/[pcre]/code/branches/pcre16/pcretest.c
ViewVC logotype

Diff of /code/branches/pcre16/pcretest.c

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 75 by nigel, Sat Feb 24 21:40:37 2007 UTC revision 734 by ph10, Wed Oct 12 16:35:29 2011 UTC
# Line 4  Line 4 
4    
5  /* This program was hacked up as a tester for PCRE. I really should have  /* This program was hacked up as a tester for PCRE. I really should have
6  written it more tidily in the first place. Will I ever learn? It has grown and  written it more tidily in the first place. Will I ever learn? It has grown and
7  been extended and consequently is now rather untidy in places.  been extended and consequently is now rather, er, *very* untidy in places.
8    
9  -----------------------------------------------------------------------------  -----------------------------------------------------------------------------
10  Redistribution and use in source and binary forms, with or without  Redistribution and use in source and binary forms, with or without
# Line 36  POSSIBILITY OF SUCH DAMAGE. Line 36  POSSIBILITY OF SUCH DAMAGE.
36  */  */
37    
38    
39    #ifdef HAVE_CONFIG_H
40    #include "config.h"
41    #endif
42    
43  #include <ctype.h>  #include <ctype.h>
44  #include <stdio.h>  #include <stdio.h>
45  #include <string.h>  #include <string.h>
# Line 44  POSSIBILITY OF SUCH DAMAGE. Line 48  POSSIBILITY OF SUCH DAMAGE.
48  #include <locale.h>  #include <locale.h>
49  #include <errno.h>  #include <errno.h>
50    
51  /* We need the internal info for displaying the results of pcre_study(). Also  #ifdef SUPPORT_LIBREADLINE
52  for getting the opcodes for showing compiled code. */  #ifdef HAVE_UNISTD_H
53    #include <unistd.h>
54    #endif
55    #include <readline/readline.h>
56    #include <readline/history.h>
57    #endif
58    
59    
60    /* A number of things vary for Windows builds. Originally, pcretest opened its
61    input and output without "b"; then I was told that "b" was needed in some
62    environments, so it was added for release 5.0 to both the input and output. (It
63    makes no difference on Unix-like systems.) Later I was told that it is wrong
64    for the input on Windows. I've now abstracted the modes into two macros that
65    are set here, to make it easier to fiddle with them, and removed "b" from the
66    input mode under Windows. */
67    
68    #if defined(_WIN32) || defined(WIN32)
69    #include <io.h>                /* For _setmode() */
70    #include <fcntl.h>             /* For _O_BINARY */
71    #define INPUT_MODE   "r"
72    #define OUTPUT_MODE  "wb"
73    
74    #ifndef isatty
75    #define isatty _isatty         /* This is what Windows calls them, I'm told, */
76    #endif                         /* though in some environments they seem to   */
77                                   /* be already defined, hence the #ifndefs.    */
78    #ifndef fileno
79    #define fileno _fileno
80    #endif
81    
82    /* A user sent this fix for Borland Builder 5 under Windows. */
83    
84    #ifdef __BORLANDC__
85    #define _setmode(handle, mode) setmode(handle, mode)
86    #endif
87    
88    /* Not Windows */
89    
90    #else
91    #include <sys/time.h>          /* These two includes are needed */
92    #include <sys/resource.h>      /* for setrlimit(). */
93    #define INPUT_MODE   "rb"
94    #define OUTPUT_MODE  "wb"
95    #endif
96    
97    
98    /* We have to include pcre_internal.h because we need the internal info for
99    displaying the results of pcre_study() and we also need to know about the
100    internal macros, structures, and other internal data values; pcretest has
101    "inside information" compared to a program that strictly follows the PCRE API.
102    
103    Although pcre_internal.h does itself include pcre.h, we explicitly include it
104    here before pcre_internal.h so that the PCRE_EXP_xxx macros get set
105    appropriately for an application, not for building PCRE. */
106    
107    #include "pcre.h"
108    #include "pcre_internal.h"
109    
110    /* We need access to some of the data tables that PCRE uses. So as not to have
111    to keep two copies, we include the source file here, changing the names of the
112    external symbols to prevent clashes. */
113    
114    #define _pcre_ucp_gentype      ucp_gentype
115    #define _pcre_ucp_typerange    ucp_typerange
116    #define _pcre_utf8_table1      utf8_table1
117    #define _pcre_utf8_table1_size utf8_table1_size
118    #define _pcre_utf8_table2      utf8_table2
119    #define _pcre_utf8_table3      utf8_table3
120    #define _pcre_utf8_table4      utf8_table4
121    #define _pcre_utf8_char_sizes  utf8_char_sizes
122    #define _pcre_utt              utt
123    #define _pcre_utt_size         utt_size
124    #define _pcre_utt_names        utt_names
125    #define _pcre_OP_lengths       OP_lengths
126    
127    #include "pcre_tables.c"
128    
129    /* We also need the pcre_printint() function for printing out compiled
130    patterns. This function is in a separate file so that it can be included in
131    pcre_compile.c when that module is compiled with debugging enabled. It needs to
132    know which case is being compiled. */
133    
134    #define COMPILING_PCRETEST
135    #include "pcre_printint.src"
136    
137    /* The definition of the macro PRINTABLE, which determines whether to print an
138    output character as-is or as a hex value when showing compiled patterns, is
139    contained in the printint.src file. We uses it here also, in cases when the
140    locale has not been explicitly changed, so as to get consistent output from
141    systems that differ in their output from isprint() even in the "C" locale. */
142    
143  #define PCRE_SPY        /* For Win32 build, import data, not export */  #define PRINTHEX(c) (locale_set? isprint(c) : PRINTABLE(c))
 #include "internal.h"  
144    
145  /* It is possible to compile this test program without including support for  /* It is possible to compile this test program without including support for
146  testing the POSIX interface, though this is not available via the standard  testing the POSIX interface, though this is not available via the standard
# Line 58  Makefile. */ Line 150  Makefile. */
150  #include "pcreposix.h"  #include "pcreposix.h"
151  #endif  #endif
152    
153    /* It is also possible, for the benefit of the version currently imported into
154    Exim, to build pcretest without support for UTF8 (define NOUTF8), without the
155    interface to the DFA matcher (NODFA), and without the doublecheck of the old
156    "info" function (define NOINFOCHECK). In fact, we automatically cut out the
157    UTF8 support if PCRE is built without it. */
158    
159    #ifndef SUPPORT_UTF8
160    #ifndef NOUTF8
161    #define NOUTF8
162    #endif
163    #endif
164    
165    
166    /* Other parameters */
167    
168  #ifndef CLOCKS_PER_SEC  #ifndef CLOCKS_PER_SEC
169  #ifdef CLK_TCK  #ifdef CLK_TCK
170  #define CLOCKS_PER_SEC CLK_TCK  #define CLOCKS_PER_SEC CLK_TCK
# Line 66  Makefile. */ Line 173  Makefile. */
173  #endif  #endif
174  #endif  #endif
175    
176  #define LOOPREPEAT 500000  /* This is the default loop count for timing. */
177    
178  #define BUFFER_SIZE 30000  #define LOOPREPEAT 500000
 #define PBUFFER_SIZE BUFFER_SIZE  
 #define DBUFFER_SIZE BUFFER_SIZE  
179    
180    /* Static variables */
181    
182  static FILE *outfile;  static FILE *outfile;
183  static int log_store = 0;  static int log_store = 0;
# Line 79  static int callout_count; Line 185  static int callout_count;
185  static int callout_extra;  static int callout_extra;
186  static int callout_fail_count;  static int callout_fail_count;
187  static int callout_fail_id;  static int callout_fail_id;
188    static int debug_lengths;
189  static int first_callout;  static int first_callout;
190    static int locale_set = 0;
191  static int show_malloc;  static int show_malloc;
192  static int use_utf8;  static int use_utf8;
193  static size_t gotten_store;  static size_t gotten_store;
194    static const unsigned char *last_callout_mark = NULL;
195    
196    /* The buffers grow automatically if very long input lines are encountered. */
197    
198    static int buffer_size = 50000;
199    static uschar *buffer = NULL;
200    static uschar *dbuffer = NULL;
201  static uschar *pbuffer = NULL;  static uschar *pbuffer = NULL;
202    
203    /* Textual explanations for runtime error codes */
204    
205    static const char *errtexts[] = {
206      NULL,  /* 0 is no error */
207      NULL,  /* NOMATCH is handled specially */
208      "NULL argument passed",
209      "bad option value",
210      "magic number missing",
211      "unknown opcode - pattern overwritten?",
212      "no more memory",
213      NULL,  /* never returned by pcre_exec() or pcre_dfa_exec() */
214      "match limit exceeded",
215      "callout error code",
216      NULL,  /* BADUTF8 is handled specially */
217      "bad UTF-8 offset",
218      NULL,  /* PARTIAL is handled specially */
219      "not used - internal error",
220      "internal error - pattern overwritten?",
221      "bad count value",
222      "item unsupported for DFA matching",
223      "backreference condition or recursion test not supported for DFA matching",
224      "match limit not supported for DFA matching",
225      "workspace size exceeded in DFA matching",
226      "too much recursion for DFA matching",
227      "recursion limit exceeded",
228      "not used - internal error",
229      "invalid combination of newline options",
230      "bad offset value",
231      NULL,  /* SHORTUTF8 is handled specially */
232      "nested recursion at the same subject position",
233      "JIT stack limit reached"
234    };
235    
236    
237  static const int utf8_table1[] = {  /*************************************************
238    0x0000007f, 0x000007ff, 0x0000ffff, 0x001fffff, 0x03ffffff, 0x7fffffff};  *         Alternate character tables             *
239    *************************************************/
240    
241    /* By default, the "tables" pointer when calling PCRE is set to NULL, thereby
242    using the default tables of the library. However, the T option can be used to
243    select alternate sets of tables, for different kinds of testing. Note also that
244    the L (locale) option also adjusts the tables. */
245    
246    /* This is the set of tables distributed as default with PCRE. It recognizes
247    only ASCII characters. */
248    
249    static const unsigned char tables0[] = {
250    
251    /* This table is a lower casing table. */
252    
253        0,  1,  2,  3,  4,  5,  6,  7,
254        8,  9, 10, 11, 12, 13, 14, 15,
255       16, 17, 18, 19, 20, 21, 22, 23,
256       24, 25, 26, 27, 28, 29, 30, 31,
257       32, 33, 34, 35, 36, 37, 38, 39,
258       40, 41, 42, 43, 44, 45, 46, 47,
259       48, 49, 50, 51, 52, 53, 54, 55,
260       56, 57, 58, 59, 60, 61, 62, 63,
261       64, 97, 98, 99,100,101,102,103,
262      104,105,106,107,108,109,110,111,
263      112,113,114,115,116,117,118,119,
264      120,121,122, 91, 92, 93, 94, 95,
265       96, 97, 98, 99,100,101,102,103,
266      104,105,106,107,108,109,110,111,
267      112,113,114,115,116,117,118,119,
268      120,121,122,123,124,125,126,127,
269      128,129,130,131,132,133,134,135,
270      136,137,138,139,140,141,142,143,
271      144,145,146,147,148,149,150,151,
272      152,153,154,155,156,157,158,159,
273      160,161,162,163,164,165,166,167,
274      168,169,170,171,172,173,174,175,
275      176,177,178,179,180,181,182,183,
276      184,185,186,187,188,189,190,191,
277      192,193,194,195,196,197,198,199,
278      200,201,202,203,204,205,206,207,
279      208,209,210,211,212,213,214,215,
280      216,217,218,219,220,221,222,223,
281      224,225,226,227,228,229,230,231,
282      232,233,234,235,236,237,238,239,
283      240,241,242,243,244,245,246,247,
284      248,249,250,251,252,253,254,255,
285    
286    /* This table is a case flipping table. */
287    
288        0,  1,  2,  3,  4,  5,  6,  7,
289        8,  9, 10, 11, 12, 13, 14, 15,
290       16, 17, 18, 19, 20, 21, 22, 23,
291       24, 25, 26, 27, 28, 29, 30, 31,
292       32, 33, 34, 35, 36, 37, 38, 39,
293       40, 41, 42, 43, 44, 45, 46, 47,
294       48, 49, 50, 51, 52, 53, 54, 55,
295       56, 57, 58, 59, 60, 61, 62, 63,
296       64, 97, 98, 99,100,101,102,103,
297      104,105,106,107,108,109,110,111,
298      112,113,114,115,116,117,118,119,
299      120,121,122, 91, 92, 93, 94, 95,
300       96, 65, 66, 67, 68, 69, 70, 71,
301       72, 73, 74, 75, 76, 77, 78, 79,
302       80, 81, 82, 83, 84, 85, 86, 87,
303       88, 89, 90,123,124,125,126,127,
304      128,129,130,131,132,133,134,135,
305      136,137,138,139,140,141,142,143,
306      144,145,146,147,148,149,150,151,
307      152,153,154,155,156,157,158,159,
308      160,161,162,163,164,165,166,167,
309      168,169,170,171,172,173,174,175,
310      176,177,178,179,180,181,182,183,
311      184,185,186,187,188,189,190,191,
312      192,193,194,195,196,197,198,199,
313      200,201,202,203,204,205,206,207,
314      208,209,210,211,212,213,214,215,
315      216,217,218,219,220,221,222,223,
316      224,225,226,227,228,229,230,231,
317      232,233,234,235,236,237,238,239,
318      240,241,242,243,244,245,246,247,
319      248,249,250,251,252,253,254,255,
320    
321    /* This table contains bit maps for various character classes. Each map is 32
322    bytes long and the bits run from the least significant end of each byte. The
323    classes that have their own maps are: space, xdigit, digit, upper, lower, word,
324    graph, print, punct, and cntrl. Other classes are built from combinations. */
325    
326      0x00,0x3e,0x00,0x00,0x01,0x00,0x00,0x00,
327      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
328      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
329      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
330    
331      0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03,
332      0x7e,0x00,0x00,0x00,0x7e,0x00,0x00,0x00,
333      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
334      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
335    
336      0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03,
337      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
338      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
339      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
340    
341      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
342      0xfe,0xff,0xff,0x07,0x00,0x00,0x00,0x00,
343      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
344      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
345    
346      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
347      0x00,0x00,0x00,0x00,0xfe,0xff,0xff,0x07,
348      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
349      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
350    
351      0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03,
352      0xfe,0xff,0xff,0x87,0xfe,0xff,0xff,0x07,
353      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
354      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
355    
356      0x00,0x00,0x00,0x00,0xfe,0xff,0xff,0xff,
357      0xff,0xff,0xff,0xff,0xff,0xff,0xff,0x7f,
358      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
359      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
360    
361      0x00,0x00,0x00,0x00,0xff,0xff,0xff,0xff,
362      0xff,0xff,0xff,0xff,0xff,0xff,0xff,0x7f,
363      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
364      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
365    
366      0x00,0x00,0x00,0x00,0xfe,0xff,0x00,0xfc,
367      0x01,0x00,0x00,0xf8,0x01,0x00,0x00,0x78,
368      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
369      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
370    
371      0xff,0xff,0xff,0xff,0x00,0x00,0x00,0x00,
372      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x80,
373      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
374      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
375    
376    /* This table identifies various classes of character by individual bits:
377      0x01   white space character
378      0x02   letter
379      0x04   decimal digit
380      0x08   hexadecimal digit
381      0x10   alphanumeric or '_'
382      0x80   regular expression metacharacter or binary zero
383    */
384    
385  static const int utf8_table2[] = {    0x80,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*   0-  7 */
386    0, 0xc0, 0xe0, 0xf0, 0xf8, 0xfc};    0x00,0x01,0x01,0x00,0x01,0x01,0x00,0x00, /*   8- 15 */
387      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  16- 23 */
388      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  24- 31 */
389      0x01,0x00,0x00,0x00,0x80,0x00,0x00,0x00, /*    - '  */
390      0x80,0x80,0x80,0x80,0x00,0x00,0x80,0x00, /*  ( - /  */
391      0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c, /*  0 - 7  */
392      0x1c,0x1c,0x00,0x00,0x00,0x00,0x00,0x80, /*  8 - ?  */
393      0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /*  @ - G  */
394      0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  H - O  */
395      0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  P - W  */
396      0x12,0x12,0x12,0x80,0x80,0x00,0x80,0x10, /*  X - _  */
397      0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /*  ` - g  */
398      0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  h - o  */
399      0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  p - w  */
400      0x12,0x12,0x12,0x80,0x80,0x00,0x00,0x00, /*  x -127 */
401      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 128-135 */
402      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 136-143 */
403      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144-151 */
404      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 152-159 */
405      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160-167 */
406      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 168-175 */
407      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 176-183 */
408      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */
409      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 192-199 */
410      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 200-207 */
411      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 208-215 */
412      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 216-223 */
413      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 224-231 */
414      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 232-239 */
415      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 240-247 */
416      0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00};/* 248-255 */
417    
418    /* This is a set of tables that came orginally from a Windows user. It seems to
419    be at least an approximation of ISO 8859. In particular, there are characters
420    greater than 128 that are marked as spaces, letters, etc. */
421    
422    static const unsigned char tables1[] = {
423    0,1,2,3,4,5,6,7,
424    8,9,10,11,12,13,14,15,
425    16,17,18,19,20,21,22,23,
426    24,25,26,27,28,29,30,31,
427    32,33,34,35,36,37,38,39,
428    40,41,42,43,44,45,46,47,
429    48,49,50,51,52,53,54,55,
430    56,57,58,59,60,61,62,63,
431    64,97,98,99,100,101,102,103,
432    104,105,106,107,108,109,110,111,
433    112,113,114,115,116,117,118,119,
434    120,121,122,91,92,93,94,95,
435    96,97,98,99,100,101,102,103,
436    104,105,106,107,108,109,110,111,
437    112,113,114,115,116,117,118,119,
438    120,121,122,123,124,125,126,127,
439    128,129,130,131,132,133,134,135,
440    136,137,138,139,140,141,142,143,
441    144,145,146,147,148,149,150,151,
442    152,153,154,155,156,157,158,159,
443    160,161,162,163,164,165,166,167,
444    168,169,170,171,172,173,174,175,
445    176,177,178,179,180,181,182,183,
446    184,185,186,187,188,189,190,191,
447    224,225,226,227,228,229,230,231,
448    232,233,234,235,236,237,238,239,
449    240,241,242,243,244,245,246,215,
450    248,249,250,251,252,253,254,223,
451    224,225,226,227,228,229,230,231,
452    232,233,234,235,236,237,238,239,
453    240,241,242,243,244,245,246,247,
454    248,249,250,251,252,253,254,255,
455    0,1,2,3,4,5,6,7,
456    8,9,10,11,12,13,14,15,
457    16,17,18,19,20,21,22,23,
458    24,25,26,27,28,29,30,31,
459    32,33,34,35,36,37,38,39,
460    40,41,42,43,44,45,46,47,
461    48,49,50,51,52,53,54,55,
462    56,57,58,59,60,61,62,63,
463    64,97,98,99,100,101,102,103,
464    104,105,106,107,108,109,110,111,
465    112,113,114,115,116,117,118,119,
466    120,121,122,91,92,93,94,95,
467    96,65,66,67,68,69,70,71,
468    72,73,74,75,76,77,78,79,
469    80,81,82,83,84,85,86,87,
470    88,89,90,123,124,125,126,127,
471    128,129,130,131,132,133,134,135,
472    136,137,138,139,140,141,142,143,
473    144,145,146,147,148,149,150,151,
474    152,153,154,155,156,157,158,159,
475    160,161,162,163,164,165,166,167,
476    168,169,170,171,172,173,174,175,
477    176,177,178,179,180,181,182,183,
478    184,185,186,187,188,189,190,191,
479    224,225,226,227,228,229,230,231,
480    232,233,234,235,236,237,238,239,
481    240,241,242,243,244,245,246,215,
482    248,249,250,251,252,253,254,223,
483    192,193,194,195,196,197,198,199,
484    200,201,202,203,204,205,206,207,
485    208,209,210,211,212,213,214,247,
486    216,217,218,219,220,221,222,255,
487    0,62,0,0,1,0,0,0,
488    0,0,0,0,0,0,0,0,
489    32,0,0,0,1,0,0,0,
490    0,0,0,0,0,0,0,0,
491    0,0,0,0,0,0,255,3,
492    126,0,0,0,126,0,0,0,
493    0,0,0,0,0,0,0,0,
494    0,0,0,0,0,0,0,0,
495    0,0,0,0,0,0,255,3,
496    0,0,0,0,0,0,0,0,
497    0,0,0,0,0,0,12,2,
498    0,0,0,0,0,0,0,0,
499    0,0,0,0,0,0,0,0,
500    254,255,255,7,0,0,0,0,
501    0,0,0,0,0,0,0,0,
502    255,255,127,127,0,0,0,0,
503    0,0,0,0,0,0,0,0,
504    0,0,0,0,254,255,255,7,
505    0,0,0,0,0,4,32,4,
506    0,0,0,128,255,255,127,255,
507    0,0,0,0,0,0,255,3,
508    254,255,255,135,254,255,255,7,
509    0,0,0,0,0,4,44,6,
510    255,255,127,255,255,255,127,255,
511    0,0,0,0,254,255,255,255,
512    255,255,255,255,255,255,255,127,
513    0,0,0,0,254,255,255,255,
514    255,255,255,255,255,255,255,255,
515    0,2,0,0,255,255,255,255,
516    255,255,255,255,255,255,255,127,
517    0,0,0,0,255,255,255,255,
518    255,255,255,255,255,255,255,255,
519    0,0,0,0,254,255,0,252,
520    1,0,0,248,1,0,0,120,
521    0,0,0,0,254,255,255,255,
522    0,0,128,0,0,0,128,0,
523    255,255,255,255,0,0,0,0,
524    0,0,0,0,0,0,0,128,
525    255,255,255,255,0,0,0,0,
526    0,0,0,0,0,0,0,0,
527    128,0,0,0,0,0,0,0,
528    0,1,1,0,1,1,0,0,
529    0,0,0,0,0,0,0,0,
530    0,0,0,0,0,0,0,0,
531    1,0,0,0,128,0,0,0,
532    128,128,128,128,0,0,128,0,
533    28,28,28,28,28,28,28,28,
534    28,28,0,0,0,0,0,128,
535    0,26,26,26,26,26,26,18,
536    18,18,18,18,18,18,18,18,
537    18,18,18,18,18,18,18,18,
538    18,18,18,128,128,0,128,16,
539    0,26,26,26,26,26,26,18,
540    18,18,18,18,18,18,18,18,
541    18,18,18,18,18,18,18,18,
542    18,18,18,128,128,0,0,0,
543    0,0,0,0,0,1,0,0,
544    0,0,0,0,0,0,0,0,
545    0,0,0,0,0,0,0,0,
546    0,0,0,0,0,0,0,0,
547    1,0,0,0,0,0,0,0,
548    0,0,18,0,0,0,0,0,
549    0,0,20,20,0,18,0,0,
550    0,20,18,0,0,0,0,0,
551    18,18,18,18,18,18,18,18,
552    18,18,18,18,18,18,18,18,
553    18,18,18,18,18,18,18,0,
554    18,18,18,18,18,18,18,18,
555    18,18,18,18,18,18,18,18,
556    18,18,18,18,18,18,18,18,
557    18,18,18,18,18,18,18,0,
558    18,18,18,18,18,18,18,18
559    };
560    
 static const int utf8_table3[] = {  
   0xff, 0x1f, 0x0f, 0x07, 0x03, 0x01};  
561    
562    
563    
564    #ifndef HAVE_STRERROR
565  /*************************************************  /*************************************************
566  *         Print compiled regex                   *  *     Provide strerror() for non-ANSI libraries  *
567  *************************************************/  *************************************************/
568    
569  /* The code for doing this is held in a separate file that is also included in  /* Some old-fashioned systems still around (e.g. SunOS4) don't have strerror()
570  pcre.c when it is compiled with the debug switch. It defines a function called  in their libraries, but can provide the same facility by this simple
571  print_internals(), which uses a table of opcode lengths defined by the macro  alternative function. */
572  OP_LENGTHS, whose name must be OP_lengths. It also uses a table that translates  
573  Unicode property names to numbers; this is kept in a separate file. */  extern int   sys_nerr;
574    extern char *sys_errlist[];
575  static uschar OP_lengths[] = { OP_LENGTHS };  
576    char *
577  #include "ucp.h"  strerror(int n)
578  #include "ucptypetable.c"  {
579  #include "printint.c"  if (n < 0 || n >= sys_nerr) return "unknown error number";
580    return sys_errlist[n];
581    }
582    #endif /* HAVE_STRERROR */
583    
584    
585    /*************************************************
586    *         JIT memory callback                    *
587    *************************************************/
588    
589    static pcre_jit_stack* jit_callback(void *arg)
590    {
591    return (pcre_jit_stack *)arg;
592    }
593    
594    
595    /*************************************************
596    *        Read or extend an input line            *
597    *************************************************/
598    
599    /* Input lines are read into buffer, but both patterns and data lines can be
600    continued over multiple input lines. In addition, if the buffer fills up, we
601    want to automatically expand it so as to be able to handle extremely large
602    lines that are needed for certain stress tests. When the input buffer is
603    expanded, the other two buffers must also be expanded likewise, and the
604    contents of pbuffer, which are a copy of the input for callouts, must be
605    preserved (for when expansion happens for a data line). This is not the most
606    optimal way of handling this, but hey, this is just a test program!
607    
608    Arguments:
609      f            the file to read
610      start        where in buffer to start (this *must* be within buffer)
611      prompt       for stdin or readline()
612    
613    Returns:       pointer to the start of new data
614                   could be a copy of start, or could be moved
615                   NULL if no data read and EOF reached
616    */
617    
618    static uschar *
619    extend_inputline(FILE *f, uschar *start, const char *prompt)
620    {
621    uschar *here = start;
622    
623    for (;;)
624      {
625      int rlen = (int)(buffer_size - (here - buffer));
626    
627      if (rlen > 1000)
628        {
629        int dlen;
630    
631        /* If libreadline support is required, use readline() to read a line if the
632        input is a terminal. Note that readline() removes the trailing newline, so
633        we must put it back again, to be compatible with fgets(). */
634    
635    #ifdef SUPPORT_LIBREADLINE
636        if (isatty(fileno(f)))
637          {
638          size_t len;
639          char *s = readline(prompt);
640          if (s == NULL) return (here == start)? NULL : start;
641          len = strlen(s);
642          if (len > 0) add_history(s);
643          if (len > rlen - 1) len = rlen - 1;
644          memcpy(here, s, len);
645          here[len] = '\n';
646          here[len+1] = 0;
647          free(s);
648          }
649        else
650    #endif
651    
652        /* Read the next line by normal means, prompting if the file is stdin. */
653    
654          {
655          if (f == stdin) printf("%s", prompt);
656          if (fgets((char *)here, rlen,  f) == NULL)
657            return (here == start)? NULL : start;
658          }
659    
660        dlen = (int)strlen((char *)here);
661        if (dlen > 0 && here[dlen - 1] == '\n') return start;
662        here += dlen;
663        }
664    
665      else
666        {
667        int new_buffer_size = 2*buffer_size;
668        uschar *new_buffer = (unsigned char *)malloc(new_buffer_size);
669        uschar *new_dbuffer = (unsigned char *)malloc(new_buffer_size);
670        uschar *new_pbuffer = (unsigned char *)malloc(new_buffer_size);
671    
672        if (new_buffer == NULL || new_dbuffer == NULL || new_pbuffer == NULL)
673          {
674          fprintf(stderr, "pcretest: malloc(%d) failed\n", new_buffer_size);
675          exit(1);
676          }
677    
678        memcpy(new_buffer, buffer, buffer_size);
679        memcpy(new_pbuffer, pbuffer, buffer_size);
680    
681        buffer_size = new_buffer_size;
682    
683        start = new_buffer + (start - buffer);
684        here = new_buffer + (here - buffer);
685    
686        free(buffer);
687        free(dbuffer);
688        free(pbuffer);
689    
690        buffer = new_buffer;
691        dbuffer = new_dbuffer;
692        pbuffer = new_pbuffer;
693        }
694      }
695    
696    return NULL;  /* Control never gets here */
697    }
698    
699    
700    
701    
702    
703    
704    
# Line 122  static uschar OP_lengths[] = { OP_LENGTH Line 708  static uschar OP_lengths[] = { OP_LENGTH
708    
709  /* We don't use strtoul() because SunOS4 doesn't have it. Rather than mess  /* We don't use strtoul() because SunOS4 doesn't have it. Rather than mess
710  around with conditional compilation, just do the job by hand. It is only used  around with conditional compilation, just do the job by hand. It is only used
711  for unpicking the -o argument, so just keep it simple.  for unpicking arguments, so just keep it simple.
712    
713  Arguments:  Arguments:
714    str           string to be converted    str           string to be converted
# Line 143  return(result); Line 729  return(result);
729    
730    
731    
 /*************************************************  
 *       Convert character value to UTF-8         *  
 *************************************************/  
   
