code/trunk/pcre_printint.src

/*************************************************
*      Perl-Compatible Regular Expressions       *
*************************************************/

/* PCRE is a library of functions to support regular expressions whose syntax
and semantics are as close as possible to those of the Perl 5 language.

                       Written by Philip Hazel
           Copyright (c) 1997-2009 University of Cambridge

-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:

    * Redistributions of source code must retain the above copyright notice,
      this list of conditions and the following disclaimer.

    * Redistributions in binary form must reproduce the above copyright
      notice, this list of conditions and the following disclaimer in the
      documentation and/or other materials provided with the distribution.

    * Neither the name of the University of Cambridge nor the names of its
      contributors may be used to endorse or promote products derived from
      this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
POSSIBILITY OF SUCH DAMAGE.
-----------------------------------------------------------------------------
*/


/* This module contains a PCRE private debugging function for printing out the
internal form of a compiled regular expression, along with some supporting
local functions. This source file is used in two places:

(1) It is #included by pcre_compile.c when it is compiled in debugging mode
(DEBUG defined in pcre_internal.h). It is not included in production compiles.

(2) It is always #included by pcretest.c, which can be asked to print out a
compiled regex for debugging purposes. */


/* Macro that decides whether a character should be output as a literal or in
hexadecimal. We don't use isprint() because that can vary from system to system
(even without the use of locales) and we want the output always to be the same,
for testing purposes. This macro is used in pcretest as well as in this file. */

#ifdef EBCDIC
#define PRINTABLE(c) ((c) >= 64 && (c) < 255)
#else
#define PRINTABLE(c) ((c) >= 32 && (c) < 127)
#endif

/* The table of operator names. */

static const char *OP_names[] = { OP_NAME_LIST };


/*************************************************
*       Print single- or multi-byte character    *
*************************************************/

static int
print_char(FILE *f, uschar *ptr, BOOL utf8)
{
int c = *ptr;

#ifndef SUPPORT_UTF8
utf8 = utf8;  /* Avoid compiler warning */
if (PRINTABLE(c)) fprintf(f, "%c", c); else fprintf(f, "\\x%02x", c);
return 0;

#else
if (!utf8 || (c & 0xc0) != 0xc0)
  {
  if (PRINTABLE(c)) fprintf(f, "%c", c); else fprintf(f, "\\x%02x", c);
  return 0;
  }
else
  {
  int i;
  int a = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */
  int s = 6*a;
  c = (c & _pcre_utf8_table3[a]) << s;
  for (i = 1; i <= a; i++)
    {
    /* This is a check for malformed UTF-8; it should only occur if the sanity
    check has been turned off. Rather than swallow random bytes, just stop if
    we hit a bad one. Print it with \X instead of \x as an indication. */

    if ((ptr[i] & 0xc0) != 0x80)
      {
      fprintf(f, "\\X{%x}", c);
      return i - 1;
      }

    /* The byte is OK */

    s -= 6;
    c |= (ptr[i] & 0x3f) << s;
    }
  if (c < 128) fprintf(f, "\\x%02x", c); else fprintf(f, "\\x{%x}", c);
  return a;
  }
#endif
}


/*************************************************
*          Find Unicode property name            *
*************************************************/

static const char *
get_ucpname(int ptype, int pvalue)
{
#ifdef SUPPORT_UCP
int i;
for (i = _pcre_utt_size - 1; i >= 0; i--)
  {
  if (ptype == _pcre_utt[i].type && pvalue == _pcre_utt[i].value) break;
  }
return (i >= 0)? _pcre_utt_names + _pcre_utt[i].name_offset : "??";
#else
/* It gets harder and harder to shut off unwanted compiler warnings. */
ptype = ptype * pvalue;
return (ptype == pvalue)? "??" : "??";
#endif
}


/*************************************************
*         Print compiled regex                   *
*************************************************/

