No need to zero the array of token values ends.

[TP_AL_C.git] / lexer / lexical_analyzer.c

/* Lexical analyzer */

#include <stdlib.h>
#include <stdbool.h>

#include "global_vars.h"
#include "print_helper.h"

wint_t c;

/* It looks silly to check for each characters but for debugging, it's just the way to go */
static bool isAlphaNum() {
    if (c == L'a' || c == L'b' || c == L'c' || c == L'd' || c == L'e' || c == L'f' || c == L'g' || \
        c == L'h' || c == L'i' || c == L'j' || c == L'k' || c == L'l' || c == L'm' || c == L'n' || \
        c == L'o' || c == L'p' || c == L'q' || c == L'r' || c == L's' || c == L't' || c == L'u' || \
        c == L'v' || c == L'w' || c == L'x' || c == L'y' || c == L'z' || \
        c == L'A' || c == L'B' || c == L'C' || c == L'D' || c == L'E' || c == L'F' || c == L'G' || \
        c == L'H' || c == L'I' || c == L'J' || c == L'K' || c == L'L' || c == L'M' || c == L'N' || \
        c == L'O' || c == L'P' || c == L'Q' || c == L'R' || c == L'S' || c == L'T' || c == L'U' || \
        c == L'V' || c == L'W' || c == L'X' || c == L'Y' || c == L'Z' || \
        c == L'.' || c == L'?' || c == L'!' || c == L',' || c == L';' || c == L':' || c == L'-' || \
        c == L'\''|| c == L'#' || \
        c == L'0' || c == L'1' || c == L'2' || c == L'3' || c == L'4' || c == L'5' || c == L'6' || \
        c == L'7' || c == L'8' || c == L'9' || \
        //FIXME: Accentued characters (aka multibytes characters) support is still buggy
        c == L'à' || c == L'â' || c == L'ç' || c == L'è' || c == L'é' || c == L'î' || c == L'ô' || \
        c == L'ù' || c == L'û' || \
        c == L'À' || c == L'Â' || c == L'Ç' || c == L'È' || c == L'É' || c == L'Î' || c == L'Ô' || \
        c == L'Ù' || c == L'Û') {
            return true;
        }
        return false;
}

static bool isSeparator() {
    if (c == L'\t' || c == L' ' || c == L'\n') {
        return true;
    }
    return false;
}

static bool isEOF() {
    if (c == WEOF) {
        return true;
    }
    return false;
}

int scanner() {
    tokenValue[0] = 0;
    unsigned int i = 0;
    wchar_t m[6];

init:
    if (c == L' ' || c == L'\t') {
        c = fgetwc(source);
        goto init;
    }
    if (c == L'\n') {
        c = fgetwc(source);
        goto initLV1;
    }
    if (c == L'>') {
        c = fgetwc(source);
        goto MC1;
    }
    if (c == L'=') {
        c = fgetwc(source);
        goto S1SS1;
    }
    if (isAlphaNum()) {
        token[tokenFound].value[i] = c;
        tokenValue[i] = c;
        i++;
        c = fgetwc(source);
        goto M1;
    }
    if (isEOF()) {
        goto FIN;
    }
    goto error;

MC1:
    if (c == L'A' && !wcscmp(fgetws(m, 6, source), L"uteur")) {
        wcscpy((wchar_t*)token[tokenFound].value, L">Auteur");
        wcscpy((wchar_t*)tokenValue, L">Auteur");
        c = fgetwc(source);
        goto MC2;
    }
    if (c == L'T' && !wcscmp(fgetws(m, 5, source), L"itre")) {
        wcscpy((wchar_t*)token[tokenFound].value, L">Titre");
        wcscpy((wchar_t*)tokenValue, L">Titre");
        c = fgetwc(source);
        goto MC2;
    }
    goto error;

S1SS1:
    if (c == L'=') {
        c = fgetwc(source);
        goto SS2;
    }
    if (isSeparator() || isEOF()) {
        goto SECTION;
    }
    goto error;

SS2:
    if (isSeparator() || isEOF()) {
        goto SSECTION;
    }
    goto error;

SECTION:
    tokenType = SECTION;
    return EXIT_SUCCESS;

SSECTION:
    tokenType = SSECTION;
    return EXIT_SUCCESS;

M1:
    if (isAlphaNum()) {
        token[tokenFound].value[i] = c;
        tokenValue[i] = c;
        i++;
        c = fgetwc(source);
        goto M1;
    }
    if (isSeparator() || isEOF()) {
        goto MOT;
    }
    goto error;

initLV1:
    if (c == L' ' || c == L'\t') {
        c = fgetwc(source);
        goto initLV1;
    }
    if (c == L'\n') {
        c = fgetwc(source);
        goto initLV1LV2;
    }
    if (isAlphaNum()) {
        token[tokenFound].value[i] = c;
        tokenValue[i] = c;
        i++;
        c = fgetwc(source);
        goto M1;
    }
    if (c == L'=') {
        c = fgetwc(source);
        goto S1SS1;
    }
    if (c == L'>') {
        c = fgetwc(source);
        goto MC1;
    }
    if (isEOF()) {
        goto FIN;
    }
    goto error;

initLV1LV2:
    if (isSeparator()) {
        c = fgetwc(source);
        goto initLV1LV2;
    }
    if (isAlphaNum()) {
        goto NPARA;
    }
    if (c == L'>') {
        c = fgetwc(source);
        goto MC1;
    }
    if (c == L'=') {
        c = fgetwc(source);
        goto S1SS1;
    }
    if (isEOF()) {
        goto FIN;
    }
    goto error;

NPARA:
    tokenType = NPARA;
    return EXIT_SUCCESS;

MOT:
    tokenValue[i] = 0;
    tokenType = MOT;
    return EXIT_SUCCESS;

MC2:
    if (isSeparator() || isEOF()) {
        goto MOTCLE;
    }
    goto error;

MOTCLE:
    tokenType = MOTCLE;
    return EXIT_SUCCESS;

FIN:
    tokenType = FIN;
    return EXIT_SUCCESS;

error:
    if (tokenType == MOT || tokenType == MOTCLE) {
        fwprintf(stderr, L"%s error with token type: %s and value: %ls\n",
                 __func__,
                 tokenTypestr[tokenType],
                 token[tokenFound].value);
    } else {
        fwprintf(stderr, L"%s error with token type: %s\n",
                 __func__,
                 tokenTypestr[tokenType]);
    }
    fflush(stderr);
    tokenType = FIN;
    exit(EXIT_FAILURE);
}