utf8proc
annotate pgsql/utf8proc_pgsql.c @ 3:4ee0d5f54af1
Version 1.0
- added the LUMP option, which lumps certain characters together (see lump.txt) (also used for the PostgreSQL "unifold" function)
- added the STRIPMARK option, which strips marking characters (or marks of composed characters)
- deprecated ruby method String#char_ary in favour of String#utf8chars
- added the LUMP option, which lumps certain characters together (see lump.txt) (also used for the PostgreSQL "unifold" function)
- added the STRIPMARK option, which strips marking characters (or marks of composed characters)
- deprecated ruby method String#char_ary in favour of String#utf8chars
author | jbe |
---|---|
date | Sun Sep 17 12:00:00 2006 +0200 (2006-09-17) |
parents | aaad485d5335 |
children | c18366878af9 |
rev | line source |
---|---|
jbe@0 | 1 /* |
jbe@0 | 2 * Copyright (c) 2006, FlexiGuided GmbH, Berlin, Germany |
jbe@0 | 3 * Author: Jan Behrens <jan.behrens@flexiguided.de> |
jbe@0 | 4 * All rights reserved. |
jbe@0 | 5 * |
jbe@0 | 6 * Redistribution and use in source and binary forms, with or without |
jbe@0 | 7 * modification, are permitted provided that the following conditions are |
jbe@0 | 8 * met: |
jbe@0 | 9 * |
jbe@0 | 10 * 1. Redistributions of source code must retain the above copyright |
jbe@0 | 11 * notice, this list of conditions and the following disclaimer. |
jbe@0 | 12 * 2. Redistributions in binary form must reproduce the above copyright |
jbe@0 | 13 * notice, this list of conditions and the following disclaimer in the |
jbe@0 | 14 * documentation and/or other materials provided with the distribution. |
jbe@0 | 15 * 3. Neither the name of the FlexiGuided GmbH nor the names of its |
jbe@0 | 16 * contributors may be used to endorse or promote products derived from |
jbe@0 | 17 * this software without specific prior written permission. |
jbe@0 | 18 * |
jbe@0 | 19 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS |
jbe@0 | 20 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT |
jbe@0 | 21 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A |
jbe@0 | 22 * PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER |
jbe@0 | 23 * OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, |
jbe@0 | 24 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, |
jbe@0 | 25 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR |
jbe@0 | 26 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF |
jbe@0 | 27 * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING |
jbe@0 | 28 * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS |
jbe@0 | 29 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
jbe@0 | 30 * |
jbe@0 | 31 */ |
jbe@0 | 32 |
jbe@0 | 33 |
jbe@0 | 34 /* |
jbe@0 | 35 * File name: pgsql/utf8proc_pgsql.c |
jbe@3 | 36 * Version: 1.0 |
jbe@3 | 37 * Last changed: 2006-09-17 |
jbe@0 | 38 * |
jbe@0 | 39 * Description: |
jbe@0 | 40 * PostgreSQL extension to provide a function 'unifold', which can be used |
