universe@27: /*
universe@34:  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS HEADER. 
universe@57:  * Copyright 2018 Mike Becker. All rights reserved.
universe@34:  * 
universe@34:  * Redistribution and use in source and binary forms, with or without
universe@34:  * modification, are permitted provided that the following conditions are met:
universe@34:  * 
universe@34:  * 1. Redistributions of source code must retain the above copyright
universe@34:  * notice, this list of conditions and the following disclaimer.
universe@34:  * 
universe@34:  * 2. Redistributions in binary form must reproduce the above copyright
universe@34:  * notice, this list of conditions and the following disclaimer in the
universe@34:  * documentation and/or other materials provided with the distribution.
universe@34:  * 
universe@34:  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
universe@34:  * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
universe@34:  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
universe@34:  * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
universe@34:  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
universe@34:  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
universe@34:  * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
universe@34:  * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
universe@34:  * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
universe@57:  * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
universe@27:  */
universe@27: 
universe@27: #include "regex_parser.h"
universe@66: #include <ctype.h>
universe@27: 
universe@27: regex_parser_t* new_regex_parser_t() {
universe@27:   regex_parser_t* ret = malloc(sizeof(regex_parser_t));
universe@27:   if (ret != NULL) {
universe@27:     ret->pattern_list = new_string_list_t();
universe@66:     ret->matched_counted = 0;
universe@27:     ret->pattern_match = 0;
universe@27:     ret->compiled_patterns = NULL;
universe@28:     ret->compiled_pattern_count = 0;
universe@66:     ret->count_chars = false;
universe@27:   }
universe@27:   return ret;
universe@27: }
universe@27: 
universe@54: void regex_parser_reset(regex_parser_t* parser) {
universe@66:   parser->pattern_match = parser->matched_counted = 0;
universe@54: }
universe@54: 
universe@28: void regex_destcomppats(regex_parser_t* parser) {
universe@28:   if (parser->compiled_patterns != NULL) {
universe@66:     for (unsigned i = 0 ; i < parser->compiled_pattern_count ; i++) {
universe@28:       if (parser->compiled_patterns[i] != NULL) {
universe@28:         free(parser->compiled_patterns[i]);
universe@28:       }
universe@28:     }
universe@28:     free(parser->compiled_patterns);
universe@28:     parser->compiled_patterns = NULL;
universe@28:     parser->compiled_pattern_count = 0;
universe@28:   }
universe@28: }
universe@28: 
universe@27: void destroy_regex_parser_t(regex_parser_t* parser) {
universe@28:   regex_destcomppats(parser);
universe@27:   destroy_string_list_t(parser->pattern_list);
universe@27:   free(parser);
universe@27: }
universe@27: 
universe@27: bool regex_parser_matching(regex_parser_t* parser) {
universe@27:   return parser->pattern_match > 0;
universe@27: }
universe@27: 
universe@66: static unsigned regex_parser_count_chars(const char* input,
universe@66:                                          unsigned start, unsigned end) {
universe@66:   unsigned ret = 0;
universe@66:   for (unsigned i = start ; i < end ; i++) {
universe@66:     ret += isspace(input[i]) ? 0 : 1;
universe@66:   }
universe@66:   return ret;
universe@66: }
universe@66: 
universe@28: int regex_parser_do(regex_parser_t* parser, char* input) {
universe@28:   int err = REG_NOMATCH;
universe@28:   if (parser->compiled_pattern_count > 0) {
universe@28:     regmatch_t match;
universe@27: 
universe@28:     if (regex_parser_matching(parser)) {
universe@66:       if (parser->count_chars) {
universe@66:         parser->matched_counted +=
universe@66:             regex_parser_count_chars(input, 0, strlen(input));
universe@66:       } else {
universe@66:         parser->matched_counted++;
universe@66:       }
universe@28: 
universe@28:       err = regexec(parser->compiled_patterns[parser->pattern_match],
universe@28:           input, 1, &match, 0);
universe@28:       if (err > 0 && err != REG_NOMATCH) {
universe@28:         fprintf(stderr, "Regex-Error: 0x%08x", err);
universe@28:       }
universe@28:       if (err == 0) {
universe@28:         parser->pattern_match = 0;
universe@66:         size_t input_len = strlen(input);
universe@66:         if (match.rm_eo < input_len) {
universe@66:           if (parser->count_chars) {
universe@66:             /* do not exclude chars that occur after pattern end */
universe@66:             parser->matched_counted -=
universe@66:                 regex_parser_count_chars(input, match.rm_eo, input_len);
universe@66:           } else {
universe@66:             /* do not exclude line, if it does not end with the pattern */
universe@66:             parser->matched_counted--;
universe@66:           }
universe@28:         }
universe@28:       }
universe@28:     } else {
universe@66:       for (unsigned i = 0 ; i < parser->compiled_pattern_count - 1 ; i += 2) {
universe@28:         err = regexec(parser->compiled_patterns[i], input, 1, &match, 0);
universe@28:         if (err > 0 && err != REG_NOMATCH) {
universe@28:           fprintf(stderr, "Regex-Error: 0x%08x", err);
universe@28:         }
universe@28:         if (err == 0) {
universe@66:           /* a start pattern matches, start counting */
universe@66:           parser->matched_counted = 0;
universe@66:           /* Check, if end pattern is also in this line */
universe@28:           parser->pattern_match = i+1;
universe@28:           regex_parser_do(parser, input);
universe@66:           /* If something was found, determine what exactly to exclude */
universe@66:           if (parser->matched_counted > 0) {
universe@66:             if (parser->count_chars) {
universe@66:               /* do not exclude the characters before the pattern */
universe@66:               parser->matched_counted -=
universe@66:                   regex_parser_count_chars(input, 0, match.rm_so);
universe@66:             } else {
universe@66:               /* do not match line, if it does not start with the pattern */
universe@66:               if (match.rm_so > 0) {
universe@66:                 parser->matched_counted--;
universe@66:               }
universe@66:             }
universe@28:           }
universe@28:           break;
universe@28:         }
universe@27:       }
universe@27:     }
universe@27:   }
universe@28:   return err;
universe@27: }
universe@28: 
universe@28: bool regex_compile_all(regex_parser_t* parser) {
universe@28:   bool success = true;
universe@28:   size_t pcount = parser->pattern_list->count;
universe@28:   if (pcount > 0) {
universe@28:     regex_destcomppats(parser);
universe@28:     parser->compiled_patterns = calloc(pcount, sizeof(regex_t));
universe@28:     parser->compiled_pattern_count = pcount;
universe@28: 
universe@28:     regex_t* re;
universe@66:     for (unsigned i = 0 ; i < pcount ; i++) {
universe@28:       re = malloc(sizeof(regex_t));
universe@28:       if (regcomp(re, parser->pattern_list->items[i], REG_EXTENDED) == 0) {
universe@28:         parser->compiled_patterns[i] = re;
universe@28:       } else {
universe@28:         fprintf(stderr, "Cannot compile pattern: %s\n",
universe@28:             (parser->pattern_list->items[i]));
universe@28:         parser->compiled_patterns[i] = NULL;
universe@28:         success = false;
universe@28:       }
universe@28:     }
universe@28:   }
universe@28:   return success;
universe@28: }