diff options
Diffstat (limited to 'src/3rdparty/pcre2/src/pcre2_xclass.c')
| -rw-r--r-- | src/3rdparty/pcre2/src/pcre2_xclass.c | 71 |
1 files changed, 54 insertions, 17 deletions
diff --git a/src/3rdparty/pcre2/src/pcre2_xclass.c b/src/3rdparty/pcre2/src/pcre2_xclass.c index 8b052be66a..5df25d2c8d 100644 --- a/src/3rdparty/pcre2/src/pcre2_xclass.c +++ b/src/3rdparty/pcre2/src/pcre2_xclass.c @@ -7,7 +7,7 @@ and semantics are as close as possible to those of the Perl 5 language. Written by Philip Hazel Original API code Copyright (c) 1997-2012 University of Cambridge - New API code Copyright (c) 2016-2019 University of Cambridge + New API code Copyright (c) 2016-2023 University of Cambridge ----------------------------------------------------------------------------- Redistribution and use in source and binary forms, with or without @@ -133,8 +133,10 @@ while ((t = *data++) != XCL_END) #ifdef SUPPORT_UNICODE else /* XCL_PROP & XCL_NOTPROP */ { + int chartype; const ucd_record *prop = GET_UCD(c); BOOL isprop = t == XCL_PROP; + BOOL ok; switch(*data) { @@ -143,8 +145,9 @@ while ((t = *data++) != XCL_END) break; case PT_LAMP: - if ((prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || - prop->chartype == ucp_Lt) == isprop) return !negated; + chartype = prop->chartype; + if ((chartype == ucp_Lu || chartype == ucp_Ll || + chartype == ucp_Lt) == isprop) return !negated; break; case PT_GC: @@ -160,9 +163,16 @@ while ((t = *data++) != XCL_END) if ((data[1] == prop->script) == isprop) return !negated; break; + case PT_SCX: + ok = (data[1] == prop->script || + MAPBIT(PRIV(ucd_script_sets) + UCD_SCRIPTX_PROP(prop), data[1]) != 0); + if (ok == isprop) return !negated; + break; + case PT_ALNUM: - if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L || - PRIV(ucp_gentype)[prop->chartype] == ucp_N) == isprop) + chartype = prop->chartype; + if ((PRIV(ucp_gentype)[chartype] == ucp_L || + PRIV(ucp_gentype)[chartype] == ucp_N) == isprop) return !negated; break; @@ -187,9 +197,10 @@ while ((t = *data++) != XCL_END) break; case PT_WORD: - if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L || - PRIV(ucp_gentype)[prop->chartype] == ucp_N || c == CHAR_UNDERSCORE) - == isprop) + chartype = prop->chartype; + if ((PRIV(ucp_gentype)[chartype] == ucp_L || + PRIV(ucp_gentype)[chartype] == ucp_N || + chartype == ucp_Mn || chartype == ucp_Pc) == isprop) return !negated; break; @@ -207,6 +218,17 @@ while ((t = *data++) != XCL_END) } break; + case PT_BIDICL: + if ((UCD_BIDICLASS_PROP(prop) == data[1]) == isprop) + return !negated; + break; + + case PT_BOOL: + ok = MAPBIT(PRIV(ucd_boolprop_sets) + + UCD_BPROPS_PROP(prop), data[1]) != 0; + if (ok == isprop) return !negated; + break; + /* The following three properties can occur only in an XCLASS, as there is no \p or \P coding for them. */ @@ -220,9 +242,10 @@ while ((t = *data++) != XCL_END) */ case PT_PXGRAPH: - if ((PRIV(ucp_gentype)[prop->chartype] != ucp_Z && - (PRIV(ucp_gentype)[prop->chartype] != ucp_C || - (prop->chartype == ucp_Cf && + chartype = prop->chartype; + if ((PRIV(ucp_gentype)[chartype] != ucp_Z && + (PRIV(ucp_gentype)[chartype] != ucp_C || + (chartype == ucp_Cf && c != 0x061c && c != 0x180e && (c < 0x2066 || c > 0x2069)) )) == isprop) return !negated; @@ -232,10 +255,11 @@ while ((t = *data++) != XCL_END) not Zl and not Zp, and U+180E. */ case PT_PXPRINT: - if ((prop->chartype != ucp_Zl && - prop->chartype != ucp_Zp && - (PRIV(ucp_gentype)[prop->chartype] != ucp_C || - (prop->chartype == ucp_Cf && + chartype = prop->chartype; + if ((chartype != ucp_Zl && + chartype != ucp_Zp && + (PRIV(ucp_gentype)[chartype] != ucp_C || + (chartype == ucp_Cf && c != 0x061c && (c < 0x2066 || c > 0x2069)) )) == isprop) return !negated; @@ -246,8 +270,21 @@ while ((t = *data++) != XCL_END) compatibility (these are $+<=>^`|~). */ case PT_PXPUNCT: - if ((PRIV(ucp_gentype)[prop->chartype] == ucp_P || - (c < 128 && PRIV(ucp_gentype)[prop->chartype] == ucp_S)) == isprop) + chartype = prop->chartype; + if ((PRIV(ucp_gentype)[chartype] == ucp_P || + (c < 128 && PRIV(ucp_gentype)[chartype] == ucp_S)) == isprop) + return !negated; + break; + + /* Perl has two sets of hex digits */ + + case PT_PXXDIGIT: + if (((c >= CHAR_0 && c <= CHAR_9) || + (c >= CHAR_A && c <= CHAR_F) || + (c >= CHAR_a && c <= CHAR_f) || + (c >= 0xff10 && c <= 0xff19) || /* Fullwidth digits */ + (c >= 0xff21 && c <= 0xff26) || /* Fullwidth letters */ + (c >= 0xff41 && c <= 0xff46)) == isprop) return !negated; break; |