5 Copyright (C) 2009-2016
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"
40 /*** global variables ****************************************************************************/
42 /*** file scope macro definitions ****************************************************************/
44 #define REPLACE_PREPARE_T_NOTHING_SPECIAL -1
45 #define REPLACE_PREPARE_T_REPLACE_FLAG -2
46 #define REPLACE_PREPARE_T_ESCAPE_SEQ -3
48 /*** file scope type declarations ****************************************************************/
52 REPLACE_T_NO_TRANSFORM
= 0,
53 REPLACE_T_UPP_TRANSFORM_CHAR
= 1,
54 REPLACE_T_LOW_TRANSFORM_CHAR
= 2,
55 REPLACE_T_UPP_TRANSFORM
= 4,
56 REPLACE_T_LOW_TRANSFORM
= 8
57 } replace_transform_type_t
;
60 /*** file scope variables ************************************************************************/
62 /*** file scope functions ************************************************************************/
65 mc_search__regex_str_append_if_special (GString
* copy_to
, const GString
* regex_str
,
68 const char *special_chars
[] = {
85 const char **spec_chr
;
87 tmp_regex_str
= &(regex_str
->str
[*offset
]);
89 for (spec_chr
= special_chars
; *spec_chr
!= NULL
; spec_chr
++)
93 spec_chr_len
= strlen (*spec_chr
);
95 if (strncmp (tmp_regex_str
, *spec_chr
, spec_chr_len
) == 0
96 && !strutils_is_char_escaped (regex_str
->str
, tmp_regex_str
))
98 if (strncmp ("\\x", *spec_chr
, spec_chr_len
) == 0)
100 if (tmp_regex_str
[spec_chr_len
] != '{')
104 while ((spec_chr_len
< regex_str
->len
- *offset
)
105 && tmp_regex_str
[spec_chr_len
] != '}')
107 if (tmp_regex_str
[spec_chr_len
] == '}')
111 g_string_append_len (copy_to
, tmp_regex_str
, spec_chr_len
);
112 *offset
+= spec_chr_len
;
120 /* --------------------------------------------------------------------------------------------- */
123 mc_search__cond_struct_new_regex_hex_add (const char *charset
, GString
* str_to
,
124 const char *one_char
, gsize str_len
)
129 upp
= mc_search__toupper_case_str (charset
, one_char
, str_len
);
130 low
= mc_search__tolower_case_str (charset
, one_char
, str_len
);
132 for (loop
= 0; loop
< upp
->len
; loop
++)
134 gchar tmp_str
[10 + 1]; /* longest content is "[\\x%02X\\x%02X]" */
137 if (loop
>= low
->len
|| upp
->str
[loop
] == low
->str
[loop
])
139 g_snprintf (tmp_str
, sizeof (tmp_str
), "\\x%02X", (unsigned char) upp
->str
[loop
]);
142 g_snprintf (tmp_str
, sizeof (tmp_str
), "[\\x%02X\\x%02X]",
143 (unsigned char) upp
->str
[loop
], (unsigned char) low
->str
[loop
]);
145 g_string_append_len (str_to
, tmp_str
, tmp_len
);
148 g_string_free (upp
, TRUE
);
149 g_string_free (low
, TRUE
);
152 /* --------------------------------------------------------------------------------------------- */
155 mc_search__cond_struct_new_regex_accum_append (const char *charset
, GString
* str_to
,
158 GString
*recoded_part
;
161 recoded_part
= g_string_sized_new (32);
163 while (loop
< str_from
->len
)
167 gboolean just_letters
;
170 mc_search__get_one_symbol (charset
, &(str_from
->str
[loop
]),
171 MIN (str_from
->len
- loop
