environments), a warning is issued if the value of fwrite() is ignored.
120: Unfortunately, casting to (void) does not suppress the warning. To get round
121: this, we use a macro that compiles a fudge. Oddly, this does not also seem to
122: apply to fprintf(). */
123:
124: #define FWRITE(a,b,c,d) if (fwrite(a,b,c,d)) {}
125:
126:
127:
128: /*************************************************
129: * Global variables *
130: *************************************************/
131:
132: /* Jeffrey Friedl has some debugging requirements that are not part of the
133: regular code. */
134:
135: #ifdef JFRIEDL_DEBUG
136: static int S_arg = -1;
137: static unsigned int jfriedl_XR = 0; /* repeat regex attempt this many times */
138: static unsigned int jfriedl_XT = 0; /* replicate text this many times */
139: static const char *jfriedl_prefix = "";
140: static const char *jfriedl_postfix = "";
141: #endif
142:
143: static int endlinetype;
144:
145: static char *colour_string = (char *)"1;31";
146: static char *colour_option = NULL;
147: static char *dee_option = NULL;
148: static char *DEE_option = NULL;
1.1.1.4 misho 149: static char *locale = NULL;
1.1 misho 150: static char *main_buffer = NULL;
151: static char *newline = NULL;
1.1.1.4 misho 152: static char *om_separator = (char *)"";
1.1 misho 153: static char *stdin_name = (char *)"(standard input)";
154:
155: static const unsigned char *pcretables = NULL;
156:
157: static int after_context = 0;
158: static int before_context = 0;
1.1.1.3 misho 159: static int binary_files = BIN_BINARY;
1.1 misho 160: static int both_context = 0;
161: static int bufthird = PCREGREP_BUFSIZE;
162: static int bufsize = 3*PCREGREP_BUFSIZE;
1.1.1.4 misho 163:
164: #if defined HAVE_WINDOWS_H && HAVE_WINDOWS_H
165: static int dee_action = dee_SKIP;
166: #else
1.1 misho 167: static int dee_action = dee_READ;
1.1.1.4 misho 168: #endif
169:
1.1 misho 170: static int DEE_action = DEE_READ;
171: static int error_count = 0;
172: static int filenames = FN_DEFAULT;
1.1.1.4 misho 173: static int pcre_options = 0;
1.1 misho 174: static int process_options = 0;
175:
176: #ifdef SUPPORT_PCREGREP_JIT
177: static int study_options = PCRE_STUDY_JIT_COMPILE;
178: #else
179: static int study_options = 0;
180: #endif
181:
182: static unsigned long int match_limit = 0;
183: static unsigned long int match_limit_recursion = 0;
184:
185: static BOOL count_only = FALSE;
186: static BOOL do_colour = FALSE;
187: static BOOL file_offsets = FALSE;
188: static BOOL hyphenpending = FALSE;
189: static BOOL invert = FALSE;
190: static BOOL line_buffered = FALSE;
191: static BOOL line_offsets = FALSE;
192: static BOOL multiline = FALSE;
193: static BOOL number = FALSE;
194: static BOOL omit_zero_count = FALSE;
195: static BOOL resource_error = FALSE;
196: static BOOL quiet = FALSE;
1.1.1.4 misho 197: static BOOL show_only_matching = FALSE;
1.1 misho 198: static BOOL silent = FALSE;
199: static BOOL utf8 = FALSE;
200:
1.1.1.4 misho 201: /* Structure for list of --only-matching capturing numbers. */
202:
203: typedef struct omstr {
204: struct omstr *next;
205: int groupnum;
206: } omstr;
207:
208: static omstr *only_matching = NULL;
209: static omstr *only_matching_last = NULL;
210:
211: /* Structure for holding the two variables that describe a number chain. */
212:
213: typedef struct omdatastr {
214: omstr **anchor;
215: omstr **lastptr;
216: } omdatastr;
217:
218: static omdatastr only_matching_data = { &only_matching, &only_matching_last };
219:
220: /* Structure for list of file names (for -f and --{in,ex}clude-from) */
221:
222: typedef struct fnstr {
223: struct fnstr *next;
224: char *name;
225: } fnstr;
226:
227: static fnstr *exclude_from = NULL;
228: static fnstr *exclude_from_last = NULL;
229: static fnstr *include_from = NULL;
230: static fnstr *include_from_last = NULL;
231:
232: static fnstr *file_lists = NULL;
233: static fnstr *file_lists_last = NULL;
234: static fnstr *pattern_files = NULL;
235: static fnstr *pattern_files_last = NULL;
236:
237: /* Structure for holding the two variables that describe a file name chain. */
238:
239: typedef struct fndatastr {
240: fnstr **anchor;
241: fnstr **lastptr;
242: } fndatastr;
243:
244: static fndatastr exclude_from_data = { &exclude_from, &exclude_from_last };
245: static fndatastr include_from_data = { &include_from, &include_from_last };
246: static fndatastr file_lists_data = { &file_lists, &file_lists_last };
247: static fndatastr pattern_files_data = { &pattern_files, &pattern_files_last };
248:
249: /* Structure for pattern and its compiled form; used for matching patterns and
250: also for include/exclude patterns. */
251:
252: typedef struct patstr {
253: struct patstr *next;
254: char *string;
255: pcre *compiled;
256: pcre_extra *hint;
257: } patstr;
258:
259: static patstr *patterns = NULL;
260: static patstr *patterns_last = NULL;
261: static patstr *include_patterns = NULL;
262: static patstr *include_patterns_last = NULL;
263: static patstr *exclude_patterns = NULL;
264: static patstr *exclude_patterns_last = NULL;
265: static patstr *include_dir_patterns = NULL;
266: static patstr *include_dir_patterns_last = NULL;
267: static patstr *exclude_dir_patterns = NULL;
268: static patstr *exclude_dir_patterns_last = NULL;
269:
270: /* Structure holding the two variables that describe a pattern chain. A pointer
271: to such structures is used for each appropriate option. */
272:
273: typedef struct patdatastr {
274: patstr **anchor;
275: patstr **lastptr;
276: } patdatastr;
277:
278: static patdatastr match_patdata = { &patterns, &patterns_last };
279: static patdatastr include_patdata = { &include_patterns, &include_patterns_last };
280: static patdatastr exclude_patdata = { &exclude_patterns, &exclude_patterns_last };
281: static patdatastr include_dir_patdata = { &include_dir_patterns, &include_dir_patterns_last };
282: static patdatastr exclude_dir_patdata = { &exclude_dir_patterns, &exclude_dir_patterns_last };
283:
284: static patstr **incexlist[4] = { &include_patterns, &exclude_patterns,
285: &include_dir_patterns, &exclude_dir_patterns };
286:
287: static const char *incexname[4] = { "--include", "--exclude",
288: "--include-dir", "--exclude-dir" };
289:
1.1 misho 290: /* Structure for options and list of them */
291:
292: enum { OP_NODATA, OP_STRING, OP_OP_STRING, OP_NUMBER, OP_LONGNUMBER,
1.1.1.4 misho 293: OP_OP_NUMBER, OP_OP_NUMBERS, OP_PATLIST, OP_FILELIST, OP_BINFILES };
1.1 misho 294:
295: typedef struct option_item {
296: int type;
297: int one_char;
298: void *dataptr;
299: const char *long_name;
300: const char *help_text;
301: } option_item;
302:
303: /* Options without a single-letter equivalent get a negative value. This can be
304: used to identify them. */
305:
306: #define N_COLOUR (-1)
307: #define N_EXCLUDE (-2)
308: #define N_EXCLUDE_DIR (-3)
309: #define N_HELP (-4)
310: #define N_INCLUDE (-5)
311: #define N_INCLUDE_DIR (-6)
312: #define N_LABEL (-7)
313: #define N_LOCALE (-8)
314: #define N_NULL (-9)
315: #define N_LOFFSETS (-10)
316: #define N_FOFFSETS (-11)
317: #define N_LBUFFER (-12)
318: #define N_M_LIMIT (-13)
319: #define N_M_LIMIT_REC (-14)
320: #define N_BUFSIZE (-15)
321: #define N_NOJIT (-16)
1.1.1.3 misho 322: #define N_FILE_LIST (-17)
323: #define N_BINARY_FILES (-18)
1.1.1.4 misho 324: #define N_EXCLUDE_FROM (-19)
325: #define N_INCLUDE_FROM (-20)
326: #define N_OM_SEPARATOR (-21)
1.1 misho 327:
328: static option_item optionlist[] = {
1.1.1.3 misho 329: { OP_NODATA, N_NULL, NULL, "", "terminate options" },
1.1 misho 330: { OP_NODATA, N_HELP, NULL, "help", "display this help and exit" },
331: { OP_NUMBER, 'A', &after_context, "after-context=number", "set number of following context lines" },
1.1.1.3 misho 332: { OP_NODATA, 'a', NULL, "text", "treat binary files as text" },
1.1 misho 333: { OP_NUMBER, 'B', &before_context, "before-context=number", "set number of prior context lines" },
1.1.1.3 misho 334: { OP_BINFILES, N_BINARY_FILES, NULL, "binary-files=word", "set treatment of binary files" },
1.1 misho 335: { OP_NUMBER, N_BUFSIZE,&bufthird, "buffer-size=number", "set processing buffer size parameter" },
336: { OP_OP_STRING, N_COLOUR, &colour_option, "color=option", "matched text color option" },
337: { OP_OP_STRING, N_COLOUR, &colour_option, "colour=option", "matched text colour option" },
338: { OP_NUMBER, 'C', &both_context, "context=number", "set number of context lines, before & after" },
339: { OP_NODATA, 'c', NULL, "count", "print only a count of matching lines per FILE" },
340: { OP_STRING, 'D', &DEE_option, "devices=action","how to handle devices, FIFOs, and sockets" },
341: { OP_STRING, 'd', &dee_option, "directories=action", "how to handle directories" },
1.1.1.4 misho 342: { OP_PATLIST, 'e', &match_patdata, "regex(p)=pattern", "specify pattern (may be used more than once)" },
1.1 misho 343: { OP_NODATA, 'F', NULL, "fixed-strings", "patterns are sets of newline-separated strings" },
1.1.1.4 misho 344: { OP_FILELIST, 'f', &pattern_files_data, "file=path", "read patterns from file" },
345: { OP_FILELIST, N_FILE_LIST, &file_lists_data, "file-list=path","read files to search from file" },
1.1 misho 346: { OP_NODATA, N_FOFFSETS, NULL, "file-offsets", "output file offsets, not text" },
347: { OP_NODATA, 'H', NULL, "with-filename", "force the prefixing filename on output" },
348: { OP_NODATA, 'h', NULL, "no-filename", "suppress the prefixing filename on output" },
1.1.1.3 misho 349: { OP_NODATA, 'I', NULL, "", "treat binary files as not matching (ignore)" },
1.1 misho 350: { OP_NODATA, 'i', NULL, "ignore-case", "ignore case distinctions" },
351: #ifdef SUPPORT_PCREGREP_JIT
352: { OP_NODATA, N_NOJIT, NULL, "no-jit", "do not use just-in-time compiler optimization" },
353: #else
354: { OP_NODATA, N_NOJIT, NULL, "no-jit", "ignored: this pcregrep does not support JIT" },
355: #endif
356: { OP_NODATA, 'l', NULL, "files-with-matches", "print only FILE names containing matches" },
357: { OP_NODATA, 'L', NULL, "files-without-match","print only FILE names not containing matches" },
358: { OP_STRING, N_LABEL, &stdin_name, "label=name", "set name for standard input" },
359: { OP_NODATA, N_LBUFFER, NULL, "line-buffered", "use line buffering" },
360: { OP_NODATA, N_LOFFSETS, NULL, "line-offsets", "output line numbers and offsets, not text" },
361: { OP_STRING, N_LOCALE, &locale, "locale=locale", "use the named locale" },
362: { OP_LONGNUMBER, N_M_LIMIT, &match_limit, "match-limit=number", "set PCRE match limit option" },
363: { OP_LONGNUMBER, N_M_LIMIT_REC, &match_limit_recursion, "recursion-limit=number", "set PCRE match recursion limit option" },
364: { OP_NODATA, 'M', NULL, "multiline", "run in multiline mode" },
365: { OP_STRING, 'N', &newline, "newline=type", "set newline type (CR, LF, CRLF, ANYCRLF or ANY)" },
366: { OP_NODATA, 'n', NULL, "line-number", "print line number with output lines" },
1.1.1.4 misho 367: { OP_OP_NUMBERS, 'o', &only_matching_data, "only-matching=n", "show only the part of the line that matched" },
368: { OP_STRING, N_OM_SEPARATOR, &om_separator, "om-separator=text", "set separator for multiple -o output" },
1.1 misho 369: { OP_NODATA, 'q', NULL, "quiet", "suppress output, just set return code" },
370: { OP_NODATA, 'r', NULL, "recursive", "recursively scan sub-directories" },
1.1.1.4 misho 371: { OP_PATLIST, N_EXCLUDE,&exclude_patdata, "exclude=pattern","exclude matching files when recursing" },
372: { OP_PATLIST, N_INCLUDE,&include_patdata, "include=pattern","include matching files when recursing" },
373: { OP_PATLIST, N_EXCLUDE_DIR,&exclude_dir_patdata, "exclude-dir=pattern","exclude matching directories when recursing" },
374: { OP_PATLIST, N_INCLUDE_DIR,&include_dir_patdata, "include-dir=pattern","include matching directories when recursing" },
375: { OP_FILELIST, N_EXCLUDE_FROM,&exclude_from_data, "exclude-from=path", "read exclude list from file" },
376: { OP_FILELIST, N_INCLUDE_FROM,&include_from_data, "include-from=path", "read include list from file" },
1.1 misho 377:
378: /* These two were accidentally implemented with underscores instead of
379: hyphens in the option names. As this was not discovered for several releases,
380: the incorrect versions are left in the table for compatibility. However, the
381: --help function misses out any option that has an underscore in its name. */
382:
1.1.1.4 misho 383: { OP_PATLIST, N_EXCLUDE_DIR,&exclude_dir_patdata, "exclude_dir=pattern","exclude matching directories when recursing" },
384: { OP_PATLIST, N_INCLUDE_DIR,&include_dir_patdata, "include_dir=pattern","include matching directories when recursing" },
1.1 misho 385:
386: #ifdef JFRIEDL_DEBUG
387: { OP_OP_NUMBER, 'S', &S_arg, "jeffS", "replace matched (sub)string with X" },
388: #endif
389: { OP_NODATA, 's', NULL, "no-messages", "suppress error messages" },
390: { OP_NODATA, 'u', NULL, "utf-8", "use UTF-8 mode" },
391: { OP_NODATA, 'V', NULL, "version", "print version information and exit" },
392: { OP_NODATA, 'v', NULL, "invert-match", "select non-matching lines" },
393: { OP_NODATA, 'w', NULL, "word-regex(p)", "force patterns to match only as words" },
394: { OP_NODATA, 'x', NULL, "line-regex(p)", "force patterns to match only whole lines" },
395: { OP_NODATA, 0, NULL, NULL, NULL }
396: };
397:
398: /* Tables for prefixing and suffixing patterns, according to the -w, -x, and -F
399: options. These set the 1, 2, and 4 bits in process_options, respectively. Note
400: that the combination of -w and -x has the same effect as -x on its own, so we
1.1.1.4 misho 401: can treat them as the same. Note that the MAXPATLEN macro assumes the longest
402: prefix+suffix is 10 characters; if anything longer is added, it must be
403: adjusted. */
1.1 misho 404:
405: static const char *prefix[] = {
406: "", "\\b", "^(?:", "^(?:", "\\Q", "\\b\\Q", "^(?:\\Q", "^(?:\\Q" };
407:
408: static const char *suffix[] = {
409: "", "\\b", ")$", ")$", "\\E", "\\E\\b", "\\E)$", "\\E)$" };
410:
411: /* UTF-8 tables - used only when the newline setting is "any". */
412:
413: const int utf8_table3[] = { 0xff, 0x1f, 0x0f, 0x07, 0x03, 0x01};
414:
415: const char utf8_table4[] = {
416: 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
417: 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
418: 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,
419: 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5 };
420:
421:
422:
423: /*************************************************
424: * Exit from the program *
425: *************************************************/
426:
427: /* If there has been a resource error, give a suitable message.