/* Make this function work for a regex with integers either byte order.
However, we assume that what we are passed is a compiled regex. The
print_lengths flag controls whether offsets and lengths of items are printed.
They can be turned off from pcretest so that automatic tests on bytecode can be
written that do not depend on the value of LINK_SIZE. */

static void
pcre_printint(pcre *external_re, FILE *f, BOOL print_lengths)
{
real_pcre *re = (real_pcre *)external_re;
uschar *codestart, *code;
BOOL utf8;

unsigned int options = re->options;
int offset = re->name_table_offset;
int count = re->name_count;
int size = re->name_entry_size;

if (re->magic_number != MAGIC_NUMBER)
  {
  offset = ((offset << 8) & 0xff00) | ((offset >> 8) & 0xff);
  count = ((count << 8) & 0xff00) | ((count >> 8) & 0xff);
  size = ((size << 8) & 0xff00) | ((size >> 8) & 0xff);
  options = ((options << 24) & 0xff000000) |
            ((options <<  8) & 0x00ff0000) |
            ((options >>  8) & 0x0000ff00) |
            ((options >> 24) & 0x000000ff);
  }

code = codestart = (uschar *)re + offset + count * size;
utf8 = (options & PCRE_UTF8) != 0;

for(;;)
  {
  uschar *ccode;
  int c;
  int extra = 0;

  if (print_lengths)
    fprintf(f, "%3d ", (int)(code - codestart));
  else
    fprintf(f, "    ");

  switch(*code)
    {
    case OP_END:
    fprintf(f, "    %s\n", OP_names[*code]);
    fprintf(f, "------------------------------------------------------------------\n");
    return;

    case OP_OPT:
    fprintf(f, " %.2x %s", code[1], OP_names[*code]);
    break;

    case OP_CHAR:
    fprintf(f, "    ");
    do
      {
      code++;
      code += 1 + print_char(f, code, utf8);
      }
    while (*code == OP_CHAR);
    fprintf(f, "\n");
    continue;

    case OP_CHARNC:
    fprintf(f, " NC ");
    do
      {
      code++;
      code += 1 + print_char(f, code, utf8);
      }
    while (*code == OP_CHARNC);
    fprintf(f, "\n");
    continue;

    case OP_CBRA:
    case OP_SCBRA:
    if (print_lengths) fprintf(f, "%3d ", GET(code, 1));
      else fprintf(f, "    ");
    fprintf(f, "%s %d", OP_names[*code], GET2(code, 1+LINK_SIZE));
    break;

    case OP_BRA:
    case OP_SBRA:
    case OP_KETRMAX:
    case OP_KETRMIN:
    case OP_ALT:
    case OP_KET:
    case OP_ASSERT:
    case OP_ASSERT_NOT:
    case OP_ASSERTBACK:
    case OP_ASSERTBACK_NOT:
    case OP_ONCE:
    case OP_COND:
    case OP_SCOND:
    case OP_REVERSE:
    if (print_lengths) fprintf(f, "%3d ", GET(code, 1));
      else fprintf(f, "    ");
    fprintf(f, "%s", OP_names[*code]);
    break;
    
    case OP_CLOSE:
    fprintf(f, "    %s %d", OP_names[*code], GET2(code, 1));
    break;   

    case OP_CREF:
    case OP_NCREF: 
    fprintf(f, "%3d %s", GET2(code,1), OP_names[*code]);
    break;

    case OP_RREF:
    c = GET2(code, 1);
    if (c == RREF_ANY)
      fprintf(f, "    Cond recurse any");
    else
      fprintf(f, "    Cond recurse %d", c);
    break;

    case OP_NRREF:
    c = GET2(code, 1);
    if (c == RREF_ANY)
      fprintf(f, "    Cond nrecurse any");
    else
      fprintf(f, "    Cond nrecurse %d", c);
    break;

    case OP_DEF:
    fprintf(f, "    Cond def");
    break;

    case OP_STAR:
    case OP_MINSTAR:
    case OP_POSSTAR:
    case OP_PLUS:
    case OP_MINPLUS:
    case OP_POSPLUS:
    case OP_QUERY:
    case OP_MINQUERY:
    case OP_POSQUERY:
    case OP_TYPESTAR:
    case OP_TYPEMINSTAR:
    case OP_TYPEPOSSTAR:
    case OP_TYPEPLUS:
    case OP_TYPEMINPLUS:
    case OP_TYPEPOSPLUS:
    case OP_TYPEQUERY:
    case OP_TYPEMINQUERY:
    case OP_TYPEPOSQUERY:
    fprintf(f, "    ");
    if (*code >= OP_TYPESTAR)
      {
      fprintf(f, "%s", OP_names[code[1]]);
      if (code[1] == OP_PROP || code[1] == OP_NOTPROP)
        {
        fprintf(f, " %s ", get_ucpname(code[2], code[3]));
        extra = 2;
        }
      }
    else extra = print_char(f, code+1, utf8);
    fprintf(f, "%s", OP_names[*code]);
    break;