, 6), &just_letters
);
172 one_char_len
= strlen (one_char
);
174 if (one_char_len
== 0)
178 loop
+= one_char_len
;
181 mc_search__cond_struct_new_regex_hex_add (charset
, recoded_part
, one_char
,
184 g_string_append_len (recoded_part
, one_char
, one_char_len
);
190 g_string_append_len (str_to
, recoded_part
->str
, recoded_part
->len
);
191 g_string_free (recoded_part
, TRUE
);
192 g_string_set_size (str_from
, 0);
195 /* --------------------------------------------------------------------------------------------- */
198 mc_search__cond_struct_new_regex_ci_str (const char *charset
, const GString
* astr
)
200 GString
*accumulator
, *spec_char
, *ret_str
;
203 ret_str
= g_string_sized_new (64);
204 accumulator
= g_string_sized_new (64);
205 spec_char
= g_string_sized_new (64);
208 while (loop
<= astr
->len
)
210 if (mc_search__regex_str_append_if_special (spec_char
, astr
, &loop
))
212 mc_search__cond_struct_new_regex_accum_append (charset
, ret_str
, accumulator
);
213 g_string_append_len (ret_str
, spec_char
->str
, spec_char
->len
);
214 g_string_set_size (spec_char
, 0);
218 if (astr
->str
[loop
] == '[' && !strutils_is_char_escaped (astr
->str
, &(astr
->str
[loop
])))
220 mc_search__cond_struct_new_regex_accum_append (charset
, ret_str
, accumulator
);
222 while (loop
< astr
->len
&& !(astr
->str
[loop
] == ']'
223 && !strutils_is_char_escaped (astr
->str
,
224 &(astr
->str
[loop
]))))
226 g_string_append_c (ret_str
, astr
->str
[loop
]);
230 g_string_append_c (ret_str
, astr
->str
[loop
]);
237 g_string_append_c (accumulator
, astr
->str
[loop
]);
240 mc_search__cond_struct_new_regex_accum_append (charset
, ret_str
, accumulator
);
242 g_string_free (accumulator
, TRUE
);
243 g_string_free (spec_char
, TRUE
);
248 /* --------------------------------------------------------------------------------------------- */
250 #ifdef SEARCH_TYPE_GLIB
251 /* A thin wrapper above g_regex_match_full that makes sure the string passed
252 * to it is valid UTF-8 (unless G_REGEX_RAW compile flag was set), as it is a
253 * requirement by glib and it might crash otherwise. See: mc ticket 3449.
254 * Be careful: there might be embedded NULs in the strings. */
256 mc_search__g_regex_match_full_safe (const GRegex
* regex
,
257 const gchar
* string
,
260 GRegexMatchFlags match_options
,
261 GMatchInfo
** match_info
, GError
** error
)
263 char *string_safe
, *p
, *end
;
266 if ((g_regex_get_compile_flags (regex
) & G_REGEX_RAW
)
267 || g_utf8_validate (string
, string_len
, NULL
))
269 return g_regex_match_full (regex
, string
, string_len
, start_position
, match_options
,
275 string_len
= strlen (string
);
278 /* Correctly handle embedded NULs while copying */
279 p
= string_safe
= g_malloc (string_len
);
280 memcpy (string_safe
, string
, string_len
);
281 end
= p
+ string_len
;
285 gunichar c
= g_utf8_get_char_validated (p
, -1);
286 if (c
!= (gunichar
) (-1) && c
!= (gunichar
) (-2))
288 p
= g_utf8_next_char (p
);
292 /* U+FFFD would be the proper choice, but then we'd have to
293 maintain mapping between old and new offsets.
