parse_text.c
author viric@llimona
Sat, 01 Sep 2007 00:50:11 +0200
changeset 15 17a66ceb774a
parent 14 a961bb8806b9
child 17 d95d9e7a2b81
permissions -rw-r--r--
Pritraktado de majuskloj per ICU.

#include <stdio.h>
#include "dictre.h"

static void give_accent_to_word(const char *word)
{
    char def[MAXDEF];
    char low[MAXWORD];
    char recased[MAXWORD];
    enum Case vcase[MAXWORD];

    /* Get case */
    get_case(vcase, word);

    /* Get lowercase version */
    get_lowcase_str(low, word);

    /* Find the lowercase version */
    find_def(low, def);
    if (def[0] != 0) /* found */
    {
        /* Print the word UNTIL a space.
         * the definition will have the form:
         *    ACCENTED_WORD NOMINATIVE1 NOMINATIVE2 ... \n */
        char *first_space;
        char spacepos;
        first_space = strchr(def, ' ');
        if (first_space != 0) /* Space found */
        {
            spacepos = first_space - def;
            def[spacepos] = 0; /* Mark an end of string */
            reapply_case(recased, def, vcase);
            printf("%s", recased);
        }
        return;
    }

    /* if first_space == 0 or word not found */
    printf("%s", word);
}

static void process_text(FILE *in, int pos, int length)
{
    unsigned char tmp[MAXWORD];
    int wordpos = 0;
    do
    {
        int c;
        /* Check pos only if length >= 0 */
        if (length >= 0 && pos >= length)
            break;
        c = fgetc(in);
        if (c == EOF)
            break;
        if (is_ASCII(c))
        {
            if (wordpos != 0)
            {
                tmp[wordpos] = 0;
                give_accent_to_word(tmp);
                wordpos = 0;
            }
            putchar(c);
        }
        else /* non-ASCII - we consider it russian */
        {
            tmp[wordpos++] = c;
        }

        pos += 1;
    } while(1);
}

int main()
{
    init_dictionary();
    process_text(stdin, 0, -1);
    end_dictionary();
}