428:
429: Argument: the return code
430: Returns: does not return
431: */
432:
433: static void
434: pcregrep_exit(int rc)
435: {
436: if (resource_error)
437: {
438: fprintf(stderr, "pcregrep: Error %d, %d or %d means that a resource limit "
439: "was exceeded.\n", PCRE_ERROR_MATCHLIMIT, PCRE_ERROR_RECURSIONLIMIT,
440: PCRE_ERROR_JIT_STACKLIMIT);
441: fprintf(stderr, "pcregrep: Check your regex for nested unlimited loops.\n");
442: }
443: exit(rc);
444: }
445:
446:
447: /*************************************************
1.1.1.4 misho 448: * Add item to chain of patterns *
449: *************************************************/
450:
451: /* Used to add an item onto a chain, or just return an unconnected item if the
452: "after" argument is NULL.
453:
454: Arguments:
455: s pattern string to add
456: after if not NULL points to item to insert after
457:
458: Returns: new pattern block
459: */
460:
461: static patstr *
462: add_pattern(char *s, patstr *after)
463: {
464: patstr *p = (patstr *)malloc(sizeof(patstr));
465: if (p == NULL)
466: {
467: fprintf(stderr, "pcregrep: malloc failed\n");
468: pcregrep_exit(2);
469: }
470: if (strlen(s) > MAXPATLEN)
471: {
472: fprintf(stderr, "pcregrep: pattern is too long (limit is %d bytes)\n",
473: MAXPATLEN);
474: return NULL;
475: }
476: p->next = NULL;
477: p->string = s;
478: p->compiled = NULL;
479: p->hint = NULL;
480:
481: if (after != NULL)
482: {
483: p->next = after->next;
484: after->next = p;
485: }
486: return p;
487: }
488:
489:
490: /*************************************************
491: * Free chain of patterns *
492: *************************************************/
493:
494: /* Used for several chains of patterns.
495:
496: Argument: pointer to start of chain
497: Returns: nothing
498: */
499:
500: static void
501: free_pattern_chain(patstr *pc)
502: {
503: while (pc != NULL)
504: {
505: patstr *p = pc;
506: pc = p->next;
507: if (p->hint != NULL) pcre_free_study(p->hint);
508: if (p->compiled != NULL) pcre_free(p->compiled);
509: free(p);
510: }
511: }
512:
513:
514: /*************************************************
515: * Free chain of file names *
516: *************************************************/
517:
518: /*
519: Argument: pointer to start of chain
520: Returns: nothing
521: */
522:
523: static void
524: free_file_chain(fnstr *fn)
525: {
526: while (fn != NULL)
527: {
528: fnstr *f = fn;
529: fn = f->next;
530: free(f);
531: }
532: }
533:
534:
535: /*************************************************
1.1 misho 536: * OS-specific functions *
537: *************************************************/
538:
1.1.1.5 ! misho 539: /* These functions are defined so that they can be made system specific.
! 540: At present there are versions for Unix-style environments, Windows, native
! 541: z/OS, and "no support". */
1.1 misho 542:
543:
1.1.1.5 ! misho 544: /************* Directory scanning Unix-style and z/OS ***********/
1.1 misho 545:
1.1.1.5 ! misho 546: #if (defined HAVE_SYS_STAT_H && defined HAVE_DIRENT_H && defined HAVE_SYS_TYPES_H) || defined NATIVE_ZOS
1.1 misho 547: #include <sys/types.h>
548: #include <sys/stat.h>
549: #include <dirent.h>
550:
1.1.1.5 ! misho 551: #if defined NATIVE_ZOS
! 552: /************* Directory and PDS/E scanning for z/OS ***********/
! 553: /************* z/OS looks mostly like Unix with USS ************/
! 554: /* However, z/OS needs the #include statements in this header */
! 555: #include "pcrzosfs.h"
! 556: /* That header is not included in the main PCRE distribution because
! 557: other apparatus is needed to compile pcregrep for z/OS. The header
! 558: can be found in the special z/OS distribution, which is available
! 559: from www.zaconsultants.net or from www.cbttape.org. */
! 560: #endif
! 561:
1.1 misho 562: typedef DIR directory_type;
1.1.1.4 misho 563: #define FILESEP '/'
1.1 misho 564:
565: static int
566: isdirectory(char *filename)
567: {
568: struct stat statbuf;
569: if (stat(filename, &statbuf) < 0)
570: return 0; /* In the expectation that opening as a file will fail */
1.1.1.4 misho 571: return (statbuf.st_mode & S_IFMT) == S_IFDIR;
1.1 misho 572: }
573:
574: static directory_type *
575: opendirectory(char *filename)
576: {
577: return opendir(filename);
578: }
579:
580: static char *
581: readdirectory(directory_type *dir)
582: {
583: for (;;)
584: {
585: struct dirent *dent = readdir(dir);
586: if (dent == NULL) return NULL;
587: if (strcmp(dent->d_name, ".") != 0 && strcmp(dent->d_name, "..") != 0)
588: return dent->d_name;
589: }
590: /* Control never reaches here */
591: }
592:
593: static void
594: closedirectory(directory_type *dir)
595: {
596: closedir(dir);
597: }
598:
599:
1.1.1.5 ! misho 600: /************* Test for regular file, Unix-style **********/
1.1 misho 601:
602: static int
603: isregfile(char *filename)
604: {
605: struct stat statbuf;
606: if (stat(filename, &statbuf) < 0)
607: return 1; /* In the expectation that opening as a file will fail */
608: return (statbuf.st_mode & S_IFMT) == S_IFREG;
609: }
610:
611:
1.1.1.5 ! misho 612: #if defined NATIVE_ZOS
! 613: /************* Test for a terminal in z/OS **********/
! 614: /* isatty() does not work in a TSO environment, so always give FALSE.*/
! 615:
! 616: static BOOL
! 617: is_stdout_tty(void)
! 618: {
! 619: return FALSE;
! 620: }
! 621:
! 622: static BOOL
! 623: is_file_tty(FILE *f)
! 624: {
! 625: return FALSE;
! 626: }
! 627:
1.1 misho 628:
1.1.1.5 ! misho 629: /************* Test for a terminal, Unix-style **********/
! 630:
! 631: #else
1.1 misho 632: static BOOL
633: is_stdout_tty(void)
634: {
635: return isatty(fileno(stdout));
636: }
637:
638: static BOOL
639: is_file_tty(FILE *f)
640: {
641: return isatty(fileno(f));
642: }
1.1.1.5 ! misho 643: #endif
! 644:
! 645: /* End of Unix-style or native z/OS environment functions. */
1.1 misho 646:
647:
1.1.1.5 ! misho 648: /************* Directory scanning in Windows ***********/
1.1 misho 649:
650: /* I (Philip Hazel) have no means of testing this code. It was contributed by
651: Lionel Fourquaux. David Burgess added a patch to define INVALID_FILE_ATTRIBUTES
652: when it did not exist. David Byron added a patch that moved the #include of
653: <windows.h> to before the INVALID_FILE_ATTRIBUTES definition rather than after.
654: The double test below stops gcc 4.4.4 grumbling that HAVE_WINDOWS_H is
655: undefined when it is indeed undefined. */
656:
657: #elif defined HAVE_WINDOWS_H && HAVE_WINDOWS_H
658:
659: #ifndef STRICT
660: # define STRICT
661: #endif
662: #ifndef WIN32_LEAN_AND_MEAN
663: # define WIN32_LEAN_AND_MEAN
664: #endif
665:
666: #include <windows.h>
667:
668: #ifndef INVALID_FILE_ATTRIBUTES
669: #define INVALID_FILE_ATTRIBUTES 0xFFFFFFFF
670: #endif
671:
672: typedef struct directory_type
673: {
674: HANDLE handle;
675: BOOL first;
676: WIN32_FIND_DATA data;
677: } directory_type;
678:
1.1.1.4 misho 679: #define FILESEP '/'
680:
1.1 misho 681: int
682: isdirectory(char *filename)
683: {
684: DWORD attr = GetFileAttributes(filename);
685: if (attr == INVALID_FILE_ATTRIBUTES)
686: return 0;
1.1.1.4 misho 687: return (attr & FILE_ATTRIBUTE_DIRECTORY) != 0;
1.1 misho 688: }
689:
690: directory_type *
691: opendirectory(char *filename)
692: {
693: size_t len;
694: char *pattern;
695: directory_type *dir;
696: DWORD err;
697: len = strlen(filename);
1.1.1.4 misho 698: pattern = (char *)malloc(len + 3);
699: dir = (directory_type *)malloc(sizeof(*dir));
1.1 misho 700: if ((pattern == NULL) || (dir == NULL))
701: {
702: fprintf(stderr, "pcregrep: malloc failed\n");
703: pcregrep_exit(2);
704: }
705: memcpy(pattern, filename, len);
706: memcpy(&(pattern[len]), "\\*", 3);
707: dir->handle = FindFirstFile(pattern, &(dir->data));
708: if (dir->handle != INVALID_HANDLE_VALUE)
709: {
710: free(pattern);
711: dir->first = TRUE;
712: return dir;
713: }
714: err = GetLastError();
715: free(pattern);
716: free(dir);
717: errno = (err == ERROR_ACCESS_DENIED) ? EACCES : ENOENT;
718: return NULL;
719: }
720:
721: char *
722: readdirectory(directory_type *dir)
723: {
724: for (;;)
725: {
726: if (!dir->first)
727: {
728: if (!FindNextFile(dir->handle, &(dir->data)))
729: return NULL;
730: }
731: else
732: {
733: dir->first = FALSE;
734: }
735: if (strcmp(dir->data.cFileName, ".") != 0 && strcmp(dir->data.cFileName, "..") != 0)
736: return dir->data.cFileName;
737: }
738: #ifndef _MSC_VER
739: return NULL; /* Keep compiler happy; never executed */
740: #endif
741: }
742:
743: void
744: closedirectory(directory_type *dir)
745: {
746: FindClose(dir->handle);
747: free(dir);
748: }
749:
750:
1.1.1.5 ! misho 751: /************* Test for regular file in Windows **********/
1.1 misho 752:
753: /* I don't know how to do this, or if it can be done; assume all paths are
754: regular if they are not directories. */
755:
756: int isregfile(char *filename)
757: {
758: return !isdirectory(filename);
759: }
760:
761:
1.1.1.5 ! misho 762: /************* Test for a terminal in Windows **********/
1.1 misho 763:
764: /* I don't know how to do this; assume never */
765:
766: static BOOL
767: is_stdout_tty(void)
768: {
769: return FALSE;
770: }
771:
772: static BOOL
773: is_file_tty(FILE *f)
774: {
775: return FALSE;
776: }
777:
1.1.1.5 ! misho 778: /* End of Windows functions */
! 779:
1.1 misho 780:
781: /************* Directory scanning when we can't do it ***********/
782:
783: /* The type is void, and apart from isdirectory(), the functions do nothing. */
784:
785: #else
786:
1.1.1.4 misho 787: #define FILESEP 0
1.1 misho 788: typedef void directory_type;
789:
790: int isdirectory(char *filename) { return 0; }
791: directory_type * opendirectory(char *filename) { return (directory_type*)0;}
792: char *readdirectory(directory_type *dir) { return (char*)0;}
793: void closedirectory(directory_type *dir) {}
794:
795:
1.1.1.5 ! misho 796: /************* Test for regular file when we can't do it **********/
1.1 misho 797:
798: /* Assume all files are regular. */
799:
800: int isregfile(char *filename) { return 1; }
801:
802:
803: /************* Test for a terminal when we can't do it **********/
804:
805: static BOOL
806: is_stdout_tty(void)
807: {
808: return FALSE;
809: }
810:
811: static BOOL
812: is_file_tty(FILE *f)
813: {
814: return FALSE;
815: }
816:
1.1.1.5 ! misho 817: #endif /* End of system-specific functions */
1.1 misho 818:
819:
820:
821: #ifndef HAVE_STRERROR
822: /*************************************************
823: * Provide strerror() for non-ANSI libraries *
824: *************************************************/
825:
826: /* Some old-fashioned systems still around (e.g. SunOS4) don't have strerror()
827: in their libraries, but can provide the same facility by this simple
828: alternative function. */
829:
830: extern int sys_nerr;
831: extern char *sys_errlist[];
832:
833: char *
834: strerror(int n)
835: {
836: if (n < 0 || n >= sys_nerr) return "unknown error number";
837: return sys_errlist[n];
838: }
839: #endif /* HAVE_STRERROR */
840:
841:
842:
843: /*************************************************
1.1.1.4 misho 844: * Usage function *
845: *************************************************/
846:
847: static int
848: usage(int rc)
849: {
850: option_item *op;
851: fprintf(stderr, "Usage: pcregrep [-");
852: for (op = optionlist; op->one_char != 0; op++)
853: {
854: if (op->one_char > 0) fprintf(stderr, "%c", op->one_char);
855: }
856: fprintf(stderr, "] [long options] [pattern] [files]\n");
857: fprintf(stderr, "Type `pcregrep --help' for more information and the long "
858: "options.\n");
859: return rc;
860: }
861:
862:
863:
864: /*************************************************
865: * Help function *
866: *************************************************/
867:
868: static void
869: help(void)
870: {
871: option_item *op;
872:
873: printf("Usage: pcregrep [OPTION]... [PATTERN] [FILE1 FILE2 ...]\n");
874: printf("Search for PATTERN in each FILE or standard input.\n");
875: printf("PATTERN must be present if neither -e nor -f is used.\n");
876: printf("\"-\" can be used as a file name to mean STDIN.\n");
877:
878: #ifdef SUPPORT_LIBZ
879: printf("Files whose names end in .gz are read using zlib.\n");
880: #endif
881:
882: #ifdef SUPPORT_LIBBZ2
883: printf("Files whose names end in .bz2 are read using bzlib2.\n");
884: #endif
885:
886: #if defined SUPPORT_LIBZ || defined SUPPORT_LIBBZ2
887: printf("Other files and the standard input are read as plain files.\n\n");
888: #else
889: printf("All files are read as plain files, without any interpretation.\n\n");
890: #endif
891:
892: printf("Example: pcregrep -i 'hello.*world' menu.h main.c\n\n");
893: printf("Options:\n");
894:
895: for (op = optionlist; op->one_char != 0; op++)
896: {
897: int n;
898: char s[4];
899:
900: /* Two options were accidentally implemented and documented with underscores
901: instead of hyphens in their names, something that was not noticed for quite a
902: few releases. When fixing this, I left the underscored versions in the list
903: in case people were using them. However, we don't want to display them in the
904: help data. There are no other options that contain underscores, and we do not
905: expect ever to implement such options. Therefore, just omit any option that
906: contains an underscore. */
907:
908: if (strchr(op->long_name, '_') != NULL) continue;
909:
910: if (op->one_char > 0 && (op->long_name)[0] == 0)
911: n = 31 - printf(" -%c", op->one_char);
912: else
913: {
914: if (op->one_char > 0) sprintf(s, "-%c,", op->one_char);
915: else strcpy(s, " ");
916: n = 31 - printf(" %s --%s", s, op->long_name);
917: }
918:
919: if (n < 1) n = 1;
920: printf("%.*s%s\n", n, " ", op->help_text);
921: }
922:
923: printf("\nNumbers may be followed by K or M, e.g. --buffer-size=100K.\n");
924: printf("The default value for --buffer-size is %d.\n", PCREGREP_BUFSIZE);
925: printf("When reading patterns or file names from a file, trailing white\n");
926: printf("space is removed and blank lines are ignored.\n");
927: printf("The maximum size of any pattern is %d bytes.\n", MAXPATLEN);
928:
929: printf("\nWith no FILEs, read standard input. If fewer than two FILEs given, assume -h.\n");
930: printf("Exit status is 0 if any matches, 1 if no matches, and 2 if trouble.\n");
931: }
932:
933:
934:
935: /*************************************************
936: * Test exclude/includes *
937: *************************************************/
938:
939: /* If any exclude pattern matches, the path is excluded. Otherwise, unless
940: there are no includes, the path must match an include pattern.