 /* This function takes an integer value in the range 0 - 0x7fffffff  
 and encodes it as a UTF-8 character in 0 to 6 bytes.  
   
 Arguments:  
   cvalue     the character value  
   buffer     pointer to buffer for result - at least 6 bytes long  
   
 Returns:     number of characters placed in the buffer  
              -1 if input character is negative  
              0 if input character is positive but too big (only when  
              int is longer than 32 bits)  
 */  
   
 static int  
 ord2utf8(int cvalue, unsigned char *buffer)  
 {  
 register int i, j;  
 for (i = 0; i < sizeof(utf8_table1)/sizeof(int); i++)  
   if (cvalue <= utf8_table1[i]) break;  
 if (i >= sizeof(utf8_table1)/sizeof(int)) return 0;  
 if (cvalue < 0) return -1;  
   
 buffer += i;  
 for (j = i; j > 0; j--)  
  {  
  *buffer-- = 0x80 | (cvalue & 0x3f);  
  cvalue >>= 6;  
  }  
 *buffer = utf8_table2[i] | cvalue;  
 return i + 1;  
 }  
   
732    
733  /*************************************************  /*************************************************
734  *            Convert UTF-8 string to value       *  *            Convert UTF-8 string to value       *
# Line 188  return i + 1; Line 738  return i + 1;
738  and returns the value of the character.  and returns the value of the character.
739    
740  Argument:  Argument:
741    buffer   a pointer to the byte vector    utf8bytes   a pointer to the byte vector
742    vptr     a pointer to an int to receive the value    vptr        a pointer to an int to receive the value
743    
744  Returns:   >  0 => the number of bytes consumed  Returns:      >  0 => the number of bytes consumed
745             -6 to 0 => malformed UTF-8 character at offset = (-return)                -6 to 0 => malformed UTF-8 character at offset = (-return)
746  */  */
747    
748    #if !defined NOUTF8
749    
750  static int  static int
751  utf82ord(unsigned char *buffer, int *vptr)  utf82ord(unsigned char *utf8bytes, int *vptr)
752  {  {
753  int c = *buffer++;  int c = *utf8bytes++;
754  int d = c;  int d = c;
755  int i, j, s;  int i, j, s;
756    
# Line 218  d = (c & utf8_table3[i]) << s; Line 770  d = (c & utf8_table3[i]) << s;
770    
771  for (j = 0; j < i; j++)  for (j = 0; j < i; j++)
772    {    {
773    c = *buffer++;    c = *utf8bytes++;
774    if ((c & 0xc0) != 0x80) return -(j+1);    if ((c & 0xc0) != 0x80) return -(j+1);
775    s -= 6;    s -= 6;
776    d |= (c & 0x3f) << s;    d |= (c & 0x3f) << s;
# Line 226  for (j = 0; j < i; j++) Line 778  for (j = 0; j < i; j++)
778    
779  /* Check that encoding was the correct unique one */  /* Check that encoding was the correct unique one */
780    
781  for (j = 0; j < sizeof(utf8_table1)/sizeof(int); j++)  for (j = 0; j < utf8_table1_size; j++)
782    if (d <= utf8_table1[j]) break;    if (d <= utf8_table1[j]) break;
783  if (j != i) return -(i+1);  if (j != i) return -(i+1);
784    
# Line 236  if (j != i) return -(i+1); Line 788  if (j != i) return -(i+1);
788  return i+1;  return i+1;
789  }  }
790    
791    #endif
792    
793    
794    
795    /*************************************************
796    *       Convert character value to UTF-8         *
797    *************************************************/
798    
799    /* This function takes an integer value in the range 0 - 0x7fffffff
800    and encodes it as a UTF-8 character in 0 to 6 bytes.
801    
802    Arguments:
803      cvalue     the character value
804      utf8bytes  pointer to buffer for result - at least 6 bytes long
805    
806    Returns:     number of characters placed in the buffer
807    */
808    
809    #if !defined NOUTF8
810    
811    static int
812    ord2utf8(int cvalue, uschar *utf8bytes)
813    {
814    register int i, j;
815    for (i = 0; i < utf8_table1_size; i++)
816      if (cvalue <= utf8_table1[i]) break;
817    utf8bytes += i;
818    for (j = i; j > 0; j--)
819     {
820     *utf8bytes-- = 0x80 | (cvalue & 0x3f);
821     cvalue >>= 6;
822     }
823    *utf8bytes = utf8_table2[i] | cvalue;
824    return i + 1;
825    }
826    
827    #endif
828    
829    
830    
831  /*************************************************  /*************************************************
# Line 248  chars without printing. */ Line 838  chars without printing. */
838    
839  static int pchars(unsigned char *p, int length, FILE *f)  static int pchars(unsigned char *p, int length, FILE *f)
840  {  {
841  int c;  int c = 0;
842  int yield = 0;  int yield = 0;
843    
844  while (length-- > 0)  while (length-- > 0)
845    {    {
846    #if !defined NOUTF8
847    if (use_utf8)    if (use_utf8)
848      {      {
849      int rc = utf82ord(p, &c);      int rc = utf82ord(p, &c);
# Line 261  while (length-- > 0) Line 852  while (length-- > 0)
852        {        {
853        length -= rc - 1;        length -= rc - 1;
854        p += rc;        p += rc;
855        if (c < 256 && isprint(c))        if (PRINTHEX(c))
856          {          {
857          if (f != NULL) fprintf(f, "%c", c);          if (f != NULL) fprintf(f, "%c", c);
858          yield++;          yield++;
859          }          }
860        else        else
861          {          {
862          int n;          int n = 4;
863          if (f != NULL) fprintf(f, "\\x{%02x}%n", c, &n);          if (f != NULL) fprintf(f, "\\x{%02x}", c);
864          yield += n;          yield += (n <= 0x000000ff)? 2 :
865                     (n <= 0x00000fff)? 3 :
866                     (n <= 0x0000ffff)? 4 :
867                     (n <= 0x000fffff)? 5 : 6;
868          }          }
869        continue;        continue;
870        }        }
871      }      }
872    #endif
873    
874     /* Not UTF-8, or malformed UTF-8  */     /* Not UTF-8, or malformed UTF-8  */
875    
876    if (isprint(c = *(p++)))    c = *p++;
877      if (PRINTHEX(c))
878      {      {
879      if (f != NULL) fprintf(f, "%c", c);      if (f != NULL) fprintf(f, "%c", c);
880      yield++;      yield++;
# Line 376  fprintf(outfile, "%.*s", (cb->next_item_ Line 972  fprintf(outfile, "%.*s", (cb->next_item_
972  fprintf(outfile, "\n");  fprintf(outfile, "\n");
973  first_callout = 0;  first_callout = 0;
974    
975    if (cb->mark != last_callout_mark)
976      {
977      fprintf(outfile, "Latest Mark: %s\n",
978        (cb->mark == NULL)? "<unset>" : (char *)(cb->mark));
979      last_callout_mark = cb->mark;
980      }
981    
982  if (cb->callout_data != NULL)  if (cb->callout_data != NULL)
983    {    {
984    int callout_data = *((int *)(cb->callout_data));    int callout_data = *((int *)(cb->callout_data));
# Line 395  return (cb->callout_number != callout_fa Line 998  return (cb->callout_number != callout_fa
998  *            Local malloc functions              *  *            Local malloc functions              *
999  *************************************************/  *************************************************/
1000    
1001  /* Alternative malloc function, to test functionality and show the size of the  /* Alternative malloc function, to test functionality and save the size of a
1002  compiled re. */  compiled re. The show_malloc variable is set only during matching. */
1003    
1004  static void *new_malloc(size_t size)  static void *new_malloc(size_t size)
1005  {  {
1006  void *block = malloc(size);  void *block = malloc(size);
1007  gotten_store = size;  gotten_store = size;
1008  if (show_malloc)  if (show_malloc)
1009    fprintf(outfile, "malloc       %3d %p\n", size, block);    fprintf(outfile, "malloc       %3d %p\n", (int)size, block);
1010  return block;  return block;
1011  }  }
1012    
# Line 414  if (show_malloc) Line 1017  if (show_malloc)
1017  free(block);  free(block);
1018  }  }
1019    
   
1020  /* For recursion malloc/free, to test stacking calls */  /* For recursion malloc/free, to test stacking calls */
1021    
1022  static void *stack_malloc(size_t size)  static void *stack_malloc(size_t size)
1023  {  {
1024  void *block = malloc(size);  void *block = malloc(size);
1025  if (show_malloc)  if (show_malloc)
1026    fprintf(outfile, "stack_malloc %3d %p\n", size, block);    fprintf(outfile, "stack_malloc %3d %p\n", (int)size, block);
1027  return block;  return block;
1028  }  }
1029    
# Line 452  if ((rc = pcre_fullinfo(re, study, optio Line 1054  if ((rc = pcre_fullinfo(re, study, optio
1054  *         Byte flipping function                 *  *         Byte flipping function                 *
1055  *************************************************/  *************************************************/
1056    
1057  static long int  static unsigned long int
1058  byteflip(long int value, int n)  byteflip(unsigned long int value, int n)
1059  {  {
1060  if (n == 2) return ((value & 0x00ff) << 8) | ((value & 0xff00) >> 8);  if (n == 2) return ((value & 0x00ff) << 8) | ((value & 0xff00) >> 8);
1061  return ((value & 0x000000ff) << 24) |  return ((value & 0x000000ff) << 24) |
# Line 466  return ((value & 0x000000ff) << 24) | Line 1068  return ((value & 0x000000ff) << 24) |
1068    
1069    
1070  /*************************************************  /*************************************************
1071    *        Check match or recursion limit          *
1072    *************************************************/
1073    
1074    static int
1075    check_match_limit(pcre *re, pcre_extra *extra, uschar *bptr, int len,
1076      int start_offset, int options, int *use_offsets, int use_size_offsets,
1077      int flag, unsigned long int *limit, int errnumber, const char *msg)
1078    {
1079    int count;
1080    int min = 0;
1081    int mid = 64;
1082    int max = -1;
1083    
1084    extra->flags |= flag;
1085    
1086    for (;;)
1087      {
1088      *limit = mid;
1089    
1090      count = pcre_exec(re, extra, (char *)bptr, len, start_offset, options,
1091        use_offsets, use_size_offsets);
1092    
1093      if (count == errnumber)
1094        {
1095        /* fprintf(outfile, "Testing %s limit = %d\n", msg, mid); */
1096        min = mid;
1097        mid = (mid == max - 1)? max : (max > 0)? (min + max)/2 : mid*2;
1098        }
1099    
1100      else if (count >= 0 || count == PCRE_ERROR_NOMATCH ||
1101                             count == PCRE_ERROR_PARTIAL)
1102        {
1103        if (mid == min + 1)
1104          {
1105          fprintf(outfile, "Minimum %s limit = %d\n", msg, mid);
1106          break;
1107          }
1108        /* fprintf(outfile, "Testing %s limit = %d\n", msg, mid); */
1109        max = mid;
1110        mid = (min + mid)/2;
1111        }
1112      else break;    /* Some other error */
1113      }
1114    
1115    extra->flags &= ~flag;
1116    return count;
1117    }
1118    
1119    
1120    
1121    /*************************************************
1122    *         Case-independent strncmp() function    *
1123    *************************************************/
1124    
1125    /*
1126    Arguments:
1127      s         first string
1128      t         second string
1129      n         number of characters to compare
1130    
1131    Returns:    < 0, = 0, or > 0, according to the comparison
1132    */
1133    
1134    static int
1135    strncmpic(uschar *s, uschar *t, int n)
1136    {
1137    while (n--)
1138      {
1139      int c = tolower(*s++) - tolower(*t++);
1140      if (c) return c;
1141      }
1142    return 0;
1143    }
1144    
1145    
1146    
1147    /*************************************************
1148    *         Check newline indicator                *
1149    *************************************************/
1150    
1151    /* This is used both at compile and run-time to check for <xxx> escapes. Print
1152    a message and return 0 if there is no match.
1153    
1154    Arguments:
1155      p           points after the leading '<'
1156      f           file for error message
1157    
1158    Returns:      appropriate PCRE_NEWLINE_xxx flags, or 0
1159    */
1160    
1161    static int
1162    check_newline(uschar *p, FILE *f)
1163    {
1164    if (strncmpic(p, (uschar *)"cr>", 3) == 0) return PCRE_NEWLINE_CR;
1165    if (strncmpic(p, (uschar *)"lf>", 3) == 0) return PCRE_NEWLINE_LF;
1166    if (strncmpic(p, (uschar *)"crlf>", 5) == 0) return PCRE_NEWLINE_CRLF;
1167    if (strncmpic(p, (uschar *)"anycrlf>", 8) == 0) return PCRE_NEWLINE_ANYCRLF;
1168    if (strncmpic(p, (uschar *)"any>", 4) == 0) return PCRE_NEWLINE_ANY;
1169    if (strncmpic(p, (uschar *)"bsr_anycrlf>", 12) == 0) return PCRE_BSR_ANYCRLF;
1170    if (strncmpic(p, (uschar *)"bsr_unicode>", 12) == 0) return PCRE_BSR_UNICODE;
1171    fprintf(f, "Unknown newline type at: <%s\n", p);
1172    return 0;
1173    }
1174    
1175    
1176    
1177    /*************************************************
1178    *             Usage function                     *
1179    *************************************************/
1180    
1181    static void
1182    usage(void)
1183    {
1184    printf("Usage:     pcretest [options] [<input file> [<output file>]]\n\n");
1185    printf("Input and output default to stdin and stdout.\n");
1186    #ifdef SUPPORT_LIBREADLINE
1187    printf("If input is a terminal, readline() is used to read from it.\n");
1188    #else
1189    printf("This version of pcretest is not linked with readline().\n");
1190    #endif
1191    printf("\nOptions:\n");
1192    printf("  -b       show compiled code (bytecode)\n");
1193    printf("  -C       show PCRE compile-time options and exit\n");
1194    printf("  -d       debug: show compiled code and information (-b and -i)\n");
1195    #if !defined NODFA
1196    printf("  -dfa     force DFA matching for all subjects\n");
1197    #endif
1198    printf("  -help    show usage information\n");
1199    printf("  -i       show information about compiled patterns\n"
1200           "  -M       find MATCH_LIMIT minimum for each subject\n"
1201           "  -m       output memory used information\n"
1202           "  -o <n>   set size of offsets vector to <n>\n");
1203    #if !defined NOPOSIX
1204    printf("  -p       use POSIX interface\n");
1205    #endif
1206    printf("  -q       quiet: do not output PCRE version number at start\n");
1207    printf("  -S <n>   set stack size to <n> megabytes\n");
1208    printf("  -s       force each pattern to be studied at basic level\n"
1209           "  -s+      force each pattern to be studied, using JIT if available\n"
1210           "  -t       time compilation and execution\n");
1211    printf("  -t <n>   time compilation and execution, repeating <n> times\n");
1212    printf("  -tm      time execution (matching) only\n");
1213    printf("  -tm <n>  time execution (matching) only, repeating <n> times\n");
1214    }
1215    
1216    
1217    
1218    /*************************************************
1219  *                Main Program                    *  *                Main Program                    *
1220  *************************************************/  *************************************************/
1221    
# Line 478  int main(int argc, char **argv) Line 1228  int main(int argc, char **argv)
1228  FILE *infile = stdin;  FILE *infile = stdin;
1229  int options = 0;  int options = 0;
1230  int study_options = 0;  int study_options = 0;
1231    int default_find_match_limit = FALSE;
1232  int op = 1;  int op = 1;
1233  int timeit = 0;  int timeit = 0;
1234    int timeitm = 0;
1235  int showinfo = 0;  int showinfo = 0;
1236  int showstore = 0;  int showstore = 0;
1237    int force_study = -1;
1238    int force_study_options = 0;
1239    int quiet = 0;
1240  int size_offsets = 45;  int size_offsets = 45;
1241  int size_offsets_max;  int size_offsets_max;
1242  int *offsets;  int *offsets = NULL;
1243  #if !defined NOPOSIX  #if !defined NOPOSIX
1244  int posix = 0;  int posix = 0;
1245  #endif  #endif
1246  int debug = 0;  int debug = 0;
1247  int done = 0;  int done = 0;
1248    int all_use_dfa = 0;
1249    int yield = 0;
1250    int stack_size;
1251    
1252    pcre_jit_stack *jit_stack = NULL;
1253    
1254    
1255    /* These vectors store, end-to-end, a list of captured substring names. Assume
1256    that 1024 is plenty long enough for the few names we'll be testing. */
1257    
1258  unsigned char *buffer;  uschar copynames[1024];
1259  unsigned char *dbuffer;  uschar getnames[1024];
1260    
1261    uschar *copynamesptr;
1262    uschar *getnamesptr;
1263    
1264  /* Get buffers from malloc() so that Electric Fence will check their misuse  /* Get buffers from malloc() so that Electric Fence will check their misuse
1265  when I am debugging. */  when I am debugging. They grow automatically when very long lines are read. */
1266    
1267  buffer = (unsigned char *)malloc(BUFFER_SIZE);  buffer = (unsigned char *)malloc(buffer_size);
1268  dbuffer = (unsigned char *)malloc(DBUFFER_SIZE);  dbuffer = (unsigned char *)malloc(buffer_size);
1269  pbuffer = (unsigned char *)malloc(PBUFFER_SIZE);  pbuffer = (unsigned char *)malloc(buffer_size);
   
