5 Copyright (C) 2009-2019
6 Free Software Foundation, Inc.
9 Slava Zanko <slavazanko@gmail.com>, 2009, 2010, 2011, 2013
10 Vitaliy Filippov <vitalif@yourcmc.ru>, 2011
11 Andrew Borodin <aborodin@vmail.ru>, 2013-2015
13 This file is part of the Midnight Commander.
15 The Midnight Commander is free software: you can redistribute it
16 and/or modify it under the terms of the GNU General Public License as
17 published by the Free Software Foundation, either version 3 of the License,
18 or (at your option) any later version.
20 The Midnight Commander is distributed in the hope that it will be useful,
21 but WITHOUT ANY WARRANTY; without even the implied warranty of
22 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
23 GNU General Public License for more details.
25 You should have received a copy of the GNU General Public License
26 along with this program. If not, see <http://www.gnu.org/licenses/>.
33 #include "lib/global.h"
34 #include "lib/strutil.h"
35 #include "lib/search.h"
36 #include "lib/strescape.h"
37 #include "lib/util.h" /* MC_PTR_FREE */
41 /*** global variables ****************************************************************************/
43 /*** file scope macro definitions ****************************************************************/
45 #define REPLACE_PREPARE_T_NOTHING_SPECIAL -1
46 #define REPLACE_PREPARE_T_REPLACE_FLAG -2
47 #define REPLACE_PREPARE_T_ESCAPE_SEQ -3
49 /*** file scope type declarations ****************************************************************/
53 REPLACE_T_NO_TRANSFORM
= 0,
54 REPLACE_T_UPP_TRANSFORM_CHAR
= 1,
55 REPLACE_T_LOW_TRANSFORM_CHAR
= 2,
56 REPLACE_T_UPP_TRANSFORM
= 4,
57 REPLACE_T_LOW_TRANSFORM
= 8
58 } replace_transform_type_t
;
61 /*** file scope variables ************************************************************************/
63 /*** file scope functions ************************************************************************/
66 mc_search__regex_str_append_if_special (GString
* copy_to
, const GString
* regex_str
,
69 const char *special_chars
[] = {
86 const char **spec_chr
;
88 tmp_regex_str
= &(regex_str
->str
[*offset
]);
90 for (spec_chr
= special_chars
; *spec_chr
!= NULL
; spec_chr
++)
94 spec_chr_len
= strlen (*spec_chr
);
96 if (strncmp (tmp_regex_str
, *spec_chr
, spec_chr_len
) == 0
97 && !strutils_is_char_escaped (regex_str
->str
, tmp_regex_str
))
99 if (strncmp ("\\x", *spec_chr
, spec_chr_len
) == 0)
101 if (tmp_regex_str
[spec_chr_len
] != '{')
105 while ((spec_chr_len
< regex_str
->len
- *offset
)
106 && tmp_regex_str
[spec_chr_len
] != '}')
108 if (tmp_regex_str
[spec_chr_len
] == '}')
112 g_string_append_len (copy_to
, tmp_regex_str
, spec_chr_len
);
113 *offset
+= spec_chr_len
;
121 /* --------------------------------------------------------------------------------------------- */
124 mc_search__cond_struct_new_regex_hex_add (const char *charset
, GString
* str_to
,
125 const char *one_char
, gsize str_len
)
130 upp
= mc_search__toupper_case_str (charset
, one_char
, str_len
);
131 low
= mc_search__tolower_case_str (charset
, one_char
, str_len
);
133 for (loop
= 0; loop
< upp
->len
; loop
++)
135 gchar tmp_str
[10 + 1]; /* longest content is "[\\x%02X\\x%02X]" */
138 if (loop
>= low
->len
|| upp
->str
[loop
] == low
->str
[loop
])
140 g_snprintf (tmp_str
, sizeof (tmp_str
), "\\x%02X", (unsigned char) upp
->str
[loop
]);
143 g_snprintf (tmp_str
, sizeof (tmp_str
), "[\\x%02X\\x%02X]",
144 (unsigned char) upp
->str
[loop
], (unsigned char) low
->str
[loop
]);
146 g_string_append_len (str_to
, tmp_str
, tmp_len
);
149 g_string_free (upp
, TRUE
);
150 g_string_free (low
, TRUE
);
153 /* --------------------------------------------------------------------------------------------- */
156 mc_search__cond_struct_new_regex_accum_append (const char *charset
, GString
* str_to
,
159 GString
*recoded_part
;
162 recoded_part
= g_string_sized_new (32);
164 while (loop
< str_from
->len
)
168 gboolean just_letters
;
171 mc_search__get_one_symbol (charset
, &(str_from
->str
[loop
]),
172 MIN (str_from
->len
- loop
, 6), &just_letters
);
173 one_char_len
= strlen (one_char
);
175 if (one_char_len
== 0)
179 loop
+= one_char_len
;
182 mc_search__cond_struct_new_regex_hex_add (charset
, recoded_part
, one_char
,
185 g_string_append_len (recoded_part
, one_char
, one_char_len
);
191 g_string_append_len (str_to
, recoded_part
->str
, recoded_part
->len
);
192 g_string_free (recoded_part
, TRUE
);
193 g_string_set_size (str_from
, 0);
196 /* --------------------------------------------------------------------------------------------- */
199 * Creates a case-insensitive version of a regex pattern.