294 So rather do a byte by byte replacement. */
300 g_regex_match_full (regex
, string_safe
, string_len
, start_position
, match_options
,
302 g_free (string_safe
);
305 #endif /* SEARCH_TYPE_GLIB */
307 /* --------------------------------------------------------------------------------------------- */
309 static mc_search__found_cond_t
310 mc_search__regex_found_cond_one (mc_search_t
* lc_mc_search
, mc_search_regex_t
* regex
,
311 GString
* search_str
)
313 #ifdef SEARCH_TYPE_GLIB
314 GError
*mcerror
= NULL
;
316 if (!mc_search__g_regex_match_full_safe
317 (regex
, search_str
->str
, search_str
->len
, 0, G_REGEX_MATCH_NEWLINE_ANY
,
318 &lc_mc_search
->regex_match_info
, &mcerror
))
320 g_match_info_free (lc_mc_search
->regex_match_info
);
321 lc_mc_search
->regex_match_info
= NULL
;
324 lc_mc_search
->error
= MC_SEARCH_E_REGEX
;
325 lc_mc_search
->error_str
=
326 str_conv_gerror_message (mcerror
, _("Regular expression error"));
327 g_error_free (mcerror
);
328 return COND__FOUND_ERROR
;
330 return COND__NOT_FOUND
;
332 lc_mc_search
->num_results
= g_match_info_get_match_count (lc_mc_search
->regex_match_info
);
333 #else /* SEARCH_TYPE_GLIB */
334 lc_mc_search
->num_results
= pcre_exec (regex
, lc_mc_search
->regex_match_info
,
335 search_str
->str
, search_str
->len
, 0, 0,
336 lc_mc_search
->iovector
, MC_SEARCH__NUM_REPLACE_ARGS
);
337 if (lc_mc_search
->num_results
< 0)
339 return COND__NOT_FOUND
;
341 #endif /* SEARCH_TYPE_GLIB */
342 return COND__FOUND_OK
;
346 /* --------------------------------------------------------------------------------------------- */
348 static mc_search__found_cond_t
349 mc_search__regex_found_cond (mc_search_t
* lc_mc_search
, GString
* search_str
)
353 for (loop1
= 0; loop1
< lc_mc_search
->conditions
->len
; loop1
++)
355 mc_search_cond_t
*mc_search_cond
;
356 mc_search__found_cond_t ret
;
358 mc_search_cond
= (mc_search_cond_t
*) g_ptr_array_index (lc_mc_search
->conditions
, loop1
);
360 if (!mc_search_cond
->regex_handle
)
364 mc_search__regex_found_cond_one (lc_mc_search
, mc_search_cond
->regex_handle
,
366 if (ret
!= COND__NOT_FOUND
)
369 return COND__NOT_ALL_FOUND
;
372 /* --------------------------------------------------------------------------------------------- */
375 mc_search_regex__get_max_num_of_replace_tokens (const gchar
* str
, gsize len
)
379 for (loop
= 0; loop
< len
- 1; loop
++)
381 if (str
[loop
] == '\\' && g_ascii_isdigit (str
[loop
+ 1]))
383 if (strutils_is_char_escaped (str
, &str
[loop
]))
385 if (max_token
< str
[loop
+ 1] - '0')
386 max_token
= str
[loop
+ 1] - '0';
389 if (str
[loop
] == '$' && str
[loop
+ 1] == '{')
393 if (strutils_is_char_escaped (str
, &str
[loop
]))
397 loop
+ tmp_len
+ 2 < len
&& (str
[loop
+ 2 + tmp_len
] & (char) 0xf0) == 0x30;
400 if (str
[loop
+ 2 + tmp_len
] == '}')
405 tmp_str
= g_strndup (&str
[loop
+ 2], tmp_len
);
406 tmp_token
= atoi (tmp_str
);
407 if (max_token
< tmp_token
)
408 max_token
= tmp_token
;
416 /* --------------------------------------------------------------------------------------------- */