941:
942: Arguments:
943: path the path to be matched
944: ip the chain of include patterns
945: ep the chain of exclude patterns
946:
947: Returns: TRUE if the path is not excluded
948: */
949:
950: static BOOL
951: test_incexc(char *path, patstr *ip, patstr *ep)
952: {
953: int plen = strlen(path);
954:
955: for (; ep != NULL; ep = ep->next)
956: {
957: if (pcre_exec(ep->compiled, NULL, path, plen, 0, 0, NULL, 0) >= 0)
958: return FALSE;
959: }
960:
961: if (ip == NULL) return TRUE;
962:
963: for (; ip != NULL; ip = ip->next)
964: {
965: if (pcre_exec(ip->compiled, NULL, path, plen, 0, 0, NULL, 0) >= 0)
966: return TRUE;
967: }
968:
969: return FALSE;
970: }
971:
972:
973:
974: /*************************************************
975: * Decode integer argument value *
976: *************************************************/
977:
978: /* Integer arguments can be followed by K or M. Avoid the use of strtoul()
979: because SunOS4 doesn't have it. This is used only for unpicking arguments, so
980: just keep it simple.
981:
982: Arguments:
983: option_data the option data string
984: op the option item (for error messages)
985: longop TRUE if option given in long form
986:
987: Returns: a long integer
988: */
989:
990: static long int
991: decode_number(char *option_data, option_item *op, BOOL longop)
992: {
993: unsigned long int n = 0;
994: char *endptr = option_data;
995: while (*endptr != 0 && isspace((unsigned char)(*endptr))) endptr++;
996: while (isdigit((unsigned char)(*endptr)))
997: n = n * 10 + (int)(*endptr++ - '0');
998: if (toupper(*endptr) == 'K')
999: {
1000: n *= 1024;
1001: endptr++;
1002: }
1003: else if (toupper(*endptr) == 'M')
1004: {
1005: n *= 1024*1024;
1006: endptr++;
1007: }
1008:
1009: if (*endptr != 0) /* Error */
1010: {
1011: if (longop)
1012: {
1013: char *equals = strchr(op->long_name, '=');
1014: int nlen = (equals == NULL)? (int)strlen(op->long_name) :
1015: (int)(equals - op->long_name);
1016: fprintf(stderr, "pcregrep: Malformed number \"%s\" after --%.*s\n",
1017: option_data, nlen, op->long_name);
1018: }
1019: else
1020: fprintf(stderr, "pcregrep: Malformed number \"%s\" after -%c\n",
1021: option_data, op->one_char);
1022: pcregrep_exit(usage(2));
1023: }
1024:
1025: return n;
1026: }
1027:
1028:
1029:
1030: /*************************************************
1031: * Add item to a chain of numbers *
1032: *************************************************/
1033:
1034: /* Used to add an item onto a chain, or just return an unconnected item if the
1035: "after" argument is NULL.
1036:
1037: Arguments:
1038: n the number to add
1039: after if not NULL points to item to insert after
1040:
1041: Returns: new number block
1042: */
1043:
1044: static omstr *
1045: add_number(int n, omstr *after)
1046: {
1047: omstr *om = (omstr *)malloc(sizeof(omstr));
1048:
1049: if (om == NULL)
1050: {
1051: fprintf(stderr, "pcregrep: malloc failed\n");
1052: pcregrep_exit(2);
1053: }
1054: om->next = NULL;
1055: om->groupnum = n;
1056:
1057: if (after != NULL)
1058: {
1059: om->next = after->next;
1060: after->next = om;
1061: }
1062: return om;
1063: }
1064:
1065:
1066:
1067: /*************************************************
1.1 misho 1068: * Read one line of input *
1069: *************************************************/
1070:
1071: /* Normally, input is read using fread() into a large buffer, so many lines may
1072: be read at once. However, doing this for tty input means that no output appears
1073: until a lot of input has been typed. Instead, tty input is handled line by
1074: line. We cannot use fgets() for this, because it does not stop at a binary
1075: zero, and therefore there is no way of telling how many characters it has read,
1076: because there may be binary zeros embedded in the data.
1077:
1078: Arguments:
1079: buffer the buffer to read into
1080: length the maximum number of characters to read
1081: f the file
1082:
1083: Returns: the number of characters read, zero at end of file
1084: */
1085:
1.1.1.2 misho 1086: static unsigned int
1.1 misho 1087: read_one_line(char *buffer, int length, FILE *f)
1088: {
1089: int c;
1090: int yield = 0;
1091: while ((c = fgetc(f)) != EOF)
1092: {
1093: buffer[yield++] = c;
1094: if (c == '\n' || yield >= length) break;
1095: }
1096: return yield;
1097: }
1098:
1099:
1100:
1101: /*************************************************
1102: * Find end of line *
1103: *************************************************/
1104:
1105: /* The length of the endline sequence that is found is set via lenptr. This may
1106: be zero at the very end of the file if there is no line-ending sequence there.
1107:
1108: Arguments:
1109: p current position in line
1110: endptr end of available data
1111: lenptr where to put the length of the eol sequence
1112:
1113: Returns: pointer after the last byte of the line,
1114: including the newline byte(s)
1115: */
1116:
1117: static char *
1118: end_of_line(char *p, char *endptr, int *lenptr)
1119: {
1120: switch(endlinetype)
1121: {
1122: default: /* Just in case */
1123: case EL_LF:
1124: while (p < endptr && *p != '\n') p++;
1125: if (p < endptr)
1126: {
1127: *lenptr = 1;
1128: return p + 1;
1129: }
1130: *lenptr = 0;
1131: return endptr;
1132:
1133: case EL_CR:
1134: while (p < endptr && *p != '\r') p++;
1135: if (p < endptr)
1136: {
1137: *lenptr = 1;
1138: return p + 1;
1139: }
1140: *lenptr = 0;
1141: return endptr;
1142:
1143: case EL_CRLF:
1144: for (;;)
1145: {
1146: while (p < endptr && *p != '\r') p++;
1147: if (++p >= endptr)
1148: {
1149: *lenptr = 0;
1150: return endptr;
1151: }
1152: if (*p == '\n')
1153: {
1154: *lenptr = 2;
1155: return p + 1;
1156: }
1157: }
1158: break;
1159:
1160: case EL_ANYCRLF:
1161: while (p < endptr)
1162: {
1163: int extra = 0;
1164: register int c = *((unsigned char *)p);
1165:
1166: if (utf8 && c >= 0xc0)
1167: {
1168: int gcii, gcss;
1169: extra = utf8_table4[c & 0x3f]; /* Number of additional bytes */
1170: gcss = 6*extra;
1171: c = (c & utf8_table3[extra]) << gcss;
1172: for (gcii = 1; gcii <= extra; gcii++)
1173: {
1174: gcss -= 6;
1175: c |= (p[gcii] & 0x3f) << gcss;
1176: }
1177: }
1178:
1179: p += 1 + extra;
1180:
1181: switch (c)
1182: {
1.1.1.4 misho 1183: case '\n':
1.1 misho 1184: *lenptr = 1;
1185: return p;
1186:
1.1.1.4 misho 1187: case '\r':
1188: if (p < endptr && *p == '\n')
1.1 misho 1189: {
1190: *lenptr = 2;
1191: p++;
1192: }
1193: else *lenptr = 1;
1194: return p;
1195:
1196: default:
1197: break;
1198: }
1199: } /* End of loop for ANYCRLF case */
1200:
1201: *lenptr = 0; /* Must have hit the end */
1202: return endptr;
1203:
1204: case EL_ANY:
1205: while (p < endptr)
1206: {
1207: int extra = 0;
1208: register int c = *((unsigned char *)p);
1209:
1210: if (utf8 && c >= 0xc0)
1211: {
1212: int gcii, gcss;
1213: extra = utf8_table4[c & 0x3f]; /* Number of additional bytes */
1214: gcss = 6*extra;
1215: c = (c & utf8_table3[extra]) << gcss;
1216: for (gcii = 1; gcii <= extra; gcii++)
1217: {
1218: gcss -= 6;
1219: c |= (p[gcii] & 0x3f) << gcss;
1220: }
1221: }
1222:
1223: p += 1 + extra;
1224:
1225: switch (c)
1226: {
1.1.1.4 misho 1227: case '\n': /* LF */
1228: case '\v': /* VT */
1229: case '\f': /* FF */
1.1 misho 1230: *lenptr = 1;
1231: return p;
1232:
1.1.1.4 misho 1233: case '\r': /* CR */
1234: if (p < endptr && *p == '\n')
1.1 misho 1235: {
1236: *lenptr = 2;
1237: p++;
1238: }
1239: else *lenptr = 1;
1240: return p;
1241:
1.1.1.4 misho 1242: #ifndef EBCDIC
1243: case 0x85: /* Unicode NEL */
1.1 misho 1244: *lenptr = utf8? 2 : 1;
1245: return p;
1246:
1.1.1.4 misho 1247: case 0x2028: /* Unicode LS */
1248: case 0x2029: /* Unicode PS */
1.1 misho 1249: *lenptr = 3;
1250: return p;
1.1.1.4 misho 1251: #endif /* Not EBCDIC */
1.1 misho 1252:
1253: default:
1254: break;
1255: }
1256: } /* End of loop for ANY case */
1257:
1258: *lenptr = 0; /* Must have hit the end */
1259: return endptr;
1260: } /* End of overall switch */
1261: }
1262:
1263:
1264:
1265: /*************************************************
1266: * Find start of previous line *
1267: *************************************************/
1268:
1269: /* This is called when looking back for before lines to print.
1270:
1271: Arguments:
1272: p start of the subsequent line
1273: startptr start of available data
1274:
1275: Returns: pointer to the start of the previous line
1276: */
1277:
1278: static char *
1279: previous_line(char *p, char *startptr)
1280: {
1281: switch(endlinetype)
1282: {
1283: default: /* Just in case */
1284: case EL_LF:
1285: p--;
1286: while (p > startptr && p[-1] != '\n') p--;
1287: return p;
1288:
1289: case EL_CR:
1290: p--;
1291: while (p > startptr && p[-1] != '\n') p--;
1292: return p;
1293:
1294: case EL_CRLF:
1295: for (;;)
1296: {
1297: p -= 2;
1298: while (p > startptr && p[-1] != '\n') p--;
1299: if (p <= startptr + 1 || p[-2] == '\r') return p;
1300: }
1301: return p; /* But control should never get here */
1302:
1303: case EL_ANY:
1304: case EL_ANYCRLF:
1305: if (*(--p) == '\n' && p > startptr && p[-1] == '\r') p--;
1306: if (utf8) while ((*p & 0xc0) == 0x80) p--;
1307:
1308: while (p > startptr)
1309: {
1.1.1.4 misho 1310: register unsigned int c;
1.1 misho 1311: char *pp = p - 1;
1312:
1313: if (utf8)
1314: {
1315: int extra = 0;
1316: while ((*pp & 0xc0) == 0x80) pp--;
1317: c = *((unsigned char *)pp);
1318: if (c >= 0xc0)
1319: {
1320: int gcii, gcss;
1321: extra = utf8_table4[c & 0x3f]; /* Number of additional bytes */
1322: gcss = 6*extra;
1323: c = (c & utf8_table3[extra]) << gcss;
1324: for (gcii = 1; gcii <= extra; gcii++)
1325: {
1326: gcss -= 6;
1327: c |= (pp[gcii] & 0x3f) << gcss;
1328: }
1329: }
1330: }
1331: else c = *((unsigned char *)pp);
1332:
1333: if (endlinetype == EL_ANYCRLF) switch (c)
1334: {
1.1.1.4 misho 1335: case '\n': /* LF */
1336: case '\r': /* CR */
1.1 misho 1337: return p;
1338:
1339: default:
1340: break;
1341: }
1342:
1343: else switch (c)
1344: {
1.1.1.4 misho 1345: case '\n': /* LF */
1346: case '\v': /* VT */
1347: case '\f': /* FF */
1348: case '\r': /* CR */
1349: #ifndef EBCDIE
1350: case 0x85: /* Unicode NEL */
1351: case 0x2028: /* Unicode LS */
1352: case 0x2029: /* Unicode PS */
1353: #endif /* Not EBCDIC */
1.1 misho 1354: return p;
1355:
1356: default:
1357: break;
1358: }
1359:
1360: p = pp; /* Back one character */
1361: } /* End of loop for ANY case */
1362:
1363: return startptr; /* Hit start of data */
1364: } /* End of overall switch */
1365: }
1366:
1367:
1368:
1369:
1370:
1371: /*************************************************
1372: * Print the previous "after" lines *
1373: *************************************************/
1374:
1375: /* This is called if we are about to lose said lines because of buffer filling,
1376: and at the end of the file. The data in the line is written using fwrite() so
1377: that a binary zero does not terminate it.
