-/* $Cambridge: exim/src/src/pcre/pcre_printint.src,v 1.1 2006/11/07 16:50:36 ph10 Exp $ */
+/* $Cambridge: exim/src/src/pcre/pcre_printint.src,v 1.4 2007/11/12 13:02:20 nm4 Exp $ */
/*************************************************
* Perl-Compatible Regular Expressions *
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2005 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
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. */
+
+#define PRINTABLE(c) ((c) >= 32 && (c) < 127)
+
+/* The table of operator names. */
+
static const char *OP_names[] = { OP_NAME_LIST };
+
/*************************************************
* Print single- or multi-byte character *
*************************************************/
{
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 (isprint(c)) fprintf(f, "%c", c); else fprintf(f, "\\x%02x", c);
+ if (PRINTABLE(c)) fprintf(f, "%c", c); else fprintf(f, "\\x%02x", c);
return 0;
}
else
if (c < 128) fprintf(f, "\\x%02x", c); else fprintf(f, "\\x{%x}", c);
return a;
}
+#endif
}
{
#ifdef SUPPORT_UCP
int i;
-for (i = _pcre_utt_size; i >= 0; 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[i].name : "??";
+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;
*************************************************/
/* Make this function work for a regex with integers either byte order.
-However, we assume that what we are passed is a compiled regex. */
+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)
+pcre_printint(pcre *external_re, FILE *f, BOOL print_lengths)
{
real_pcre *re = (real_pcre *)external_re;
uschar *codestart, *code;
int c;
int extra = 0;
- fprintf(f, "%3d ", (int)(code - codestart));
-
- if (*code >= OP_BRA)
- {
- if (*code - OP_BRA > EXTRACT_BASIC_MAX)
- fprintf(f, "%3d Bra extra\n", GET(code, 1));
- else
- fprintf(f, "%3d Bra %d\n", GET(code, 1), *code - OP_BRA);
- code += _pcre_OP_lengths[OP_BRA];
- continue;
- }
+ if (print_lengths)
+ fprintf(f, "%3d ", (int)(code - codestart));
+ else
+ fprintf(f, " ");
switch(*code)
{
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_ASSERTBACK_NOT:
case OP_ONCE:
case OP_COND:
+ case OP_SCOND:
case OP_REVERSE:
- fprintf(f, "%3d %s", GET(code, 1), OP_names[*code]);
+ if (print_lengths) fprintf(f, "%3d ", GET(code, 1));
+ else fprintf(f, " ");
+ fprintf(f, "%s", OP_names[*code]);
break;
- case OP_BRANUMBER:
- printf("%3d %s", GET2(code, 1), OP_names[*code]);
+ case OP_CREF:
+ fprintf(f, "%3d %s", GET2(code,1), OP_names[*code]);
break;
- case OP_CREF:
- if (GET2(code, 1) == CREF_RECURSE)
- fprintf(f, " Cond recurse");
+ case OP_RREF:
+ c = GET2(code, 1);
+ if (c == RREF_ANY)
+ fprintf(f, " Cond recurse any");
else
- fprintf(f, "%3d %s", GET2(code,1), OP_names[*code]);
+ fprintf(f, " Cond recurse %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)
{
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, ",");
+ 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)
{
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:
- if (isprint(c = code[1])) fprintf(f, " [^%c]", c);
+ 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:
- if (isprint(c = code[1])) fprintf(f, " [^%c]", c);
+ 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:
- if (isprint(c = code[3])) fprintf(f, " [^%c]{", c);
+ 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:
- fprintf(f, "%3d %s", GET(code, 1), OP_names[*code]);
+ if (print_lengths) fprintf(f, "%3d ", GET(code, 1));
+ else fprintf(f, " ");
+ fprintf(f, "%s", OP_names[*code]);
break;
case OP_REF:
for (j = i+1; j < 256; j++)
if ((ccode[j/8] & (1 << (j&7))) == 0) break;
if (i == '-' || i == ']') fprintf(f, "\\");
- if (isprint(i)) fprintf(f, "%c", i); else fprintf(f, "\\x%02x", i);
+ 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 (isprint(j)) fprintf(f, "%c", j); else fprintf(f, "\\x%02x", j);
+ if (PRINTABLE(j)) fprintf(f, "%c", j);
+ else fprintf(f, "\\x%02x", j);
}
i = j;
}
git://git.exim.org / exim.git / blobdiff