419 mc_search_regex__get_token_by_num (const mc_search_t
* lc_mc_search
, gsize lc_index
)
421 int fnd_start
= 0, fnd_end
= 0;
423 #ifdef SEARCH_TYPE_GLIB
424 g_match_info_fetch_pos (lc_mc_search
->regex_match_info
, lc_index
, &fnd_start
, &fnd_end
);
425 #else /* SEARCH_TYPE_GLIB */
426 fnd_start
= lc_mc_search
->iovector
[lc_index
* 2 + 0];
427 fnd_end
= lc_mc_search
->iovector
[lc_index
* 2 + 1];
428 #endif /* SEARCH_TYPE_GLIB */
430 if (fnd_end
== fnd_start
)
431 return g_strdup ("");
433 return g_strndup (lc_mc_search
->regex_buffer
->str
+ fnd_start
, fnd_end
- fnd_start
);
437 /* --------------------------------------------------------------------------------------------- */
440 mc_search_regex__replace_handle_esc_seq (const GString
* replace_str
, const gsize current_pos
,
441 gsize
* skip_len
, int *ret
)
443 char *curr_str
= &(replace_str
->str
[current_pos
]);
444 char c
= curr_str
[1];
446 if (replace_str
->len
> current_pos
+ 2)
450 for (*skip_len
= 2; /* \{ */
451 current_pos
+ *skip_len
< replace_str
->len
&& curr_str
[*skip_len
] >= '0'
452 && curr_str
[*skip_len
] <= '7'; (*skip_len
)++)
455 if (current_pos
+ *skip_len
< replace_str
->len
&& curr_str
[*skip_len
] == '}')
458 *ret
= REPLACE_PREPARE_T_ESCAPE_SEQ
;
463 *ret
= REPLACE_PREPARE_T_NOTHING_SPECIAL
;
470 *skip_len
= 2; /* \x */
474 for (*skip_len
= 3; /* \x{ */
475 current_pos
+ *skip_len
< replace_str
->len
476 && g_ascii_isxdigit ((guchar
) curr_str
[*skip_len
]); (*skip_len
)++)
479 if (current_pos
+ *skip_len
< replace_str
->len
&& curr_str
[*skip_len
] == '}')
482 *ret
= REPLACE_PREPARE_T_ESCAPE_SEQ
;
487 *ret
= REPLACE_PREPARE_T_NOTHING_SPECIAL
;
491 else if (!g_ascii_isxdigit ((guchar
) c
))
493 *skip_len
= 2; /* \x without number behind */
494 *ret
= REPLACE_PREPARE_T_NOTHING_SPECIAL
;
500 if (!g_ascii_isxdigit ((guchar
) c
))
501 *skip_len
= 3; /* \xH */
503 *skip_len
= 4; /* \xHH */
504 *ret
= REPLACE_PREPARE_T_ESCAPE_SEQ
;
510 if (strchr ("ntvbrfa", c
) != NULL
)
513 *ret
= REPLACE_PREPARE_T_ESCAPE_SEQ
;
519 /* --------------------------------------------------------------------------------------------- */
522 mc_search_regex__process_replace_str (const GString
* replace_str
, const gsize current_pos
,
523 gsize
* skip_len
, replace_transform_type_t
* replace_flags
)
526 const char *curr_str
= &(replace_str
->str
[current_pos
]);
528 if (current_pos
> replace_str
->len
)
529 return REPLACE_PREPARE_T_NOTHING_SPECIAL
;
533 if (replace_str
->len
> current_pos
+ 2 && curr_str
[0] == '$' && curr_str
[1] == '{'
534 && (curr_str
[2] & (char) 0xf0) == 0x30)
538 if (strutils_is_char_escaped (replace_str
->str
, curr_str
))
541 return REPLACE_PREPARE_T_NOTHING_SPECIAL
;
545 current_pos
+ *skip_len
+ 2 < replace_str
->len
546 && (curr_str
[2 + *skip_len
] & (char) 0xf0) == 0x30; (*skip_len
)++)
549 if (curr_str
[2 + *skip_len
] != '}')
550 return REPLACE_PREPARE_T_NOTHING_SPECIAL
;
552 tmp_str
= g_strndup (curr_str
+ 2, *skip_len
);
554 return REPLACE_PREPARE_T_NOTHING_SPECIAL