201 * For example (assuming ASCII charset): given "\\bHello!\\xAB", returns
202 * "\\b[Hh][Ee][Ll][Ll][Oo]!\\xAB" (this example is for easier reading; in
203 * reality hex codes are used instead of letters).
205 * This function knows not to ruin special regex symbols.
207 * This function is used when working with non-UTF-8 charsets: GLib's
208 * regex engine doesn't understand such charsets and therefore can't do
212 mc_search__cond_struct_new_regex_ci_str (const char *charset
, const GString
* astr
)
214 GString
*accumulator
, *spec_char
, *ret_str
;
217 ret_str
= g_string_sized_new (64);
218 accumulator
= g_string_sized_new (64);
219 spec_char
= g_string_sized_new (64);
222 while (loop
<= astr
->len
)
224 if (mc_search__regex_str_append_if_special (spec_char
, astr
, &loop
))
226 mc_search__cond_struct_new_regex_accum_append (charset
, ret_str
, accumulator
);
227 g_string_append_len (ret_str
, spec_char
->str
, spec_char
->len
);
228 g_string_set_size (spec_char
, 0);
232 if (astr
->str
[loop
] == '[' && !strutils_is_char_escaped (astr
->str
, &(astr
->str
[loop
])))
234 mc_search__cond_struct_new_regex_accum_append (charset
, ret_str
, accumulator
);
236 while (loop
< astr
->len
&& !(astr
->str
[loop
] == ']'
237 && !strutils_is_char_escaped (astr
->str
,
238 &(astr
->str
[loop
]))))
240 g_string_append_c (ret_str
, astr
->str
[loop
]);
244 g_string_append_c (ret_str
, astr
->str
[loop
]);
251 g_string_append_c (accumulator
, astr
->str
[loop
]);
254 mc_search__cond_struct_new_regex_accum_append (charset
, ret_str
, accumulator
);
256 g_string_free (accumulator
, TRUE
);
257 g_string_free (spec_char
, TRUE
);
262 /* --------------------------------------------------------------------------------------------- */
264 #ifdef SEARCH_TYPE_GLIB
265 /* A thin wrapper above g_regex_match_full that makes sure the string passed
266 * to it is valid UTF-8 (unless G_REGEX_RAW compile flag was set), as it is a
267 * requirement by glib and it might crash otherwise. See: mc ticket 3449.
268 * Be careful: there might be embedded NULs in the strings. */
270 mc_search__g_regex_match_full_safe (const GRegex
* regex
,
271 const gchar
* string
,
274 GRegexMatchFlags match_options
,
275 GMatchInfo
** match_info
, GError
** error
)
277 char *string_safe
, *p
, *end
;
281 string_len
= strlen (string
);
283 if ((g_regex_get_compile_flags (regex
) & G_REGEX_RAW
)
284 || g_utf8_validate (string
, string_len
, NULL
))
286 return g_regex_match_full (regex
, string
, string_len
, start_position
, match_options
,
290 /* Correctly handle embedded NULs while copying */
291 p
= string_safe
= g_malloc (string_len
);
292 memcpy (string_safe
, string
, string_len
);
293 end
= p
+ string_len
;
297 gunichar c
= g_utf8_get_char_validated (p
, -1);
298 if (c
!= (gunichar
) (-1) && c
!= (gunichar
) (-2))
300 p
= g_utf8_next_char (p
);
304 /* U+FFFD would be the proper choice, but then we'd have to
305 maintain mapping between old and new offsets.