 /* The outfile variable is static so that new_malloc can use it. The _setmode()  
 stuff is some magic that I don't understand, but which apparently does good  
 things in Windows. It's related to line terminations.  */  
1270    
1271  #if defined(_WIN32) || defined(WIN32)  /* The outfile variable is static so that new_malloc can use it. */
 _setmode( _fileno( stdout ), 0x8000 );  
 #endif  /* defined(_WIN32) || defined(WIN32) */  
1272    
1273  outfile = stdout;  outfile = stdout;
1274    
1275    /* The following  _setmode() stuff is some Windows magic that tells its runtime
1276    library to translate CRLF into a single LF character. At least, that's what
1277    I've been told: never having used Windows I take this all on trust. Originally
1278    it set 0x8000, but then I was advised that _O_BINARY was better. */
1279    
1280    #if defined(_WIN32) || defined(WIN32)
1281    _setmode( _fileno( stdout ), _O_BINARY );
1282    #endif
1283    
1284  /* Scan options */  /* Scan options */
1285    
1286  while (argc > 1 && argv[op][0] == '-')  while (argc > 1 && argv[op][0] == '-')
1287    {    {
1288    unsigned char *endptr;    unsigned char *endptr;
1289    
1290    if (strcmp(argv[op], "-s") == 0 || strcmp(argv[op], "-m") == 0)    if (strcmp(argv[op], "-m") == 0) showstore = 1;
1291      showstore = 1;    else if (strcmp(argv[op], "-s") == 0) force_study = 0;
1292    else if (strcmp(argv[op], "-t") == 0) timeit = 1;    else if (strcmp(argv[op], "-s+") == 0)
1293        {
1294        force_study = 1;
1295        force_study_options = PCRE_STUDY_JIT_COMPILE;
1296        }
1297      else if (strcmp(argv[op], "-q") == 0) quiet = 1;
1298      else if (strcmp(argv[op], "-b") == 0) debug = 1;
1299    else if (strcmp(argv[op], "-i") == 0) showinfo = 1;    else if (strcmp(argv[op], "-i") == 0) showinfo = 1;
1300    else if (strcmp(argv[op], "-d") == 0) showinfo = debug = 1;    else if (strcmp(argv[op], "-d") == 0) showinfo = debug = 1;
1301      else if (strcmp(argv[op], "-M") == 0) default_find_match_limit = TRUE;
1302    #if !defined NODFA
1303      else if (strcmp(argv[op], "-dfa") == 0) all_use_dfa = 1;
1304    #endif
1305    else if (strcmp(argv[op], "-o") == 0 && argc > 2 &&    else if (strcmp(argv[op], "-o") == 0 && argc > 2 &&
1306        ((size_offsets = get_value((unsigned char *)argv[op+1], &endptr)),        ((size_offsets = get_value((unsigned char *)argv[op+1], &endptr)),
1307          *endptr == 0))          *endptr == 0))
# Line 529  while (argc > 1 && argv[op][0] == '-') Line 1309  while (argc > 1 && argv[op][0] == '-')
1309      op++;      op++;
1310      argc--;      argc--;
1311      }      }
1312      else if (strcmp(argv[op], "-t") == 0 || strcmp(argv[op], "-tm") == 0)
1313        {
1314        int both = argv[op][2] == 0;
1315        int temp;
1316        if (argc > 2 && (temp = get_value((unsigned char *)argv[op+1], &endptr),
1317                         *endptr == 0))
1318          {
1319          timeitm = temp;
1320          op++;
1321          argc--;
1322          }
1323        else timeitm = LOOPREPEAT;
1324        if (both) timeit = timeitm;
1325        }
1326      else if (strcmp(argv[op], "-S") == 0 && argc > 2 &&
1327          ((stack_size = get_value((unsigned char *)argv[op+1], &endptr)),
1328            *endptr == 0))
1329        {
1330    #if defined(_WIN32) || defined(WIN32) || defined(__minix)
1331        printf("PCRE: -S not supported on this OS\n");
1332        exit(1);
1333    #else
1334        int rc;
1335        struct rlimit rlim;
1336        getrlimit(RLIMIT_STACK, &rlim);
1337        rlim.rlim_cur = stack_size * 1024 * 1024;
1338        rc = setrlimit(RLIMIT_STACK, &rlim);
1339        if (rc != 0)
1340          {
1341        printf("PCRE: setrlimit() failed with error %d\n", rc);
1342        exit(1);
1343          }
1344        op++;
1345        argc--;
1346    #endif
1347        }
1348  #if !defined NOPOSIX  #if !defined NOPOSIX
1349    else if (strcmp(argv[op], "-p") == 0) posix = 1;    else if (strcmp(argv[op], "-p") == 0) posix = 1;
1350  #endif  #endif
1351    else if (strcmp(argv[op], "-C") == 0)    else if (strcmp(argv[op], "-C") == 0)
1352      {      {
1353      int rc;      int rc;
1354        unsigned long int lrc;
1355      printf("PCRE version %s\n", pcre_version());      printf("PCRE version %s\n", pcre_version());
1356      printf("Compiled with\n");      printf("Compiled with\n");
1357      (void)pcre_config(PCRE_CONFIG_UTF8, &rc);      (void)pcre_config(PCRE_CONFIG_UTF8, &rc);
1358      printf("  %sUTF-8 support\n", rc? "" : "No ");      printf("  %sUTF-8 support\n", rc? "" : "No ");
1359      (void)pcre_config(PCRE_CONFIG_UNICODE_PROPERTIES, &rc);      (void)pcre_config(PCRE_CONFIG_UNICODE_PROPERTIES, &rc);
1360      printf("  %sUnicode properties support\n", rc? "" : "No ");      printf("  %sUnicode properties support\n", rc? "" : "No ");
1361        (void)pcre_config(PCRE_CONFIG_JIT, &rc);
1362        if (rc)
1363          printf("  Just-in-time compiler support\n");
1364        else
1365          printf("  No just-in-time compiler support\n");
1366      (void)pcre_config(PCRE_CONFIG_NEWLINE, &rc);      (void)pcre_config(PCRE_CONFIG_NEWLINE, &rc);
1367      printf("  Newline character is %s\n", (rc == '\r')? "CR" : "LF");      /* Note that these values are always the ASCII values, even
1368        in EBCDIC environments. CR is 13 and NL is 10. */
1369        printf("  Newline sequence is %s\n", (rc == 13)? "CR" :
1370          (rc == 10)? "LF" : (rc == (13<<8 | 10))? "CRLF" :
1371          (rc == -2)? "ANYCRLF" :
1372          (rc == -1)? "ANY" : "???");
1373        (void)pcre_config(PCRE_CONFIG_BSR, &rc);
1374        printf("  \\R matches %s\n", rc? "CR, LF, or CRLF only" :
1375                                         "all Unicode newlines");
1376      (void)pcre_config(PCRE_CONFIG_LINK_SIZE, &rc);      (void)pcre_config(PCRE_CONFIG_LINK_SIZE, &rc);
1377      printf("  Internal link size = %d\n", rc);      printf("  Internal link size = %d\n", rc);
1378      (void)pcre_config(PCRE_CONFIG_POSIX_MALLOC_THRESHOLD, &rc);      (void)pcre_config(PCRE_CONFIG_POSIX_MALLOC_THRESHOLD, &rc);
1379      printf("  POSIX malloc threshold = %d\n", rc);      printf("  POSIX malloc threshold = %d\n", rc);
1380      (void)pcre_config(PCRE_CONFIG_MATCH_LIMIT, &rc);      (void)pcre_config(PCRE_CONFIG_MATCH_LIMIT, &lrc);
1381      printf("  Default match limit = %d\n", rc);      printf("  Default match limit = %ld\n", lrc);
1382        (void)pcre_config(PCRE_CONFIG_MATCH_LIMIT_RECURSION, &lrc);
1383        printf("  Default recursion depth limit = %ld\n", lrc);
1384      (void)pcre_config(PCRE_CONFIG_STACKRECURSE, &rc);      (void)pcre_config(PCRE_CONFIG_STACKRECURSE, &rc);
1385      printf("  Match recursion uses %s\n", rc? "stack" : "heap");      printf("  Match recursion uses %s\n", rc? "stack" : "heap");
1386      exit(0);      goto EXIT;
1387        }
1388      else if (strcmp(argv[op], "-help") == 0 ||
1389               strcmp(argv[op], "--help") == 0)
1390        {
1391        usage();
1392        goto EXIT;
1393      }      }
1394    else    else
1395      {      {
1396      printf("** Unknown or malformed option %s\n", argv[op]);      printf("** Unknown or malformed option %s\n", argv[op]);
1397      printf("Usage:   pcretest [-d] [-i] [-o <n>] [-p] [-s] [-t] [<input> [<output>]]\n");      usage();
1398      printf("  -C     show PCRE compile-time options and exit\n");      yield = 1;
1399      printf("  -d     debug: show compiled code; implies -i\n"      goto EXIT;
            "  -i     show information about compiled pattern\n"  
            "  -m     output memory used information\n"  
            "  -o <n> set size of offsets vector to <n>\n");  
 #if !defined NOPOSIX  
     printf("  -p     use POSIX interface\n");  
 #endif  
     printf("  -s     output store (memory) used information\n"  
            "  -t     time compilation and execution\n");  
     return 1;  
1400      }      }
1401    op++;    op++;
1402    argc--;    argc--;
# Line 580  offsets = (int *)malloc(size_offsets_max Line 1409  offsets = (int *)malloc(size_offsets_max
1409  if (offsets == NULL)  if (offsets == NULL)
1410    {    {
1411    printf("** Failed to get %d bytes of memory for offsets vector\n",    printf("** Failed to get %d bytes of memory for offsets vector\n",
1412      size_offsets_max * sizeof(int));      (int)(size_offsets_max * sizeof(int)));
1413    return 1;    yield = 1;
1414      goto EXIT;
1415    }    }
1416    
1417  /* Sort out the input and output files */  /* Sort out the input and output files */
1418    
1419  if (argc > 1)  if (argc > 1)
1420    {    {
1421    infile = fopen(argv[op], "rb");    infile = fopen(argv[op], INPUT_MODE);
1422    if (infile == NULL)    if (infile == NULL)
1423      {      {
1424      printf("** Failed to open %s\n", argv[op]);      printf("** Failed to open %s\n", argv[op]);
1425      return 1;      yield = 1;
1426        goto EXIT;
1427      }      }
1428    }    }
1429    
1430  if (argc > 2)  if (argc > 2)
1431    {    {
1432    outfile = fopen(argv[op+1], "wb");    outfile = fopen(argv[op+1], OUTPUT_MODE);
1433    if (outfile == NULL)    if (outfile == NULL)
1434      {      {
1435      printf("** Failed to open %s\n", argv[op+1]);      printf("** Failed to open %s\n", argv[op+1]);
1436      return 1;      yield = 1;
1437        goto EXIT;
1438      }      }
1439    }    }
1440    
# Line 613  pcre_free = new_free; Line 1445  pcre_free = new_free;
1445  pcre_stack_malloc = stack_malloc;  pcre_stack_malloc = stack_malloc;
1446  pcre_stack_free = stack_free;  pcre_stack_free = stack_free;
1447    
1448  /* Heading line, then prompt for first regex if stdin */  /* Heading line unless quiet, then prompt for first regex if stdin */
1449    
1450  fprintf(outfile, "PCRE version %s\n\n", pcre_version());  if (!quiet) fprintf(outfile, "PCRE version %s\n\n", pcre_version());
1451    
1452  /* Main loop */  /* Main loop */
1453    
# Line 630  while (!done) Line 1462  while (!done)
1462  #endif  #endif
1463    
1464    const char *error;    const char *error;
1465      unsigned char *markptr;
1466    unsigned char *p, *pp, *ppp;    unsigned char *p, *pp, *ppp;
1467    unsigned char *to_file = NULL;    unsigned char *to_file = NULL;
1468    const unsigned char *tables = NULL;    const unsigned char *tables = NULL;
1469    unsigned long int true_size, true_study_size = 0;    unsigned long int true_size, true_study_size = 0;
1470    size_t size, regex_gotten_store;    size_t size, regex_gotten_store;
1471      int do_allcaps = 0;
1472      int do_mark = 0;
1473    int do_study = 0;    int do_study = 0;
1474      int no_force_study = 0;
1475    int do_debug = debug;    int do_debug = debug;
1476    int do_G = 0;    int do_G = 0;
1477    int do_g = 0;    int do_g = 0;
1478    int do_showinfo = showinfo;    int do_showinfo = showinfo;
1479    int do_showrest = 0;    int do_showrest = 0;
1480      int do_showcaprest = 0;
1481    int do_flip = 0;    int do_flip = 0;
1482    int erroroffset, len, delimiter;    int erroroffset, len, delimiter, poffset;
1483    
1484    use_utf8 = 0;    use_utf8 = 0;
1485      debug_lengths = 1;
1486    
1487    if (infile == stdin) printf("  re> ");    if (extend_inputline(infile, buffer, "  re> ") == NULL) break;
   if (fgets((char *)buffer, BUFFER_SIZE, infile) == NULL) break;  
1488    if (infile != stdin) fprintf(outfile, "%s", (char *)buffer);    if (infile != stdin) fprintf(outfile, "%s", (char *)buffer);
1489    fflush(outfile);    fflush(outfile);
1490    
# Line 659  while (!done) Line 1496  while (!done)
1496    
1497    if (*p == '<' && strchr((char *)(p+1), '<') == NULL)    if (*p == '<' && strchr((char *)(p+1), '<') == NULL)
1498      {      {
1499      unsigned long int magic;      unsigned long int magic, get_options;
1500      uschar sbuf[8];      uschar sbuf[8];
1501      FILE *f;      FILE *f;
1502    
# Line 702  while (!done) Line 1539  while (!done)
1539          }          }
1540        }        }
1541    
1542      fprintf(outfile, "Compiled regex%s loaded from %s\n",      fprintf(outfile, "Compiled pattern%s loaded from %s\n",
1543        do_flip? " (byte-inverted)" : "", p);        do_flip? " (byte-inverted)" : "", p);
1544    
1545      /* Need to know if UTF-8 for printing data strings */      /* Need to know if UTF-8 for printing data strings */
1546    
1547      new_info(re, NULL, PCRE_INFO_OPTIONS, &options);      new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options);
1548      use_utf8 = (options & PCRE_UTF8) != 0;      use_utf8 = (get_options & PCRE_UTF8) != 0;
1549    
1550      /* Now see if there is any following study data */      /* Now see if there is any following study data. */
1551    
1552      if (true_study_size != 0)      if (true_study_size != 0)
1553        {        {
# Line 726  while (!done) Line 1563  while (!done)
1563          {          {
1564          FAIL_READ:          FAIL_READ:
1565          fprintf(outfile, "Failed to read data from %s\n", p);          fprintf(outfile, "Failed to read data from %s\n", p);
1566          if (extra != NULL) new_free(extra);          if (extra != NULL) pcre_free_study(extra);
1567          if (re != NULL) new_free(re);          if (re != NULL) new_free(re);
1568          fclose(f);          fclose(f);
1569          continue;          continue;
# Line 747  while (!done) Line 1584  while (!done)
1584    
1585    if (isalnum(delimiter) || delimiter == '\\')    if (isalnum(delimiter) || delimiter == '\\')
1586      {      {
1587      fprintf(outfile, "** Delimiter must not be alphameric or \\\n");      fprintf(outfile, "** Delimiter must not be alphanumeric or \\\n");
1588      goto SKIP_DATA;      goto SKIP_DATA;
1589      }      }
1590    
1591    pp = p;    pp = p;
1592      poffset = (int)(p - buffer);
1593    
1594    for(;;)    for(;;)
1595      {      {
# Line 762  while (!done) Line 1600  while (!done)
1600        pp++;        pp++;
1601        }        }
1602      if (*pp != 0) break;      if (*pp != 0) break;
1603        if ((pp = extend_inputline(infile, pp, "    > ")) == NULL)
     len = BUFFER_SIZE - (pp - buffer);  
     if (len < 256)  
       {  
       fprintf(outfile, "** Expression too long - missing delimiter?\n");  
       goto SKIP_DATA;  
       }  
   