;
556 ret
= atoi (tmp_str
);
559 *skip_len
+= 3; /* ${} */
560 return ret
; /* capture buffer index >= 0 */
563 if (curr_str
[0] == '\\' && replace_str
->len
> current_pos
+ 1)
565 if (strutils_is_char_escaped (replace_str
->str
, curr_str
))
568 return REPLACE_PREPARE_T_NOTHING_SPECIAL
;
571 if (g_ascii_isdigit (curr_str
[1]))
573 ret
= g_ascii_digit_value (curr_str
[1]); /* capture buffer index >= 0 */
574 *skip_len
= 2; /* \\ and one digit */
578 if (!mc_search_regex__replace_handle_esc_seq (replace_str
, current_pos
, skip_len
, &ret
))
581 ret
= REPLACE_PREPARE_T_REPLACE_FLAG
;
587 *replace_flags
|= REPLACE_T_UPP_TRANSFORM
;
588 *replace_flags
&= ~REPLACE_T_LOW_TRANSFORM
;
591 *replace_flags
|= REPLACE_T_UPP_TRANSFORM_CHAR
;
594 *replace_flags
|= REPLACE_T_LOW_TRANSFORM
;
595 *replace_flags
&= ~REPLACE_T_UPP_TRANSFORM
;
598 *replace_flags
|= REPLACE_T_LOW_TRANSFORM_CHAR
;
601 *replace_flags
= REPLACE_T_NO_TRANSFORM
;
604 ret
= REPLACE_PREPARE_T_NOTHING_SPECIAL
;
611 /* --------------------------------------------------------------------------------------------- */
614 mc_search_regex__process_append_str (GString
* dest_str
, const char *from
, gsize len
,
615 replace_transform_type_t
* replace_flags
)
620 if (len
== (gsize
) (-1))
623 if (*replace_flags
== REPLACE_T_NO_TRANSFORM
)
625 g_string_append_len (dest_str
, from
, len
);
629 for (loop
= 0; loop
< len
; loop
+= char_len
)
631 GString
*tmp_string
= NULL
;
634 tmp_str
= mc_search__get_one_symbol (NULL
, from
+ loop
, len
- loop
, NULL
);
635 char_len
= strlen (tmp_str
);
637 if ((*replace_flags
& REPLACE_T_UPP_TRANSFORM_CHAR
) != 0)
639 *replace_flags
&= ~REPLACE_T_UPP_TRANSFORM_CHAR
;
640 tmp_string
= mc_search__toupper_case_str (NULL
, tmp_str
, char_len
);
641 g_string_append_len (dest_str
, tmp_string
->str
, tmp_string
->len
);
642 g_string_free (tmp_string
, TRUE
);
644 else if ((*replace_flags
& REPLACE_T_LOW_TRANSFORM_CHAR
) != 0)
646 *replace_flags
&= ~REPLACE_T_LOW_TRANSFORM_CHAR
;
647 tmp_string
= mc_search__tolower_case_str (NULL
, tmp_str
, char_len
);
648 g_string_append_len (dest_str
, tmp_string
->str
, tmp_string
->len
);
649 g_string_free (tmp_string
, TRUE
);
651 else if ((*replace_flags
& REPLACE_T_UPP_TRANSFORM
) != 0)
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
) != 0)
659 tmp_string
= mc_search__tolower_case_str (NULL
, tmp_str
, char_len
);
660 g_string_append_len (dest_str
, tmp_string
->str
, tmp_string
->len
);
661 g_string_free (tmp_string
, TRUE
);
668 /* --------------------------------------------------------------------------------------------- */
671 mc_search_regex__process_escape_sequence (GString
* dest_str
, const char *from
, gsize len
,
672 replace_transform_type_t
* replace_flags
,
679 if (len
== (gsize
) (-1))
692 if (i
< len
&& from
[i
] == '{')
696 if (from
[i
] >= '0' && from
[i
] <= '9')
697 c
= c
* 16 + from
[i
] - '0';
698 else if (from
[i
] >= 'a' && from
[i
] <= 'f')
699 c
= c
* 16 + 10 + from
[i
] - 'a';
700 else if (from
[i
] >= 'A' && from
[i
] <= 'F')