306 So rather do a byte by byte replacement. */
312 g_regex_match_full (regex
, string_safe
, string_len
, start_position
, match_options
,
314 g_free (string_safe
);
317 #endif /* SEARCH_TYPE_GLIB */
319 /* --------------------------------------------------------------------------------------------- */
321 static mc_search__found_cond_t
322 mc_search__regex_found_cond_one (mc_search_t
* lc_mc_search
, mc_search_regex_t
* regex
,
323 GString
* search_str
)
325 #ifdef SEARCH_TYPE_GLIB
326 GError
*mcerror
= NULL
;
328 if (!mc_search__g_regex_match_full_safe
329 (regex
, search_str
->str
, search_str
->len
, 0, G_REGEX_MATCH_NEWLINE_ANY
,
330 &lc_mc_search
->regex_match_info
, &mcerror
))
332 g_match_info_free (lc_mc_search
->regex_match_info
);
333 lc_mc_search
->regex_match_info
= NULL
;
336 lc_mc_search
->error
= MC_SEARCH_E_REGEX
;
337 g_free (lc_mc_search
->error_str
);
338 lc_mc_search
->error_str
=
339 str_conv_gerror_message (mcerror
, _("Regular expression error"));
340 g_error_free (mcerror
);
341 return COND__FOUND_ERROR
;
343 return COND__NOT_FOUND
;
345 lc_mc_search
->num_results
= g_match_info_get_match_count (lc_mc_search
->regex_match_info
);
346 #else /* SEARCH_TYPE_GLIB */
347 lc_mc_search
->num_results
= pcre_exec (regex
, lc_mc_search
->regex_match_info
,
348 search_str
->str
, search_str
->len
, 0, 0,
349 lc_mc_search
->iovector
, MC_SEARCH__NUM_REPLACE_ARGS
);
350 if (lc_mc_search
->num_results
< 0)
352 return COND__NOT_FOUND
;
354 #endif /* SEARCH_TYPE_GLIB */
355 return COND__FOUND_OK
;
359 /* --------------------------------------------------------------------------------------------- */
361 static mc_search__found_cond_t
362 mc_search__regex_found_cond (mc_search_t
* lc_mc_search
, GString
* search_str
)
366 for (loop1
= 0; loop1
< lc_mc_search
->conditions
->len
; loop1
++)
368 mc_search_cond_t
*mc_search_cond
;
369 mc_search__found_cond_t ret
;
371 mc_search_cond
= (mc_search_cond_t
*) g_ptr_array_index (lc_mc_search
->conditions
, loop1
);
373 if (!mc_search_cond
->regex_handle
)
377 mc_search__regex_found_cond_one (lc_mc_search
, mc_search_cond
->regex_handle
,
379 if (ret
!= COND__NOT_FOUND
)
382 return COND__NOT_ALL_FOUND
;
385 /* --------------------------------------------------------------------------------------------- */
388 mc_search_regex__get_max_num_of_replace_tokens (const gchar
* str
, gsize len
)
392 for (loop
= 0; loop
< len
- 1; loop
++)
394 if (str
[loop
] == '\\' && g_ascii_isdigit (str
[loop
+ 1]))
396 if (strutils_is_char_escaped (str
, &str
[loop
]))
398 if (max_token
< str
[loop
+ 1] - '0')
399 max_token
= str
[loop
+ 1] - '0';
402 if (str
[loop
] == '$' && str
[loop
+ 1] == '{')
406 if (strutils_is_char_escaped (str
, &str
[loop
]))
410 loop
+ tmp_len
+ 2 < len
&& (str
[loop
+ 2 + tmp_len
] & (char) 0xf0) == 0x30;
413 if (str
[loop
+ 2 + tmp_len
] == '}')
418 tmp_str
= g_strndup (&str
[loop
+ 2], tmp_len
);
419 tmp_token
= atoi (tmp_str
);
420 if (max_token
< tmp_token
)
421 max_token
= tmp_token
;
429 /* --------------------------------------------------------------------------------------------- */