    case OP_EXACT:
    case OP_UPTO:
    case OP_MINUPTO:
    case OP_POSUPTO:
    fprintf(f, "    ");
    extra = print_char(f, code+3, utf8);
    fprintf(f, "{");
    if (*code != OP_EXACT) fprintf(f, "0,");
    fprintf(f, "%d}", GET2(code,1));
    if (*code == OP_MINUPTO) fprintf(f, "?");
      else if (*code == OP_POSUPTO) fprintf(f, "+");
    break;

    case OP_TYPEEXACT:
    case OP_TYPEUPTO:
    case OP_TYPEMINUPTO:
    case OP_TYPEPOSUPTO:
    fprintf(f, "    %s", OP_names[code[3]]);
    if (code[3] == OP_PROP || code[3] == OP_NOTPROP)
      {
      fprintf(f, " %s ", get_ucpname(code[4], code[5]));
      extra = 2;
      }
    fprintf(f, "{");
    if (*code != OP_TYPEEXACT) fprintf(f, "0,");
    fprintf(f, "%d}", GET2(code,1));
    if (*code == OP_TYPEMINUPTO) fprintf(f, "?");
      else if (*code == OP_TYPEPOSUPTO) fprintf(f, "+");
    break;

    case OP_NOT:
    c = code[1];
    if (PRINTABLE(c)) fprintf(f, "    [^%c]", c);
      else fprintf(f, "    [^\\x%02x]", c);
    break;

    case OP_NOTSTAR:
    case OP_NOTMINSTAR:
    case OP_NOTPOSSTAR:
    case OP_NOTPLUS:
    case OP_NOTMINPLUS:
    case OP_NOTPOSPLUS:
    case OP_NOTQUERY:
    case OP_NOTMINQUERY:
    case OP_NOTPOSQUERY:
    c = code[1];
    if (PRINTABLE(c)) fprintf(f, "    [^%c]", c);
      else fprintf(f, "    [^\\x%02x]", c);
    fprintf(f, "%s", OP_names[*code]);
    break;

    case OP_NOTEXACT:
    case OP_NOTUPTO:
    case OP_NOTMINUPTO:
    case OP_NOTPOSUPTO:
    c = code[3];
    if (PRINTABLE(c)) fprintf(f, "    [^%c]{", c);
      else fprintf(f, "    [^\\x%02x]{", c);
    if (*code != OP_NOTEXACT) fprintf(f, "0,");
    fprintf(f, "%d}", GET2(code,1));
    if (*code == OP_NOTMINUPTO) fprintf(f, "?");
      else if (*code == OP_NOTPOSUPTO) fprintf(f, "+");
    break;

    case OP_RECURSE:
    if (print_lengths) fprintf(f, "%3d ", GET(code, 1));
      else fprintf(f, "    ");
    fprintf(f, "%s", OP_names[*code]);
    break;

    case OP_REF:
    fprintf(f, "    \\%d", GET2(code,1));
    ccode = code + _pcre_OP_lengths[*code];
    goto CLASS_REF_REPEAT;

    case OP_CALLOUT:
    fprintf(f, "    %s %d %d %d", OP_names[*code], code[1], GET(code,2),
      GET(code, 2 + LINK_SIZE));
    break;

    case OP_PROP:
    case OP_NOTPROP:
    fprintf(f, "    %s %s", OP_names[*code], get_ucpname(code[1], code[2]));
    break;

    /* OP_XCLASS can only occur in UTF-8 mode. However, there's no harm in
    having this code always here, and it makes it less messy without all those
    #ifdefs. */

    case OP_CLASS:
    case OP_NCLASS:
    case OP_XCLASS:
      {
      int i, min, max;
      BOOL printmap;

      fprintf(f, "    [");

      if (*code == OP_XCLASS)
        {
        extra = GET(code, 1);
        ccode = code + LINK_SIZE + 1;
        printmap = (*ccode & XCL_MAP) != 0;
        if ((*ccode++ & XCL_NOT) != 0) fprintf(f, "^");
        }
      else
        {
        printmap = TRUE;
        ccode = code + 1;
        }

      /* Print a bit map */

      if (printmap)
        {
        for (i = 0; i < 256; i++)
          {
          if ((ccode[i/8] & (1 << (i&7))) != 0)
            {
            int j;
            for (j = i+1; j < 256; j++)
              if ((ccode[j/8] & (1 << (j&7))) == 0) break;
            if (i == '-' || i == ']') fprintf(f, "\\");
            if (PRINTABLE(i)) fprintf(f, "%c", i);
              else fprintf(f, "\\x%02x", i);
            if (--j > i)
              {
              if (j != i + 1) fprintf(f, "-");
              if (j == '-' || j == ']') fprintf(f, "\\");
              if (PRINTABLE(j)) fprintf(f, "%c", j);
                else fprintf(f, "\\x%02x", j);
              }
            i = j;
            }
          }
        ccode += 32;
        }