701 c
= c
* 16 + 10 + from
[i
] - 'A';
706 else if (from
[i
] >= '0' && from
[i
] <= '7')
707 for (; i
< len
&& from
[i
] >= '0' && from
[i
] <= '7'; i
++)
708 c
= c
* 8 + from
[i
] - '0';
735 mc_search_regex__process_append_str (dest_str
, from
, len
, replace_flags
);
740 if (c
< 0x80 || !is_utf8
)
741 g_string_append_c (dest_str
, (char) c
);
745 g_string_append_c (dest_str
, b
);
746 b
= 0x80 | (c
& 0x3F);
747 g_string_append_c (dest_str
, b
);
749 else if (c
< 0x10000)
751 b
= 0xE0 | (c
>> 12);
752 g_string_append_c (dest_str
, b
);
753 b
= 0x80 | ((c
>> 6) & 0x3F);
754 g_string_append_c (dest_str
, b
);
755 b
= 0x80 | (c
& 0x3F);
756 g_string_append_c (dest_str
, b
);
758 else if (c
< 0x10FFFF)
760 b
= 0xF0 | (c
>> 16);
761 g_string_append_c (dest_str
, b
);
762 b
= 0x80 | ((c
>> 12) & 0x3F);
763 g_string_append_c (dest_str
, b
);
764 b
= 0x80 | ((c
>> 6) & 0x3F);
765 g_string_append_c (dest_str
, b
);
766 b
= 0x80 | (c
& 0x3F);
767 g_string_append_c (dest_str
, b
);
771 /* --------------------------------------------------------------------------------------------- */
772 /*** public functions ****************************************************************************/
773 /* --------------------------------------------------------------------------------------------- */
776 mc_search__cond_struct_new_init_regex (const char *charset
, mc_search_t
* lc_mc_search
,
777 mc_search_cond_t
* mc_search_cond
)
779 if (lc_mc_search
->whole_words
&& !lc_mc_search
->is_entire_line
)
781 /* NOTE: \b as word boundary doesn't allow search
782 * whole words with non-ASCII symbols.
783 * Update: Is it still true nowadays? Probably not. #2396, #3524 */
784 g_string_prepend (mc_search_cond
->str
, "(?<![\\p{L}\\p{N}_])");
785 g_string_append (mc_search_cond
->str
, "(?![\\p{L}\\p{N}_])");
789 #ifdef SEARCH_TYPE_GLIB
790 GError
*mcerror
= NULL
;
791 GRegexCompileFlags g_regex_options
= G_REGEX_OPTIMIZE
| G_REGEX_DOTALL
;
793 if (str_isutf8 (charset
) && mc_global
.utf8_display
)
795 if (!lc_mc_search
->is_case_sensitive
)
796 g_regex_options
|= G_REGEX_CASELESS
;
800 g_regex_options
|= G_REGEX_RAW
;
802 if (!lc_mc_search
->is_case_sensitive
)
806 tmp
= mc_search_cond
->str
;
807 mc_search_cond
->str
= mc_search__cond_struct_new_regex_ci_str (charset
, tmp
);
808 g_string_free (tmp
, TRUE
);
813 mc_search_cond
->regex_handle
=
814 g_regex_new (mc_search_cond
->str
->str
, g_regex_options
, 0, &mcerror
);
818 lc_mc_search
->error
= MC_SEARCH_E_REGEX_COMPILE
;
819 lc_mc_search
->error_str
=
820 str_conv_gerror_message (mcerror
, _("Regular expression error"));
821 g_error_free (mcerror
);
824 #else /* SEARCH_TYPE_GLIB */
827 int pcre_options
= PCRE_EXTRA
| PCRE_MULTILINE
;
829 if (str_isutf8 (charset
) && mc_global
.utf8_display
)
831 pcre_options
|= PCRE_UTF8
;
832 if (!lc_mc_search
->is_case_sensitive
)
833 pcre_options
|= PCRE_CASELESS
;
837 if (!lc_mc_search
->is_case_sensitive
)
841 tmp
= mc_search_cond
->str
;
842 mc_search_cond
->str
= mc_search__cond_struct_new_regex_ci_str (charset
, tmp