432 mc_search_regex__get_token_by_num (const mc_search_t
* lc_mc_search
, gsize lc_index
)
434 int fnd_start
= 0, fnd_end
= 0;
436 #ifdef SEARCH_TYPE_GLIB
437 g_match_info_fetch_pos (lc_mc_search
->regex_match_info
, lc_index
, &fnd_start
, &fnd_end
);
438 #else /* SEARCH_TYPE_GLIB */
439 fnd_start
= lc_mc_search
->iovector
[lc_index
* 2 + 0];
440 fnd_end
= lc_mc_search
->iovector
[lc_index
* 2 + 1];
441 #endif /* SEARCH_TYPE_GLIB */
443 if (fnd_end
== fnd_start
)
444 return g_strdup ("");
446 return g_strndup (lc_mc_search
->regex_buffer
->str
+ fnd_start
, fnd_end
- fnd_start
);
450 /* --------------------------------------------------------------------------------------------- */
453 mc_search_regex__replace_handle_esc_seq (const GString
* replace_str
, const gsize current_pos
,
454 gsize
* skip_len
, int *ret
)
456 char *curr_str
= &(replace_str
->str
[current_pos
]);
457 char c
= curr_str
[1];
459 if (replace_str
->len
> current_pos
+ 2)
463 for (*skip_len
= 2; /* \{ */
464 current_pos
+ *skip_len
< replace_str
->len
&& curr_str
[*skip_len
] >= '0'
465 && curr_str
[*skip_len
] <= '7'; (*skip_len
)++)
468 if (current_pos
+ *skip_len
< replace_str
->len
&& curr_str
[*skip_len
] == '}')
471 *ret
= REPLACE_PREPARE_T_ESCAPE_SEQ
;
476 *ret
= REPLACE_PREPARE_T_NOTHING_SPECIAL
;
483 *skip_len
= 2; /* \x */
487 for (*skip_len
= 3; /* \x{ */
488 current_pos
+ *skip_len
< replace_str
->len
489 && g_ascii_isxdigit ((guchar
) curr_str
[*skip_len
]); (*skip_len
)++)
492 if (current_pos
+ *skip_len
< replace_str
->len
&& curr_str
[*skip_len
] == '}')
495 *ret
= REPLACE_PREPARE_T_ESCAPE_SEQ
;
500 *ret
= REPLACE_PREPARE_T_NOTHING_SPECIAL
;
504 else if (!g_ascii_isxdigit ((guchar
) c
))
506 *skip_len
= 2; /* \x without number behind */
507 *ret
= REPLACE_PREPARE_T_NOTHING_SPECIAL
;
513 if (!g_ascii_isxdigit ((guchar
) c
))
514 *skip_len
= 3; /* \xH */
516 *skip_len
= 4; /* \xHH */
517 *ret
= REPLACE_PREPARE_T_ESCAPE_SEQ
;
523 if (strchr ("ntvbrfa", c
) != NULL
)
526 *ret
= REPLACE_PREPARE_T_ESCAPE_SEQ
;
532 /* --------------------------------------------------------------------------------------------- */
535 mc_search_regex__process_replace_str (const GString
* replace_str
, const gsize current_pos
,
536 gsize
* skip_len
, replace_transform_type_t
* replace_flags
)
539 const char *curr_str
= &(replace_str
->str
[current_pos
]);
541 if (current_pos
> replace_str
->len
)
542 return REPLACE_PREPARE_T_NOTHING_SPECIAL
;
546 if (replace_str
->len
> current_pos
+ 2 && curr_str
[0] == '$' && curr_str
[1] == '{'
547 && (curr_str
[2] & (char) 0xf0) == 0x30)
551 if (strutils_is_char_escaped (replace_str
->str
, curr_str
))
554 return REPLACE_PREPARE_T_NOTHING_SPECIAL
;
558 current_pos
+ *skip_len
+ 2 < replace_str
->len
559 && (curr_str
[2 + *skip_len
] & (char) 0xf0) == 0x30; (*skip_len
)++)
562 if (curr_str
[2 + *skip_len
] != '}')
563 return REPLACE_PREPARE_T_NOTHING_SPECIAL
;
565 tmp_str
= g_strndup (curr_str
+ 2, *skip_len
);
567 return REPLACE_PREPARE_T_NOTHING_SPECIAL
;