1378:
1379: Arguments:
1380: lastmatchnumber the number of the last matching line, plus one
1381: lastmatchrestart where we restarted after the last match
1382: endptr end of available data
1383: printname filename for printing
1384:
1385: Returns: nothing
1386: */
1387:
1.1.1.4 misho 1388: static void
1389: do_after_lines(int lastmatchnumber, char *lastmatchrestart, char *endptr,
1390: char *printname)
1.1 misho 1391: {
1392: if (after_context > 0 && lastmatchnumber > 0)
1393: {
1394: int count = 0;
1395: while (lastmatchrestart < endptr && count++ < after_context)
1396: {
1397: int ellength;
1398: char *pp = lastmatchrestart;
1399: if (printname != NULL) fprintf(stdout, "%s-", printname);
1400: if (number) fprintf(stdout, "%d-", lastmatchnumber++);
1401: pp = end_of_line(pp, endptr, &ellength);
1402: FWRITE(lastmatchrestart, 1, pp - lastmatchrestart, stdout);
1403: lastmatchrestart = pp;
1404: }
1405: hyphenpending = TRUE;
1406: }
1407: }
1408:
1409:
1410:
1411: /*************************************************
1412: * Apply patterns to subject till one matches *
1413: *************************************************/
1414:
1415: /* This function is called to run through all patterns, looking for a match. It
1416: is used multiple times for the same subject when colouring is enabled, in order
1417: to find all possible matches.
1418:
1419: Arguments:
1420: matchptr the start of the subject
1421: length the length of the subject to match
1.1.1.4 misho 1422: options options for pcre_exec
1.1 misho 1423: startoffset where to start matching
1424: offsets the offets vector to fill in
1425: mrc address of where to put the result of pcre_exec()
1426:
1427: Returns: TRUE if there was a match
1428: FALSE if there was no match
1429: invert if there was a non-fatal error
1430: */
1431:
1432: static BOOL
1.1.1.4 misho 1433: match_patterns(char *matchptr, size_t length, unsigned int options,
1434: int startoffset, int *offsets, int *mrc)
1.1 misho 1435: {
1436: int i;
1437: size_t slen = length;
1.1.1.4 misho 1438: patstr *p = patterns;
1.1 misho 1439: const char *msg = "this text:\n\n";
1.1.1.4 misho 1440:
1.1 misho 1441: if (slen > 200)
1442: {
1443: slen = 200;
1444: msg = "text that starts:\n\n";
1445: }
1.1.1.4 misho 1446: for (i = 1; p != NULL; p = p->next, i++)
1.1 misho 1447: {
1.1.1.4 misho 1448: *mrc = pcre_exec(p->compiled, p->hint, matchptr, (int)length,
1449: startoffset, options, offsets, OFFSET_SIZE);
1.1 misho 1450: if (*mrc >= 0) return TRUE;
1451: if (*mrc == PCRE_ERROR_NOMATCH) continue;
1452: fprintf(stderr, "pcregrep: pcre_exec() gave error %d while matching ", *mrc);
1.1.1.4 misho 1453: if (patterns->next != NULL) fprintf(stderr, "pattern number %d to ", i);
1.1 misho 1454: fprintf(stderr, "%s", msg);
1455: FWRITE(matchptr, 1, slen, stderr); /* In case binary zero included */
1456: fprintf(stderr, "\n\n");
1457: if (*mrc == PCRE_ERROR_MATCHLIMIT || *mrc == PCRE_ERROR_RECURSIONLIMIT ||
1458: *mrc == PCRE_ERROR_JIT_STACKLIMIT)
1459: resource_error = TRUE;
1460: if (error_count++ > 20)
1461: {
1462: fprintf(stderr, "pcregrep: Too many errors - abandoned.\n");
1463: pcregrep_exit(2);
1464: }
1465: return invert; /* No more matching; don't show the line again */
1466: }
1467:
1468: return FALSE; /* No match, no errors */
1469: }
1470:
1471:
1472:
1473: /*************************************************
1474: * Grep an individual file *
1475: *************************************************/
1476:
1477: /* This is called from grep_or_recurse() below. It uses a buffer that is three
1478: times the value of bufthird. The matching point is never allowed to stray into
1479: the top third of the buffer, thus keeping more of the file available for
1480: context printing or for multiline scanning. For large files, the pointer will
1481: be in the middle third most of the time, so the bottom third is available for
1482: "before" context printing.
1483:
1484: Arguments:
1485: handle the fopened FILE stream for a normal file
1486: the gzFile pointer when reading is via libz
1487: the BZFILE pointer when reading is via libbz2
1488: frtype FR_PLAIN, FR_LIBZ, or FR_LIBBZ2
1489: filename the file name or NULL (for errors)
1490: printname the file name if it is to be printed for each match
1491: or NULL if the file name is not to be printed
1492: it cannot be NULL if filenames[_nomatch]_only is set
1493:
1494: Returns: 0 if there was at least one match
1495: 1 otherwise (no matches)
1496: 2 if an overlong line is encountered
1497: 3 if there is a read error on a .bz2 file
1498: */
1499:
1500: static int
1501: pcregrep(void *handle, int frtype, char *filename, char *printname)
1502: {
1503: int rc = 1;
1504: int linenumber = 1;
1505: int lastmatchnumber = 0;
1506: int count = 0;
1507: int filepos = 0;
1508: int offsets[OFFSET_SIZE];
1509: char *lastmatchrestart = NULL;
1510: char *ptr = main_buffer;
1511: char *endptr;
1512: size_t bufflength;
1.1.1.3 misho 1513: BOOL binary = FALSE;
1.1 misho 1514: BOOL endhyphenpending = FALSE;
1515: BOOL input_line_buffered = line_buffered;
1516: FILE *in = NULL; /* Ensure initialized */
1517:
1518: #ifdef SUPPORT_LIBZ
1519: gzFile ingz = NULL;
1520: #endif
1521:
1522: #ifdef SUPPORT_LIBBZ2
1523: BZFILE *inbz2 = NULL;
1524: #endif
1525:
1526:
1527: /* Do the first read into the start of the buffer and set up the pointer to end
1528: of what we have. In the case of libz, a non-zipped .gz file will be read as a
1529: plain file. However, if a .bz2 file isn't actually bzipped, the first read will
1530: fail. */
1531:
1.1.1.4 misho 1532: (void)frtype;
1533:
1.1 misho 1534: #ifdef SUPPORT_LIBZ
1535: if (frtype == FR_LIBZ)
1536: {
1537: ingz = (gzFile)handle;
1538: bufflength = gzread (ingz, main_buffer, bufsize);
1539: }
1540: else
1541: #endif
1542:
1543: #ifdef SUPPORT_LIBBZ2
1544: if (frtype == FR_LIBBZ2)
1545: {
1546: inbz2 = (BZFILE *)handle;
1547: bufflength = BZ2_bzread(inbz2, main_buffer, bufsize);
1548: if ((int)bufflength < 0) return 2; /* Gotcha: bufflength is size_t; */
1549: } /* without the cast it is unsigned. */
1550: else
1551: #endif
1552:
1553: {
1554: in = (FILE *)handle;
1555: if (is_file_tty(in)) input_line_buffered = TRUE;
1556: bufflength = input_line_buffered?
1557: read_one_line(main_buffer, bufsize, in) :
1558: fread(main_buffer, 1, bufsize, in);
1559: }
1560:
1561: endptr = main_buffer + bufflength;
1562:
1.1.1.3 misho 1563: /* Unless binary-files=text, see if we have a binary file. This uses the same
1564: rule as GNU grep, namely, a search for a binary zero byte near the start of the
1565: file. */
1566:
1567: if (binary_files != BIN_TEXT)
1568: {
1569: binary =
1570: memchr(main_buffer, 0, (bufflength > 1024)? 1024 : bufflength) != NULL;
1571: if (binary && binary_files == BIN_NOMATCH) return 1;
1572: }
1573:
1.1 misho 1574: /* Loop while the current pointer is not at the end of the file. For large
1575: files, endptr will be at the end of the buffer when we are in the middle of the
1576: file, but ptr will never get there, because as soon as it gets over 2/3 of the
1577: way, the buffer is shifted left and re-filled. */
1578:
1579: while (ptr < endptr)
1580: {
1581: int endlinelength;
1582: int mrc = 0;
1583: int startoffset = 0;
1.1.1.4 misho 1584: unsigned int options = 0;
1.1 misho 1585: BOOL match;
1586: char *matchptr = ptr;
1587: char *t = ptr;
1588: size_t length, linelength;
1589:
1590: /* At this point, ptr is at the start of a line. We need to find the length
1591: of the subject string to pass to pcre_exec(). In multiline mode, it is the
1592: length remainder of the data in the buffer. Otherwise, it is the length of
1593: the next line, excluding the terminating newline. After matching, we always
1594: advance by the length of the next line. In multiline mode the PCRE_FIRSTLINE
1595: option is used for compiling, so that any match is constrained to be in the
1596: first line. */
1597:
1598: t = end_of_line(t, endptr, &endlinelength);
1599: linelength = t - ptr - endlinelength;
1600: length = multiline? (size_t)(endptr - ptr) : linelength;
1601:
1602: /* Check to see if the line we are looking at extends right to the very end
1603: of the buffer without a line terminator. This means the line is too long to
1604: handle. */
1605:
1606: if (endlinelength == 0 && t == main_buffer + bufsize)
1607: {
1608: fprintf(stderr, "pcregrep: line %d%s%s is too long for the internal buffer\n"
1609: "pcregrep: check the --buffer-size option\n",
1610: linenumber,
1611: (filename == NULL)? "" : " of file ",
1612: (filename == NULL)? "" : filename);
1613: return 2;
1614: }
1615:
1616: /* Extra processing for Jeffrey Friedl's debugging. */
1617:
1618: #ifdef JFRIEDL_DEBUG
1619: if (jfriedl_XT || jfriedl_XR)
1620: {
1.1.1.4 misho 1621: # include <sys/time.h>
1622: # include <time.h>
1.1 misho 1623: struct timeval start_time, end_time;
1624: struct timezone dummy;
1625: int i;
1626:
1627: if (jfriedl_XT)
1628: {
1629: unsigned long newlen = length * jfriedl_XT + strlen(jfriedl_prefix) + strlen(jfriedl_postfix);
1630: const char *orig = ptr;
1631: ptr = malloc(newlen + 1);
1632: if (!ptr) {
1633: printf("out of memory");
1634: pcregrep_exit(2);
1635: }
1636: endptr = ptr;
1637: strcpy(endptr, jfriedl_prefix); endptr += strlen(jfriedl_prefix);
1638: for (i = 0; i < jfriedl_XT; i++) {
1639: strncpy(endptr, orig, length);
1640: endptr += length;
1641: }
1642: strcpy(endptr, jfriedl_postfix); endptr += strlen(jfriedl_postfix);
1643: length = newlen;
1644: }
1645:
1646: if (gettimeofday(&start_time, &dummy) != 0)
1647: perror("bad gettimeofday");
1648:
1649:
1650: for (i = 0; i < jfriedl_XR; i++)
1.1.1.4 misho 1651: match = (pcre_exec(patterns->compiled, patterns->hint, ptr, length, 0,
1.1 misho 1652: PCRE_NOTEMPTY, offsets, OFFSET_SIZE) >= 0);
1653:
1654: if (gettimeofday(&end_time, &dummy) != 0)
1655: perror("bad gettimeofday");
1656:
1657: double delta = ((end_time.tv_sec + (end_time.tv_usec / 1000000.0))
1658: -
1659: (start_time.tv_sec + (start_time.tv_usec / 1000000.0)));
1660:
1661: printf("%s TIMER[%.4f]\n", match ? "MATCH" : "FAIL", delta);
1662: return 0;
1663: }
1664: #endif
1665:
1.1.1.4 misho 1666: /* We come back here after a match when show_only_matching is set, in order
1667: to find any further matches in the same line. This applies to
1668: --only-matching, --file-offsets, and --line-offsets. */
1.1 misho 1669:
1670: ONLY_MATCHING_RESTART:
1671:
1672: /* Run through all the patterns until one matches or there is an error other
1673: than NOMATCH. This code is in a subroutine so that it can be re-used for
1.1.1.4 misho 1674: finding subsequent matches when colouring matched lines. After finding one
1675: match, set PCRE_NOTEMPTY to disable any further matches of null strings in
1676: this line. */
1.1 misho 1677:
1.1.1.4 misho 1678: match = match_patterns(matchptr, length, options, startoffset, offsets, &mrc);
1679: options = PCRE_NOTEMPTY;
1.1 misho 1680:
1681: /* If it's a match or a not-match (as required), do what's wanted. */
1682:
1683: if (match != invert)
1684: {
1685: BOOL hyphenprinted = FALSE;
1686:
1687: /* We've failed if we want a file that doesn't have any matches. */
1688:
1689: if (filenames == FN_NOMATCH_ONLY) return 1;
1690:
1691: /* Just count if just counting is wanted. */
1692:
1693: if (count_only) count++;
1694:
1.1.1.3 misho 1695: /* When handling a binary file and binary-files==binary, the "binary"
1696: variable will be set true (it's false in all other cases). In this
1697: situation we just want to output the file name. No need to scan further. */
1698:
1699: else if (binary)
1700: {
1701: fprintf(stdout, "Binary file %s matches\n", filename);
1702: return 0;
1703: }
1704:
1.1 misho 1705: /* If all we want is a file name, there is no need to scan any more lines
1706: in the file. */
1707:
1708: else if (filenames == FN_MATCH_ONLY)
1709: {
1710: fprintf(stdout, "%s\n", printname);
1711: return 0;
1712: }
1713:
1714: /* Likewise, if all we want is a yes/no answer. */
1715:
1716: else if (quiet) return 0;
1717:
1.1.1.4 misho 1718: /* The --only-matching option prints just the substring that matched,
1719: and/or one or more captured portions of it, as long as these strings are
1720: not empty. The --file-offsets and --line-offsets options output offsets for
1721: the matching substring (all three set show_only_matching). None of these
1722: mutually exclusive options prints any context. Afterwards, adjust the start
1723: and then jump back to look for further matches in the same line. If we are
1724: in invert mode, however, nothing is printed and we do not restart - this
1725: could still be useful because the return code is set. */
1.1 misho 1726:
1.1.1.4 misho 1727: else if (show_only_matching)
1.1 misho 1728: {
1729: if (!invert)
1730: {
1731: if (printname != NULL) fprintf(stdout, "%s:", printname);
1732: if (number) fprintf(stdout, "%d:", linenumber);
1.1.1.4 misho 1733:
1734: /* Handle --line-offsets */
1735:
1.1 misho 1736: if (line_offsets)
1737: fprintf(stdout, "%d,%d\n", (int)(matchptr + offsets[0] - ptr),