);
843 g_string_free (tmp
, TRUE
);
847 mc_search_cond
->regex_handle
=
848 pcre_compile (mc_search_cond
->str
->str
, pcre_options
, &error
, &erroffset
, NULL
);
849 if (mc_search_cond
->regex_handle
== NULL
)
851 lc_mc_search
->error
= MC_SEARCH_E_REGEX_COMPILE
;
852 lc_mc_search
->error_str
= g_strdup (error
);
855 lc_mc_search
->regex_match_info
= pcre_study (mc_search_cond
->regex_handle
, 0, &error
);
856 if (lc_mc_search
->regex_match_info
== NULL
)
860 lc_mc_search
->error
= MC_SEARCH_E_REGEX_COMPILE
;
861 lc_mc_search
->error_str
= g_strdup (error
);
862 g_free (mc_search_cond
->regex_handle
);
863 mc_search_cond
->regex_handle
= NULL
;
867 #endif /* SEARCH_TYPE_GLIB */
870 lc_mc_search
->is_utf8
= str_isutf8 (charset
);
873 /* --------------------------------------------------------------------------------------------- */
876 mc_search__run_regex (mc_search_t
* lc_mc_search
, const void *user_data
,
877 gsize start_search
, gsize end_search
, gsize
* found_len
)
879 mc_search_cbret_t ret
= MC_SEARCH_CB_ABORT
;
880 gsize current_pos
, virtual_pos
;
884 if (lc_mc_search
->regex_buffer
!= NULL
)
885 g_string_free (lc_mc_search
->regex_buffer
, TRUE
);
887 lc_mc_search
->regex_buffer
= g_string_sized_new (64);
889 virtual_pos
= current_pos
= start_search
;
890 while (virtual_pos
<= end_search
)
892 g_string_set_size (lc_mc_search
->regex_buffer
, 0);
893 lc_mc_search
->start_buffer
= current_pos
;
895 if (lc_mc_search
->search_fn
!= NULL
)
899 int current_chr
= '\n'; /* stop search symbol */
901 ret
= lc_mc_search
->search_fn (user_data
, current_pos
, ¤t_chr
);
903 if (ret
== MC_SEARCH_CB_ABORT
)
906 if (ret
== MC_SEARCH_CB_INVALID
)
911 if (ret
== MC_SEARCH_CB_SKIP
)
916 g_string_append_c (lc_mc_search
->regex_buffer
, (char) current_chr
);
918 if ((char) current_chr
== '\n' || virtual_pos
> end_search
)
924 /* optimization for standard case (for search from file manager)
925 * where there is no MC_SEARCH_CB_INVALID or MC_SEARCH_CB_SKIP
926 * return codes, so we can copy line at regex buffer all at once
930 const char current_chr
= ((const char *) user_data
)[current_pos
];
932 if (current_chr
== '\0')
937 if (current_chr
== '\n' || current_pos
> end_search
)
941 /* use virtual_pos as index of start of current chunk */
942 g_string_append_len (lc_mc_search
->regex_buffer
, (const char *) user_data
+ virtual_pos
,
943 current_pos
- virtual_pos
);
944 virtual_pos
= current_pos
;
947 switch (mc_search__regex_found_cond (lc_mc_search
, lc_mc_search
->regex_buffer
))
950 #ifdef SEARCH_TYPE_GLIB
951 g_match_info_fetch_pos (lc_mc_search
->regex_match_info
, 0, &start_pos
, &end_pos
);
952 #else /* SEARCH_TYPE_GLIB */
953 start_pos
= lc_mc_search
->iovector
[0];
954 end_pos
= lc_mc_search
->iovector
[1];
955 #endif /* SEARCH_TYPE_GLIB */
956 if (found_len
!= NULL
)
957 *found_len
= end_pos
- start_pos
;
958 lc_mc_search
->normal_offset
= lc_mc_search
->start_buffer
+ start_pos
;
960 case COND__NOT_ALL_FOUND
:
963 g_string_free (lc_mc_search
->regex_buffer
, TRUE
);
964 lc_mc_search