      /* For an XCLASS there is always some additional data */

      if (*code == OP_XCLASS)
        {
        int ch;
        while ((ch = *ccode++) != XCL_END)
          {
          if (ch == XCL_PROP)
            {
            int ptype = *ccode++;
            int pvalue = *ccode++;
            fprintf(f, "\\p{%s}", get_ucpname(ptype, pvalue));
            }
          else if (ch == XCL_NOTPROP)
            {
            int ptype = *ccode++;
            int pvalue = *ccode++;
            fprintf(f, "\\P{%s}", get_ucpname(ptype, pvalue));
            }
          else
            {
            ccode += 1 + print_char(f, ccode, TRUE);
            if (ch == XCL_RANGE)
              {
              fprintf(f, "-");
              ccode += 1 + print_char(f, ccode, TRUE);
              }
            }
          }
        }

      /* Indicate a non-UTF8 class which was created by negation */

      fprintf(f, "]%s", (*code == OP_NCLASS)? " (neg)" : "");

      /* Handle repeats after a class or a back reference */

      CLASS_REF_REPEAT:
      switch(*ccode)
        {
        case OP_CRSTAR:
        case OP_CRMINSTAR:
        case OP_CRPLUS:
        case OP_CRMINPLUS:
        case OP_CRQUERY:
        case OP_CRMINQUERY:
        fprintf(f, "%s", OP_names[*ccode]);
        extra += _pcre_OP_lengths[*ccode];
        break;

        case OP_CRRANGE:
        case OP_CRMINRANGE:
        min = GET2(ccode,1);
        max = GET2(ccode,3);
        if (max == 0) fprintf(f, "{%d,}", min);
        else fprintf(f, "{%d,%d}", min, max);
        if (*ccode == OP_CRMINRANGE) fprintf(f, "?");
        extra += _pcre_OP_lengths[*ccode];
        break;

        /* Do nothing if it's not a repeat; this code stops picky compilers
        warning about the lack of a default code path. */

        default:
        break;
        }
      }
    break;