1738: offsets[1] - offsets[0]);
1.1.1.4 misho 1739:
1740: /* Handle --file-offsets */
1741:
1.1 misho 1742: else if (file_offsets)
1743: fprintf(stdout, "%d,%d\n",
1744: (int)(filepos + matchptr + offsets[0] - ptr),
1745: offsets[1] - offsets[0]);
1.1.1.4 misho 1746:
1747: /* Handle --only-matching, which may occur many times */
1748:
1749: else
1.1 misho 1750: {
1.1.1.4 misho 1751: BOOL printed = FALSE;
1752: omstr *om;
1753:
1754: for (om = only_matching; om != NULL; om = om->next)
1.1 misho 1755: {
1.1.1.4 misho 1756: int n = om->groupnum;
1757: if (n < mrc)
1758: {
1759: int plen = offsets[2*n + 1] - offsets[2*n];
1760: if (plen > 0)
1761: {
1762: if (printed) fprintf(stdout, "%s", om_separator);
1763: if (do_colour) fprintf(stdout, "%c[%sm", 0x1b, colour_string);
1764: FWRITE(matchptr + offsets[n*2], 1, plen, stdout);
1765: if (do_colour) fprintf(stdout, "%c[00m", 0x1b);
1766: printed = TRUE;
1767: }
1768: }
1.1 misho 1769: }
1.1.1.4 misho 1770:
1771: if (printed || printname != NULL || number) fprintf(stdout, "\n");
1.1 misho 1772: }
1.1.1.4 misho 1773:
1774: /* Prepare to repeat to find the next match */
1775:
1.1 misho 1776: match = FALSE;
1777: if (line_buffered) fflush(stdout);
1778: rc = 0; /* Had some success */
1779: startoffset = offsets[1]; /* Restart after the match */
1780: goto ONLY_MATCHING_RESTART;
1781: }
1782: }
1783:
1784: /* This is the default case when none of the above options is set. We print
1785: the matching lines(s), possibly preceded and/or followed by other lines of
1786: context. */
1787:
1788: else
1789: {
1790: /* See if there is a requirement to print some "after" lines from a
1791: previous match. We never print any overlaps. */
1792:
1793: if (after_context > 0 && lastmatchnumber > 0)
1794: {
1795: int ellength;
1796: int linecount = 0;
1797: char *p = lastmatchrestart;
1798:
1799: while (p < ptr && linecount < after_context)
1800: {
1801: p = end_of_line(p, ptr, &ellength);
1802: linecount++;
1803: }
1804:
1805: /* It is important to advance lastmatchrestart during this printing so
1806: that it interacts correctly with any "before" printing below. Print
1807: each line's data using fwrite() in case there are binary zeroes. */
1808:
1809: while (lastmatchrestart < p)
1810: {
1811: char *pp = lastmatchrestart;
1812: if (printname != NULL) fprintf(stdout, "%s-", printname);
1813: if (number) fprintf(stdout, "%d-", lastmatchnumber++);
1814: pp = end_of_line(pp, endptr, &ellength);
1815: FWRITE(lastmatchrestart, 1, pp - lastmatchrestart, stdout);
1816: lastmatchrestart = pp;
1817: }
1818: if (lastmatchrestart != ptr) hyphenpending = TRUE;
1819: }
1820:
1821: /* If there were non-contiguous lines printed above, insert hyphens. */
1822:
1823: if (hyphenpending)
1824: {
1825: fprintf(stdout, "--\n");
1826: hyphenpending = FALSE;
1827: hyphenprinted = TRUE;
1828: }
1829:
1830: /* See if there is a requirement to print some "before" lines for this
1831: match. Again, don't print overlaps. */
1832:
1833: if (before_context > 0)
1834: {
1835: int linecount = 0;
1836: char *p = ptr;
1837:
1838: while (p > main_buffer && (lastmatchnumber == 0 || p > lastmatchrestart) &&
1839: linecount < before_context)
1840: {
1841: linecount++;
1842: p = previous_line(p, main_buffer);
1843: }
1844:
1845: if (lastmatchnumber > 0 && p > lastmatchrestart && !hyphenprinted)
1846: fprintf(stdout, "--\n");
1847:
1848: while (p < ptr)
1849: {
1850: int ellength;
1851: char *pp = p;
1852: if (printname != NULL) fprintf(stdout, "%s-", printname);
1853: if (number) fprintf(stdout, "%d-", linenumber - linecount--);
1854: pp = end_of_line(pp, endptr, &ellength);
1855: FWRITE(p, 1, pp - p, stdout);
1856: p = pp;
1857: }
1858: }
1859:
1860: /* Now print the matching line(s); ensure we set hyphenpending at the end
1861: of the file if any context lines are being output. */
1862:
1863: if (after_context > 0 || before_context > 0)
1864: endhyphenpending = TRUE;
1865:
1866: if (printname != NULL) fprintf(stdout, "%s:", printname);
1867: if (number) fprintf(stdout, "%d:", linenumber);
1868:
1869: /* In multiline mode, we want to print to the end of the line in which
1870: the end of the matched string is found, so we adjust linelength and the
1871: line number appropriately, but only when there actually was a match
1872: (invert not set). Because the PCRE_FIRSTLINE option is set, the start of
1873: the match will always be before the first newline sequence. */
1874:
1875: if (multiline & !invert)
1876: {
1877: char *endmatch = ptr + offsets[1];
1878: t = ptr;
1.1.1.5 ! misho 1879: while (t <= endmatch)
1.1 misho 1880: {
1881: t = end_of_line(t, endptr, &endlinelength);
1882: if (t < endmatch) linenumber++; else break;
1883: }
1884: linelength = t - ptr - endlinelength;
1885: }
1886:
1887: /*** NOTE: Use only fwrite() to output the data line, so that binary
1888: zeroes are treated as just another data character. */
1889:
1890: /* This extra option, for Jeffrey Friedl's debugging requirements,
1891: replaces the matched string, or a specific captured string if it exists,
1892: with X. When this happens, colouring is ignored. */
1893:
1894: #ifdef JFRIEDL_DEBUG
1895: if (S_arg >= 0 && S_arg < mrc)
1896: {
1897: int first = S_arg * 2;
1898: int last = first + 1;
1899: FWRITE(ptr, 1, offsets[first], stdout);
1900: fprintf(stdout, "X");
1901: FWRITE(ptr + offsets[last], 1, linelength - offsets[last], stdout);
1902: }
1903: else
1904: #endif
1905:
1906: /* We have to split the line(s) up if colouring, and search for further
1907: matches, but not of course if the line is a non-match. */
1908:
1909: if (do_colour && !invert)
1910: {
1911: int plength;
1912: FWRITE(ptr, 1, offsets[0], stdout);
1913: fprintf(stdout, "%c[%sm", 0x1b, colour_string);
1914: FWRITE(ptr + offsets[0], 1, offsets[1] - offsets[0], stdout);
1915: fprintf(stdout, "%c[00m", 0x1b);
1916: for (;;)
1917: {
1918: startoffset = offsets[1];
1919: if (startoffset >= (int)linelength + endlinelength ||
1.1.1.4 misho 1920: !match_patterns(matchptr, length, options, startoffset, offsets,
1921: &mrc))
1.1 misho 1922: break;
1923: FWRITE(matchptr + startoffset, 1, offsets[0] - startoffset, stdout);
1924: fprintf(stdout, "%c[%sm", 0x1b, colour_string);
1925: FWRITE(matchptr + offsets[0], 1, offsets[1] - offsets[0], stdout);
1926: fprintf(stdout, "%c[00m", 0x1b);
1927: }
1928:
1929: /* In multiline mode, we may have already printed the complete line
1930: and its line-ending characters (if they matched the pattern), so there
1931: may be no more to print. */
1932:
1933: plength = (int)((linelength + endlinelength) - startoffset);
1934: if (plength > 0) FWRITE(ptr + startoffset, 1, plength, stdout);
1935: }
1936:
1937: /* Not colouring; no need to search for further matches */
1938:
1939: else FWRITE(ptr, 1, linelength + endlinelength, stdout);
1940: }
1941:
1942: /* End of doing what has to be done for a match. If --line-buffered was
1943: given, flush the output. */
1944:
1945: if (line_buffered) fflush(stdout);
1946: rc = 0; /* Had some success */
1947:
1948: /* Remember where the last match happened for after_context. We remember
1949: where we are about to restart, and that line's number. */
1950:
1951: lastmatchrestart = ptr + linelength + endlinelength;
1952: lastmatchnumber = linenumber + 1;
1953: }
1954:
1955: /* For a match in multiline inverted mode (which of course did not cause
1956: anything to be printed), we have to move on to the end of the match before
1957: proceeding. */
1958:
1959: if (multiline && invert && match)
1960: {
1961: int ellength;
1962: char *endmatch = ptr + offsets[1];
1963: t = ptr;
1964: while (t < endmatch)
1965: {
1966: t = end_of_line(t, endptr, &ellength);
1967: if (t <= endmatch) linenumber++; else break;
1968: }
1969: endmatch = end_of_line(endmatch, endptr, &ellength);
1970: linelength = endmatch - ptr - ellength;
1971: }
1972:
1973: /* Advance to after the newline and increment the line number. The file
1974: offset to the current line is maintained in filepos. */
1975:
1976: ptr += linelength + endlinelength;
1977: filepos += (int)(linelength + endlinelength);
1978: linenumber++;
1979:
1980: /* If input is line buffered, and the buffer is not yet full, read another
1981: line and add it into the buffer. */
1982:
1983: if (input_line_buffered && bufflength < (size_t)bufsize)
1984: {
1985: int add = read_one_line(ptr, bufsize - (int)(ptr - main_buffer), in);
1986: bufflength += add;
1987: endptr += add;
1988: }
1989:
1990: /* If we haven't yet reached the end of the file (the buffer is full), and
1991: the current point is in the top 1/3 of the buffer, slide the buffer down by
1992: 1/3 and refill it. Before we do this, if some unprinted "after" lines are
1993: about to be lost, print them. */
1994:
1995: if (bufflength >= (size_t)bufsize && ptr > main_buffer + 2*bufthird)
1996: {
1997: if (after_context > 0 &&
1998: lastmatchnumber > 0 &&
1999: lastmatchrestart < main_buffer + bufthird)
2000: {
2001: do_after_lines(lastmatchnumber, lastmatchrestart, endptr, printname);
2002: lastmatchnumber = 0;
2003: }
2004:
2005: /* Now do the shuffle */
2006:
2007: memmove(main_buffer, main_buffer + bufthird, 2*bufthird);
2008: ptr -= bufthird;
2009:
2010: #ifdef SUPPORT_LIBZ
2011: if (frtype == FR_LIBZ)
2012: bufflength = 2*bufthird +
2013: gzread (ingz, main_buffer + 2*bufthird, bufthird);
2014: else
2015: #endif
2016:
2017: #ifdef SUPPORT_LIBBZ2
2018: if (frtype == FR_LIBBZ2)
2019: bufflength = 2*bufthird +
2020: BZ2_bzread(inbz2, main_buffer + 2*bufthird, bufthird);
2021: else
2022: #endif
2023:
2024: bufflength = 2*bufthird +
2025: (input_line_buffered?
2026: read_one_line(main_buffer + 2*bufthird, bufthird, in) :
2027: fread(main_buffer + 2*bufthird, 1, bufthird, in));
2028: endptr = main_buffer + bufflength;
2029:
2030: /* Adjust any last match point */
2031:
2032: if (lastmatchnumber > 0) lastmatchrestart -= bufthird;
2033: }
2034: } /* Loop through the whole file */
2035:
2036: /* End of file; print final "after" lines if wanted; do_after_lines sets
2037: hyphenpending if it prints something. */
2038:
1.1.1.4 misho 2039: if (!show_only_matching && !count_only)
1.1 misho 2040: {
2041: do_after_lines(lastmatchnumber, lastmatchrestart, endptr, printname);
2042: hyphenpending |= endhyphenpending;
2043: }
2044:
2045: /* Print the file name if we are looking for those without matches and there
2046: were none. If we found a match, we won't have got this far. */
2047:
2048: if (filenames == FN_NOMATCH_ONLY)
2049: {
2050: fprintf(stdout, "%s\n", printname);
2051: return 0;
2052: }
2053:
2054: /* Print the match count if wanted */
2055:
2056: if (count_only)
2057: {
2058: if (count > 0 || !omit_zero_count)
2059: {
2060: if (printname != NULL && filenames != FN_NONE)
2061: fprintf(stdout, "%s:", printname);
2062: fprintf(stdout, "%d\n", count);
2063: }
2064: }
2065:
2066: return rc;
2067: }
2068:
2069:
2070:
2071: /*************************************************
2072: * Grep a file or recurse into a directory *
2073: *************************************************/
2074:
2075: /* Given a path name, if it's a directory, scan all the files if we are
2076: recursing; if it's a file, grep it.
2077:
2078: Arguments:
2079: pathname the path to investigate
2080: dir_recurse TRUE if recursing is wanted (-r or -drecurse)
2081: only_one_at_top TRUE if the path is the only one at toplevel
2082:
1.1.1.4 misho 2083: Returns: -1 the file/directory was skipped
2084: 0 if there was at least one match
1.1 misho 2085: 1 if there were no matches
2086: 2 there was some kind of error
2087:
2088: However, file opening failures are suppressed if "silent" is set.
2089: */
2090:
2091: static int
2092: grep_or_recurse(char *pathname, BOOL dir_recurse, BOOL only_one_at_top)
2093: {
2094: int rc = 1;
2095: int frtype;
2096: void *handle;
1.1.1.4 misho 2097: char *lastcomp;
1.1 misho 2098: FILE *in = NULL; /* Ensure initialized */
2099:
2100: #ifdef SUPPORT_LIBZ
2101: gzFile ingz = NULL;
2102: #endif
2103:
2104: #ifdef SUPPORT_LIBBZ2
2105: BZFILE *inbz2 = NULL;
2106: #endif
2107:
1.1.1.3 misho 2108: #if defined SUPPORT_LIBZ || defined SUPPORT_LIBBZ2
1.1.1.2 misho 2109: int pathlen;
2110: #endif
2111:
1.1.1.5 ! misho 2112: #if defined NATIVE_ZOS
! 2113: int zos_type;
! 2114: FILE *zos_test_file;
! 2115: #endif
! 2116:
1.1 misho 2117: /* If the file name is "-" we scan stdin */
2118:
2119: if (strcmp(pathname, "-") == 0)
2120: {
2121: return pcregrep(stdin, FR_PLAIN, stdin_name,
2122: (filenames > FN_DEFAULT || (filenames == FN_DEFAULT && !only_one_at_top))?
2123: stdin_name : NULL);
2124: }
2125:
1.1.1.4 misho 2126: /* Inclusion and exclusion: --include-dir and --exclude-dir apply only to
2127: directories, whereas --include and --exclude apply to everything else. The test
2128: is against the final component of the path. */
2129:
2130: lastcomp = strrchr(pathname, FILESEP);
2131: lastcomp = (lastcomp == NULL)? pathname : lastcomp + 1;
2132:
2133: /* If the file is a directory, skip if not recursing or if explicitly excluded.
