src/er.cc

   1 /*
   2  * Copyright (C) 2004 Jesus Gimenez, Lluis Marquez and Senen Moya
   3  *
   4  * This library is free software; you can redistribute it and/or
   5  * modify it under the terms of the GNU Lesser General Public
   6  * License as published by the Free Software Foundation; either
   7  * version 2.1 of the License, or (at your option) any later version.
   8  *
   9  * This library is distributed in the hope that it will be useful,
  10  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  11  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  12  * Lesser General Public License for more details.
  13  *
  14  * You should have received a copy of the GNU Lesser General Public
  15  * License along with this library; if not, write to the Free Software
  16  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  17  */
  18
  19 #include "stdio.h"
  20 #include "er.h"
  21
  22 /*****************************************************************
  23  * Regular expressions
  24  *****************************************************************/
  25
  26 regex_t  erCard,erCardPunct,erCardSeps,erCardSuffix;
  27 regex_t  erMultiWord,erContainNum,erStartCap,erStartLower,erStartNumber,
  28 erAllUp,erAllLow,erContainCap,erContainCaps,erContainPeriod,erContainComma;
  29
  30 /**************************************************/
  31
  32 #define ER_STARTCAP "^[A-Z\307\321\301\311\315\323\332\300\310\314\322\331\304\313\317\326\334].*$"
  33 #define ER_STARTLOWER   "^[a-z\347\361\341\351\355\363\372\340\350\354\362\371\344\353\357\366\374].*$"
  34 #define ER_STARTNUMBER  "^[0-9].*$"
  35 #define ER_ALLUP    "^[A-Z\307\321\301\311\315\323\332\300\310\314\322\331\304\313\317\326\334]+$"
  36 #define ER_ALLLOW   "^[a-z\347\361\341\351\355\363\372\340\350\354\362\371\344\353\357\366\374]+$"
  37 #define ER_CONTAINCAP   "^.+[A-Z\307\321\301\311\315\323\332\300\310\314\322\331\304\313\317\326\334].*$"
  38 #define ER_CONTAINCAPS  "^.*[A-Z\307\321\301\311\315\323\332\300\310\314\322\331\304\313\317\326\334].*[A-Z\307\321\301\311\315\323\332\300\310\314\322\331\304\313\317\326\334].*$"
  39 #define ER_CONTAINPERIOD "^.*[.].*$"
  40 #define ER_CONTAINCOMMA "^.*[,].*$"
  41 #define ER_CONTAINNUM   "^.*[0-9].*$"
  42 #define ER_MULTIWORD    "^.*[-].*$"
  43 #define ER_CARD     "^[0-9]+$"
  44 #define ER_CARDPUNCT    "^[0-9]+[,!?:.]+$"
  45 #define ER_CARDSEPS "^[0-9]+[-,:\\/.][0-9,:\\/.-]+$"
  46 #define ER_CARDSUFFIX   "^[0-9]+[^0-9]+.*$"
  47
  48 /**************************************************/
  49
  50 void erCompRegExp()
  51 {
  52         regcomp (&erCard,ER_CARD,REG_EXTENDED);
  53         regcomp (&erCardPunct,ER_CARDPUNCT,REG_EXTENDED);
  54         regcomp (&erCardSeps,ER_CARDSEPS,REG_EXTENDED);
  55         regcomp (&erCardSuffix,ER_CARDSUFFIX,REG_EXTENDED);
  56
  57         regcomp (&erStartCap,ER_STARTCAP,REG_EXTENDED);
  58         regcomp (&erStartNumber,ER_STARTNUMBER,REG_EXTENDED);
  59         regcomp (&erStartLower,ER_STARTLOWER,REG_EXTENDED);
  60         regcomp (&erAllUp,ER_ALLUP,REG_EXTENDED);
  61         regcomp (&erAllLow,ER_ALLLOW,REG_EXTENDED);
  62         regcomp (&erContainCap,ER_CONTAINCAP,REG_EXTENDED);
  63         regcomp (&erContainCaps,ER_CONTAINCAPS,REG_EXTENDED);
  64         regcomp (&erContainPeriod,ER_CONTAINPERIOD,REG_EXTENDED);
  65         regcomp (&erContainComma,ER_CONTAINCOMMA,REG_EXTENDED);
  66         regcomp (&erContainNum,ER_CONTAINNUM,REG_EXTENDED);
  67         regcomp (&erMultiWord,ER_MULTIWORD,REG_EXTENDED);
  68 }
  69
  70
  71 /**************************************************/
  72
  73 void erFreeRegExp()
  74 {
  75         regfree(&erCard);
  76         regfree(&erCardSuffix);
  77         regfree(&erCardSeps);
  78         regfree(&erCardPunct);
  79
  80         regfree(&erStartCap);
  81         regfree(&erStartLower);
  82         regfree(&erStartNumber);
  83         regfree(&erAllUp);
  84         regfree(&erAllLow);
  85         regfree(&erContainCap);
  86         regfree(&erContainCaps);
  87         regfree(&erContainComma);
  88         regfree(&erContainPeriod);
  89         regfree(&erContainNum);
  90         regfree(&erMultiWord);
  91 }
  92
  93
  94 /**************************************************/
  95
  96 /*
  97  * return 1 if str is like the regular expression
  98  * in other case return 0
  99  */
 100 int erLookRegExp2(void *er,char * str)
 101 {
 102         int ret=0;
 103
 104         if (!regexec ((regex_t *)er,str,0,NULL,0)) return  1;
 105
 106         return 0;
 107 }
 108
 109
 110 /**************************************************/
 111
 112 int erLookRegExp(char *m)
 113 {
 114         int ret=-1;
 115
 116         if (!regexec (&erCardPunct,m,0,NULL,0)) ret=CARDPUNCT;
 117         else if (!regexec (&erCardSeps,m,0,NULL,0)) ret=CARDSEPS;
 118         else if (!regexec (&erCardSuffix,m,0,NULL,0)) ret=CARDSUFFIX;
 119         else if (!regexec (&erCard,m,0,NULL,0))  ret=CARD;
 120
 121         return ret;
 122 }