lexer/lexical_analyzer.c

   1 /* Lexical analyzer */
   2
   3 #include <stdlib.h>
   4 #include <stdbool.h>
   5
   6 #include "global_vars.h"
   7 #include "print_helper.h"
   8
   9 wint_t c;
  10
  11 /* It looks silly to check for each characters but for debugging, it's just the way to go */
  12 static bool isAlphaNum() {
  13     if (c == L'a' || c == L'b' || c == L'c' || c == L'd' || c == L'e' || c == L'f' || c == L'g' || \
  14         c == L'h' || c == L'i' || c == L'j' || c == L'k' || c == L'l' || c == L'm' || c == L'n' || \
  15         c == L'o' || c == L'p' || c == L'q' || c == L'r' || c == L's' || c == L't' || c == L'u' || \
  16         c == L'v' || c == L'w' || c == L'x' || c == L'y' || c == L'z' || \
  17         c == L'A' || c == L'B' || c == L'C' || c == L'D' || c == L'E' || c == L'F' || c == L'G' || \
  18         c == L'H' || c == L'I' || c == L'J' || c == L'K' || c == L'L' || c == L'M' || c == L'N' || \
  19         c == L'O' || c == L'P' || c == L'Q' || c == L'R' || c == L'S' || c == L'T' || c == L'U' || \
  20         c == L'V' || c == L'W' || c == L'X' || c == L'Y' || c == L'Z' || \
  21         c == L'.' || c == L'?' || c == L'!' || c == L',' || c == L';' || c == L':' || c == L'-' || \
  22         c == L'\''|| c == L'#' || \
  23         c == L'0' || c == L'1' || c == L'2' || c == L'3' || c == L'4' || c == L'5' || c == L'6' || \
  24         c == L'7' || c == L'8' || c == L'9' || \
  25         // FIXME: Accentued characters (aka multibytes characters) support is still buggy
  26         c == L'à' || c == L'â' || c == L'ç' || c == L'è' || c == L'é' || c == L'î' || c == L'ô' || \
  27         c == L'ù' || c == L'û' || \
  28         c == L'À' || c == L'Â' || c == L'Ç' || c == L'È' || c == L'É' || c == L'Î' || c == L'Ô' || \
  29         c == L'Ù' || c == L'Û') {
  30             return true;
  31         }
  32         return false;
  33 }
  34
  35 static bool isSeparator() {
  36     if (c == L'\t' || c == L' ' || c == L'\n') {
  37         return true;
  38     }
  39     return false;
  40 }
  41
  42 static bool isEOF() {
  43     if (c == WEOF) {
  44         return true;
  45     }
  46     return false;
  47 }
  48
  49 int scanner() {
  50     unsigned int i = 0;
  51     wchar_t m[6];
  52
  53 init:
  54     if (c == L' ' || c == L'\t') {
  55         c = fgetwc(source);
  56         goto init;
  57     }
  58     if (c == L'\n') {
  59         c = fgetwc(source);
  60         goto initLV1;
  61     }
  62     if (c == L'>') {
  63         c = fgetwc(source);
  64         goto MC1;
  65     }
  66     if (c == L'=') {
  67         c = fgetwc(source);
  68         goto S1SS1;
  69     }
  70     if (isAlphaNum()) {
  71         token[tokenFound].value[i] = c;
  72         i++;
  73         c = fgetwc(source);
  74         goto M1;
  75     }
  76     if (isEOF()) {
  77         goto FIN;
  78     }
  79     goto error;
  80
  81 MC1:
  82     if (c == L'A' && !wcscmp(fgetws(m, 6, source), L"uteur")) {
  83         wcscpy((wchar_t*)token[tokenFound].value, L">Auteur");
  84         c = fgetwc(source);
  85         goto MC2;
  86     }
  87     if (c == L'T' && !wcscmp(fgetws(m, 5, source), L"itre")) {
  88         wcscpy((wchar_t*)token[tokenFound].value, L">Titre");
  89         c = fgetwc(source);
  90         goto MC2;
  91     }
  92     goto error;
  93
  94 S1SS1:
  95     if (c == L'=') {
  96         c = fgetwc(source);
  97         goto SS2;
  98     }
  99     if (isSeparator() || isEOF()) {
 100         goto SECTION;
 101     }
 102     goto error;
 103
 104 SS2:
 105     if (isSeparator() || isEOF()) {
 106         goto SSECTION;
 107     }
 108     goto error;
 109
 110 SECTION:
 111     tokenType = SECTION;
 112     return EXIT_SUCCESS;
 113
 114 SSECTION:
 115     tokenType = SSECTION;
 116     return EXIT_SUCCESS;
 117
 118 M1:
 119     if (isAlphaNum()) {
 120         token[tokenFound].value[i] = c;
 121         i++;
 122         c = fgetwc(source);
 123         goto M1;
 124     }
 125     if (isSeparator() || isEOF()) {
 126         goto MOT;
 127     }
 128     goto error;
 129
 130 initLV1:
 131     if (c == L' ' || c == L'\t') {
 132         c = fgetwc(source);
 133         goto initLV1;
 134     }
 135     if (c == L'\n') {
 136         c = fgetwc(source);
 137         goto initLV1LV2;
 138     }
 139     if (isAlphaNum()) {
 140         token[tokenFound].value[i] = c;
 141         i++;
 142         c = fgetwc(source);
 143         goto M1;
 144     }
 145     if (c == L'=') {
 146         c = fgetwc(source);
 147         goto S1SS1;
 148     }
 149     if (c == L'>') {
 150         c = fgetwc(source);
 151         goto MC1;
 152     }
 153     if (isEOF()) {
 154         goto FIN;
 155     }
 156     goto error;
 157
 158 initLV1LV2:
 159     if (isSeparator()) {
 160         c = fgetwc(source);
 161         goto initLV1LV2;
 162     }
 163     if (isAlphaNum()) {
 164         goto NPARA;
 165     }
 166     if (c == L'>') {
 167         c = fgetwc(source);
 168         goto MC1;
 169     }
 170     if (c == L'=') {
 171         c = fgetwc(source);
 172         goto S1SS1;
 173     }
 174     if (isEOF()) {
 175         goto FIN;
 176     }
 177     goto error;
 178
 179 NPARA:
 180     tokenType = NPARA;
 181     return EXIT_SUCCESS;
 182
 183 MOT:
 184     tokenType = MOT;
 185     return EXIT_SUCCESS;
 186
 187 MC2:
 188     if (isSeparator() || isEOF()) {
 189         goto MOTCLE;
 190     }
 191     goto error;
 192
 193 MOTCLE:
 194     tokenType = MOTCLE;
 195     return EXIT_SUCCESS;
 196
 197 FIN:
 198     tokenType = FIN;
 199     return EXIT_SUCCESS;
 200
 201 error:
 202     if (tokenType == MOT || tokenType == MOTCLE) {
 203         fwprintf(stderr, L"%s error with token type: %s and value: %ls\n",
 204                   __func__,
 205                   tokenTypestr[tokenType],
 206                   token[tokenFound].value);
 207     } else {
 208         fwprintf(stderr, L"%s error with token type: %s\n",
 209                   __func__,
 210                   tokenTypestr[tokenType]);
 211     }
 212     tokenType = FIN;
 213     exit(EXIT_FAILURE);
 214 }