2134: Otherwise, scan the directory and recurse for each path within it. The scanning
2135: code is localized so it can be made system-specific. */
2136:
1.1.1.5 ! misho 2137:
! 2138: /* For z/OS, determine the file type. */
! 2139:
! 2140: #if defined NATIVE_ZOS
! 2141: zos_test_file = fopen(pathname,"rb");
! 2142:
! 2143: if (zos_test_file == NULL)
! 2144: {
! 2145: if (!silent) fprintf(stderr, "pcregrep: failed to test next file %s\n",
! 2146: pathname, strerror(errno));
! 2147: return -1;
! 2148: }
! 2149: zos_type = identifyzosfiletype (zos_test_file);
! 2150: fclose (zos_test_file);
! 2151:
! 2152: /* Handle a PDS in separate code */
! 2153:
! 2154: if (zos_type == __ZOS_PDS || zos_type == __ZOS_PDSE)
! 2155: {
! 2156: return travelonpdsdir (pathname, only_one_at_top);
! 2157: }
! 2158:
! 2159: /* Deal with regular files in the normal way below. These types are:
! 2160: zos_type == __ZOS_PDS_MEMBER
! 2161: zos_type == __ZOS_PS
! 2162: zos_type == __ZOS_VSAM_KSDS
! 2163: zos_type == __ZOS_VSAM_ESDS
! 2164: zos_type == __ZOS_VSAM_RRDS
! 2165: */
! 2166:
! 2167: /* Handle a z/OS directory using common code. */
! 2168:
! 2169: else if (zos_type == __ZOS_HFS)
! 2170: {
! 2171: #endif /* NATIVE_ZOS */
! 2172:
! 2173:
! 2174: /* Handle directories: common code for all OS */
! 2175:
1.1.1.4 misho 2176: if (isdirectory(pathname))
2177: {
2178: if (dee_action == dee_SKIP ||
2179: !test_incexc(lastcomp, include_dir_patterns, exclude_dir_patterns))
2180: return -1;
1.1 misho 2181:
2182: if (dee_action == dee_RECURSE)
2183: {
2184: char buffer[1024];
2185: char *nextfile;
2186: directory_type *dir = opendirectory(pathname);
2187:
2188: if (dir == NULL)
2189: {
2190: if (!silent)
2191: fprintf(stderr, "pcregrep: Failed to open directory %s: %s\n", pathname,
2192: strerror(errno));
2193: return 2;
2194: }
2195:
2196: while ((nextfile = readdirectory(dir)) != NULL)
2197: {
1.1.1.4 misho 2198: int frc;
2199: sprintf(buffer, "%.512s%c%.128s", pathname, FILESEP, nextfile);
1.1 misho 2200: frc = grep_or_recurse(buffer, dir_recurse, FALSE);
2201: if (frc > 1) rc = frc;
2202: else if (frc == 0 && rc == 1) rc = 0;
2203: }
2204:
2205: closedirectory(dir);
2206: return rc;
2207: }
2208: }
2209:
1.1.1.5 ! misho 2210: #if defined NATIVE_ZOS
! 2211: }
! 2212: #endif
1.1 misho 2213:
1.1.1.5 ! misho 2214: /* If the file is not a directory, check for a regular file, and if it is not,
! 2215: skip it if that's been requested. Otherwise, check for an explicit inclusion or
! 2216: exclusion. */
! 2217:
! 2218: else if (
! 2219: #if defined NATIVE_ZOS
! 2220: (zos_type == __ZOS_NOFILE && DEE_action == DEE_SKIP) ||
! 2221: #else /* all other OS */
! 2222: (!isregfile(pathname) && DEE_action == DEE_SKIP) ||
! 2223: #endif
! 2224: !test_incexc(lastcomp, include_patterns, exclude_patterns))
! 2225: return -1; /* File skipped */
1.1 misho 2226:
2227: /* Control reaches here if we have a regular file, or if we have a directory
2228: and recursion or skipping was not requested, or if we have anything else and
2229: skipping was not requested. The scan proceeds. If this is the first and only
2230: argument at top level, we don't show the file name, unless we are only showing
2231: the file name, or the filename was forced (-H). */
2232:
1.1.1.3 misho 2233: #if defined SUPPORT_LIBZ || defined SUPPORT_LIBBZ2
1.1 misho 2234: pathlen = (int)(strlen(pathname));
1.1.1.2 misho 2235: #endif
1.1 misho 2236:
2237: /* Open using zlib if it is supported and the file name ends with .gz. */
2238:
2239: #ifdef SUPPORT_LIBZ
2240: if (pathlen > 3 && strcmp(pathname + pathlen - 3, ".gz") == 0)
2241: {
2242: ingz = gzopen(pathname, "rb");
2243: if (ingz == NULL)
2244: {
2245: if (!silent)
2246: fprintf(stderr, "pcregrep: Failed to open %s: %s\n", pathname,
2247: strerror(errno));
2248: return 2;
2249: }
2250: handle = (void *)ingz;
2251: frtype = FR_LIBZ;
2252: }
2253: else
2254: #endif
2255:
2256: /* Otherwise open with bz2lib if it is supported and the name ends with .bz2. */
2257:
2258: #ifdef SUPPORT_LIBBZ2
2259: if (pathlen > 4 && strcmp(pathname + pathlen - 4, ".bz2") == 0)
2260: {
2261: inbz2 = BZ2_bzopen(pathname, "rb");
2262: handle = (void *)inbz2;
2263: frtype = FR_LIBBZ2;
2264: }
2265: else
2266: #endif
2267:
2268: /* Otherwise use plain fopen(). The label is so that we can come back here if
2269: an attempt to read a .bz2 file indicates that it really is a plain file. */
2270:
2271: #ifdef SUPPORT_LIBBZ2
2272: PLAIN_FILE:
2273: #endif
2274: {
2275: in = fopen(pathname, "rb");
2276: handle = (void *)in;
2277: frtype = FR_PLAIN;
2278: }
2279:
2280: /* All the opening methods return errno when they fail. */
2281:
2282: if (handle == NULL)
2283: {
2284: if (!silent)
2285: fprintf(stderr, "pcregrep: Failed to open %s: %s\n", pathname,
2286: strerror(errno));
2287: return 2;
2288: }
2289:
2290: /* Now grep the file */
2291:
2292: rc = pcregrep(handle, frtype, pathname, (filenames > FN_DEFAULT ||
2293: (filenames == FN_DEFAULT && !only_one_at_top))? pathname : NULL);
2294:
2295: /* Close in an appropriate manner. */
2296:
2297: #ifdef SUPPORT_LIBZ
2298: if (frtype == FR_LIBZ)
2299: gzclose(ingz);
2300: else
2301: #endif
2302:
2303: /* If it is a .bz2 file and the result is 3, it means that the first attempt to
2304: read failed. If the error indicates that the file isn't in fact bzipped, try
2305: again as a normal file. */
2306:
2307: #ifdef SUPPORT_LIBBZ2
2308: if (frtype == FR_LIBBZ2)
2309: {
2310: if (rc == 3)
2311: {
2312: int errnum;
2313: const char *err = BZ2_bzerror(inbz2, &errnum);
2314: if (errnum == BZ_DATA_ERROR_MAGIC)
2315: {
2316: BZ2_bzclose(inbz2);
2317: goto PLAIN_FILE;
2318: }
2319: else if (!silent)
2320: fprintf(stderr, "pcregrep: Failed to read %s using bzlib: %s\n",
2321: pathname, err);
2322: rc = 2; /* The normal "something went wrong" code */
2323: }
2324: BZ2_bzclose(inbz2);
2325: }
2326: else
2327: #endif
2328:
2329: /* Normal file close */
2330:
2331: fclose(in);
2332:
2333: /* Pass back the yield from pcregrep(). */
2334:
2335: return rc;
2336: }
2337:
2338:
2339:
2340: /*************************************************
2341: * Handle a single-letter, no data option *
2342: *************************************************/
2343:
2344: static int
2345: handle_option(int letter, int options)
2346: {
2347: switch(letter)
2348: {
2349: case N_FOFFSETS: file_offsets = TRUE; break;
2350: case N_HELP: help(); pcregrep_exit(0);
2351: case N_LBUFFER: line_buffered = TRUE; break;
2352: case N_LOFFSETS: line_offsets = number = TRUE; break;
2353: case N_NOJIT: study_options &= ~PCRE_STUDY_JIT_COMPILE; break;
1.1.1.3 misho 2354: case 'a': binary_files = BIN_TEXT; break;
1.1 misho 2355: case 'c': count_only = TRUE; break;
2356: case 'F': process_options |= PO_FIXED_STRINGS; break;
2357: case 'H': filenames = FN_FORCE; break;
1.1.1.3 misho 2358: case 'I': binary_files = BIN_NOMATCH; break;
1.1 misho 2359: case 'h': filenames = FN_NONE; break;
2360: case 'i': options |= PCRE_CASELESS; break;
2361: case 'l': omit_zero_count = TRUE; filenames = FN_MATCH_ONLY; break;
2362: case 'L': filenames = FN_NOMATCH_ONLY; break;
2363: case 'M': multiline = TRUE; options |= PCRE_MULTILINE|PCRE_FIRSTLINE; break;
2364: case 'n': number = TRUE; break;
1.1.1.4 misho 2365:
2366: case 'o':
2367: only_matching_last = add_number(0, only_matching_last);
2368: if (only_matching == NULL) only_matching = only_matching_last;
2369: break;
2370:
1.1 misho 2371: case 'q': quiet = TRUE; break;
2372: case 'r': dee_action = dee_RECURSE; break;
2373: case 's': silent = TRUE; break;
2374: case 'u': options |= PCRE_UTF8; utf8 = TRUE; break;
2375: case 'v': invert = TRUE; break;
2376: case 'w': process_options |= PO_WORD_MATCH; break;
2377: case 'x': process_options |= PO_LINE_MATCH; break;
2378:
2379: case 'V':
1.1.1.4 misho 2380: fprintf(stdout, "pcregrep version %s\n", pcre_version());
1.1 misho 2381: pcregrep_exit(0);
2382: break;
2383:
2384: default:
2385: fprintf(stderr, "pcregrep: Unknown option -%c\n", letter);
2386: pcregrep_exit(usage(2));
2387: }
2388:
2389: return options;
2390: }
2391:
2392:
2393:
2394:
2395: /*************************************************
2396: * Construct printed ordinal *
2397: *************************************************/
2398:
2399: /* This turns a number into "1st", "3rd", etc. */
2400:
2401: static char *
2402: ordin(int n)
2403: {
2404: static char buffer[8];
2405: char *p = buffer;
2406: sprintf(p, "%d", n);
2407: while (*p != 0) p++;
2408: switch (n%10)
2409: {
2410: case 1: strcpy(p, "st"); break;
2411: case 2: strcpy(p, "nd"); break;
2412: case 3: strcpy(p, "rd"); break;
2413: default: strcpy(p, "th"); break;
2414: }
2415: return buffer;
2416: }
2417:
2418:
2419:
2420: /*************************************************
2421: * Compile a single pattern *
2422: *************************************************/
2423:
1.1.1.4 misho 2424: /* Do nothing if the pattern has already been compiled. This is the case for
2425: include/exclude patterns read from a file.
2426:
2427: When the -F option has been used, each "pattern" may be a list of strings,
2428: separated by line breaks. They will be matched literally. We split such a
2429: string and compile the first substring, inserting an additional block into the
2430: pattern chain.
1.1 misho 2431:
2432: Arguments:
1.1.1.4 misho 2433: p points to the pattern block
1.1 misho 2434: options the PCRE options
1.1.1.4 misho 2435: popts the processing options
2436: fromfile TRUE if the pattern was read from a file
2437: fromtext file name or identifying text (e.g. "include")
1.1 misho 2438: count 0 if this is the only command line pattern, or
2439: number of the command line pattern, or
2440: linenumber for a pattern from a file
2441:
2442: Returns: TRUE on success, FALSE after an error
2443: */
2444:
2445: static BOOL
1.1.1.4 misho 2446: compile_pattern(patstr *p, int options, int popts, int fromfile,
2447: const char *fromtext, int count)
1.1 misho 2448: {
2449: char buffer[PATBUFSIZE];
2450: const char *error;
1.1.1.4 misho 2451: char *ps = p->string;
2452: int patlen = strlen(ps);
1.1 misho 2453: int errptr;
2454:
1.1.1.4 misho 2455: if (p->compiled != NULL) return TRUE;
1.1 misho 2456:
1.1.1.4 misho 2457: if ((popts & PO_FIXED_STRINGS) != 0)
1.1 misho 2458: {
1.1.1.4 misho 2459: int ellength;
2460: char *eop = ps + patlen;
2461: char *pe = end_of_line(ps, eop, &ellength);
2462:
2463: if (ellength != 0)
2464: {
2465: if (add_pattern(pe, p) == NULL) return FALSE;
2466: patlen = (int)(pe - ps - ellength);
2467: }
1.1 misho 2468: }
2469:
1.1.1.4 misho 2470: sprintf(buffer, "%s%.*s%s", prefix[popts], patlen, ps, suffix[popts]);
2471: p->compiled = pcre_compile(buffer, options, &error, &errptr, pcretables);
2472: if (p->compiled != NULL) return TRUE;
2473:
1.1 misho 2474: /* Handle compile errors */
2475:
1.1.1.4 misho 2476: errptr -= (int)strlen(prefix[popts]);
2477: if (errptr > patlen) errptr = patlen;
1.1 misho 2478:
1.1.1.4 misho 2479: if (fromfile)
1.1 misho 2480: {
1.1.1.4 misho 2481: fprintf(stderr, "pcregrep: Error in regex in line %d of %s "
2482: "at offset %d: %s\n", count, fromtext, errptr, error);
1.1 misho 2483: }
2484: else
2485: {
1.1.1.4 misho 2486: if (count == 0)
2487: fprintf(stderr, "pcregrep: Error in %s regex at offset %d: %s\n",
2488: fromtext, errptr, error);
2489: else
2490: fprintf(stderr, "pcregrep: Error in %s %s regex at offset %d: %s\n",
2491: ordin(count), fromtext, errptr, error);
1.1 misho 2492: }
2493:
2494: return FALSE;
2495: }
2496:
2497:
2498:
2499: /*************************************************
1.1.1.4 misho 2500: * Read and compile a file of patterns *
1.1 misho 2501: *************************************************/
2502:
1.1.1.4 misho 2503: /* This is used for --filelist, --include-from, and --exclude-from.