569 ret
= atoi (tmp_str
);
572 *skip_len
+= 3; /* ${} */
573 return ret
; /* capture buffer index >= 0 */
576 if (curr_str
[0] == '\\' && replace_str
->len
> current_pos
+ 1)
578 if (strutils_is_char_escaped (replace_str
->str
, curr_str
))
581 return REPLACE_PREPARE_T_NOTHING_SPECIAL
;
584 if (g_ascii_isdigit (curr_str
[1]))
586 ret
= g_ascii_digit_value (curr_str
[1]); /* capture buffer index >= 0 */
587 *skip_len
= 2; /* \\ and one digit */
591 if (!mc_search_regex__replace_handle_esc_seq (replace_str
, current_pos
, skip_len
, &ret
))
594 ret
= REPLACE_PREPARE_T_REPLACE_FLAG
;
600 *replace_flags
|= REPLACE_T_UPP_TRANSFORM
;
601 *replace_flags
&= ~REPLACE_T_LOW_TRANSFORM
;
604 *replace_flags
|= REPLACE_T_UPP_TRANSFORM_CHAR
;
607 *replace_flags
|= REPLACE_T_LOW_TRANSFORM
;
608 *replace_flags
&= ~REPLACE_T_UPP_TRANSFORM
;
611 *replace_flags
|= REPLACE_T_LOW_TRANSFORM_CHAR
;
614 *replace_flags
= REPLACE_T_NO_TRANSFORM
;
617 ret
= REPLACE_PREPARE_T_NOTHING_SPECIAL
;
624 /* --------------------------------------------------------------------------------------------- */
627 mc_search_regex__process_append_str (GString
* dest_str
, const char *from
, gsize len
,
628 replace_transform_type_t
* replace_flags
)
633 if (len
== (gsize
) (-1))
636 if (*replace_flags
== REPLACE_T_NO_TRANSFORM
)
638 g_string_append_len (dest_str
, from
, len
);
642 for (loop
= 0; loop
< len
; loop
+= char_len
)
644 GString
*tmp_string
= NULL
;
647 tmp_str
= mc_search__get_one_symbol (NULL
, from
+ loop
, len
- loop
, NULL
);
648 char_len
= strlen (tmp_str
);
650 if ((*replace_flags
& REPLACE_T_UPP_TRANSFORM_CHAR
) != 0)
652 *replace_flags
&= ~REPLACE_T_UPP_TRANSFORM_CHAR
;
653 tmp_string
= mc_search__toupper_case_str (NULL
, tmp_str
, char_len
);
654 g_string_append_len (dest_str
, tmp_string
->str
, tmp_string
->len
);
655 g_string_free (tmp_string
, TRUE
);
657 else if ((*replace_flags
& REPLACE_T_LOW_TRANSFORM_CHAR
) != 0)
659 *replace_flags
&= ~REPLACE_T_LOW_TRANSFORM_CHAR
;
660 tmp_string
= mc_search__tolower_case_str (NULL
, tmp_str
, char_len
);
661 g_string_append_len (dest_str
, tmp_string
->str
, tmp_string
->len
);
662 g_string_free (tmp_string
, TRUE
);
664 else if ((*replace_flags
& REPLACE_T_UPP_TRANSFORM
) != 0)
666 tmp_string
= mc_search__toupper_case_str (NULL
, tmp_str
, char_len
);
667 g_string_append_len (dest_str
, tmp_string
->str
, tmp_string
->len
);
668 g_string_free (tmp_string
, TRUE
);
670 else if ((*replace_flags
& REPLACE_T_LOW_TRANSFORM
) != 0)
672 tmp_string
= mc_search__tolower_case_str (NULL
, tmp_str
, char_len
);
673 g_string_append_len (dest_str
, tmp_string
->str
, tmp_string
->len
);
674 g_string_free (tmp_string
, TRUE
);
681 /* --------------------------------------------------------------------------------------------- */
684 mc_search_regex__process_escape_sequence (GString
* dest_str
, const char *from
, gsize len
,
685 replace_transform_type_t
* replace_flags
,
692 if (len
== (gsize
) (-1))
705 if (i
< len
&& from
[i
] == '{')
709 if (from
[i
] >= '0' && from
[i
] <= '9')
710 c
= c
* 16 + from
