src/lib/str-sanitize.c

178N/A/* Copyright (c) 2004-2018 Dovecot authors, see the included COPYING file */
178N/A
178N/A#include "lib.h"
178N/A#include "unichar.h"
178N/A#include "str.h"
178N/A#include "str-sanitize.h"
178N/A
178N/Astatic size_t str_sanitize_skip_start(const char *src, size_t max_bytes)
178N/A{
178N/A    unichar_t chr;
178N/A    size_t i;
178N/A
178N/A    for (i = 0; i < max_bytes && src[i] != '\0'; ) {
178N/A        int len = uni_utf8_get_char_n(src+i, max_bytes-i, &chr);
178N/A        if (len <= 0)
178N/A            break;
178N/A        if ((unsigned char)src[i] < 32)
178N/A            break;
178N/A        i += len;
178N/A    }
178N/A    i_assert(i <= max_bytes);
178N/A    return i;
178N/A}
178N/A
178N/Astatic void str_sanitize_truncate_char(string_t *dest, unsigned int initial_pos)
178N/A{
178N/A    const unsigned char *data = str_data(dest);
178N/A    size_t len = str_len(dest);
178N/A
178N/A    if (len == initial_pos)
178N/A        return;
178N/A
178N/A    i_assert(len > 0);
178N/A    if ((data[len-1] & 0x80) == 0) {
2015N/A        str_truncate(dest, len-1);
178N/A        return;
178N/A    }
178N/A    /* truncate UTF-8 sequence. */
178N/A    while (len > 0 && (data[len-1] & 0xc0) == 0x80)
178N/A        len--;
178N/A    if (len > 0 && (data[len-1] & 0xc0) == 0xc0)
178N/A        len--;
178N/A    if (len >= initial_pos)
178N/A        str_truncate(dest, len);
178N/A}
178N/A
178N/Avoid str_sanitize_append(string_t *dest, const char *src, size_t max_bytes)
178N/A{
178N/A    size_t initial_pos = str_len(dest);
178N/A    unichar_t chr;
178N/A    size_t i;
178N/A
178N/A    for (i = 0; i < max_bytes && src[i] != '\0'; ) {
178N/A        int len = uni_utf8_get_char_n(src+i, max_bytes-i, &chr);
178N/A        if (len == 0)
178N/A            break; /* input ended too early */
178N/A
178N/A        if (len < 0) {
178N/A            /* invalid UTF-8 */
178N/A            str_append_c(dest, '?');
178N/A            i++;
            continue;
        }
        if ((unsigned char)src[i] < 32)
            str_append_c(dest, '?');
        else
            str_append_n(dest, src+i, len);
        i += len;
    }

    if (src[i] != '\0') {
        if (max_bytes < 3)
            str_truncate(dest, initial_pos);
        else {
            while (str_len(dest) - initial_pos > max_bytes-3)
                str_sanitize_truncate_char(dest, initial_pos);
        }
        str_append(dest, "...");
    }
}

const char *str_sanitize(const char *src, size_t max_bytes)
{
    string_t *str;
    size_t i;

    if (src == NULL)
        return NULL;

    i = str_sanitize_skip_start(src, max_bytes);
    if (src[i] == '\0')
        return src;

    str = t_str_new(I_MIN(max_bytes, 256));
    str_sanitize_append(str, src, max_bytes);
    return str_c(str);
}