1.1 misho 2504:
2505: Arguments:
1.1.1.4 misho 2506: name the name of the file; "-" is stdin
2507: patptr pointer to the pattern chain anchor
2508: patlastptr pointer to the last pattern pointer
2509: popts the process options to pass to pattern_compile()
1.1 misho 2510:
1.1.1.4 misho 2511: Returns: TRUE if all went well
1.1 misho 2512: */
2513:
2514: static BOOL
1.1.1.4 misho 2515: read_pattern_file(char *name, patstr **patptr, patstr **patlastptr, int popts)
1.1 misho 2516: {
1.1.1.4 misho 2517: int linenumber = 0;
2518: FILE *f;
2519: char *filename;
2520: char buffer[PATBUFSIZE];
2521:
2522: if (strcmp(name, "-") == 0)
1.1 misho 2523: {
1.1.1.4 misho 2524: f = stdin;
2525: filename = stdin_name;
2526: }
2527: else
2528: {
2529: f = fopen(name, "r");
2530: if (f == NULL)
2531: {
2532: fprintf(stderr, "pcregrep: Failed to open %s: %s\n", name, strerror(errno));
2533: return FALSE;
2534: }
2535: filename = name;
2536: }
2537:
2538: while (fgets(buffer, PATBUFSIZE, f) != NULL)
2539: {
2540: char *s = buffer + (int)strlen(buffer);
2541: while (s > buffer && isspace((unsigned char)(s[-1]))) s--;
2542: *s = 0;
2543: linenumber++;
2544: if (buffer[0] == 0) continue; /* Skip blank lines */
2545:
2546: /* Note: this call to add_pattern() puts a pointer to the local variable
2547: "buffer" into the pattern chain. However, that pointer is used only when
2548: compiling the pattern, which happens immediately below, so we flatten it
2549: afterwards, as a precaution against any later code trying to use it. */
2550:
2551: *patlastptr = add_pattern(buffer, *patlastptr);
2552: if (*patlastptr == NULL) return FALSE;
2553: if (*patptr == NULL) *patptr = *patlastptr;
2554:
2555: /* This loop is needed because compiling a "pattern" when -F is set may add
2556: on additional literal patterns if the original contains a newline. In the
2557: common case, it never will, because fgets() stops at a newline. However,
2558: the -N option can be used to give pcregrep a different newline setting. */
2559:
1.1 misho 2560: for(;;)
2561: {
1.1.1.4 misho 2562: if (!compile_pattern(*patlastptr, pcre_options, popts, TRUE, filename,
2563: linenumber))
1.1 misho 2564: return FALSE;
1.1.1.4 misho 2565: (*patlastptr)->string = NULL; /* Insurance */
2566: if ((*patlastptr)->next == NULL) break;
2567: *patlastptr = (*patlastptr)->next;
1.1 misho 2568: }
2569: }
1.1.1.4 misho 2570:
2571: if (f != stdin) fclose(f);
2572: return TRUE;
1.1 misho 2573: }
2574:
2575:
2576:
2577: /*************************************************
2578: * Main program *
2579: *************************************************/
2580:
2581: /* Returns 0 if something matched, 1 if nothing matched, 2 after an error. */
2582:
2583: int
2584: main(int argc, char **argv)
2585: {
2586: int i, j;
2587: int rc = 1;
2588: BOOL only_one_at_top;
1.1.1.4 misho 2589: patstr *cp;
2590: fnstr *fn;
1.1 misho 2591: const char *locale_from = "--locale";
2592: const char *error;
2593:
2594: #ifdef SUPPORT_PCREGREP_JIT
2595: pcre_jit_stack *jit_stack = NULL;
2596: #endif
2597:
2598: /* Set the default line ending value from the default in the PCRE library;
2599: "lf", "cr", "crlf", and "any" are supported. Anything else is treated as "lf".
2600: Note that the return values from pcre_config(), though derived from the ASCII
2601: codes, are the same in EBCDIC environments, so we must use the actual values
2602: rather than escapes such as as '\r'. */
2603:
2604: (void)pcre_config(PCRE_CONFIG_NEWLINE, &i);
2605: switch(i)
2606: {
2607: default: newline = (char *)"lf"; break;
2608: case 13: newline = (char *)"cr"; break;
2609: case (13 << 8) | 10: newline = (char *)"crlf"; break;
2610: case -1: newline = (char *)"any"; break;
2611: case -2: newline = (char *)"anycrlf"; break;
2612: }
2613:
2614: /* Process the options */
2615:
2616: for (i = 1; i < argc; i++)
2617: {
2618: option_item *op = NULL;
2619: char *option_data = (char *)""; /* default to keep compiler happy */
2620: BOOL longop;
2621: BOOL longopwasequals = FALSE;
2622:
2623: if (argv[i][0] != '-') break;
2624:
2625: /* If we hit an argument that is just "-", it may be a reference to STDIN,
2626: but only if we have previously had -e or -f to define the patterns. */
2627:
2628: if (argv[i][1] == 0)
2629: {
1.1.1.4 misho 2630: if (pattern_files != NULL || patterns != NULL) break;
1.1 misho 2631: else pcregrep_exit(usage(2));
2632: }
2633:
2634: /* Handle a long name option, or -- to terminate the options */
2635:
2636: if (argv[i][1] == '-')
2637: {
2638: char *arg = argv[i] + 2;
2639: char *argequals = strchr(arg, '=');
2640:
2641: if (*arg == 0) /* -- terminates options */
2642: {
2643: i++;
2644: break; /* out of the options-handling loop */
2645: }
2646:
2647: longop = TRUE;
2648:
2649: /* Some long options have data that follows after =, for example file=name.
2650: Some options have variations in the long name spelling: specifically, we
2651: allow "regexp" because GNU grep allows it, though I personally go along
2652: with Jeffrey Friedl and Larry Wall in preferring "regex" without the "p".
2653: These options are entered in the table as "regex(p)". Options can be in
2654: both these categories. */
2655:
2656: for (op = optionlist; op->one_char != 0; op++)
2657: {
2658: char *opbra = strchr(op->long_name, '(');
2659: char *equals = strchr(op->long_name, '=');
2660:
2661: /* Handle options with only one spelling of the name */
2662:
2663: if (opbra == NULL) /* Does not contain '(' */
2664: {
2665: if (equals == NULL) /* Not thing=data case */
2666: {
2667: if (strcmp(arg, op->long_name) == 0) break;
2668: }
2669: else /* Special case xxx=data */
2670: {
2671: int oplen = (int)(equals - op->long_name);
2672: int arglen = (argequals == NULL)?
2673: (int)strlen(arg) : (int)(argequals - arg);
2674: if (oplen == arglen && strncmp(arg, op->long_name, oplen) == 0)
2675: {
2676: option_data = arg + arglen;
2677: if (*option_data == '=')
2678: {
2679: option_data++;
2680: longopwasequals = TRUE;
2681: }
2682: break;
2683: }
2684: }
2685: }
2686:
2687: /* Handle options with an alternate spelling of the name */
2688:
2689: else
2690: {
2691: char buff1[24];
2692: char buff2[24];
2693:
2694: int baselen = (int)(opbra - op->long_name);
2695: int fulllen = (int)(strchr(op->long_name, ')') - op->long_name + 1);
2696: int arglen = (argequals == NULL || equals == NULL)?
2697: (int)strlen(arg) : (int)(argequals - arg);
2698:
2699: sprintf(buff1, "%.*s", baselen, op->long_name);
2700: sprintf(buff2, "%s%.*s", buff1, fulllen - baselen - 2, opbra + 1);
2701:
2702: if (strncmp(arg, buff1, arglen) == 0 ||
2703: strncmp(arg, buff2, arglen) == 0)
2704: {
2705: if (equals != NULL && argequals != NULL)
2706: {
2707: option_data = argequals;
2708: if (*option_data == '=')
2709: {
2710: option_data++;
2711: longopwasequals = TRUE;
2712: }
2713: }
2714: break;
2715: }
2716: }
2717: }
2718:
2719: if (op->one_char == 0)
2720: {
2721: fprintf(stderr, "pcregrep: Unknown option %s\n", argv[i]);
2722: pcregrep_exit(usage(2));
2723: }
2724: }
2725:
2726: /* Jeffrey Friedl's debugging harness uses these additional options which
2727: are not in the right form for putting in the option table because they use
2728: only one hyphen, yet are more than one character long. By putting them
2729: separately here, they will not get displayed as part of the help() output,
2730: but I don't think Jeffrey will care about that. */
2731:
2732: #ifdef JFRIEDL_DEBUG
2733: else if (strcmp(argv[i], "-pre") == 0) {
2734: jfriedl_prefix = argv[++i];
2735: continue;
2736: } else if (strcmp(argv[i], "-post") == 0) {
2737: jfriedl_postfix = argv[++i];
2738: continue;
2739: } else if (strcmp(argv[i], "-XT") == 0) {
2740: sscanf(argv[++i], "%d", &jfriedl_XT);
2741: continue;
2742: } else if (strcmp(argv[i], "-XR") == 0) {
2743: sscanf(argv[++i], "%d", &jfriedl_XR);
2744: continue;
2745: }
2746: #endif
2747:
2748:
2749: /* One-char options; many that have no data may be in a single argument; we
2750: continue till we hit the last one or one that needs data. */
2751:
2752: else
2753: {
2754: char *s = argv[i] + 1;
2755: longop = FALSE;
1.1.1.4 misho 2756:
1.1 misho 2757: while (*s != 0)
2758: {
2759: for (op = optionlist; op->one_char != 0; op++)
2760: {
2761: if (*s == op->one_char) break;
2762: }
2763: if (op->one_char == 0)
2764: {
2765: fprintf(stderr, "pcregrep: Unknown option letter '%c' in \"%s\"\n",
2766: *s, argv[i]);
2767: pcregrep_exit(usage(2));
2768: }
2769:
1.1.1.4 misho 2770: option_data = s+1;
2771:
2772: /* Break out if this is the last character in the string; it's handled
2773: below like a single multi-char option. */
2774:
2775: if (*option_data == 0) break;
2776:
2777: /* Check for a single-character option that has data: OP_OP_NUMBER(S)
2778: are used for ones that either have a numerical number or defaults, i.e.
2779: the data is optional. If a digit follows, there is data; if not, carry on
1.1 misho 2780: with other single-character options in the same string. */
2781:
1.1.1.4 misho 2782: if (op->type == OP_OP_NUMBER || op->type == OP_OP_NUMBERS)
1.1 misho 2783: {
2784: if (isdigit((unsigned char)s[1])) break;
2785: }
1.1.1.4 misho 2786: else /* Check for an option with data */
1.1 misho 2787: {
1.1.1.4 misho 2788: if (op->type != OP_NODATA) break;
1.1 misho 2789: }
2790:
2791: /* Handle a single-character option with no data, then loop for the
2792: next character in the string. */
2793:
2794: pcre_options = handle_option(*s++, pcre_options);
2795: }
2796: }
2797:
2798: /* At this point we should have op pointing to a matched option. If the type
2799: is NO_DATA, it means that there is no data, and the option might set
2800: something in the PCRE options. */
2801:
2802: if (op->type == OP_NODATA)
2803: {
2804: pcre_options = handle_option(op->one_char, pcre_options);
2805: continue;
2806: }
2807:
1.1.1.4 misho 2808: /* If the option type is OP_OP_STRING or OP_OP_NUMBER(S), it's an option that
1.1 misho 2809: either has a value or defaults to something. It cannot have data in a
2810: separate item. At the moment, the only such options are "colo(u)r",
2811: "only-matching", and Jeffrey Friedl's special -S debugging option. */
2812:
2813: if (*option_data == 0 &&
1.1.1.4 misho 2814: (op->type == OP_OP_STRING || op->type == OP_OP_NUMBER ||
2815: op->type == OP_OP_NUMBERS))
1.1 misho 2816: {
2817: switch (op->one_char)
2818: {
2819: case N_COLOUR:
2820: colour_option = (char *)"auto";
2821: break;
2822:
2823: case 'o':
1.1.1.4 misho 2824: only_matching_last = add_number(0, only_matching_last);
2825: if (only_matching == NULL) only_matching = only_matching_last;
1.1 misho 2826: break;
2827:
2828: #ifdef JFRIEDL_DEBUG
2829: case 'S':
2830: S_arg = 0;
2831: break;
2832: #endif
2833: }
2834: continue;
2835: }
2836:
2837: /* Otherwise, find the data string for the option. */
2838:
2839: if (*option_data == 0)
2840: {
2841: if (i >= argc - 1 || longopwasequals)
2842: {
2843: fprintf(stderr, "pcregrep: Data missing after %s\n", argv[i]);
2844: pcregrep_exit(usage(2));
2845: }
2846: option_data = argv[++i];
2847: }
2848:
1.1.1.4 misho 2849: /* If the option type is OP_OP_NUMBERS, the value is a number that is to be
2850: added to a chain of numbers. */
2851:
2852: if (op->type == OP_OP_NUMBERS)
2853: {
2854: unsigned long int n = decode_number(option_data, op, longop);
2855: omdatastr *omd = (omdatastr *)op->dataptr;
2856: *(omd->lastptr) = add_number((int)n, *(omd->lastptr));
2857: if (*(omd->anchor) == NULL) *(omd->anchor) = *(omd->lastptr);
2858: }
2859:
2860: /* If the option type is OP_PATLIST, it's the -e option, or one of the
2861: include/exclude options, which can be called multiple times to create lists
2862: of patterns. */
1.1 misho 2863:
1.1.1.4 misho 2864: else if (op->type == OP_PATLIST)
1.1 misho 2865: {
1.1.1.4 misho 2866: patdatastr *pd = (patdatastr *)op->dataptr;
2867: *(pd->lastptr) = add_pattern(option_data, *(pd->lastptr));
2868: if (*(pd->lastptr) == NULL) goto EXIT2;
2869: if (*(pd->anchor) == NULL) *(pd->anchor) = *(pd->lastptr);
2870: }
2871:
2872: /* If the option type is OP_FILELIST, it's one of the options that names a
2873: file. */
2874:
2875: else if (op->type == OP_FILELIST)
2876: {
2877: fndatastr *fd = (fndatastr *)op->dataptr;
2878: fn = (fnstr *)malloc(sizeof(fnstr));
2879: if (fn == NULL)
1.1 misho 2880: {
1.1.1.4 misho 2881: fprintf(stderr, "pcregrep: malloc failed\n");
2882: goto EXIT2;
1.1 misho 2883: }
1.1.1.4 misho 2884: fn->next = NULL;
2885: fn->name = option_data;
2886: if (*(fd->anchor) == NULL)
2887: *(fd->anchor) = fn;
2888: else
2889: (*(fd->lastptr))->next = fn;
2890: *(fd->lastptr) = fn;
1.1 misho 2891: }
2892:
1.1.1.3 misho 2893: /* Handle OP_BINARY_FILES */
2894:
2895: else if (op->type == OP_BINFILES)
2896: {
2897: if (strcmp(option_data, "binary") == 0)
2898: binary_files = BIN_BINARY;
2899: else if (strcmp(option_data, "without-match") == 0)
2900: binary_files = BIN_NOMATCH;
2901: else if (strcmp(option_data, "text") == 0)
2902: binary_files = BIN_TEXT;
2903: else
2904: {
2905: fprintf(stderr, "pcregrep: unknown value \"%s\" for binary-files\n",
2906: option_data);
2907: pcregrep_exit(usage(2));
2908: }
2909: }
2910:
1.1.1.4 misho 2911: /* Otherwise, deal with a single string or numeric data value. */
1.1 misho 2912:
2913: else if (op->type != OP_NUMBER && op->type != OP_LONGNUMBER &&
2914: op->type != OP_OP_NUMBER)
2915: {
2916: *((char **)op->dataptr) = option_data;
2917: }
2918: else
2919: {
1.1.1.4 misho 2920: unsigned long int n = decode_number(option_data, op, longop);
2921: if (op->type == OP_LONGNUMBER) *((unsigned long int *)op->dataptr) = n;
2922: else *((int *)op->dataptr) = n;
1.1 misho 2923: }
2924: }
2925:
2926: /* Options have been decoded. If -C was used, its value is used as a default
2927: for -A and -B. */
2928:
2929: if (both_context > 0)
2930: {
2931: if (after_context == 0) after_context = both_context;
2932: if (before_context == 0) before_context = both_context;
2933: }
2934:
2935: /* Only one of --only-matching, --file-offsets, or --line-offsets is permitted.