     if (infile == stdin) printf("    > ");  
     if (fgets((char *)pp, len, infile) == NULL)  
1604        {        {
1605        fprintf(outfile, "** Unexpected EOF\n");        fprintf(outfile, "** Unexpected EOF\n");
1606        done = 1;        done = 1;
# Line 780  while (!done) Line 1609  while (!done)
1609      if (infile != stdin) fprintf(outfile, "%s", (char *)pp);      if (infile != stdin) fprintf(outfile, "%s", (char *)pp);
1610      }      }
1611    
1612      /* The buffer may have moved while being extended; reset the start of data
1613      pointer to the correct relative point in the buffer. */
1614    
1615      p = buffer + poffset;
1616    
1617    /* If the first character after the delimiter is backslash, make    /* If the first character after the delimiter is backslash, make
1618    the pattern end with backslash. This is purely to provide a way    the pattern end with backslash. This is purely to provide a way
1619    of testing for the error message when a pattern ends with backslash. */    of testing for the error message when a pattern ends with backslash. */
# Line 795  while (!done) Line 1629  while (!done)
1629    /* Look for options after final delimiter */    /* Look for options after final delimiter */
1630    
1631    options = 0;    options = 0;
   study_options = 0;  
1632    log_store = showstore;  /* default from command line */    log_store = showstore;  /* default from command line */
1633    
1634    while (*pp != 0)    while (*pp != 0)
1635      {      {
1636      switch (*pp++)      switch (*pp++)
1637        {        {
1638          case 'f': options |= PCRE_FIRSTLINE; break;
1639        case 'g': do_g = 1; break;        case 'g': do_g = 1; break;
1640        case 'i': options |= PCRE_CASELESS; break;        case 'i': options |= PCRE_CASELESS; break;
1641        case 'm': options |= PCRE_MULTILINE; break;        case 'm': options |= PCRE_MULTILINE; break;
1642        case 's': options |= PCRE_DOTALL; break;        case 's': options |= PCRE_DOTALL; break;
1643        case 'x': options |= PCRE_EXTENDED; break;        case 'x': options |= PCRE_EXTENDED; break;
1644    
1645        case '+': do_showrest = 1; break;        case '+':
1646          if (do_showrest) do_showcaprest = 1; else do_showrest = 1;
1647          break;
1648    
1649          case '=': do_allcaps = 1; break;
1650        case 'A': options |= PCRE_ANCHORED; break;        case 'A': options |= PCRE_ANCHORED; break;
1651          case 'B': do_debug = 1; break;
1652        case 'C': options |= PCRE_AUTO_CALLOUT; break;        case 'C': options |= PCRE_AUTO_CALLOUT; break;
1653        case 'D': do_debug = do_showinfo = 1; break;        case 'D': do_debug = do_showinfo = 1; break;
1654        case 'E': options |= PCRE_DOLLAR_ENDONLY; break;        case 'E': options |= PCRE_DOLLAR_ENDONLY; break;
1655        case 'F': do_flip = 1; break;        case 'F': do_flip = 1; break;
1656        case 'G': do_G = 1; break;        case 'G': do_G = 1; break;
1657        case 'I': do_showinfo = 1; break;        case 'I': do_showinfo = 1; break;
1658          case 'J': options |= PCRE_DUPNAMES; break;
1659          case 'K': do_mark = 1; break;
1660        case 'M': log_store = 1; break;        case 'M': log_store = 1; break;
1661        case 'N': options |= PCRE_NO_AUTO_CAPTURE; break;        case 'N': options |= PCRE_NO_AUTO_CAPTURE; break;
1662    
# Line 823  while (!done) Line 1664  while (!done)
1664        case 'P': do_posix = 1; break;        case 'P': do_posix = 1; break;
1665  #endif  #endif
1666    
1667        case 'S': do_study = 1; break;        case 'S':
1668          if (do_study == 0)
1669            {
1670            do_study = 1;
1671            if (*pp == '+')
1672              {
1673              study_options |= PCRE_STUDY_JIT_COMPILE;
1674              pp++;
1675              }
1676            }
1677          else
1678            {
1679            do_study = 0;
1680            no_force_study = 1;
1681            }
1682          break;
1683    
1684        case 'U': options |= PCRE_UNGREEDY; break;        case 'U': options |= PCRE_UNGREEDY; break;
1685          case 'W': options |= PCRE_UCP; break;
1686        case 'X': options |= PCRE_EXTRA; break;        case 'X': options |= PCRE_EXTRA; break;
1687          case 'Y': options |= PCRE_NO_START_OPTIMISE; break;
1688          case 'Z': debug_lengths = 0; break;
1689        case '8': options |= PCRE_UTF8; use_utf8 = 1; break;        case '8': options |= PCRE_UTF8; use_utf8 = 1; break;
1690        case '?': options |= PCRE_NO_UTF8_CHECK; break;        case '?': options |= PCRE_NO_UTF8_CHECK; break;
1691    
1692          case 'T':
1693          switch (*pp++)
1694            {
1695            case '0': tables = tables0; break;
1696            case '1': tables = tables1; break;
1697    
1698            case '\r':
1699            case '\n':
1700            case ' ':
1701            case 0:
1702            fprintf(outfile, "** Missing table number after /T\n");
1703            goto SKIP_DATA;
1704    
1705            default:
1706            fprintf(outfile, "** Bad table number \"%c\" after /T\n", pp[-1]);
1707            goto SKIP_DATA;
1708            }
1709          break;
1710    
1711        case 'L':        case 'L':
1712        ppp = pp;        ppp = pp;
1713        while (*ppp != '\n' && *ppp != ' ') ppp++;        /* The '\r' test here is so that it works on Windows. */
1714          /* The '0' test is just in case this is an unterminated line. */
1715          while (*ppp != 0 && *ppp != '\n' && *ppp != '\r' && *ppp != ' ') ppp++;
1716        *ppp = 0;        *ppp = 0;
1717        if (setlocale(LC_CTYPE, (const char *)pp) == NULL)        if (setlocale(LC_CTYPE, (const char *)pp) == NULL)
1718          {          {
1719          fprintf(outfile, "** Failed to set locale \"%s\"\n", pp);          fprintf(outfile, "** Failed to set locale \"%s\"\n", pp);
1720          goto SKIP_DATA;          goto SKIP_DATA;
1721          }          }
1722          locale_set = 1;
1723        tables = pcre_maketables();        tables = pcre_maketables();
1724        pp = ppp;        pp = ppp;
1725        break;        break;
# Line 849  while (!done) Line 1731  while (!done)
1731        *pp = 0;        *pp = 0;
1732        break;        break;
1733    
1734        case '\n': case ' ': break;        case '<':
1735            {
1736            if (strncmpic(pp, (uschar *)"JS>", 3) == 0)
1737              {
1738              options |= PCRE_JAVASCRIPT_COMPAT;
1739              pp += 3;
1740              }
1741            else
1742              {
1743              int x = check_newline(pp, outfile);
1744              if (x == 0) goto SKIP_DATA;
1745              options |= x;
1746              while (*pp++ != '>');
1747              }
1748            }
1749          break;
1750    
1751          case '\r':                      /* So that it works in Windows */
1752          case '\n':
1753          case ' ':
1754          break;
1755    
1756        default:        default:
1757        fprintf(outfile, "** Unknown option '%c'\n", pp[-1]);        fprintf(outfile, "** Unknown option '%c'\n", pp[-1]);
# Line 869  while (!done) Line 1771  while (!done)
1771    
1772      if ((options & PCRE_CASELESS) != 0) cflags |= REG_ICASE;      if ((options & PCRE_CASELESS) != 0) cflags |= REG_ICASE;
1773      if ((options & PCRE_MULTILINE) != 0) cflags |= REG_NEWLINE;      if ((options & PCRE_MULTILINE) != 0) cflags |= REG_NEWLINE;
1774        if ((options & PCRE_DOTALL) != 0) cflags |= REG_DOTALL;
1775        if ((options & PCRE_NO_AUTO_CAPTURE) != 0) cflags |= REG_NOSUB;
1776        if ((options & PCRE_UTF8) != 0) cflags |= REG_UTF8;
1777        if ((options & PCRE_UCP) != 0) cflags |= REG_UCP;
1778        if ((options & PCRE_UNGREEDY) != 0) cflags |= REG_UNGREEDY;
1779    
1780      rc = regcomp(&preg, (char *)p, cflags);      rc = regcomp(&preg, (char *)p, cflags);
1781    
1782      /* Compilation failed; go back for another re, skipping to blank line      /* Compilation failed; go back for another re, skipping to blank line
# Line 876  while (!done) Line 1784  while (!done)
1784    
1785      if (rc != 0)      if (rc != 0)
1786        {        {
1787        (void)regerror(rc, &preg, (char *)buffer, BUFFER_SIZE);        (void)regerror(rc, &preg, (char *)buffer, buffer_size);
1788        fprintf(outfile, "Failed: POSIX code %d: %s\n", rc, buffer);        fprintf(outfile, "Failed: POSIX code %d: %s\n", rc, buffer);
1789        goto SKIP_DATA;        goto SKIP_DATA;
1790        }        }
# Line 888  while (!done) Line 1796  while (!done)
1796  #endif  /* !defined NOPOSIX */  #endif  /* !defined NOPOSIX */
1797    
1798      {      {
1799      if (timeit)      unsigned long int get_options;
1800    
1801        if (timeit > 0)
1802        {        {
1803        register int i;        register int i;
1804        clock_t time_taken;        clock_t time_taken;
1805        clock_t start_time = clock();        clock_t start_time = clock();
1806        for (i = 0; i < LOOPREPEAT; i++)        for (i = 0; i < timeit; i++)
1807          {          {
1808          re = pcre_compile((char *)p, options, &error, &erroroffset, tables);          re = pcre_compile((char *)p, options, &error, &erroroffset, tables);
1809          if (re != NULL) free(re);          if (re != NULL) free(re);
1810          }          }
1811        time_taken = clock() - start_time;        time_taken = clock() - start_time;
1812        fprintf(outfile, "Compile time %.3f milliseconds\n",        fprintf(outfile, "Compile time %.4f milliseconds\n",
1813          (((double)time_taken * 1000.0) / (double)LOOPREPEAT) /          (((double)time_taken * 1000.0) / (double)timeit) /
1814            (double)CLOCKS_PER_SEC);            (double)CLOCKS_PER_SEC);
1815        }        }
1816    
# Line 917  while (!done) Line 1827  while (!done)
1827          {          {
1828          for (;;)          for (;;)
1829            {            {
1830            if (fgets((char *)buffer, BUFFER_SIZE, infile) == NULL)            if (extend_inputline(infile, buffer, NULL) == NULL)
1831              {              {
1832              done = 1;              done = 1;
1833              goto CONTINUE;              goto CONTINUE;
# Line 931  while (!done) Line 1841  while (!done)
1841        goto CONTINUE;        goto CONTINUE;
1842        }        }
1843    
1844      /* Compilation succeeded; print data if required. There are now two      /* Compilation succeeded. It is now possible to set the UTF-8 option from
1845      info-returning functions. The old one has a limited interface and      within the regex; check for this so that we know how to process the data
1846      returns only limited data. Check that it agrees with the newer one. */      lines. */
1847    
1848        new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options);
1849        if ((get_options & PCRE_UTF8) != 0) use_utf8 = 1;
1850    
1851        /* Print information if required. There are now two info-returning
1852        functions. The old one has a limited interface and returns only limited
1853        data. Check that it agrees with the newer one. */
1854    
1855      if (log_store)      if (log_store)
1856        fprintf(outfile, "Memory allocation (code space): %d\n",        fprintf(outfile, "Memory allocation (code space): %d\n",
# Line 947  while (!done) Line 1864  while (!done)
1864      true_size = ((real_pcre *)re)->size;      true_size = ((real_pcre *)re)->size;
1865      regex_gotten_store = gotten_store;      regex_gotten_store = gotten_store;
1866    
1867      /* If /S was present, study the regexp to generate additional info to      /* If -s or /S was present, study the regex to generate additional info to
1868      help with the matching. */      help with the matching, unless the pattern has the SS option, which
1869        suppresses the effect of /S (used for a few test patterns where studying is
1870        never sensible). */
1871    
1872      if (do_study)      if (do_study || (force_study >= 0 && !no_force_study))
1873        {        {
1874        if (timeit)        if (timeit > 0)
1875          {          {
1876          register int i;          register int i;
1877          clock_t time_taken;          clock_t time_taken;
1878          clock_t start_time = clock();          clock_t start_time = clock();
1879          for (i = 0; i < LOOPREPEAT; i++)          for (i = 0; i < timeit; i++)
1880            extra = pcre_study(re, study_options, &error);            extra = pcre_study(re, study_options | force_study_options, &error);
1881          time_taken = clock() - start_time;          time_taken = clock() - start_time;
1882          if (extra != NULL) free(extra);          if (extra != NULL) pcre_free_study(extra);
1883          fprintf(outfile, "  Study time %.3f milliseconds\n",          fprintf(outfile, "  Study time %.4f milliseconds\n",
1884            (((double)time_taken * 1000.0) / (double)LOOPREPEAT) /            (((double)time_taken * 1000.0) / (double)timeit) /
1885              (double)CLOCKS_PER_SEC);              (double)CLOCKS_PER_SEC);
1886          }          }
1887        extra = pcre_study(re, study_options, &error);        extra = pcre_study(re, study_options | force_study_options, &error);
1888        if (error != NULL)        if (error != NULL)
1889          fprintf(outfile, "Failed to study: %s\n", error);          fprintf(outfile, "Failed to study: %s\n", error);
1890        else if (extra != NULL)        else if (extra != NULL)
1891          true_study_size = ((pcre_study_data *)(extra->study_data))->size;          true_study_size = ((pcre_study_data *)(extra->study_data))->size;
1892        }        }
1893    
1894        /* If /K was present, we set up for handling MARK data. */
1895    
1896        if (do_mark)
1897          {
1898          if (extra == NULL)
1899            {
1900            extra = (pcre_extra *)malloc(sizeof(pcre_extra));
1901            extra->flags = 0;
1902            }
1903          extra->mark = &markptr;
1904          extra->flags |= PCRE_EXTRA_MARK;
1905          }
1906    
1907      /* If the 'F' option was present, we flip the bytes of all the integer      /* If the 'F' option was present, we flip the bytes of all the integer
1908      fields in the regex data block and the study block. This is to make it      fields in the regex data block and the study block. This is to make it
1909      possible to test PCRE's handling of byte-flipped patterns, e.g. those      possible to test PCRE's handling of byte-flipped patterns, e.g. those
# Line 980  while (!done) Line 1912  while (!done)
1912      if (do_flip)      if (do_flip)
1913        {        {
1914        real_pcre *rre = (real_pcre *)re;        real_pcre *rre = (real_pcre *)re;
1915        rre->magic_number = byteflip(rre->magic_number, sizeof(rre->magic_number));        rre->magic_number =
1916            byteflip(rre->magic_number, sizeof(rre->magic_number));
1917        rre->size = byteflip(rre->size, sizeof(rre->size));        rre->size = byteflip(rre->size, sizeof(rre->size));
1918        rre->options = byteflip(rre->options, sizeof(rre->options));        rre->options = byteflip(rre->options, sizeof(rre->options));
1919        rre->top_bracket = byteflip(rre->top_bracket, sizeof(rre->top_bracket));        rre->flags = (pcre_uint16)byteflip(rre->flags, sizeof(rre->flags));
1920        rre->top_backref = byteflip(rre->top_backref, sizeof(rre->top_backref));        rre->top_bracket =
1921        rre->first_byte = byteflip(rre->first_byte, sizeof(rre->first_byte));          (pcre_uint16)byteflip(rre->top_bracket, sizeof(rre->top_bracket));
1922        rre->req_byte = byteflip(rre->req_byte, sizeof(rre->req_byte));        rre->top_backref =
1923        rre->name_table_offset = byteflip(rre->name_table_offset,          (pcre_uint16)byteflip(rre->top_backref, sizeof(rre->top_backref));
1924          rre->first_byte =
1925            (pcre_uint16)byteflip(rre->first_byte, sizeof(rre->first_byte));
1926          rre->req_byte =
1927            (pcre_uint16)byteflip(rre->req_byte, sizeof(rre->req_byte));
1928          rre->name_table_offset = (pcre_uint16)byteflip(rre->name_table_offset,
1929          sizeof(rre->name_table_offset));          sizeof(rre->name_table_offset));
1930        rre->name_entry_size = byteflip(rre->name_entry_size,        rre->name_entry_size = (pcre_uint16)byteflip(rre->name_entry_size,
1931          sizeof(rre->name_entry_size));          sizeof(rre->name_entry_size));
1932        rre->name_count = byteflip(rre->name_count, sizeof(rre->name_count));        rre->name_count = (pcre_uint16)byteflip(rre->name_count,
1933            sizeof(rre->name_count));
1934    
1935        if (extra != NULL)        if (extra != NULL)
1936          {          {
1937          pcre_study_data *rsd = (pcre_study_data *)(extra->study_data);          pcre_study_data *rsd = (pcre_study_data *)(extra->study_data);
1938          rsd->size = byteflip(rsd->size, sizeof(rsd->size));          rsd->size = byteflip(rsd->size, sizeof(rsd->size));
1939          rsd->options = byteflip(rsd->options, sizeof(rsd->options));          rsd->flags = byteflip(rsd->flags, sizeof(rsd->flags));
1940            rsd->minlength = byteflip(rsd->minlength, sizeof(rsd->minlength));
1941          }          }
1942        }        }
1943    
# Line 1005  while (!done) Line 1945  while (!done)
1945    
1946      SHOW_INFO:      SHOW_INFO:
1947    
1948        if (do_debug)
1949          {
1950          fprintf(outfile, "------------------------------------------------------------------\n");
1951          pcre_printint(re, outfile, debug_lengths);
1952          }
1953    
1954        /* We already have the options in get_options (see above) */
1955    
1956      if (do_showinfo)      if (do_showinfo)
1957        {        {
1958        unsigned long int get_options, all_options;        unsigned long int all_options;
1959    #if !defined NOINFOCHECK
1960        int old_first_char, old_options, old_count;        int old_first_char, old_options, old_count;
1961        int count, backrefmax, first_char, need_char;  #endif
1962          int count, backrefmax, first_char, need_char, okpartial, jchanged,
1963            hascrorlf;
1964        int nameentrysize, namecount;        int nameentrysize, namecount;
1965        const uschar *nametable;        const uschar *nametable;
1966    
       if (do_debug)  
         {  
         fprintf(outfile, "------------------------------------------------------------------\n");  
         print_internals(re, outfile);  
         }  
   
       new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options);  
1967        new_info(re, NULL, PCRE_INFO_SIZE, &size);        new_info(re, NULL, PCRE_INFO_SIZE, &size);
1968        new_info(re, NULL, PCRE_INFO_CAPTURECOUNT, &count);        new_info(re, NULL, PCRE_INFO_CAPTURECOUNT, &count);
1969        new_info(re, NULL, PCRE_INFO_BACKREFMAX, &backrefmax);        new_info(re, NULL, PCRE_INFO_BACKREFMAX, &backrefmax);
# Line 1028  while (!done) Line 1972  while (!done)
1972        new_info(re, NULL, PCRE_INFO_NAMEENTRYSIZE, &nameentrysize);        new_info(re, NULL, PCRE_INFO_NAMEENTRYSIZE, &nameentrysize);
1973        new_info(re, NULL, PCRE_INFO_NAMECOUNT, &namecount);        new_info(re, NULL, PCRE_INFO_NAMECOUNT, &namecount);
1974        new_info(re, NULL, PCRE_INFO_NAMETABLE, (void *)&nametable);        new_info(re, NULL, PCRE_INFO_NAMETABLE, (void *)&nametable);
1975          new_info(re, NULL, PCRE_INFO_OKPARTIAL, &okpartial);
1976          new_info(re, NULL, PCRE_INFO_JCHANGED, &jchanged);
1977          new_info(re, NULL, PCRE_INFO_HASCRORLF, &hascrorlf);
1978    
1979    #if !defined NOINFOCHECK
1980        old_count = pcre_info(re, &old_options, &old_first_char);        old_count = pcre_info(re, &old_options, &old_first_char);
1981        if (count < 0) fprintf(outfile,        if (count < 0) fprintf(outfile,
1982          "Error %d from pcre_info()\n", count);          "Error %d from pcre_info()\n", count);
# Line 1046  while (!done) Line 1994  while (!done)
1994            "Options disagreement: pcre_fullinfo=%ld pcre_info=%d\n",            "Options disagreement: pcre_fullinfo=%ld pcre_info=%d\n",
1995              get_options, old_options);              get_options, old_options);
1996          }          }
1997    #endif
1998    
1999        if (size != regex_gotten_store) fprintf(outfile,        if (size != regex_gotten_store) fprintf(outfile,
2000          "Size disagreement: pcre_fullinfo=%d call to malloc for %d\n",          "Size disagreement: pcre_fullinfo=%d call to malloc for %d\n",
2001          size, regex_gotten_store);          (int)size, (int)regex_gotten_store);
2002    
2003        fprintf(outfile, "Capturing subpattern count = %d\n", count);        fprintf(outfile, "Capturing subpattern count = %d\n", count);
2004        if (backrefmax > 0)        if (backrefmax > 0)
# Line 1067  while (!done) Line 2016  while (!done)
2016            }            }
2017          }          }
2018    
2019        /* The NOPARTIAL bit is a private bit in the options, so we have        if (!okpartial) fprintf(outfile, "Partial matching not supported\n");
2020        to fish it out via out back door */        if (hascrorlf) fprintf(outfile, "Contains explicit CR or LF match\n");
2021    
2022        all_options = ((real_pcre *)re)->options;        all_options = ((real_pcre *)re)->options;
2023        if (do_flip)        if (do_flip) all_options = byteflip(all_options, sizeof(all_options));
         {  
         all_options = byteflip(all_options, sizeof(all_options));  
         }  
   