    /* Anything else is just an item with no data*/

    default:
    fprintf(f, "    %s", OP_names[*code]);
    break;
    }

  code += _pcre_OP_lengths[*code] + extra;
  fprintf(f, "\n");
  }
}

/* End of pcre_printint.src */
1	nigel	85	/*************************************************
2			* Perl-Compatible Regular Expressions *
3			*************************************************/
4
5			/* PCRE is a library of functions to support regular expressions whose syntax
6			and semantics are as close as possible to those of the Perl 5 language.
7
8			Written by Philip Hazel
9	ph10	391	Copyright (c) 1997-2009 University of Cambridge
10	nigel	85
11			-----------------------------------------------------------------------------
12			Redistribution and use in source and binary forms, with or without
13			modification, are permitted provided that the following conditions are met:
14
15			* Redistributions of source code must retain the above copyright notice,
16			this list of conditions and the following disclaimer.
17
18			* Redistributions in binary form must reproduce the above copyright
19			notice, this list of conditions and the following disclaimer in the
20			documentation and/or other materials provided with the distribution.
21
22			* Neither the name of the University of Cambridge nor the names of its
23			contributors may be used to endorse or promote products derived from
24			this software without specific prior written permission.
25
26			THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
27			AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
28			IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
29			ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
30			LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
31			CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
32			SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
33			INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
34			CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
35			ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
36			POSSIBILITY OF SUCH DAMAGE.
37			-----------------------------------------------------------------------------
38			*/
39
40
41			/* This module contains a PCRE private debugging function for printing out the
42			internal form of a compiled regular expression, along with some supporting
43			local functions. This source file is used in two places:
44
45			(1) It is #included by pcre_compile.c when it is compiled in debugging mode
46			(DEBUG defined in pcre_internal.h). It is not included in production compiles.
47
48			(2) It is always #included by pcretest.c, which can be asked to print out a
49			compiled regex for debugging purposes. */
50
51
52	nigel	93	/* Macro that decides whether a character should be output as a literal or in
53			hexadecimal. We don't use isprint() because that can vary from system to system
54			(even without the use of locales) and we want the output always to be the same,
55			for testing purposes. This macro is used in pcretest as well as in this file. */
56
57	ph10	391	#ifdef EBCDIC
58			#define PRINTABLE(c) ((c) >= 64 && (c) < 255)
59			#else
60	nigel	93	#define PRINTABLE(c) ((c) >= 32 && (c) < 127)
61	ph10	391	#endif
62	nigel	93
63			/* The table of operator names. */
64
65	nigel	85	static const char *OP_names[] = { OP_NAME_LIST };
66
67
68	nigel	93
69	nigel	85	/*************************************************
70			* Print single- or multi-byte character *
71			*************************************************/
72
73			static int
74			print_char(FILE f, uschar ptr, BOOL utf8)
75			{
76			int c = *ptr;
77
78	ph10	107	#ifndef SUPPORT_UTF8
79			utf8 = utf8; /* Avoid compiler warning */
80			if (PRINTABLE(c)) fprintf(f, "%c", c); else fprintf(f, "\\x%02x", c);
81			return 0;
82
83			#else
84	nigel	85	if (!utf8 \|\| (c & 0xc0) != 0xc0)
85			{
86	nigel	93	if (PRINTABLE(c)) fprintf(f, "%c", c); else fprintf(f, "\\x%02x", c);
87	nigel	85	return 0;
88			}
89			else
90			{
91			int i;
92			int a = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */
93			int s = 6*a;