jbe@0 | 41 * to case-fold and normalize index fields. |
jbe@0 | 42 */ |
jbe@0 | 43 |
jbe@0 | 44 |
jbe@0 | 45 #include "../utf8proc.c" |
jbe@0 | 46 |
jbe@0 | 47 #include <postgres.h> |
jbe@0 | 48 #include <utils/elog.h> |
jbe@0 | 49 #include <fmgr.h> |
jbe@1 | 50 #include <string.h> |
jbe@0 | 51 #include <unistd.h> |
jbe@0 | 52 #include <utils/builtins.h> |
jbe@0 | 53 |
jbe@2 | 54 #define UTF8PROC_PGSQL_OPTS ( UTF8PROC_REJECTNA | UTF8PROC_COMPAT | \ |
jbe@2 | 55 UTF8PROC_COMPOSE | UTF8PROC_STABLE | UTF8PROC_IGNORE | UTF8PROC_STRIPCC | \ |
jbe@3 | 56 UTF8PROC_NLF2LF | UTF8PROC_CASEFOLD | UTF8PROC_LUMP ) |
jbe@1 | 57 |
jbe@0 | 58 PG_FUNCTION_INFO_V1(utf8proc_pgsql_unifold); |
jbe@0 | 59 Datum utf8proc_pgsql_unifold(PG_FUNCTION_ARGS) { |
jbe@1 | 60 text *input_string; |
jbe@1 | 61 text *output_string = NULL; |
jbe@0 | 62 ssize_t result; |
jbe@1 | 63 input_string = PG_GETARG_TEXT_P(0); |
jbe@1 | 64 do { |
jbe@1 | 65 result = utf8proc_decompose( |
jbe@1 | 66 VARDATA(input_string), VARSIZE(input_string) - VARHDRSZ, |
jbe@1 | 67 NULL, 0, UTF8PROC_PGSQL_OPTS |
jbe@1 | 68 ); |
jbe@1 | 69 if (result < 0) break; |
jbe@1 | 70 if (result > (SIZE_MAX-1-VARHDRSZ)/sizeof(int32_t)) { |
jbe@1 | 71 result = UTF8PROC_ERROR_OVERFLOW; |
jbe@1 | 72 break; |
jbe@1 | 73 } |
jbe@1 | 74 output_string = palloc(result * sizeof(int32_t) + 1 + VARHDRSZ); |
jbe@1 | 75 // reserve one extra byte for termination |
jbe@1 | 76 if (!output_string) { |
jbe@1 | 77 result = UTF8PROC_ERROR_NOMEM; |
jbe@1 | 78 break; |
jbe@1 | 79 } |
jbe@1 | 80 result = utf8proc_decompose( |
jbe@1 | 81 VARDATA(input_string), VARSIZE(input_string) - VARHDRSZ, |
jbe@1 | 82 (int32_t *)VARDATA(output_string), result, UTF8PROC_PGSQL_OPTS); |
jbe@1 | 83 if (result < 0) break; |
jbe@1 | 84 result = utf8proc_reencode((int32_t *)VARDATA(output_string), result, |
jbe@1 | 85 UTF8PROC_PGSQL_OPTS); |
jbe@1 | 86 } while (0); |
jbe@1 | 87 PG_FREE_IF_COPY(input_string, 0); |
jbe@0 | 88 if (result < 0) { |
jbe@0 | 89 int sqlerrcode; |
jbe@1 | 90 if (output_string) pfree(output_string); |
jbe@0 | 91 switch(result) { |
jbe@0 | 92 case UTF8PROC_ERROR_NOMEM: |
jbe@0 | 93 sqlerrcode = ERRCODE_OUT_OF_MEMORY; break; |
jbe@0 | 94 case UTF8PROC_ERROR_OVERFLOW: |
jbe@0 | 95 sqlerrcode = ERRCODE_PROGRAM_LIMIT_EXCEEDED; break; |
jbe@0 | 96 case UTF8PROC_ERROR_INVALIDUTF8: |
jbe@0 | 97 case UTF8PROC_ERROR_NOTASSIGNED: |
jbe@1 | 98 PG_RETURN_NULL(); |
jbe@0 | 99 default: |
jbe@0 | 100 sqlerrcode = ERRCODE_INTERNAL_ERROR; |
jbe@0 | 101 } |
jbe@0 | 102 ereport(ERROR, ( |
jbe@0 | 103 errcode(sqlerrcode), |
jbe@0 | 104 errmsg("%s", utf8proc_errmsg(result)) |
jbe@0 | 105 )); |
jbe@1 | 106 } else { |
jbe@1 | 107 VARATT_SIZEP(output_string) = result + VARHDRSZ; |
jbe@1 | 108 PG_RETURN_TEXT_P(output_string); |
jbe@0 | 109 } |
jbe@1 | 110 PG_RETURN_NULL(); // prohibit compiler warning |
jbe@0 | 111 } |
jbe@0 | 112 |
jbe@0 | 113 |