       if ((all_options & PCRE_NOPARTIAL) != 0)  
         fprintf(outfile, "Partial matching not supported\n");  
2024    
2025        if (get_options == 0) fprintf(outfile, "No options\n");        if (get_options == 0) fprintf(outfile, "No options\n");
2026          else fprintf(outfile, "Options:%s%s%s%s%s%s%s%s%s%s\n",          else fprintf(outfile, "Options:%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s\n",
2027            ((get_options & PCRE_ANCHORED) != 0)? " anchored" : "",            ((get_options & PCRE_ANCHORED) != 0)? " anchored" : "",
2028            ((get_options & PCRE_CASELESS) != 0)? " caseless" : "",            ((get_options & PCRE_CASELESS) != 0)? " caseless" : "",
2029            ((get_options & PCRE_EXTENDED) != 0)? " extended" : "",            ((get_options & PCRE_EXTENDED) != 0)? " extended" : "",
2030            ((get_options & PCRE_MULTILINE) != 0)? " multiline" : "",            ((get_options & PCRE_MULTILINE) != 0)? " multiline" : "",
2031              ((get_options & PCRE_FIRSTLINE) != 0)? " firstline" : "",
2032            ((get_options & PCRE_DOTALL) != 0)? " dotall" : "",            ((get_options & PCRE_DOTALL) != 0)? " dotall" : "",
2033              ((get_options & PCRE_BSR_ANYCRLF) != 0)? " bsr_anycrlf" : "",
2034              ((get_options & PCRE_BSR_UNICODE) != 0)? " bsr_unicode" : "",
2035            ((get_options & PCRE_DOLLAR_ENDONLY) != 0)? " dollar_endonly" : "",            ((get_options & PCRE_DOLLAR_ENDONLY) != 0)? " dollar_endonly" : "",
2036            ((get_options & PCRE_EXTRA) != 0)? " extra" : "",            ((get_options & PCRE_EXTRA) != 0)? " extra" : "",
2037            ((get_options & PCRE_UNGREEDY) != 0)? " ungreedy" : "",            ((get_options & PCRE_UNGREEDY) != 0)? " ungreedy" : "",
2038              ((get_options & PCRE_NO_AUTO_CAPTURE) != 0)? " no_auto_capture" : "",
2039            ((get_options & PCRE_UTF8) != 0)? " utf8" : "",            ((get_options & PCRE_UTF8) != 0)? " utf8" : "",
2040            ((get_options & PCRE_NO_UTF8_CHECK) != 0)? " no_utf8_check" : "");            ((get_options & PCRE_UCP) != 0)? " ucp" : "",
2041              ((get_options & PCRE_NO_UTF8_CHECK) != 0)? " no_utf8_check" : "",
2042              ((get_options & PCRE_NO_START_OPTIMIZE) != 0)? " no_start_optimize" : "",
2043              ((get_options & PCRE_DUPNAMES) != 0)? " dupnames" : "");
2044    
2045          if (jchanged) fprintf(outfile, "Duplicate name status changes\n");
2046    
2047          switch (get_options & PCRE_NEWLINE_BITS)
2048            {
2049            case PCRE_NEWLINE_CR:
2050            fprintf(outfile, "Forced newline sequence: CR\n");
2051            break;
2052    
2053        if (((((real_pcre *)re)->options) & PCRE_ICHANGED) != 0)          case PCRE_NEWLINE_LF:
2054          fprintf(outfile, "Case state changes\n");          fprintf(outfile, "Forced newline sequence: LF\n");
2055            break;
2056    
2057            case PCRE_NEWLINE_CRLF:
2058            fprintf(outfile, "Forced newline sequence: CRLF\n");
2059            break;
2060    
2061            case PCRE_NEWLINE_ANYCRLF:
2062            fprintf(outfile, "Forced newline sequence: ANYCRLF\n");
2063            break;
2064    
2065            case PCRE_NEWLINE_ANY:
2066            fprintf(outfile, "Forced newline sequence: ANY\n");
2067            break;
2068    
2069            default:
2070            break;
2071            }
2072    
2073        if (first_char == -1)        if (first_char == -1)
2074          {          {
2075          fprintf(outfile, "First char at start or follows \\n\n");          fprintf(outfile, "First char at start or follows newline\n");
2076          }          }
2077        else if (first_char < 0)        else if (first_char < 0)
2078          {          {
# Line 1108  while (!done) Line 2083  while (!done)
2083          int ch = first_char & 255;          int ch = first_char & 255;
2084          const char *caseless = ((first_char & REQ_CASELESS) == 0)?          const char *caseless = ((first_char & REQ_CASELESS) == 0)?
2085            "" : " (caseless)";            "" : " (caseless)";
2086          if (isprint(ch))          if (PRINTHEX(ch))
2087            fprintf(outfile, "First char = \'%c\'%s\n", ch, caseless);            fprintf(outfile, "First char = \'%c\'%s\n", ch, caseless);
2088          else          else
2089            fprintf(outfile, "First char = %d%s\n", ch, caseless);            fprintf(outfile, "First char = %d%s\n", ch, caseless);
# Line 1123  while (!done) Line 2098  while (!done)
2098          int ch = need_char & 255;          int ch = need_char & 255;
2099          const char *caseless = ((need_char & REQ_CASELESS) == 0)?          const char *caseless = ((need_char & REQ_CASELESS) == 0)?
2100            "" : " (caseless)";            "" : " (caseless)";
2101          if (isprint(ch))          if (PRINTHEX(ch))
2102            fprintf(outfile, "Need char = \'%c\'%s\n", ch, caseless);            fprintf(outfile, "Need char = \'%c\'%s\n", ch, caseless);
2103          else          else
2104            fprintf(outfile, "Need char = %d%s\n", ch, caseless);            fprintf(outfile, "Need char = %d%s\n", ch, caseless);
# Line 1132  while (!done) Line 2107  while (!done)
2107        /* Don't output study size; at present it is in any case a fixed        /* Don't output study size; at present it is in any case a fixed
2108        value, but it varies, depending on the computer architecture, and        value, but it varies, depending on the computer architecture, and
2109        so messes up the test suite. (And with the /F option, it might be        so messes up the test suite. (And with the /F option, it might be
2110        flipped.) */        flipped.) If study was forced by an external -s, don't show this
2111          information unless -i or -d was also present. This means that, except
2112          when auto-callouts are involved, the output from runs with and without
2113          -s should be identical. */
2114    
2115        if (do_study)        if (do_study || (force_study >= 0 && showinfo && !no_force_study))
2116          {          {
2117          if (extra == NULL)          if (extra == NULL)
2118            fprintf(outfile, "Study returned NULL\n");            fprintf(outfile, "Study returned NULL\n");
2119          else          else
2120            {            {
2121            uschar *start_bits = NULL;            uschar *start_bits = NULL;
2122            new_info(re, extra, PCRE_INFO_FIRSTTABLE, &start_bits);            int minlength;
2123    
2124              new_info(re, extra, PCRE_INFO_MINLENGTH, &minlength);
2125              fprintf(outfile, "Subject length lower bound = %d\n", minlength);
2126    
2127              new_info(re, extra, PCRE_INFO_FIRSTTABLE, &start_bits);
2128            if (start_bits == NULL)            if (start_bits == NULL)
2129              fprintf(outfile, "No starting byte set\n");              fprintf(outfile, "No set of starting bytes\n");
2130            else            else
2131              {              {
2132              int i;              int i;
# Line 1159  while (!done) Line 2141  while (!done)
2141                    fprintf(outfile, "\n  ");                    fprintf(outfile, "\n  ");
2142                    c = 2;                    c = 2;
2143                    }                    }
2144                  if (isprint(i) && i != ' ')                  if (PRINTHEX(i) && i != ' ')
2145                    {                    {
2146                    fprintf(outfile, "%c ", i);                    fprintf(outfile, "%c ", i);
2147                    c += 2;                    c += 2;
# Line 1174  while (!done) Line 2156  while (!done)
2156              fprintf(outfile, "\n");              fprintf(outfile, "\n");
2157              }              }
2158            }            }
2159    
2160            /* Show this only if the JIT was set by /S, not by -s. */
2161    
2162            if ((study_options & PCRE_STUDY_JIT_COMPILE) != 0)
2163              {
2164              int jit;
2165              new_info(re, extra, PCRE_INFO_JIT, &jit);
2166              if (jit)
2167                fprintf(outfile, "JIT study was successful\n");
2168              else
2169    #ifdef SUPPORT_JIT
2170                fprintf(outfile, "JIT study was not successful\n");
2171    #else
2172                fprintf(outfile, "JIT support is not available in this version of PCRE\n");
2173    #endif
2174              }
2175          }          }
2176        }        }
2177    
# Line 1191  while (!done) Line 2189  while (!done)
2189        else        else
2190          {          {
2191          uschar sbuf[8];          uschar sbuf[8];
2192          sbuf[0] = (true_size >> 24)  & 255;          sbuf[0] = (uschar)((true_size >> 24) & 255);
2193          sbuf[1] = (true_size >> 16)  & 255;          sbuf[1] = (uschar)((true_size >> 16) & 255);
2194          sbuf[2] = (true_size >>  8)  & 255;          sbuf[2] = (uschar)((true_size >>  8) & 255);
2195          sbuf[3] = (true_size)  & 255;          sbuf[3] = (uschar)((true_size) & 255);
2196    
2197          sbuf[4] = (true_study_size >> 24)  & 255;          sbuf[4] = (uschar)((true_study_size >> 24) & 255);
2198          sbuf[5] = (true_study_size >> 16)  & 255;          sbuf[5] = (uschar)((true_study_size >> 16) & 255);
2199          sbuf[6] = (true_study_size >>  8)  & 255;          sbuf[6] = (uschar)((true_study_size >>  8) & 255);
2200          sbuf[7] = (true_study_size)  & 255;          sbuf[7] = (uschar)((true_study_size) & 255);
2201    
2202          if (fwrite(sbuf, 1, 8, f) < 8 ||          if (fwrite(sbuf, 1, 8, f) < 8 ||
2203              fwrite(re, 1, true_size, f) < true_size)              fwrite(re, 1, true_size, f) < true_size)
# Line 1208  while (!done) Line 2206  while (!done)
2206            }            }
2207          else          else
2208            {            {
2209            fprintf(outfile, "Compiled regex written to %s\n", to_file);            fprintf(outfile, "Compiled pattern written to %s\n", to_file);
2210    
2211              /* If there is study data, write it. */
2212    
2213            if (extra != NULL)            if (extra != NULL)
2214              {              {
2215              if (fwrite(extra->study_data, 1, true_study_size, f) <              if (fwrite(extra->study_data, 1, true_study_size, f) <
# Line 1222  while (!done) Line 2223  while (!done)
2223            }            }
2224          fclose(f);          fclose(f);
2225          }          }
2226    
2227          new_free(re);
2228          if (extra != NULL) pcre_free_study(extra);
2229          if (locale_set)
2230            {
2231            new_free((void *)tables);
2232            setlocale(LC_CTYPE, "C");
2233            locale_set = 0;
2234            }
2235        continue;  /* With next regex */        continue;  /* With next regex */
2236        }        }
2237      }        /* End of non-POSIX compile */      }        /* End of non-POSIX compile */
# Line 1230  while (!done) Line 2240  while (!done)
2240    
2241    for (;;)    for (;;)
2242      {      {
2243      unsigned char *q;      uschar *q;
2244      unsigned char *bptr = dbuffer;      uschar *bptr;
2245      int *use_offsets = offsets;      int *use_offsets = offsets;
2246      int use_size_offsets = size_offsets;      int use_size_offsets = size_offsets;
2247      int callout_data = 0;      int callout_data = 0;
2248      int callout_data_set = 0;      int callout_data_set = 0;
2249      int count, c;      int count, c;
2250      int copystrings = 0;      int copystrings = 0;
2251      int find_match_limit = 0;      int find_match_limit = default_find_match_limit;
2252      int getstrings = 0;      int getstrings = 0;
2253      int getlist = 0;      int getlist = 0;
2254      int gmatched = 0;      int gmatched = 0;
2255      int start_offset = 0;      int start_offset = 0;
2256        int start_offset_sign = 1;
2257      int g_notempty = 0;      int g_notempty = 0;
2258        int use_dfa = 0;
2259    
2260      options = 0;      options = 0;
2261    
2262        *copynames = 0;
2263        *getnames = 0;
2264    
2265        copynamesptr = copynames;
2266        getnamesptr = getnames;
2267    
2268      pcre_callout = callout;      pcre_callout = callout;
2269      first_callout = 1;      first_callout = 1;
2270        last_callout_mark = NULL;
2271      callout_extra = 0;      callout_extra = 0;
2272      callout_count = 0;      callout_count = 0;
2273      callout_fail_count = 999999;      callout_fail_count = 999999;
2274      callout_fail_id = -1;      callout_fail_id = -1;
2275      show_malloc = 0;      show_malloc = 0;
2276    
2277      if (infile == stdin) printf("data> ");      if (extra != NULL) extra->flags &=
2278      if (fgets((char *)buffer, BUFFER_SIZE, infile) == NULL)        ~(PCRE_EXTRA_MATCH_LIMIT|PCRE_EXTRA_MATCH_LIMIT_RECURSION);
2279    
2280        len = 0;
2281        for (;;)
2282        {        {
2283        done = 1;        if (extend_inputline(infile, buffer + len, "data> ") == NULL)
2284        goto CONTINUE;          {
2285            if (len > 0)    /* Reached EOF without hitting a newline */
2286              {
2287              fprintf(outfile, "\n");
2288              break;
2289              }
2290            done = 1;
2291            goto CONTINUE;
2292            }
2293          if (infile != stdin) fprintf(outfile, "%s", (char *)buffer);
2294          len = (int)strlen((char *)buffer);
2295          if (buffer[len-1] == '\n') break;
2296        }        }
     if (infile != stdin) fprintf(outfile, "%s", (char *)buffer);  
2297    
     len = (int)strlen((char *)buffer);  
2298      while (len > 0 && isspace(buffer[len-1])) len--;      while (len > 0 && isspace(buffer[len-1])) len--;
2299      buffer[len] = 0;      buffer[len] = 0;
2300      if (len == 0) break;      if (len == 0) break;
# Line 1271  while (!done) Line 2302  while (!done)
2302      p = buffer;      p = buffer;
2303      while (isspace(*p)) p++;      while (isspace(*p)) p++;
2304    
2305      q = dbuffer;      bptr = q = dbuffer;
2306      while ((c = *p++) != 0)      while ((c = *p++) != 0)
2307        {        {
2308        int i = 0;        int i = 0;
# Line 1293  while (!done) Line 2324  while (!done)
2324          c -= '0';          c -= '0';
2325          while (i++ < 2 && isdigit(*p) && *p != '8' && *p != '9')          while (i++ < 2 && isdigit(*p) && *p != '8' && *p != '9')
2326            c = c * 8 + *p++ - '0';            c = c * 8 + *p++ - '0';
2327    
2328    #if !defined NOUTF8
2329            if (use_utf8 && c > 255)
2330              {
2331              unsigned char buff8[8];
2332              int ii, utn;
2333              utn = ord2utf8(c, buff8);
2334              for (ii = 0; ii < utn - 1; ii++) *q++ = buff8[ii];
2335              c = buff8[ii];   /* Last byte */
2336              }
2337    #endif
2338          break;          break;
2339    
2340          case 'x':          case 'x':
2341    
2342          /* Handle \x{..} specially - new Perl thing for utf8 */          /* Handle \x{..} specially - new Perl thing for utf8 */
2343    
2344    #if !defined NOUTF8
2345          if (*p == '{')          if (*p == '{')
2346            {            {
2347            unsigned char *pt = p;            unsigned char *pt = p;
2348            c = 0;            c = 0;