94			c = (c & _pcre_utf8_table3[a]) << s;
95			for (i = 1; i <= a; i++)
96			{
97			/* This is a check for malformed UTF-8; it should only occur if the sanity
98			check has been turned off. Rather than swallow random bytes, just stop if
99			we hit a bad one. Print it with \X instead of \x as an indication. */
100
101			if ((ptr[i] & 0xc0) != 0x80)
102			{
103			fprintf(f, "\\X{%x}", c);
104			return i - 1;
105			}
106
107			/* The byte is OK */
108
109			s -= 6;
110			c \|= (ptr[i] & 0x3f) << s;
111			}
112			if (c < 128) fprintf(f, "\\x%02x", c); else fprintf(f, "\\x{%x}", c);
113			return a;
114			}
115	ph10	111	#endif
116	nigel	85	}
117
118
119
120			/*************************************************
121			* Find Unicode property name *
122			*************************************************/
123
124			static const char *
125	nigel	87	get_ucpname(int ptype, int pvalue)
126	nigel	85	{
127			#ifdef SUPPORT_UCP
128			int i;
129	ph10	217	for (i = _pcre_utt_size - 1; i >= 0; i--)
130	nigel	85	{
131	nigel	87	if (ptype == _pcre_utt[i].type && pvalue == _pcre_utt[i].value) break;
132	nigel	85	}
133	ph10	240	return (i >= 0)? _pcre_utt_names + _pcre_utt[i].name_offset : "??";
134	nigel	85	#else
135	nigel	91	/* It gets harder and harder to shut off unwanted compiler warnings. */
136			ptype = ptype * pvalue;
137			return (ptype == pvalue)? "??" : "??";
138	nigel	85	#endif
139			}
140
141
142
143			/*************************************************
144			* Print compiled regex *
145			*************************************************/
146
147			/* Make this function work for a regex with integers either byte order.
148	ph10	116	However, we assume that what we are passed is a compiled regex. The
149	ph10	123	print_lengths flag controls whether offsets and lengths of items are printed.
150	ph10	116	They can be turned off from pcretest so that automatic tests on bytecode can be
151			written that do not depend on the value of LINK_SIZE. */
152	nigel	85
153			static void
154	ph10	116	pcre_printint(pcre external_re, FILE f, BOOL print_lengths)
155	nigel	85	{
156			real_pcre re = (real_pcre )external_re;
157			uschar codestart, code;
158			BOOL utf8;
159
160			unsigned int options = re->options;
161			int offset = re->name_table_offset;
162			int count = re->name_count;
163			int size = re->name_entry_size;
164
165			if (re->magic_number != MAGIC_NUMBER)
166			{
167			offset = ((offset << 8) & 0xff00) \| ((offset >> 8) & 0xff);
168			count = ((count << 8) & 0xff00) \| ((count >> 8) & 0xff);
169			size = ((size << 8) & 0xff00) \| ((size >> 8) & 0xff);
170			options = ((options << 24) & 0xff000000) \|
171			((options << 8) & 0x00ff0000) \|
172			((options >> 8) & 0x0000ff00) \|
173			((options >> 24) & 0x000000ff);
174			}
175
176			code = codestart = (uschar )re + offset + count size;
177			utf8 = (options & PCRE_UTF8) != 0;
178
179			for(;;)
180			{
181			uschar *ccode;
182			int c;
183			int extra = 0;
184
185	ph10	116	if (print_lengths)
186			fprintf(f, "%3d ", (int)(code - codestart));
187			else
188	ph10	123	fprintf(f, " ");
189	nigel	85
190			switch(*code)
191			{
192			case OP_END:
193			fprintf(f, " %s\n", OP_names[*code]);
194			fprintf(f, "------------------------------------------------------------------\n");
195			return;
196
197			case OP_OPT:
198			fprintf(f, " %.2x %s", code[1], OP_names[*code]);
199			break;
200
201			case OP_CHAR:
202	nigel	91	fprintf(f, " ");
203			do
204	nigel	85	{
205	nigel	91	code++;
206			code += 1 + print_char(f, code, utf8);
207	nigel	85	}
208	nigel	91	while (*code == OP_CHAR);
209			fprintf(f, "\n");
210			continue;
211	nigel	85
212			case OP_CHARNC:
213	nigel	91	fprintf(f, " NC ");
214			do
215	nigel	85	{
216	nigel	91	code++;
217			code += 1 + print_char(f, code, utf8);
218	nigel	85	}
219	nigel	91	while (*code == OP_CHARNC);
220			fprintf(f, "\n");
221			continue;
222	nigel	85
223	nigel	93	case OP_CBRA:
224			case OP_SCBRA:
225	ph10	116	if (print_lengths) fprintf(f, "%3d ", GET(code, 1));
226	ph10	123	else fprintf(f, " ");
227	ph10	116	fprintf(f, "%s %d", OP_names[*code], GET2(code, 1+LINK_SIZE));
