--- code/trunk/pcretest.c 2007/03/12 15:19:06 123 +++ code/trunk/pcretest.c 2007/04/16 13:24:37 147 @@ -67,13 +67,16 @@ #endif -#define PCRE_SPY /* For Win32 build, import data, not export */ - -/* We include pcre_internal.h because we need the internal info for displaying -the results of pcre_study() and we also need to know about the internal -macros, structures, and other internal data values; pcretest has "inside -information" compared to a program that strictly follows the PCRE API. */ +/* We have to include pcre_internal.h because we need the internal info for +displaying the results of pcre_study() and we also need to know about the +internal macros, structures, and other internal data values; pcretest has +"inside information" compared to a program that strictly follows the PCRE API. + +Although pcre_internal.h does itself include pcre.h, we explicitly include it +here before pcre_internal.h so that the PCRE_EXP_xxx macros get set +appropriately for an application, not for building PCRE. */ +#include "pcre.h" #include "pcre_internal.h" /* We need access to the data tables that PCRE uses. So as not to have to keep @@ -1135,7 +1138,7 @@ case 'S': do_study = 1; break; case 'U': options |= PCRE_UNGREEDY; break; case 'X': options |= PCRE_EXTRA; break; - case 'Z': debug_lengths = 0; + case 'Z': debug_lengths = 0; break; case '8': options |= PCRE_UTF8; use_utf8 = 1; break; case '?': options |= PCRE_NO_UTF8_CHECK; break; @@ -1591,7 +1594,7 @@ for (;;) { uschar *q; - uschar *bptr = dbuffer; + uschar *bptr; int *use_offsets = offsets; int use_size_offsets = size_offsets; int callout_data = 0; @@ -1647,7 +1650,7 @@ p = buffer; while (isspace(*p)) p++; - q = dbuffer; + bptr = q = dbuffer; while ((c = *p++) != 0) { int i = 0; @@ -2211,19 +2214,41 @@ } /* Failed to match. If this is a /g or /G loop and we previously set - g_notempty after a null match, this is not necessarily the end. - We want to advance the start offset, and continue. In the case of UTF-8 - matching, the advance must be one character, not one byte. Fudge the - offset values to achieve this. We won't be at the end of the string - - that was checked before setting g_notempty. */ + g_notempty after a null match, this is not necessarily the end. We want + to advance the start offset, and continue. We won't be at the end of the + string - that was checked before setting g_notempty. + + Complication arises in the case when the newline option is "any". + If the previous match was at the end of a line terminated by CRLF, an + advance of one character just passes the \r, whereas we should prefer the + longer newline sequence, as does the code in pcre_exec(). Fudge the + offset value to achieve this. + + Otherwise, in the case of UTF-8 matching, the advance must be one + character, not one byte. */ else { if (g_notempty != 0) { int onechar = 1; + unsigned int obits = ((real_pcre *)re)->options; use_offsets[0] = start_offset; - if (use_utf8) + if ((obits & PCRE_NEWLINE_BITS) == 0) + { + int d; + (void)pcre_config(PCRE_CONFIG_NEWLINE, &d); + obits = (d == '\r')? PCRE_NEWLINE_CR : + (d == '\n')? PCRE_NEWLINE_LF : + (d == ('\r'<<8 | '\n'))? PCRE_NEWLINE_CRLF : + (d == -1)? PCRE_NEWLINE_ANY : 0; + } + if ((obits & PCRE_NEWLINE_BITS) == PCRE_NEWLINE_ANY && + start_offset < len - 1 && + bptr[start_offset] == '\r' && + bptr[start_offset+1] == '\n') + onechar++; + else if (use_utf8) { while (start_offset + onechar < len) { @@ -2258,6 +2283,7 @@ character. */ g_notempty = 0; + if (use_offsets[0] == use_offsets[1]) { if (use_offsets[0] == len) break;