2349            while (isxdigit(*(++pt)))            while (isxdigit(*(++pt)))
2350              c = c * 16 + tolower(*pt) - ((isdigit(*pt))? '0' : 'W');              c = c * 16 + tolower(*pt) - ((isdigit(*pt))? '0' : 'a' - 10);
2351            if (*pt == '}')            if (*pt == '}')
2352              {              {
2353              unsigned char buff8[8];              unsigned char buff8[8];
2354              int ii, utn;              int ii, utn;
2355              utn = ord2utf8(c, buff8);              if (use_utf8)
2356              for (ii = 0; ii < utn - 1; ii++) *q++ = buff8[ii];                {
2357              c = buff8[ii];   /* Last byte */                utn = ord2utf8(c, buff8);
2358                  for (ii = 0; ii < utn - 1; ii++) *q++ = buff8[ii];
2359                  c = buff8[ii];   /* Last byte */
2360                  }
2361                else
2362                 {
2363                 if (c > 255)
2364                   fprintf(outfile, "** Character \\x{%x} is greater than 255 and "
2365                     "UTF-8 mode is not enabled.\n"
2366                     "** Truncation will probably give the wrong result.\n", c);
2367                 }
2368              p = pt + 1;              p = pt + 1;
2369              break;              break;
2370              }              }
2371            /* Not correct form; fall through */            /* Not correct form; fall through */
2372            }            }
2373    #endif
2374    
2375          /* Ordinary \x */          /* Ordinary \x */
2376    
2377          c = 0;          c = 0;
2378          while (i++ < 2 && isxdigit(*p))          while (i++ < 2 && isxdigit(*p))
2379            {            {
2380            c = c * 16 + tolower(*p) - ((isdigit(*p))? '0' : 'W');            c = c * 16 + tolower(*p) - ((isdigit(*p))? '0' : 'a' - 10);
2381            p++;            p++;
2382            }            }
2383          break;          break;
# Line 1333  while (!done) Line 2387  while (!done)
2387          continue;          continue;
2388    
2389          case '>':          case '>':
2390            if (*p == '-')
2391              {
2392              start_offset_sign = -1;
2393              p++;
2394              }
2395          while(isdigit(*p)) start_offset = start_offset * 10 + *p++ - '0';          while(isdigit(*p)) start_offset = start_offset * 10 + *p++ - '0';
2396            start_offset *= start_offset_sign;
2397          continue;          continue;
2398    
2399          case 'A':  /* Option setting */          case 'A':  /* Option setting */
# Line 1352  while (!done) Line 2412  while (!done)
2412            }            }
2413          else if (isalnum(*p))          else if (isalnum(*p))
2414            {            {
2415            uschar name[256];            uschar *npp = copynamesptr;
           uschar *npp = name;  
2416            while (isalnum(*p)) *npp++ = *p++;            while (isalnum(*p)) *npp++ = *p++;
2417              *npp++ = 0;
2418            *npp = 0;            *npp = 0;
2419            n = pcre_get_stringnumber(re, (char *)name);            n = pcre_get_stringnumber(re, (char *)copynamesptr);
2420            if (n < 0)            if (n < 0)
2421              fprintf(outfile, "no parentheses with name \"%s\"\n", name);              fprintf(outfile, "no parentheses with name \"%s\"\n", copynamesptr);
2422            else copystrings |= 1 << n;            copynamesptr = npp;
2423            }            }
2424          else if (*p == '+')          else if (*p == '+')
2425            {            {
# Line 1397  while (!done) Line 2457  while (!done)
2457            }            }
2458          continue;          continue;
2459    
2460    #if !defined NODFA
2461            case 'D':
2462    #if !defined NOPOSIX
2463            if (posix || do_posix)
2464              printf("** Can't use dfa matching in POSIX mode: \\D ignored\n");
2465            else
2466    #endif
2467              use_dfa = 1;
2468            continue;
2469    #endif
2470    
2471    #if !defined NODFA
2472            case 'F':
2473            options |= PCRE_DFA_SHORTEST;
2474            continue;
2475    #endif
2476    
2477          case 'G':          case 'G':
2478          if (isdigit(*p))          if (isdigit(*p))
2479            {            {
# Line 1405  while (!done) Line 2482  while (!done)
2482            }            }
2483          else if (isalnum(*p))          else if (isalnum(*p))
2484            {            {
2485            uschar name[256];            uschar *npp = getnamesptr;
           uschar *npp = name;  
2486            while (isalnum(*p)) *npp++ = *p++;            while (isalnum(*p)) *npp++ = *p++;
2487              *npp++ = 0;
2488            *npp = 0;            *npp = 0;
2489            n = pcre_get_stringnumber(re, (char *)name);            n = pcre_get_stringnumber(re, (char *)getnamesptr);
2490            if (n < 0)            if (n < 0)
2491              fprintf(outfile, "no parentheses with name \"%s\"\n", name);              fprintf(outfile, "no parentheses with name \"%s\"\n", getnamesptr);
2492            else getstrings |= 1 << n;            getnamesptr = npp;
2493              }
2494            continue;
2495    
2496            case 'J':
2497            while(isdigit(*p)) n = n * 10 + *p++ - '0';
2498            if (extra != NULL
2499                && (extra->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0
2500                && extra->executable_jit != NULL)
2501              {
2502              if (jit_stack != NULL) pcre_jit_stack_free(jit_stack);
2503              jit_stack = pcre_jit_stack_alloc(1, n * 1024);
2504              pcre_assign_jit_stack(extra, jit_callback, jit_stack);
2505            }            }
2506          continue;          continue;
2507    
# Line 1425  while (!done) Line 2514  while (!done)
2514          continue;          continue;
2515    
2516          case 'N':          case 'N':
2517          options |= PCRE_NOTEMPTY;          if ((options & PCRE_NOTEMPTY) != 0)
2518              options = (options & ~PCRE_NOTEMPTY) | PCRE_NOTEMPTY_ATSTART;
2519            else
2520              options |= PCRE_NOTEMPTY;
2521          continue;          continue;
2522    
2523          case 'O':          case 'O':
# Line 1438  while (!done) Line 2530  while (!done)
2530            if (offsets == NULL)            if (offsets == NULL)
2531              {              {
2532              printf("** Failed to get %d bytes of memory for offsets vector\n",              printf("** Failed to get %d bytes of memory for offsets vector\n",
2533                size_offsets_max * sizeof(int));                (int)(size_offsets_max * sizeof(int)));
2534              return 1;              yield = 1;
2535                goto EXIT;
2536              }              }
2537            }            }
2538          use_size_offsets = n;          use_size_offsets = n;
# Line 1447  while (!done) Line 2540  while (!done)
2540          continue;          continue;
2541    
2542          case 'P':          case 'P':
2543          options |= PCRE_PARTIAL;          options |= ((options & PCRE_PARTIAL_SOFT) == 0)?
2544              PCRE_PARTIAL_SOFT : PCRE_PARTIAL_HARD;
2545            continue;
2546    
2547            case 'Q':
2548            while(isdigit(*p)) n = n * 10 + *p++ - '0';
2549            if (extra == NULL)
2550              {
2551              extra = (pcre_extra *)malloc(sizeof(pcre_extra));
2552              extra->flags = 0;
2553              }
2554            extra->flags |= PCRE_EXTRA_MATCH_LIMIT_RECURSION;
2555            extra->match_limit_recursion = n;
2556            continue;
2557    
2558            case 'q':
2559            while(isdigit(*p)) n = n * 10 + *p++ - '0';
2560            if (extra == NULL)
2561              {
2562              extra = (pcre_extra *)malloc(sizeof(pcre_extra));
2563              extra->flags = 0;
2564              }
2565            extra->flags |= PCRE_EXTRA_MATCH_LIMIT;
2566            extra->match_limit = n;
2567            continue;
2568    
2569    #if !defined NODFA
2570            case 'R':
2571            options |= PCRE_DFA_RESTART;
2572          continue;          continue;
2573    #endif
2574    
2575          case 'S':          case 'S':
2576          show_malloc = 1;          show_malloc = 1;
2577          continue;          continue;
2578    
2579            case 'Y':
2580            options |= PCRE_NO_START_OPTIMIZE;
2581            continue;
2582    
2583          case 'Z':          case 'Z':
2584          options |= PCRE_NOTEOL;          options |= PCRE_NOTEOL;
2585          continue;          continue;
# Line 1461  while (!done) Line 2587  while (!done)
2587          case '?':          case '?':
2588          options |= PCRE_NO_UTF8_CHECK;          options |= PCRE_NO_UTF8_CHECK;
2589          continue;          continue;
2590    
2591            case '<':
2592              {
2593              int x = check_newline(p, outfile);
2594              if (x == 0) goto NEXT_DATA;
2595              options |= x;
2596              while (*p++ != '>');
2597              }
2598            continue;
2599          }          }
2600        *q++ = c;        *q++ = c;
2601        }        }
2602      *q = 0;      *q = 0;
2603      len = q - dbuffer;      len = (int)(q - dbuffer);
2604    
2605        /* Move the data to the end of the buffer so that a read over the end of
2606        the buffer will be seen by valgrind, even if it doesn't cause a crash. If
2607        we are using the POSIX interface, we must include the terminating zero. */
2608    
2609    #if !defined NOPOSIX
2610        if (posix || do_posix)
2611          {
2612          memmove(bptr + buffer_size - len - 1, bptr, len + 1);
2613          bptr += buffer_size - len - 1;
2614          }
2615        else
2616    #endif
2617          {
2618          memmove(bptr + buffer_size - len, bptr, len);
2619          bptr += buffer_size - len;
2620          }
2621    
2622        if ((all_use_dfa || use_dfa) && find_match_limit)
2623          {
2624          printf("**Match limit not relevant for DFA matching: ignored\n");
2625          find_match_limit = 0;
2626          }
2627    
2628      /* Handle matching via the POSIX interface, which does not      /* Handle matching via the POSIX interface, which does not
2629      support timing or playing with the match limit or callout data. */      support timing or playing with the match limit or callout data. */
# Line 1480  while (!done) Line 2638  while (!done)
2638          pmatch = (regmatch_t *)malloc(sizeof(regmatch_t) * use_size_offsets);          pmatch = (regmatch_t *)malloc(sizeof(regmatch_t) * use_size_offsets);
2639        if ((options & PCRE_NOTBOL) != 0) eflags |= REG_NOTBOL;        if ((options & PCRE_NOTBOL) != 0) eflags |= REG_NOTBOL;
2640        if ((options & PCRE_NOTEOL) != 0) eflags |= REG_NOTEOL;        if ((options & PCRE_NOTEOL) != 0) eflags |= REG_NOTEOL;
2641          if ((options & PCRE_NOTEMPTY) != 0) eflags |= REG_NOTEMPTY;
2642    
2643        rc = regexec(&preg, (const char *)bptr, use_size_offsets, pmatch, eflags);        rc = regexec(&preg, (const char *)bptr, use_size_offsets, pmatch, eflags);
2644    
2645        if (rc != 0)        if (rc != 0)
2646          {          {
2647          (void)regerror(rc, &preg, (char *)buffer, BUFFER_SIZE);          (void)regerror(rc, &preg, (char *)buffer, buffer_size);
2648          fprintf(outfile, "No match: POSIX code %d: %s\n", rc, buffer);          fprintf(outfile, "No match: POSIX code %d: %s\n", rc, buffer);
2649          }          }
2650          else if ((((const pcre *)preg.re_pcre)->options & PCRE_NO_AUTO_CAPTURE)
2651                  != 0)
2652            {
2653            fprintf(outfile, "Matched with REG_NOSUB\n");
2654            }
2655        else        else
2656          {          {
2657          size_t i;          size_t i;
# Line 1499  while (!done) Line 2663  while (!done)
2663              (void)pchars(dbuffer + pmatch[i].rm_so,              (void)pchars(dbuffer + pmatch[i].rm_so,
2664                pmatch[i].rm_eo - pmatch[i].rm_so, outfile);                pmatch[i].rm_eo - pmatch[i].rm_so, outfile);
2665              fprintf(outfile, "\n");              fprintf(outfile, "\n");
2666              if (i == 0 && do_showrest)              if (do_showcaprest || (i == 0 && do_showrest))
2667                {                {
2668                fprintf(outfile, " 0+ ");                fprintf(outfile, "%2d+ ", (int)i);
2669                (void)pchars(dbuffer + pmatch[i].rm_eo, len - pmatch[i].rm_eo,                (void)pchars(dbuffer + pmatch[i].rm_eo, len - pmatch[i].rm_eo,
2670                  outfile);                  outfile);
2671                fprintf(outfile, "\n");                fprintf(outfile, "\n");
# Line 1519  while (!done) Line 2683  while (!done)
2683    
2684      for (;; gmatched++)    /* Loop for /g or /G */      for (;; gmatched++)    /* Loop for /g or /G */
2685        {        {
2686        if (timeit)        markptr = NULL;
2687    
2688          if (timeitm > 0)
2689          {          {
2690          register int i;          register int i;
2691          clock_t time_taken;          clock_t time_taken;
2692          clock_t start_time = clock();          clock_t start_time = clock();
2693          for (i = 0; i < LOOPREPEAT; i++)  
2694    #if !defined NODFA
2695            if (all_use_dfa || use_dfa)
2696              {
2697              int workspace[1000];
2698              for (i = 0; i < timeitm; i++)
2699                count = pcre_dfa_exec(re, extra, (char *)bptr, len, start_offset,
2700                  options | g_notempty, use_offsets, use_size_offsets, workspace,
2701                  sizeof(workspace)/sizeof(int));
2702              }
2703            else
2704    #endif
2705    
2706            for (i = 0; i < timeitm; i++)
2707            count = pcre_exec(re, extra, (char *)bptr, len,            count = pcre_exec(re, extra, (char *)bptr, len,
2708              start_offset, options | g_notempty, use_offsets, use_size_offsets);              start_offset, options | g_notempty, use_offsets, use_size_offsets);
2709    
2710          time_taken = clock() - start_time;          time_taken = clock() - start_time;
2711          fprintf(outfile, "Execute time %.3f milliseconds\n",          fprintf(outfile, "Execute time %.4f milliseconds\n",
2712            (((double)time_taken * 1000.0) / (double)LOOPREPEAT) /            (((double)time_taken * 1000.0) / (double)timeitm) /
2713              (double)CLOCKS_PER_SEC);              (double)CLOCKS_PER_SEC);
2714          }          }
2715    
2716        /* If find_match_limit is set, we want to do repeated matches with        /* If find_match_limit is set, we want to do repeated matches with
2717        varying limits in order to find the minimum value. */        varying limits in order to find the minimum value for the match limit and
2718          for the recursion limit. The match limits are relevant only to the normal
2719          running of pcre_exec(), so disable the JIT optimization. This makes it
2720          possible to run the same set of tests with and without JIT externally
2721          requested. */
2722    
2723        if (find_match_limit)        if (find_match_limit)
2724          {          {
         int min = 0;  
         int mid = 64;  
         int max = -1;  
   