228	nigel	93	break;
229
230			case OP_BRA:
231			case OP_SBRA:
232	nigel	85	case OP_KETRMAX:
233			case OP_KETRMIN:
234			case OP_ALT:
235			case OP_KET:
236			case OP_ASSERT:
237			case OP_ASSERT_NOT:
238			case OP_ASSERTBACK:
239			case OP_ASSERTBACK_NOT:
240			case OP_ONCE:
241			case OP_COND:
242	nigel	93	case OP_SCOND:
243	nigel	85	case OP_REVERSE:
244	ph10	116	if (print_lengths) fprintf(f, "%3d ", GET(code, 1));
245	ph10	123	else fprintf(f, " ");
246	ph10	116	fprintf(f, "%s", OP_names[*code]);
247	nigel	85	break;
248	ph10	447
249			case OP_CLOSE:
250			fprintf(f, " %s %d", OP_names[*code], GET2(code, 1));
251			break;
252	nigel	85
253	nigel	93	case OP_CREF:
254	ph10	459	case OP_NCREF:
255	nigel	93	fprintf(f, "%3d %s", GET2(code,1), OP_names[*code]);
256	nigel	85	break;
257
258	nigel	93	case OP_RREF:
259			c = GET2(code, 1);
260			if (c == RREF_ANY)
261			fprintf(f, " Cond recurse any");
262	nigel	85	else
263	nigel	93	fprintf(f, " Cond recurse %d", c);
264	nigel	85	break;
265
266	ph10	459	case OP_NRREF:
267			c = GET2(code, 1);
268			if (c == RREF_ANY)
269			fprintf(f, " Cond nrecurse any");
270			else
271			fprintf(f, " Cond nrecurse %d", c);
272			break;
273
274	nigel	93	case OP_DEF:
275			fprintf(f, " Cond def");
276			break;
277
278	nigel	85	case OP_STAR:
279			case OP_MINSTAR:
280	nigel	93	case OP_POSSTAR:
281	nigel	85	case OP_PLUS:
282			case OP_MINPLUS:
283	nigel	93	case OP_POSPLUS:
284	nigel	85	case OP_QUERY:
285			case OP_MINQUERY:
286	nigel	93	case OP_POSQUERY:
287	nigel	85	case OP_TYPESTAR:
288			case OP_TYPEMINSTAR:
289	nigel	93	case OP_TYPEPOSSTAR:
290	nigel	85	case OP_TYPEPLUS:
291			case OP_TYPEMINPLUS:
292	nigel	93	case OP_TYPEPOSPLUS:
293	nigel	85	case OP_TYPEQUERY:
294			case OP_TYPEMINQUERY:
295	nigel	93	case OP_TYPEPOSQUERY:
296	nigel	85	fprintf(f, " ");
297			if (*code >= OP_TYPESTAR)
298			{
299			fprintf(f, "%s", OP_names[code[1]]);
300			if (code[1] == OP_PROP \|\| code[1] == OP_NOTPROP)
301			{
302	nigel	87	fprintf(f, " %s ", get_ucpname(code[2], code[3]));
303			extra = 2;
304	nigel	85	}
305			}
306			else extra = print_char(f, code+1, utf8);
307			fprintf(f, "%s", OP_names[*code]);
308			break;
309
310			case OP_EXACT:
311			case OP_UPTO:
312			case OP_MINUPTO:
313	nigel	93	case OP_POSUPTO:
314	nigel	85	fprintf(f, " ");
315			extra = print_char(f, code+3, utf8);
316			fprintf(f, "{");
317	nigel	93	if (*code != OP_EXACT) fprintf(f, "0,");
318	nigel	85	fprintf(f, "%d}", GET2(code,1));
319			if (*code == OP_MINUPTO) fprintf(f, "?");
320	nigel	93	else if (*code == OP_POSUPTO) fprintf(f, "+");
321	nigel	85	break;
322
323			case OP_TYPEEXACT:
324			case OP_TYPEUPTO:
325			case OP_TYPEMINUPTO:
326	nigel	93	case OP_TYPEPOSUPTO:
327	nigel	85	fprintf(f, " %s", OP_names[code[3]]);
328			if (code[3] == OP_PROP \|\| code[3] == OP_NOTPROP)
329			{
330	nigel	87	fprintf(f, " %s ", get_ucpname(code[4], code[5]));
331			extra = 2;
332	nigel	85	}
333			fprintf(f, "{");
334			if (*code != OP_TYPEEXACT) fprintf(f, "0,");
335			fprintf(f, "%d}", GET2(code,1));
336			if (*code == OP_TYPEMINUPTO) fprintf(f, "?");
337	nigel	93	else if (*code == OP_TYPEPOSUPTO) fprintf(f, "+");
338	nigel	85	break;
339
340			case OP_NOT:
341	nigel	93	c = code[1];
342			if (PRINTABLE(c)) fprintf(f, " [^%c]", c);
343	nigel	85	else fprintf(f, " [^\\x%02x]", c);
344			break;
345
346			case OP_NOTSTAR:
347			case OP_NOTMINSTAR:
348	nigel	93	case OP_NOTPOSSTAR:
349	nigel	85	case OP_NOTPLUS:
350			case OP_NOTMINPLUS:
351	nigel	93	case OP_NOTPOSPLUS:
352	nigel	85	case OP_NOTQUERY:
353			case OP_NOTMINQUERY:
354	nigel	93	case OP_NOTPOSQUERY:
355			c = code[1];
356			if (PRINTABLE(c)) fprintf(f, " [^%c]", c);
357	nigel	85	else fprintf(f, " [^\\x%02x]", c);
358			fprintf(f, "%s", OP_names[*code]);
359			break;
360
361			case OP_NOTEXACT:
362			case OP_NOTUPTO:
363			case OP_NOTMINUPTO:
364	nigel	93	case OP_NOTPOSUPTO:
365			c = code[3];
366			if (PRINTABLE(c)) fprintf(f, " [^%c]{", c);
367	nigel	85	else fprintf(f, " [^\\x%02x]{", c);
368			if (*code != OP_NOTEXACT) fprintf(f, "0,");
369			fprintf(f, "%d}", GET2(code,1));
370			if (*code == OP_NOTMINUPTO) fprintf(f, "?");