->regex_buffer
= NULL
;
968 if ((lc_mc_search
->update_fn
!= NULL
) &&
969 ((lc_mc_search
->update_fn
) (user_data
, current_pos
) == MC_SEARCH_CB_ABORT
))
970 ret
= MC_SEARCH_CB_ABORT
;
972 if (ret
== MC_SEARCH_CB_ABORT
)
976 g_string_free (lc_mc_search
->regex_buffer
, TRUE
);
977 lc_mc_search
->regex_buffer
= NULL
;
978 lc_mc_search
->error
= MC_SEARCH_E_NOTFOUND
;
980 if (ret
!= MC_SEARCH_CB_ABORT
)
981 lc_mc_search
->error_str
= g_strdup (_(STR_E_NOTFOUND
));
983 lc_mc_search
->error_str
= NULL
;
988 /* --------------------------------------------------------------------------------------------- */
991 mc_search_regex_prepare_replace_str (mc_search_t
* lc_mc_search
, GString
* replace_str
)
995 int num_replace_tokens
;
998 replace_transform_type_t replace_flags
= REPLACE_T_NO_TRANSFORM
;
1000 num_replace_tokens
=
1001 mc_search_regex__get_max_num_of_replace_tokens (replace_str
->str
, replace_str
->len
);
1003 if (lc_mc_search
->num_results
< 0)
1004 return g_string_new_len (replace_str
->str
, replace_str
->len
);
1006 if (num_replace_tokens
> lc_mc_search
->num_results
- 1
1007 || num_replace_tokens
> MC_SEARCH__NUM_REPLACE_ARGS
)
1009 lc_mc_search
->error
= MC_SEARCH_E_REGEX_REPLACE
;
1010 lc_mc_search
->error_str
= g_strdup (_(STR_E_RPL_NOT_EQ_TO_FOUND
));
1014 ret
= g_string_sized_new (64);
1016 for (loop
= 0; loop
< replace_str
->len
- 1; loop
++)
1022 lc_index
= mc_search_regex__process_replace_str (replace_str
, loop
, &len
, &replace_flags
);
1024 if (lc_index
== REPLACE_PREPARE_T_NOTHING_SPECIAL
)
1028 mc_search_regex__process_append_str (ret
, replace_str
->str
+ prev
, loop
- prev
,
1030 mc_search_regex__process_append_str (ret
, replace_str
->str
+ loop
+ 1, len
- 1,
1033 loop
= prev
- 1; /* prepare to loop++ */
1039 if (lc_index
== REPLACE_PREPARE_T_REPLACE_FLAG
)
1042 mc_search_regex__process_append_str (ret
, replace_str
->str
+ prev
, loop
- prev
,
1045 loop
= prev
- 1; /* prepare to loop++ */
1049 /* escape sequence */
1050 if (lc_index
== REPLACE_PREPARE_T_ESCAPE_SEQ
)
1052 mc_search_regex__process_append_str (ret
, replace_str
->str
+ prev
, loop
- prev
,
1054 /* call process_escape_sequence without starting '\\' */
1055 mc_search_regex__process_escape_sequence (ret
, replace_str
->str
+ loop
+ 1, len
- 1,
1056 &replace_flags
, lc_mc_search
->is_utf8
);
1058 loop
= prev
- 1; /* prepare to loop++ */
1062 /* invalid capture buffer number */
1063 if (lc_index
> lc_mc_search
->num_results
)
1065 g_string_free (ret
, TRUE
);
1066 lc_mc_search
->error
= MC_SEARCH_E_REGEX_REPLACE
;
1067 lc_mc_search
->error_str
= g_strdup_printf (_(STR_E_RPL_INVALID_TOKEN
), lc_index
);
1071 tmp_str
= mc_search_regex__get_token_by_num (lc_mc_search
, lc_index
);
1074 mc_search_regex__process_append_str (ret
, replace_str
->str
+ prev
, loop
- prev
,
1077 mc_search_regex__process_append_str (ret
, tmp_str
, -1, &replace_flags
);
1081 loop
= prev
- 1; /* prepare to loop++ */
1084 mc_search_regex__process_append_str (ret
, replace_str
->str
+ prev
, replace_str
->len
- prev
,