2725          if (extra == NULL)          if (extra == NULL)
2726            {            {
2727            extra = (pcre_extra *)malloc(sizeof(pcre_extra));            extra = (pcre_extra *)malloc(sizeof(pcre_extra));
2728            extra->flags = 0;            extra->flags = 0;
2729            }            }
2730          extra->flags |= PCRE_EXTRA_MATCH_LIMIT;          else extra->flags &= ~PCRE_EXTRA_EXECUTABLE_JIT;
2731    
2732          for (;;)          (void)check_match_limit(re, extra, bptr, len, start_offset,
2733            {            options|g_notempty, use_offsets, use_size_offsets,
2734            extra->match_limit = mid;            PCRE_EXTRA_MATCH_LIMIT, &(extra->match_limit),
2735            count = pcre_exec(re, extra, (char *)bptr, len, start_offset,            PCRE_ERROR_MATCHLIMIT, "match()");
2736              options | g_notempty, use_offsets, use_size_offsets);  
2737            if (count == PCRE_ERROR_MATCHLIMIT)          count = check_match_limit(re, extra, bptr, len, start_offset,
2738              {            options|g_notempty, use_offsets, use_size_offsets,
2739              /* fprintf(outfile, "Testing match limit = %d\n", mid); */            PCRE_EXTRA_MATCH_LIMIT_RECURSION, &(extra->match_limit_recursion),
2740              min = mid;            PCRE_ERROR_RECURSIONLIMIT, "match() recursion");
             mid = (mid == max - 1)? max : (max > 0)? (min + max)/2 : mid*2;  
             }  
           else if (count >= 0 || count == PCRE_ERROR_NOMATCH ||  
                                  count == PCRE_ERROR_PARTIAL)  
             {  
             if (mid == min + 1)  
               {  
               fprintf(outfile, "Minimum match limit = %d\n", mid);  
               break;  
               }  
             /* fprintf(outfile, "Testing match limit = %d\n", mid); */  
             max = mid;  
             mid = (min + mid)/2;  
             }  
           else break;    /* Some other error */  
           }  
   
