X-Git-Url: http://ftp.carnet.hr/carnet-debian/scm?a=blobdiff_plain;ds=sidebyside;f=src%2Fexternal%2Fpcre2-10.32%2Fsrc%2Fpcre2_tables.c;fp=src%2Fexternal%2Fpcre2-10.32%2Fsrc%2Fpcre2_tables.c;h=83d6f9de55ed679f5835770396afbe2922e2f688;hb=3f728675941dc69d4e544d3a880a56240a6e394a;hp=0000000000000000000000000000000000000000;hpb=927951d1c1ad45ba9e7325f07d996154a91c911b;p=ossec-hids.git diff --git a/src/external/pcre2-10.32/src/pcre2_tables.c b/src/external/pcre2-10.32/src/pcre2_tables.c new file mode 100644 index 0000000..83d6f9d --- /dev/null +++ b/src/external/pcre2-10.32/src/pcre2_tables.c @@ -0,0 +1,827 @@ +/************************************************* +* 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 + Original API code Copyright (c) 1997-2012 University of Cambridge + New API code Copyright (c) 2016-2018 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 some fixed tables that are used by more than one of the +PCRE2 code modules. The tables are also #included by the pcre2test program, +which uses macros to change their names from _pcre2_xxx to xxxx, thereby +avoiding name clashes with the library. In this case, PCRE2_PCRE2TEST is +defined. */ + +#ifndef PCRE2_PCRE2TEST /* We're compiling the library */ +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif +#include "pcre2_internal.h" +#endif /* PCRE2_PCRE2TEST */ + + +/* Table of sizes for the fixed-length opcodes. It's defined in a macro so that +the definition is next to the definition of the opcodes in pcre2_internal.h. +This is mode-dependent, so is skipped when this file is included by pcre2test. */ + +#ifndef PCRE2_PCRE2TEST +const uint8_t PRIV(OP_lengths)[] = { OP_LENGTHS }; +#endif + +/* Tables of horizontal and vertical whitespace characters, suitable for +adding to classes. */ + +const uint32_t PRIV(hspace_list)[] = { HSPACE_LIST }; +const uint32_t PRIV(vspace_list)[] = { VSPACE_LIST }; + +/* These tables are the pairs of delimiters that are valid for callout string +arguments. For each starting delimiter there must be a matching ending +delimiter, which in fact is different only for bracket-like delimiters. */ + +const uint32_t PRIV(callout_start_delims)[] = { + CHAR_GRAVE_ACCENT, CHAR_APOSTROPHE, CHAR_QUOTATION_MARK, + CHAR_CIRCUMFLEX_ACCENT, CHAR_PERCENT_SIGN, CHAR_NUMBER_SIGN, + CHAR_DOLLAR_SIGN, CHAR_LEFT_CURLY_BRACKET, 0 }; + +const uint32_t PRIV(callout_end_delims[]) = { + CHAR_GRAVE_ACCENT, CHAR_APOSTROPHE, CHAR_QUOTATION_MARK, + CHAR_CIRCUMFLEX_ACCENT, CHAR_PERCENT_SIGN, CHAR_NUMBER_SIGN, + CHAR_DOLLAR_SIGN, CHAR_RIGHT_CURLY_BRACKET, 0 }; + + +/************************************************* +* Tables for UTF-8 support * +*************************************************/ + +/* These tables are required by pcre2test in 16- or 32-bit mode, as well +as for the library in 8-bit mode, because pcre2test uses UTF-8 internally for +handling wide characters. */ + +#if defined PCRE2_PCRE2TEST || \ + (defined SUPPORT_UNICODE && \ + defined PCRE2_CODE_UNIT_WIDTH && \ + PCRE2_CODE_UNIT_WIDTH == 8) + +/* These are the breakpoints for different numbers of bytes in a UTF-8 +character. */ + +const int PRIV(utf8_table1)[] = + { 0x7f, 0x7ff, 0xffff, 0x1fffff, 0x3ffffff, 0x7fffffff}; + +const int PRIV(utf8_table1_size) = sizeof(PRIV(utf8_table1)) / sizeof(int); + +/* These are the indicator bits and the mask for the data bits to set in the +first byte of a character, indexed by the number of additional bytes. */ + +const int PRIV(utf8_table2)[] = { 0, 0xc0, 0xe0, 0xf0, 0xf8, 0xfc}; +const int PRIV(utf8_table3)[] = { 0xff, 0x1f, 0x0f, 0x07, 0x03, 0x01}; + +/* Table of the number of extra bytes, indexed by the first byte masked with +0x3f. The highest number for a valid UTF-8 first byte is in fact 0x3d. */ + +const uint8_t PRIV(utf8_table4)[] = { + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, + 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5 }; + +#endif /* UTF-8 support needed */ + + +#ifdef SUPPORT_UNICODE + +/* Table to translate from particular type value to the general value. */ + +const uint32_t PRIV(ucp_gentype)[] = { + ucp_C, ucp_C, ucp_C, ucp_C, ucp_C, /* Cc, Cf, Cn, Co, Cs */ + ucp_L, ucp_L, ucp_L, ucp_L, ucp_L, /* Ll, Lu, Lm, Lo, Lt */ + ucp_M, ucp_M, ucp_M, /* Mc, Me, Mn */ + ucp_N, ucp_N, ucp_N, /* Nd, Nl, No */ + ucp_P, ucp_P, ucp_P, ucp_P, ucp_P, /* Pc, Pd, Pe, Pf, Pi */ + ucp_P, ucp_P, /* Ps, Po */ + ucp_S, ucp_S, ucp_S, ucp_S, /* Sc, Sk, Sm, So */ + ucp_Z, ucp_Z, ucp_Z /* Zl, Zp, Zs */ +}; + +/* This table encodes the rules for finding the end of an extended grapheme +cluster. Every code point has a grapheme break property which is one of the +ucp_gbXX values defined in pcre2_ucp.h. These changed between Unicode versions +10 and 11. The 2-dimensional table is indexed by the properties of two adjacent +code points. The left property selects a word from the table, and the right +property selects a bit from that word like this: + + PRIV(ucp_gbtable)[left-property] & (1 << right-property) + +The value is non-zero if a grapheme break is NOT permitted between the relevant +two code points. The breaking rules are as follows: + +1. Break at the start and end of text (pretty obviously). + +2. Do not break between a CR and LF; otherwise, break before and after + controls. + +3. Do not break Hangul syllable sequences, the rules for which are: + + L may be followed by L, V, LV or LVT + LV or V may be followed by V or T + LVT or T may be followed by T + +4. Do not break before extending characters or zero-width-joiner (ZWJ). + +The following rules are only for extended grapheme clusters (but that's what we +are implementing). + +5. Do not break before SpacingMarks. + +6. Do not break after Prepend characters. + +7. Do not break within emoji modifier sequences or emoji zwj sequences. That + is, do not break between characters with the Extended_Pictographic property. + Extend and ZWJ characters are allowed between the characters; this cannot be + represented in this table, the code has to deal with it. + +8. Do not break within emoji flag sequences. That is, do not break between + regional indicator (RI) symbols if there are an odd number of RI characters + before the break point. This table encodes "join RI characters"; the code + has to deal with checking for previous adjoining RIs. + +9. Otherwise, break everywhere. +*/ + +#define ESZ (1<