lib/search/hex.c

lib/search/hex.c
Manual pages: mc • mcdiff • mcedit • mcview
/* */
This source file includes following definitions.
mc_search__hex_translate_to_regex
mc_search__cond_struct_new_init_hex
mc_search__run_hex
mc_search_hex_prepare_replace_str
   1 /*
   2    Search text engine.
   3    HEX-style pattern matching
   4 
   5    Copyright (C) 2009-2025
   6    Free Software Foundation, Inc.
   7 
   8    Written by:
   9    Slava Zanko <slavazanko@gmail.com>, 2009.
  10 
  11    This file is part of the Midnight Commander.
  12 
  13    The Midnight Commander is free software: you can redistribute it
  14    and/or modify it under the terms of the GNU General Public License as
  15    published by the Free Software Foundation, either version 3 of the License,
  16    or (at your option) any later version.
  17 
  18    The Midnight Commander is distributed in the hope that it will be useful,
  19    but WITHOUT ANY WARRANTY; without even the implied warranty of
  20    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  21    GNU General Public License for more details.
  22 
  23    You should have received a copy of the GNU General Public License
  24    along with this program.  If not, see <https://www.gnu.org/licenses/>.
  25  */
  26 
  27 #include <config.h>
  28 
  29 #include <stdio.h>
  30 
  31 #include "lib/global.h"
  32 #include "lib/strutil.h"
  33 #include "lib/search.h"
  34 
  35 #include "internal.h"
  36 
  37 /*** global variables ****************************************************************************/
  38 
  39 /*** file scope macro definitions ****************************************************************/
  40 
  41 typedef enum
  42 {
  43     MC_SEARCH_HEX_E_OK,
  44     MC_SEARCH_HEX_E_NUM_OUT_OF_RANGE,
  45     MC_SEARCH_HEX_E_INVALID_CHARACTER,
  46     MC_SEARCH_HEX_E_UNMATCHED_QUOTES
  47 } mc_search_hex_parse_error_t;
  48 
  49 /*** file scope type declarations ****************************************************************/
  50 
  51 /*** forward declarations (file scope functions) *************************************************/
  52 
  53 /*** file scope variables ************************************************************************/
  54 
  55 /* --------------------------------------------------------------------------------------------- */
  56 /*** file scope functions ************************************************************************/
  57 /* --------------------------------------------------------------------------------------------- */
  58 
  59 static GString *
  60 mc_search__hex_translate_to_regex (const GString *astr, mc_search_hex_parse_error_t *error_ptr,
     /*   */
  61                                    int *error_pos_ptr)
  62 {
  63     GString *buff;
  64     const char *str;
  65     gsize str_len;
  66     gsize loop = 0;
  67     mc_search_hex_parse_error_t error = MC_SEARCH_HEX_E_OK;
  68 
  69     buff = g_string_sized_new (64);
  70     str = astr->str;
  71     str_len = astr->len;
  72 
  73     while (loop < str_len && error == MC_SEARCH_HEX_E_OK)
  74     {
  75         unsigned int val;
  76         int ptr;
  77 
  78         if (g_ascii_isspace (str[loop]))
  79         {
  80             // Eat-up whitespace between tokens.
  81             while (g_ascii_isspace (str[loop]))
  82                 loop++;
  83         }
  84         else if (sscanf (str + loop, "%x%n", &val, &ptr) == 1)
  85         {
  86             if (val > 255)
  87                 error = MC_SEARCH_HEX_E_NUM_OUT_OF_RANGE;
  88             else
  89             {
  90                 g_string_append_printf (buff, "\\x%02X", val);
  91                 loop += ptr;
  92             }
  93         }
  94         else if (str[loop] == '"')
  95         {
  96             gsize loop2;
  97 
  98             loop2 = loop + 1;
  99 
 100             while (loop2 < str_len)
 101             {
 102                 if (str[loop2] == '"')
 103                     break;
 104                 if (str[loop2] == '\\' && loop2 + 1 < str_len)
 105                     loop2++;
 106                 g_string_append_c (buff, str[loop2]);
 107                 loop2++;
 108             }
 109 
 110             if (str[loop2] == '\0')
 111                 error = MC_SEARCH_HEX_E_UNMATCHED_QUOTES;
 112             else
 113                 loop = loop2 + 1;
 114         }
 115         else
 116             error = MC_SEARCH_HEX_E_INVALID_CHARACTER;
 117     }
 118 
 119     if (error != MC_SEARCH_HEX_E_OK)
 120     {
 121         g_string_free (buff, TRUE);
 122         if (error_ptr != NULL)
 123             *error_ptr = error;
 124         if (error_pos_ptr != NULL)
 125             *error_pos_ptr = loop;
 126         return NULL;
 127     }
 128 
 129     return buff;
 130 }
 131 
 132 /* --------------------------------------------------------------------------------------------- */
 133 /*** public functions ****************************************************************************/
 134 /* --------------------------------------------------------------------------------------------- */
 135 
 136 void
 137 mc_search__cond_struct_new_init_hex (const char *charset, mc_search_t *lc_mc_search,
     /*   */
 138                                      mc_search_cond_t *mc_search_cond)
 139 {
 140     GString *tmp;
 141     mc_search_hex_parse_error_t error = MC_SEARCH_HEX_E_OK;
 142     int error_pos = 0;
 143 
 144     /*
 145      * We may be searching in binary data, which is often invalid UTF-8.