[i
] - '0';
711 else if (from
[i
] >= 'a' && from
[i
] <= 'f')
712 c
= c
* 16 + 10 + from
[i
] - 'a';
713 else if (from
[i
] >= 'A' && from
[i
] <= 'F')
714 c
= c
* 16 + 10 + from
[i
] - 'A';
719 else if (from
[i
] >= '0' && from
[i
] <= '7')
720 for (; i
< len
&& from
[i
] >= '0' && from
[i
] <= '7'; i
++)
721 c
= c
* 8 + from
[i
] - '0';
748 mc_search_regex__process_append_str (dest_str
, from
, len
, replace_flags
);
753 if (c
< 0x80 || !is_utf8
)
754 g_string_append_c (dest_str
, (char) c
);
758 g_string_append_c (dest_str
, b
);
759 b
= 0x80 | (c
& 0x3F);
760 g_string_append_c (dest_str
, b
);
762 else if (c
< 0x10000)
764 b
= 0xE0 | (c
>> 12);
765 g_string_append_c (dest_str
, b
);
766 b
= 0x80 | ((c
>> 6) & 0x3F);
767 g_string_append_c (dest_str
, b
);
768 b
= 0x80 | (c
& 0x3F);
769 g_string_append_c (dest_str
, b
);
771 else if (c
< 0x10FFFF)
773 b
= 0xF0 | (c
>> 16);
774 g_string_append_c (dest_str
, b
);
775 b
= 0x80 | ((c
>> 12) & 0x3F);
776 g_string_append_c (dest_str
, b
);
777 b
= 0x80 | ((c
>> 6) & 0x3F);
778 g_string_append_c (dest_str
, b
);
779 b
= 0x80 | (c
& 0x3F);
780 g_string_append_c (dest_str
, b
);
784 /* --------------------------------------------------------------------------------------------- */
785 /*** public functions ****************************************************************************/
786 /* --------------------------------------------------------------------------------------------- */
789 mc_search__cond_struct_new_init_regex (const char *charset
, mc_search_t
* lc_mc_search
,
790 mc_search_cond_t
* mc_search_cond
)
792 if (lc_mc_search
->whole_words
&& !lc_mc_search
->is_entire_line
)
794 /* NOTE: \b as word boundary doesn't allow search
795 * whole words with non-ASCII symbols.
796 * Update: Is it still true nowadays? Probably not. #2396, #3524 */
797 g_string_prepend (mc_search_cond
->str
, "(?<![\\p{L}\\p{N}_])");
798 g_string_append (mc_search_cond
->str
, "(?![\\p{L}\\p{N}_])");
802 #ifdef SEARCH_TYPE_GLIB
803 GError
*mcerror
= NULL
;
804 GRegexCompileFlags g_regex_options
= G_REGEX_OPTIMIZE
| G_REGEX_DOTALL
;
806 if (str_isutf8 (charset
) && mc_global
.utf8_display
)
808 if (!lc_mc_search
->is_case_sensitive
)
809 g_regex_options
|= G_REGEX_CASELESS
;
813 g_regex_options
|= G_REGEX_RAW
;
815 if (!lc_mc_search
->is_case_sensitive
)
819 tmp
= mc_search_cond
->str
;
820 mc_search_cond
->str
= mc_search__cond_struct_new_regex_ci_str (charset
, tmp
);
821 g_string_free (tmp
, TRUE
);
826 mc_search_cond
->regex_handle
=
827 g_regex_new (mc_search_cond
->str
->str
, g_regex_options
, 0, &mcerror
);
831 lc_mc_search
->error
= MC_SEARCH_E_REGEX_COMPILE
;
832 g_free (lc_mc_search
->error_str
);
833 lc_mc_search
->error_str
=
834 str_conv_gerror_message (mcerror
, _("Regular expression error"));
835 g_error_free (mcerror
);
838 #else /* SEARCH_TYPE_GLIB */
841 int pcre_options
= PCRE_EXTRA
| PCRE_MULTILINE
;
843 if (str_isutf8 (charset
) && mc_global
.utf8_display
)
845 pcre_options
|= PCRE_UTF8
;
846 if (!lc_mc_search
->is_case_sensitive
)
847 pcre_options