         extra->flags &= ~PCRE_EXTRA_MATCH_LIMIT;  
2741          }          }
2742    
2743        /* If callout_data is set, use the interface with additional data */        /* If callout_data is set, use the interface with additional data */
# Line 1597  while (!done) Line 2759  while (!done)
2759        /* The normal case is just to do the match once, with the default        /* The normal case is just to do the match once, with the default
2760        value of match_limit. */        value of match_limit. */
2761    
2762        else  #if !defined NODFA
2763          else if (all_use_dfa || use_dfa)
2764          {          {
2765          count = pcre_exec(re, extra, (char *)bptr, len,          int workspace[1000];
2766            start_offset, options | g_notempty, use_offsets, use_size_offsets);          count = pcre_dfa_exec(re, extra, (char *)bptr, len, start_offset,
2767              options | g_notempty, use_offsets, use_size_offsets, workspace,
2768              sizeof(workspace)/sizeof(int));
2769            if (count == 0)
2770              {
2771              fprintf(outfile, "Matched, but too many subsidiary matches\n");
2772              count = use_size_offsets/2;
2773              }
2774          }          }
2775    #endif
2776    
2777        if (count == 0)        else
2778          {          {
2779          fprintf(outfile, "Matched, but too many substrings\n");          count = pcre_exec(re, extra, (char *)bptr, len,
2780          count = use_size_offsets/3;            start_offset, options | g_notempty, use_offsets, use_size_offsets);
2781            if (count == 0)
2782              {
2783              fprintf(outfile, "Matched, but too many substrings\n");
2784              count = use_size_offsets/3;
2785              }
2786          }          }
2787    
2788        /* Matched */        /* Matched */
2789    
2790        if (count >= 0)        if (count >= 0)
2791          {          {
2792          int i;          int i, maxcount;
2793    
2794    #if !defined NODFA
2795            if (all_use_dfa || use_dfa) maxcount = use_size_offsets/2; else
2796    #endif
2797              maxcount = use_size_offsets/3;
2798    
2799            /* This is a check against a lunatic return value. */
2800    
2801            if (count > maxcount)
2802              {
2803              fprintf(outfile,
2804                "** PCRE error: returned count %d is too big for offset size %d\n",
2805                count, use_size_offsets);
2806              count = use_size_offsets/3;
2807              if (do_g || do_G)
2808                {
2809                fprintf(outfile, "** /%c loop abandoned\n", do_g? 'g' : 'G');
2810                do_g = do_G = FALSE;        /* Break g/G loop */
2811                }
2812              }
2813    
2814            /* do_allcaps requests showing of all captures in the pattern, to check
2815            unset ones at the end. */
2816    
2817            if (do_allcaps)
2818              {
2819              new_info(re, NULL, PCRE_INFO_CAPTURECOUNT, &count);
2820              count++;   /* Allow for full match */
2821              if (count * 2 > use_size_offsets) count = use_size_offsets/2;
2822              }
2823    
2824            /* Output the captured substrings */
2825    
2826          for (i = 0; i < count * 2; i += 2)          for (i = 0; i < count * 2; i += 2)
2827            {            {
2828            if (use_offsets[i] < 0)            if (use_offsets[i] < 0)
2829                {
2830                if (use_offsets[i] != -1)
2831                  fprintf(outfile, "ERROR: bad negative value %d for offset %d\n",
2832                    use_offsets[i], i);
2833                if (use_offsets[i+1] != -1)
2834                  fprintf(outfile, "ERROR: bad negative value %d for offset %d\n",
2835                    use_offsets[i+1], i+1);
2836              fprintf(outfile, "%2d: <unset>\n", i/2);              fprintf(outfile, "%2d: <unset>\n", i/2);
2837                }
2838            else            else
2839              {              {
2840              fprintf(outfile, "%2d: ", i/2);              fprintf(outfile, "%2d: ", i/2);
2841              (void)pchars(bptr + use_offsets[i],              (void)pchars(bptr + use_offsets[i],
2842                use_offsets[i+1] - use_offsets[i], outfile);                use_offsets[i+1] - use_offsets[i], outfile);
2843              fprintf(outfile, "\n");              fprintf(outfile, "\n");
2844              if (i == 0)              if (do_showcaprest || (i == 0 && do_showrest))
2845                {                {
2846                if (do_showrest)                fprintf(outfile, "%2d+ ", i/2);
2847                  {                (void)pchars(bptr + use_offsets[i+1], len - use_offsets[i+1],
2848                  fprintf(outfile, " 0+ ");                  outfile);
2849                  (void)pchars(bptr + use_offsets[i+1], len - use_offsets[i+1],                fprintf(outfile, "\n");
                   outfile);  
                 fprintf(outfile, "\n");  
                 }  
2850                }                }
2851              }              }
2852            }            }
2853    
2854            if (markptr != NULL) fprintf(outfile, "MK: %s\n", markptr);
2855    
2856          for (i = 0; i < 32; i++)          for (i = 0; i < 32; i++)
2857            {            {
2858            if ((copystrings & (1 << i)) != 0)            if ((copystrings & (1 << i)) != 0)
2859              {              {
2860              char copybuffer[16];              char copybuffer[256];
2861              int rc = pcre_copy_substring((char *)bptr, use_offsets, count,              int rc = pcre_copy_substring((char *)bptr, use_offsets, count,
2862                i, copybuffer, sizeof(copybuffer));                i, copybuffer, sizeof(copybuffer));
2863              if (rc < 0)              if (rc < 0)
# Line 1651  while (!done) Line 2867  while (!done)
2867              }              }
2868            }            }
2869    
2870            for (copynamesptr = copynames;
2871                 *copynamesptr != 0;
2872                 copynamesptr += (int)strlen((char*)copynamesptr) + 1)
2873              {
2874              char copybuffer[256];
2875              int rc = pcre_copy_named_substring(re, (char *)bptr, use_offsets,
2876                count, (char *)copynamesptr, copybuffer, sizeof(copybuffer));
2877              if (rc < 0)
2878                fprintf(outfile, "copy substring %s failed %d\n", copynamesptr, rc);
2879              else
2880                fprintf(outfile, "  C %s (%d) %s\n", copybuffer, rc, copynamesptr);
2881              }
2882    
2883          for (i = 0; i < 32; i++)          for (i = 0; i < 32; i++)
2884            {            {
2885            if ((getstrings & (1 << i)) != 0)            if ((getstrings & (1 << i)) != 0)
# Line 1663  while (!done) Line 2892  while (!done)
2892              else              else
2893                {                {
2894                fprintf(outfile, "%2dG %s (%d)\n", i, substring, rc);                fprintf(outfile, "%2dG %s (%d)\n", i, substring, rc);
               /* free((void *)substring); */  
2895                pcre_free_substring(substring);                pcre_free_substring(substring);
2896                }                }
2897              }              }
2898            }            }
2899    
2900            for (getnamesptr = getnames;
2901                 *getnamesptr != 0;
2902                 getnamesptr += (int)strlen((char*)getnamesptr) + 1)
2903              {
2904              const char *substring;
2905              int rc = pcre_get_named_substring(re, (char *)bptr, use_offsets,
2906                count, (char *)getnamesptr, &substring);
2907              if (rc < 0)
2908                fprintf(outfile, "copy substring %s failed %d\n", getnamesptr, rc);
2909              else
2910                {
2911                fprintf(outfile, "  G %s (%d) %s\n", substring, rc, getnamesptr);
2912                pcre_free_substring(substring);
2913                }
2914              }
2915    
2916          if (getlist)          if (getlist)
2917            {            {
2918            const char **stringlist;            const char **stringlist;
# Line 1682  while (!done) Line 2926  while (!done)
2926                fprintf(outfile, "%2dL %s\n", i, stringlist[i]);                fprintf(outfile, "%2dL %s\n", i, stringlist[i]);
2927              if (stringlist[i] != NULL)              if (stringlist[i] != NULL)
2928                fprintf(outfile, "string list not terminated by NULL\n");                fprintf(outfile, "string list not terminated by NULL\n");
             /* free((void *)stringlist); */  
2929              pcre_free_substring_list(stringlist);              pcre_free_substring_list(stringlist);
2930              }              }
2931            }            }
# Line 1692  while (!done) Line 2935  while (!done)
2935    
2936        else if (count == PCRE_ERROR_PARTIAL)        else if (count == PCRE_ERROR_PARTIAL)
2937          {          {
2938          fprintf(outfile, "Partial match\n");          if (markptr == NULL) fprintf(outfile, "Partial match");
2939              else fprintf(outfile, "Partial match, mark=%s", markptr);
2940            if (use_size_offsets > 1)
2941              {
2942              fprintf(outfile, ": ");
2943              pchars(bptr + use_offsets[0], use_offsets[1] - use_offsets[0],
2944                outfile);
2945              }
2946            fprintf(outfile, "\n");
2947          break;  /* Out of the /g loop */          break;  /* Out of the /g loop */
2948          }          }
2949    
2950        /* Failed to match. If this is a /g or /G loop and we previously set        /* Failed to match. If this is a /g or /G loop and we previously set
2951        g_notempty after a null match, this is not necessarily the end.        g_notempty after a null match, this is not necessarily the end. We want
2952        We want to advance the start offset, and continue. In the case of UTF-8        to advance the start offset, and continue. We won't be at the end of the
2953        matching, the advance must be one character, not one byte. Fudge the        string - that was checked before setting g_notempty.
2954        offset values to achieve this. We won't be at the end of the string -  
2955        that was checked before setting g_notempty. */        Complication arises in the case when the newline convention is "any",
2956          "crlf", or "anycrlf". If the previous match was at the end of a line
2957          terminated by CRLF, an advance of one character just passes the \r,
2958          whereas we should prefer the longer newline sequence, as does the code in
2959          pcre_exec(). Fudge the offset value to achieve this. We check for a
2960          newline setting in the pattern; if none was set, use pcre_config() to
2961          find the default.
2962    
2963          Otherwise, in the case of UTF-8 matching, the advance must be one
2964          character, not one byte. */
2965    
2966        else        else
2967          {          {
2968          if (g_notempty != 0)          if (g_notempty != 0)
2969            {            {
2970            int onechar = 1;            int onechar = 1;
2971              unsigned int obits = ((real_pcre *)re)->options;
2972            use_offsets[0] = start_offset;            use_offsets[0] = start_offset;
2973            if (use_utf8)            if ((obits & PCRE_NEWLINE_BITS) == 0)
2974                {
2975                int d;
2976                (void)pcre_config(PCRE_CONFIG_NEWLINE, &d);
2977                /* Note that these values are always the ASCII ones, even in
2978                EBCDIC environments. CR = 13, NL = 10. */
2979                obits = (d == 13)? PCRE_NEWLINE_CR :
2980                        (d == 10)? PCRE_NEWLINE_LF :
2981                        (d == (13<<8 | 10))? PCRE_NEWLINE_CRLF :
2982                        (d == -2)? PCRE_NEWLINE_ANYCRLF :
2983                        (d == -1)? PCRE_NEWLINE_ANY : 0;
2984                }
2985              if (((obits & PCRE_NEWLINE_BITS) == PCRE_NEWLINE_ANY ||
2986                   (obits & PCRE_NEWLINE_BITS) == PCRE_NEWLINE_CRLF ||
2987                   (obits & PCRE_NEWLINE_BITS) == PCRE_NEWLINE_ANYCRLF)
2988                  &&
2989                  start_offset < len - 1 &&
2990                  bptr[start_offset] == '\r' &&
2991                  bptr[start_offset+1] == '\n')
2992                onechar++;
2993              else if (use_utf8)
2994              {              {
2995              while (start_offset + onechar < len)              while (start_offset + onechar < len)
2996                {                {
2997                int tb = bptr[start_offset+onechar];                if ((bptr[start_offset+onechar] & 0xc0) != 0x80) break;
2998                if (tb <= 127) break;                onechar++;
               tb &= 0xc0;  
               if (tb != 0 && tb != 0xc0) onechar++;  
2999                }                }
3000              }              }
3001            use_offsets[1] = start_offset + onechar;            use_offsets[1] = start_offset + onechar;
3002            }            }
3003          else          else
3004            {            {
3005            if (count == PCRE_ERROR_NOMATCH)            switch(count)
3006              {              {
3007              if (gmatched == 0) fprintf(outfile, "No match\n");              case PCRE_ERROR_NOMATCH:
3008                if (gmatched == 0)
3009                  {
3010                  if (markptr == NULL) fprintf(outfile, "No match\n");
3011                    else fprintf(outfile, "No match, mark = %s\n", markptr);
3012                  }
3013                break;
3014    
3015                case PCRE_ERROR_BADUTF8:
3016                case PCRE_ERROR_SHORTUTF8:
3017                fprintf(outfile, "Error %d (%s UTF-8 string)", count,
3018                  (count == PCRE_ERROR_BADUTF8)? "bad" : "short");
3019                if (use_size_offsets >= 2)
3020                  fprintf(outfile, " offset=%d reason=%d", use_offsets[0],
3021                    use_offsets[1]);
3022                fprintf(outfile, "\n");
3023                break;
3024    
3025                default:
3026                if (count < 0 && (-count) < sizeof(errtexts)/sizeof(const char *))
3027                  fprintf(outfile, "Error %d (%s)\n", count, errtexts[-count]);
3028                else
3029                  fprintf(outfile, "Error %d (Unexpected value)\n", count);
3030                break;
3031              }              }
3032            else fprintf(outfile, "Error %d\n", count);  
3033            break;  /* Out of the /g loop */            break;  /* Out of the /g loop */
3034            }            }
3035          }          }
# Line 1737  while (!done) Line 3039  while (!done)
3039        if (!do_g && !do_G) break;        if (!do_g && !do_G) break;
3040    
3041        /* If we have matched an empty string, first check to see if we are at        /* If we have matched an empty string, first check to see if we are at
3042        the end of the subject. If so, the /g loop is over. Otherwise, mimic        the end of the subject. If so, the /g loop is over. Otherwise, mimic what
3043        what Perl's /g options does. This turns out to be rather cunning. First        Perl's /g options does. This turns out to be rather cunning. First we set
3044        we set PCRE_NOTEMPTY and PCRE_ANCHORED and try the match again at the        PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED and try the match again at the
3045        same point. If this fails (picked up above) we advance to the next        same point. If this fails (picked up above) we advance to the next
3046        character. */        character. */
3047    
3048        g_notempty = 0;        g_notempty = 0;
3049    
3050        if (use_offsets[0] == use_offsets[1])        if (use_offsets[0] == use_offsets[1])
3051          {          {
3052          if (use_offsets[0] == len) break;          if (use_offsets[0] == len) break;
3053          g_notempty = PCRE_NOTEMPTY | PCRE_ANCHORED;          g_notempty = PCRE_NOTEMPTY_ATSTART | PCRE_ANCHORED;
3054          }          }
3055    
3056        /* For /g, update the start offset, leaving the rest alone */        /* For /g, update the start offset, leaving the rest alone */
# Line 1762  while (!done) Line 3065  while (!done)
3065          len -= use_offsets[1];          len -= use_offsets[1];
3066          }          }
3067        }  /* End of loop for /g and /G */        }  /* End of loop for /g and /G */
3068    
3069        NEXT_DATA: continue;
3070      }    /* End of loop for data lines */      }    /* End of loop for data lines */
3071    
3072    CONTINUE:    CONTINUE:
# Line 1770  while (!done) Line 3075  while (!done)
3075    if (posix || do_posix) regfree(&preg);    if (posix || do_posix) regfree(&preg);
3076  #endif  #endif
3077    
3078    if (re != NULL) free(re);    if (re != NULL) new_free(re);
3079    if (extra != NULL) free(extra);    if (extra != NULL) pcre_free_study(extra);
3080    if (tables != NULL)    if (locale_set)
3081      {      {
3082      free((void *)tables);      new_free((void *)tables);
3083      setlocale(LC_CTYPE, "C");      setlocale(LC_CTYPE, "C");
3084        locale_set = 0;
3085        }
3086      if (jit_stack != NULL)
3087        {
3088        pcre_jit_stack_free(jit_stack);
3089        jit_stack = NULL;
3090      }      }
3091    }    }
3092    
3093  if (infile == stdin) fprintf(outfile, "\n");  if (infile == stdin) fprintf(outfile, "\n");
3094  return 0;  
3095    EXIT:
3096    
3097    if (infile != NULL && infile != stdin) fclose(infile);
3098    if (outfile != NULL && outfile != stdout) fclose(outfile);
3099    
3100    free(buffer);
3101    free(dbuffer);
3102    free(pbuffer);
3103    free(offsets);
3104    
3105    return yield;
3106  }  }
3107    
3108  /* End */  /* End of pcretest.c */

Legend:
Removed from v.75  
changed lines
  Added in v.734

  ViewVC Help
Powered by ViewVC 1.1.5