371	nigel	93	else if (*code == OP_NOTPOSUPTO) fprintf(f, "+");
372	nigel	85	break;
373
374			case OP_RECURSE:
375	ph10	116	if (print_lengths) fprintf(f, "%3d ", GET(code, 1));
376	ph10	123	else fprintf(f, " ");
377	ph10	116	fprintf(f, "%s", OP_names[*code]);
378	nigel	85	break;
379
380			case OP_REF:
381			fprintf(f, " \\%d", GET2(code,1));
382			ccode = code + _pcre_OP_lengths[*code];
383			goto CLASS_REF_REPEAT;
384
385			case OP_CALLOUT:
386			fprintf(f, " %s %d %d %d", OP_names[*code], code[1], GET(code,2),
387			GET(code, 2 + LINK_SIZE));
388			break;
389
390			case OP_PROP:
391			case OP_NOTPROP:
392	nigel	87	fprintf(f, " %s %s", OP_names[*code], get_ucpname(code[1], code[2]));
393	nigel	85	break;
394
395			/* OP_XCLASS can only occur in UTF-8 mode. However, there's no harm in
396			having this code always here, and it makes it less messy without all those
397			#ifdefs. */
398
399			case OP_CLASS:
400			case OP_NCLASS:
401			case OP_XCLASS:
402			{
403			int i, min, max;
404			BOOL printmap;
405
406			fprintf(f, " [");
407
408			if (*code == OP_XCLASS)
409			{
410			extra = GET(code, 1);
411			ccode = code + LINK_SIZE + 1;
412			printmap = (*ccode & XCL_MAP) != 0;
413			if ((*ccode++ & XCL_NOT) != 0) fprintf(f, "^");
414			}
415			else
416			{
417			printmap = TRUE;
418			ccode = code + 1;
419			}
420
421			/* Print a bit map */
422
423			if (printmap)
424			{
425			for (i = 0; i < 256; i++)
426			{
427			if ((ccode[i/8] & (1 << (i&7))) != 0)
428			{
429			int j;
430			for (j = i+1; j < 256; j++)
431			if ((ccode[j/8] & (1 << (j&7))) == 0) break;
432			if (i == '-' \|\| i == ']') fprintf(f, "\\");
433	nigel	93	if (PRINTABLE(i)) fprintf(f, "%c", i);
434			else fprintf(f, "\\x%02x", i);
435	nigel	85	if (--j > i)
436			{
437			if (j != i + 1) fprintf(f, "-");
438			if (j == '-' \|\| j == ']') fprintf(f, "\\");
439	nigel	93	if (PRINTABLE(j)) fprintf(f, "%c", j);
440			else fprintf(f, "\\x%02x", j);
441	nigel	85	}
442			i = j;
443			}
444			}
445			ccode += 32;
446			}
447
448			/* For an XCLASS there is always some additional data */
449
450			if (*code == OP_XCLASS)
451			{
452			int ch;
453			while ((ch = *ccode++) != XCL_END)
454			{
455			if (ch == XCL_PROP)
456			{
457	nigel	87	int ptype = *ccode++;
458			int pvalue = *ccode++;
459			fprintf(f, "\\p{%s}", get_ucpname(ptype, pvalue));
460	nigel	85	}
461			else if (ch == XCL_NOTPROP)
462			{
463	nigel	87	int ptype = *ccode++;
464			int pvalue = *ccode++;
465			fprintf(f, "\\P{%s}", get_ucpname(ptype, pvalue));
466	nigel	85	}
467			else
468			{
469			ccode += 1 + print_char(f, ccode, TRUE);
470			if (ch == XCL_RANGE)
471			{
472			fprintf(f, "-");
473			ccode += 1 + print_char(f, ccode, TRUE);
474			}
475			}
476			}
477			}
478
479			/* Indicate a non-UTF8 class which was created by negation */
480
481			fprintf(f, "]%s", (*code == OP_NCLASS)? " (neg)" : "");
482
483			/* Handle repeats after a class or a back reference */
484
485			CLASS_REF_REPEAT:
486			switch(*ccode)
487			{
488			case OP_CRSTAR:
489			case OP_CRMINSTAR:
490			case OP_CRPLUS:
491			case OP_CRMINPLUS:
492			case OP_CRQUERY:
493			case OP_CRMINQUERY:
494			fprintf(f, "%s", OP_names[*ccode]);
495			extra += _pcre_OP_lengths[*ccode];
496			break;
497
498			case OP_CRRANGE:
499			case OP_CRMINRANGE:
500			min = GET2(ccode,1);
501			max = GET2(ccode,3);
502			if (max == 0) fprintf(f, "{%d,}", min);
503			else fprintf(f, "{%d,%d}", min, max);
504			if (*ccode == OP_CRMINRANGE) fprintf(f, "?");
505			extra += _pcre_OP_lengths[*ccode];
506			break;
507	nigel	87
508			/* Do nothing if it's not a repeat; this code stops picky compilers
509			warning about the lack of a default code path. */
510
511			default:
512			break;
513	nigel	85	}
514			}
515			break;
516
517			/* Anything else is just an item with no data*/
518
519			default:
520			fprintf(f, " %s", OP_names[*code]);
521			break;
522			}
523
524			code += _pcre_OP_lengths[*code] + extra;
525			fprintf(f, "\n");
526			}
527			}
528
529			/* End of pcre_printint.src */
Name	Value
svn:eol-style	native
svn:keywords	"Author Date Id Revision Url"