|= PCRE_CASELESS
;
851 if (!lc_mc_search
->is_case_sensitive
)
855 tmp
= mc_search_cond
->str
;
856 mc_search_cond
->str
= mc_search__cond_struct_new_regex_ci_str (charset
, tmp
);
857 g_string_free (tmp
, TRUE
);
861 mc_search_cond
->regex_handle
=
862 pcre_compile (mc_search_cond
->str
->str
, pcre_options
, &error
, &erroffset
, NULL
);
863 if (mc_search_cond
->regex_handle
== NULL
)
865 mc_search_set_error (lc_mc_search
, MC_SEARCH_E_REGEX_COMPILE
, "%s", error
);
868 lc_mc_search
->regex_match_info
= pcre_study (mc_search_cond
->regex_handle
, 0, &error
);
869 if (lc_mc_search
->regex_match_info
== NULL
&& error
!= NULL
)
871 mc_search_set_error (lc_mc_search
, MC_SEARCH_E_REGEX_COMPILE
, "%s", error
);
872 MC_PTR_FREE (mc_search_cond
->regex_handle
);
875 #endif /* SEARCH_TYPE_GLIB */
878 lc_mc_search
->is_utf8
= str_isutf8 (charset
);
881 /* --------------------------------------------------------------------------------------------- */
884 mc_search__run_regex (mc_search_t
* lc_mc_search
, const void *user_data
,
885 gsize start_search
, gsize end_search
, gsize
* found_len
)
887 mc_search_cbret_t ret
= MC_SEARCH_CB_NOTFOUND
;
888 gsize current_pos
, virtual_pos
;
892 if (lc_mc_search
->regex_buffer
!= NULL
)
893 g_string_set_size (lc_mc_search
->regex_buffer
, 0);
895 lc_mc_search
->regex_buffer
= g_string_sized_new (64);
897 virtual_pos
= current_pos
= start_search
;
898 while (virtual_pos
<= end_search
)
900 g_string_set_size (lc_mc_search
->regex_buffer
, 0);
901 lc_mc_search
->start_buffer
= current_pos
;
903 if (lc_mc_search
->search_fn
!= NULL
)
907 int current_chr
= '\n'; /* stop search symbol */
909 ret
= lc_mc_search
->search_fn (user_data
, current_pos
, ¤t_chr
);
911 if (ret
== MC_SEARCH_CB_ABORT
)
914 if (ret
== MC_SEARCH_CB_INVALID
)
919 if (ret
== MC_SEARCH_CB_SKIP
)
924 g_string_append_c (lc_mc_search
->regex_buffer
, (char) current_chr
);
926 if ((char) current_chr
== '\n' || virtual_pos
> end_search
)
932 /* optimization for standard case (for search from file manager)
933 * where there is no MC_SEARCH_CB_INVALID or MC_SEARCH_CB_SKIP
934 * return codes, so we can copy line at regex buffer all at once
938 const char current_chr
= ((const char *) user_data
)[current_pos
];
940 if (current_chr
== '\0')
945 if (current_chr
== '\n' || current_pos
> end_search
)
949 /* use virtual_pos as index of start of current chunk */
950 g_string_append_len (lc_mc_search
->regex_buffer
, (const char *) user_data
+ virtual_pos
,
951 current_pos
- virtual_pos
);
952 virtual_pos
= current_pos
;
955 switch (mc_search__regex_found_cond (lc_mc_search
, lc_mc_search
->regex_buffer
))
958 #ifdef SEARCH_TYPE_GLIB
959 g_match_info_fetch_pos (lc_mc_search
->regex_match_info
, 0, &start_pos
, &end_pos
);
960 #else /* SEARCH_TYPE_GLIB */
961 start_pos
= lc_mc_search
->iovector
[0];
962 end_pos
= lc_mc_search
->iovector
[1];
963 #endif /* SEARCH_TYPE_GLIB */
964 if (found_len
!= NULL
)
965 *found_len
= end_pos
- start_pos
;
966 lc_mc_search
->normal_offset
= lc_mc_search
->start_buffer
+ start_pos
;
968 case COND__NOT_ALL_FOUND