1.1.1.4 misho 2936: However, all three set show_only_matching because they display, each in their
2937: own way, only the data that has matched. */
1.1 misho 2938:
1.1.1.4 misho 2939: if ((only_matching != NULL && (file_offsets || line_offsets)) ||
1.1 misho 2940: (file_offsets && line_offsets))
2941: {
2942: fprintf(stderr, "pcregrep: Cannot mix --only-matching, --file-offsets "
2943: "and/or --line-offsets\n");
2944: pcregrep_exit(usage(2));
2945: }
2946:
1.1.1.4 misho 2947: if (only_matching != NULL || file_offsets || line_offsets)
2948: show_only_matching = TRUE;
1.1 misho 2949:
2950: /* If a locale has not been provided as an option, see if the LC_CTYPE or
2951: LC_ALL environment variable is set, and if so, use it. */
2952:
2953: if (locale == NULL)
2954: {
2955: locale = getenv("LC_ALL");
2956: locale_from = "LCC_ALL";
2957: }
2958:
2959: if (locale == NULL)
2960: {
2961: locale = getenv("LC_CTYPE");
2962: locale_from = "LC_CTYPE";
2963: }
2964:
2965: /* If a locale has been provided, set it, and generate the tables the PCRE
2966: needs. Otherwise, pcretables==NULL, which causes the use of default tables. */
2967:
2968: if (locale != NULL)
2969: {
2970: if (setlocale(LC_CTYPE, locale) == NULL)
2971: {
2972: fprintf(stderr, "pcregrep: Failed to set locale %s (obtained from %s)\n",
2973: locale, locale_from);
2974: return 2;
2975: }
2976: pcretables = pcre_maketables();
2977: }
2978:
2979: /* Sort out colouring */
2980:
2981: if (colour_option != NULL && strcmp(colour_option, "never") != 0)
2982: {
2983: if (strcmp(colour_option, "always") == 0) do_colour = TRUE;
2984: else if (strcmp(colour_option, "auto") == 0) do_colour = is_stdout_tty();
2985: else
2986: {
2987: fprintf(stderr, "pcregrep: Unknown colour setting \"%s\"\n",
2988: colour_option);
2989: return 2;
2990: }
2991: if (do_colour)
2992: {
2993: char *cs = getenv("PCREGREP_COLOUR");
2994: if (cs == NULL) cs = getenv("PCREGREP_COLOR");
2995: if (cs != NULL) colour_string = cs;
2996: }
2997: }
2998:
2999: /* Interpret the newline type; the default settings are Unix-like. */
3000:
3001: if (strcmp(newline, "cr") == 0 || strcmp(newline, "CR") == 0)
3002: {
3003: pcre_options |= PCRE_NEWLINE_CR;
3004: endlinetype = EL_CR;
3005: }
3006: else if (strcmp(newline, "lf") == 0 || strcmp(newline, "LF") == 0)
3007: {
3008: pcre_options |= PCRE_NEWLINE_LF;
3009: endlinetype = EL_LF;
3010: }
3011: else if (strcmp(newline, "crlf") == 0 || strcmp(newline, "CRLF") == 0)
3012: {
3013: pcre_options |= PCRE_NEWLINE_CRLF;
3014: endlinetype = EL_CRLF;
3015: }
3016: else if (strcmp(newline, "any") == 0 || strcmp(newline, "ANY") == 0)
3017: {
3018: pcre_options |= PCRE_NEWLINE_ANY;
3019: endlinetype = EL_ANY;
3020: }
3021: else if (strcmp(newline, "anycrlf") == 0 || strcmp(newline, "ANYCRLF") == 0)
3022: {
3023: pcre_options |= PCRE_NEWLINE_ANYCRLF;
3024: endlinetype = EL_ANYCRLF;
3025: }
3026: else
3027: {
3028: fprintf(stderr, "pcregrep: Invalid newline specifier \"%s\"\n", newline);
3029: return 2;
3030: }
3031:
3032: /* Interpret the text values for -d and -D */
3033:
3034: if (dee_option != NULL)
3035: {
3036: if (strcmp(dee_option, "read") == 0) dee_action = dee_READ;
3037: else if (strcmp(dee_option, "recurse") == 0) dee_action = dee_RECURSE;
3038: else if (strcmp(dee_option, "skip") == 0) dee_action = dee_SKIP;
3039: else
3040: {
3041: fprintf(stderr, "pcregrep: Invalid value \"%s\" for -d\n", dee_option);
3042: return 2;
3043: }
3044: }
3045:
3046: if (DEE_option != NULL)
3047: {
3048: if (strcmp(DEE_option, "read") == 0) DEE_action = DEE_READ;
3049: else if (strcmp(DEE_option, "skip") == 0) DEE_action = DEE_SKIP;
3050: else
3051: {
3052: fprintf(stderr, "pcregrep: Invalid value \"%s\" for -D\n", DEE_option);
3053: return 2;
3054: }
3055: }
3056:
3057: /* Check the values for Jeffrey Friedl's debugging options. */
3058:
3059: #ifdef JFRIEDL_DEBUG
3060: if (S_arg > 9)
3061: {
3062: fprintf(stderr, "pcregrep: bad value for -S option\n");
3063: return 2;
3064: }
3065: if (jfriedl_XT != 0 || jfriedl_XR != 0)
3066: {
3067: if (jfriedl_XT == 0) jfriedl_XT = 1;
3068: if (jfriedl_XR == 0) jfriedl_XR = 1;
3069: }
3070: #endif
3071:
1.1.1.4 misho 3072: /* Get memory for the main buffer. */
1.1 misho 3073:
3074: bufsize = 3*bufthird;
3075: main_buffer = (char *)malloc(bufsize);
3076:
1.1.1.4 misho 3077: if (main_buffer == NULL)
1.1 misho 3078: {
3079: fprintf(stderr, "pcregrep: malloc failed\n");
3080: goto EXIT2;
3081: }
3082:
1.1.1.4 misho 3083: /* If no patterns were provided by -e, and there are no files provided by -f,
1.1 misho 3084: the first argument is the one and only pattern, and it must exist. */
3085:
1.1.1.4 misho 3086: if (patterns == NULL && pattern_files == NULL)
1.1 misho 3087: {
3088: if (i >= argc) return usage(2);
1.1.1.4 misho 3089: patterns = patterns_last = add_pattern(argv[i++], NULL);
3090: if (patterns == NULL) goto EXIT2;
1.1 misho 3091: }
3092:
3093: /* Compile the patterns that were provided on the command line, either by
1.1.1.4 misho 3094: multiple uses of -e or as a single unkeyed pattern. We cannot do this until
3095: after all the command-line options are read so that we know which PCRE options
3096: to use. When -F is used, compile_pattern() may add another block into the
3097: chain, so we must not access the next pointer till after the compile. */
1.1 misho 3098:
1.1.1.4 misho 3099: for (j = 1, cp = patterns; cp != NULL; j++, cp = cp->next)
1.1 misho 3100: {
1.1.1.4 misho 3101: if (!compile_pattern(cp, pcre_options, process_options, FALSE, "command-line",
3102: (j == 1 && patterns->next == NULL)? 0 : j))
1.1 misho 3103: goto EXIT2;
3104: }
3105:
1.1.1.4 misho 3106: /* Read and compile the regular expressions that are provided in files. */
1.1 misho 3107:
1.1.1.4 misho 3108: for (fn = pattern_files; fn != NULL; fn = fn->next)
1.1 misho 3109: {
1.1.1.4 misho 3110: if (!read_pattern_file(fn->name, &patterns, &patterns_last, process_options))
3111: goto EXIT2;
3112: }
1.1 misho 3113:
1.1.1.4 misho 3114: /* Study the regular expressions, as we will be running them many times. If an
3115: extra block is needed for a limit, set PCRE_STUDY_EXTRA_NEEDED so that one is
3116: returned, even if studying produces no data. */
1.1 misho 3117:
1.1.1.4 misho 3118: if (match_limit > 0 || match_limit_recursion > 0)
3119: study_options |= PCRE_STUDY_EXTRA_NEEDED;
1.1 misho 3120:
1.1.1.4 misho 3121: /* Unless JIT has been explicitly disabled, arrange a stack for it to use. */
1.1 misho 3122:
3123: #ifdef SUPPORT_PCREGREP_JIT
3124: if ((study_options & PCRE_STUDY_JIT_COMPILE) != 0)
3125: jit_stack = pcre_jit_stack_alloc(32*1024, 1024*1024);
3126: #endif
3127:
1.1.1.4 misho 3128: for (j = 1, cp = patterns; cp != NULL; j++, cp = cp->next)
1.1 misho 3129: {
1.1.1.4 misho 3130: cp->hint = pcre_study(cp->compiled, study_options, &error);
1.1 misho 3131: if (error != NULL)
3132: {
3133: char s[16];
1.1.1.4 misho 3134: if (patterns->next == NULL) s[0] = 0; else sprintf(s, " number %d", j);
1.1 misho 3135: fprintf(stderr, "pcregrep: Error while studying regex%s: %s\n", s, error);
3136: goto EXIT2;
3137: }
3138: #ifdef SUPPORT_PCREGREP_JIT
1.1.1.4 misho 3139: if (jit_stack != NULL && cp->hint != NULL)
3140: pcre_assign_jit_stack(cp->hint, NULL, jit_stack);
1.1 misho 3141: #endif
3142: }
3143:
3144: /* If --match-limit or --recursion-limit was set, put the value(s) into the
1.1.1.4 misho 3145: pcre_extra block for each pattern. There will always be an extra block because
3146: of the use of PCRE_STUDY_EXTRA_NEEDED above. */
1.1 misho 3147:
1.1.1.4 misho 3148: for (cp = patterns; cp != NULL; cp = cp->next)
1.1 misho 3149: {
1.1.1.4 misho 3150: if (match_limit > 0)
1.1 misho 3151: {
1.1.1.4 misho 3152: cp->hint->flags |= PCRE_EXTRA_MATCH_LIMIT;
3153: cp->hint->match_limit = match_limit;
3154: }
3155:
3156: if (match_limit_recursion > 0)
3157: {
3158: cp->hint->flags |= PCRE_EXTRA_MATCH_LIMIT_RECURSION;
3159: cp->hint->match_limit_recursion = match_limit_recursion;
1.1 misho 3160: }
3161: }
3162:
1.1.1.4 misho 3163: /* If there are include or exclude patterns read from the command line, compile
3164: them. -F, -w, and -x do not apply, so the third argument of compile_pattern is
3165: 0. */
1.1 misho 3166:
1.1.1.4 misho 3167: for (j = 0; j < 4; j++)
1.1 misho 3168: {
1.1.1.4 misho 3169: int k;
3170: for (k = 1, cp = *(incexlist[j]); cp != NULL; k++, cp = cp->next)
1.1 misho 3171: {
1.1.1.4 misho 3172: if (!compile_pattern(cp, pcre_options, 0, FALSE, incexname[j],
3173: (k == 1 && cp->next == NULL)? 0 : k))
3174: goto EXIT2;
1.1 misho 3175: }
3176: }
3177:
1.1.1.4 misho 3178: /* Read and compile include/exclude patterns from files. */
3179:
3180: for (fn = include_from; fn != NULL; fn = fn->next)
1.1 misho 3181: {
1.1.1.4 misho 3182: if (!read_pattern_file(fn->name, &include_patterns, &include_patterns_last, 0))
1.1 misho 3183: goto EXIT2;
3184: }
3185:
1.1.1.4 misho 3186: for (fn = exclude_from; fn != NULL; fn = fn->next)
1.1 misho 3187: {
1.1.1.4 misho 3188: if (!read_pattern_file(fn->name, &exclude_patterns, &exclude_patterns_last, 0))
1.1 misho 3189: goto EXIT2;
3190: }
3191:
1.1.1.4 misho 3192: /* If there are no files that contain lists of files to search, and there are
3193: no file arguments, search stdin, and then exit. */
3194:
3195: if (file_lists == NULL && i >= argc)
1.1 misho 3196: {
1.1.1.4 misho 3197: rc = pcregrep(stdin, FR_PLAIN, stdin_name,
3198: (filenames > FN_DEFAULT)? stdin_name : NULL);
3199: goto EXIT;
1.1 misho 3200: }
3201:
1.1.1.4 misho 3202: /* If any files that contains a list of files to search have been specified,
3203: read them line by line and search the given files. */
1.1.1.3 misho 3204:
1.1.1.4 misho 3205: for (fn = file_lists; fn != NULL; fn = fn->next)
1.1.1.3 misho 3206: {
3207: char buffer[PATBUFSIZE];
3208: FILE *fl;
1.1.1.4 misho 3209: if (strcmp(fn->name, "-") == 0) fl = stdin; else
1.1.1.3 misho 3210: {
1.1.1.4 misho 3211: fl = fopen(fn->name, "rb");
1.1.1.3 misho 3212: if (fl == NULL)
3213: {
1.1.1.4 misho 3214: fprintf(stderr, "pcregrep: Failed to open %s: %s\n", fn->name,
1.1.1.3 misho 3215: strerror(errno));
3216: goto EXIT2;
3217: }
3218: }
3219: while (fgets(buffer, PATBUFSIZE, fl) != NULL)
3220: {
3221: int frc;
3222: char *end = buffer + (int)strlen(buffer);
3223: while (end > buffer && isspace(end[-1])) end--;
3224: *end = 0;
3225: if (*buffer != 0)
3226: {
3227: frc = grep_or_recurse(buffer, dee_action == dee_RECURSE, FALSE);
3228: if (frc > 1) rc = frc;
3229: else if (frc == 0 && rc == 1) rc = 0;
3230: }
3231: }
1.1.1.4 misho 3232: if (fl != stdin) fclose(fl);
1.1.1.3 misho 3233: }
3234:
1.1.1.4 misho 3235: /* After handling file-list, work through remaining arguments. Pass in the fact
3236: that there is only one argument at top level - this suppresses the file name if
3237: the argument is not a directory and filenames are not otherwise forced. */
1.1 misho 3238:
1.1.1.4 misho 3239: only_one_at_top = i == argc - 1 && file_lists == NULL;
1.1 misho 3240:
3241: for (; i < argc; i++)
3242: {
3243: int frc = grep_or_recurse(argv[i], dee_action == dee_RECURSE,
3244: only_one_at_top);
3245: if (frc > 1) rc = frc;
3246: else if (frc == 0 && rc == 1) rc = 0;
3247: }
3248:
3249: EXIT:
3250: #ifdef SUPPORT_PCREGREP_JIT
3251: if (jit_stack != NULL) pcre_jit_stack_free(jit_stack);
3252: #endif
1.1.1.4 misho 3253:
1.1 misho 3254: if (main_buffer != NULL) free(main_buffer);
1.1.1.4 misho 3255:
3256: free_pattern_chain(patterns);
3257: free_pattern_chain(include_patterns);
3258: free_pattern_chain(include_dir_patterns);
3259: free_pattern_chain(exclude_patterns);
3260: free_pattern_chain(exclude_dir_patterns);
3261:
3262: free_file_chain(exclude_from);
3263: free_file_chain(include_from);
3264: free_file_chain(pattern_files);
3265: free_file_chain(file_lists);
3266:
3267: while (only_matching != NULL)
3268: {
3269: omstr *this = only_matching;
3270: only_matching = this->next;
3271: free(this);
1.1 misho 3272: }
1.1.1.4 misho 3273:
1.1 misho 3274: pcregrep_exit(rc);
3275:
3276: EXIT2:
3277: rc = 2;
3278: goto EXIT;
3279: }
3280:
3281: /* End of pcregrep */
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>