 146      *
 147      * We have to create a non UTF-8 regex (that is, G_REGEX_RAW) or else, as
 148      * the data is invalid UTF-8, both GLib's PCRE and our
 149      * mc_search__g_regex_match_full_safe() are going to fail us. The former by
 150      * not finding all bytes, the latter by overwriting the supposedly invalid
 151      * UTF-8 with NULs.
 152      *
 153      * To do this, we specify "ASCII" as the charset.
 154      *
 155      * In fact, we can specify any charset other than "UTF-8": any such charset
 156      * will trigger G_REGEX_RAW (see [1]). The output of [2] will be the same
 157      * for all charsets because it skips the \xXX symbols
 158      * mc_search__hex_translate_to_regex() outputs.
 159      *
 160      * But "ASCII" is the best choice because a hex pattern may contain a
 161      * quoted string: this way we know [2] will ignore any characters outside
 162      * ASCII letters range (these ignored chars will be copied verbatim to the
 163      * output and will match as-is; in other words, in a case-sensitive manner;
 164      * If the user is interested in case-insensitive searches of international
 165      * text, he shouldn't be using hex search in the first place.)
 166      *
 167      * Switching out of UTF-8 has another advantage:
 168      *
 169      * When doing case-insensitive searches, GLib treats \xXX symbols as normal
 170      * letters and therefore matches both "a" and "A" for the hex pattern
 171      * "0x61". When we switch out of UTF-8, we're switching to using [2], which
 172      * doesn't have this issue.
 173      *
 174      * [1] mc_search__cond_struct_new_init_regex
 175      * [2] mc_search__cond_struct_new_regex_ci_str
 176      */
 177     if (str_isutf8 (charset))
 178         charset = "ASCII";
 179 
 180     tmp = mc_search__hex_translate_to_regex (mc_search_cond->str, &error, &error_pos);
 181     if (tmp != NULL)
 182     {
 183         g_string_free (mc_search_cond->str, TRUE);
 184         mc_search_cond->str = tmp;
 185         mc_search__cond_struct_new_init_regex (charset, lc_mc_search, mc_search_cond);
 186     }
 187     else
 188     {
 189         const char *desc;
 190 
 191         switch (error)
 192         {
 193         case MC_SEARCH_HEX_E_NUM_OUT_OF_RANGE:
 194             desc = _ (
 195                 "Number out of range (should be in byte range, 0 <= n <= 0xFF, expressed in hex)");
 196             break;
 197         case MC_SEARCH_HEX_E_INVALID_CHARACTER:
 198             desc = _ ("Invalid character");
 199             break;
 200         case MC_SEARCH_HEX_E_UNMATCHED_QUOTES:
 201             desc = _ ("Unmatched quotes character");
 202             break;
 203         default:
 204             desc = "";
 205         }
 206 
 207         lc_mc_search->error = MC_SEARCH_E_INPUT;
 208         lc_mc_search->error_str =
 209             g_strdup_printf (_ ("Hex pattern error at position %d:\n%s."), error_pos + 1, desc);
 210     }
 211 }
 212 
 213 /* --------------------------------------------------------------------------------------------- */
 214 
 215 gboolean
 216 mc_search__run_hex (mc_search_t *lc_mc_search, const void *user_data, off_t start_search,
     /*   */
 217                     off_t end_search, gsize *found_len)
 218 {
 219     return mc_search__run_regex (lc_mc_search, user_data, start_search, end_search, found_len);
 220 }
 221 
 222 /* --------------------------------------------------------------------------------------------- */
 223 
 224 GString *
 225 mc_search_hex_prepare_replace_str (mc_search_t *lc_mc_search, GString *replace_str)
     /*   */
 226 {
 227     (void) lc_mc_search;
 228 
 229     return mc_g_string_dup (replace_str);
 230 }
 231 
 232 /* --------------------------------------------------------------------------------------------- */
/* */
root/lib/search/hex.c

DEFINITIONS