:
971 g_string_free (lc_mc_search
->regex_buffer
, TRUE
);
972 lc_mc_search
->regex_buffer
= NULL
;
976 if ((lc_mc_search
->update_fn
!= NULL
) &&
977 ((lc_mc_search
->update_fn
) (user_data
, current_pos
) == MC_SEARCH_CB_ABORT
))
978 ret
= MC_SEARCH_CB_ABORT
;
980 if (ret
== MC_SEARCH_CB_ABORT
|| ret
== MC_SEARCH_CB_NOTFOUND
)
984 g_string_free (lc_mc_search
->regex_buffer
, TRUE
);
985 lc_mc_search
->regex_buffer
= NULL
;
987 MC_PTR_FREE (lc_mc_search
->error_str
);
988 lc_mc_search
->error
= ret
== MC_SEARCH_CB_ABORT
? MC_SEARCH_E_ABORT
: MC_SEARCH_E_NOTFOUND
;
993 /* --------------------------------------------------------------------------------------------- */
996 mc_search_regex_prepare_replace_str (mc_search_t
* lc_mc_search
, GString
* replace_str
)
1000 int num_replace_tokens
;
1003 replace_transform_type_t replace_flags
= REPLACE_T_NO_TRANSFORM
;
1005 num_replace_tokens
=
1006 mc_search_regex__get_max_num_of_replace_tokens (replace_str
->str
, replace_str
->len
);
1008 if (lc_mc_search
->num_results
< 0)
1009 return g_string_new_len (replace_str
->str
, replace_str
->len
);
1011 if (num_replace_tokens
> lc_mc_search
->num_results
- 1
1012 || num_replace_tokens
> MC_SEARCH__NUM_REPLACE_ARGS
)
1014 mc_search_set_error (lc_mc_search
, MC_SEARCH_E_REGEX_REPLACE
, "%s",
1015 _(STR_E_RPL_NOT_EQ_TO_FOUND
));
1019 ret
= g_string_sized_new (64);
1021 for (loop
= 0; loop
< replace_str
->len
- 1; loop
++)
1027 lc_index
= mc_search_regex__process_replace_str (replace_str
, loop
, &len
, &replace_flags
);
1029 if (lc_index
== REPLACE_PREPARE_T_NOTHING_SPECIAL
)
1033 mc_search_regex__process_append_str (ret
, replace_str
->str
+ prev
, loop
- prev
,
1035 mc_search_regex__process_append_str (ret
, replace_str
->str
+ loop
+ 1, len
- 1,
1038 loop
= prev
- 1; /* prepare to loop++ */
1044 if (lc_index
== REPLACE_PREPARE_T_REPLACE_FLAG
)
1047 mc_search_regex__process_append_str (ret
, replace_str
->str
+ prev
, loop
- prev
,
1050 loop
= prev
- 1; /* prepare to loop++ */
1054 /* escape sequence */
1055 if (lc_index
== REPLACE_PREPARE_T_ESCAPE_SEQ
)
1057 mc_search_regex__process_append_str (ret
, replace_str
->str
+ prev
, loop
- prev
,
1059 /* call process_escape_sequence without starting '\\' */
1060 mc_search_regex__process_escape_sequence (ret
, replace_str
->str
+ loop
+ 1, len
- 1,
1061 &replace_flags
, lc_mc_search
->is_utf8
);
1063 loop
= prev
- 1; /* prepare to loop++ */
1067 /* invalid capture buffer number */
1068 if (lc_index
> lc_mc_search
->num_results
)
1070 g_string_free (ret
, TRUE
);
1071 mc_search_set_error (lc_mc_search
, MC_SEARCH_E_REGEX_REPLACE
,
1072 _(STR_E_RPL_INVALID_TOKEN
), lc_index
);
1076 tmp_str
= mc_search_regex__get_token_by_num (lc_mc_search
, lc_index
);
1079 mc_search_regex__process_append_str (ret
, replace_str
->str
+ prev
, loop
- prev
,
1082 mc_search_regex__process_append_str (ret
, tmp_str
, -1, &replace_flags
);
1086 loop
= prev
- 1; /* prepare to loop++ */
1089 mc_search_regex__process_append_str (ret
, replace_str
->str
+ prev
, replace_str
->len
- prev
,