Mercurial > octave
annotate libinterp/corefcn/strfind.cc @ 19386:f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
* NEWS: Announce change.
* strfind.cc (Fstrfind): Add note to docstring about change. Return Matrix ()
if pattern is empty.
author | Rik <rik@octave.org> |
---|---|
date | Fri, 05 Dec 2014 22:17:09 -0800 |
parents | 65554f5847ac |
children | db92e7e28e1f |
rev | line source |
---|---|
10022 | 1 /* |
2 | |
17744
d63878346099
maint: Update copyright notices for release.
John W. Eaton <jwe@octave.org>
parents:
17398
diff
changeset
|
3 Copyright (C) 2009-2013 Jaroslav Hajek |
11523 | 4 Copyright (C) 2009-2010 VZLU Prague |
10022 | 5 |
6 This file is part of Octave. | |
7 | |
8 Octave is free software; you can redistribute it and/or modify it | |
9 under the terms of the GNU General Public License as published by the | |
10 Free Software Foundation; either version 3 of the License, or (at your | |
11 option) any later version. | |
12 | |
13 Octave is distributed in the hope that it will be useful, but WITHOUT | |
14 ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or | |
15 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License | |
16 for more details. | |
17 | |
18 You should have received a copy of the GNU General Public License | |
19 along with Octave; see the file COPYING. If not, see | |
20 <http://www.gnu.org/licenses/>. | |
21 | |
22 */ | |
23 | |
24 #ifdef HAVE_CONFIG_H | |
25 #include <config.h> | |
26 #endif | |
27 | |
28 #include <algorithm> | |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
29 #include <deque> |
15215
9020dddc925a
use std::numeric_limits for integer max and min values
John W. Eaton <jwe@octave.org>
parents:
15195
diff
changeset
|
30 #include <limits> |
9020dddc925a
use std::numeric_limits for integer max and min values
John W. Eaton <jwe@octave.org>
parents:
15195
diff
changeset
|
31 #include <string> |
10022 | 32 |
19269
65554f5847ac
don't include oct-locbuf.h in header files unnecessarily
John W. Eaton <jwe@octave.org>
parents:
19062
diff
changeset
|
33 #include "oct-locbuf.h" |
65554f5847ac
don't include oct-locbuf.h in header files unnecessarily
John W. Eaton <jwe@octave.org>
parents:
19062
diff
changeset
|
34 |
10022 | 35 #include "Cell.h" |
36 #include "ov.h" | |
15039
e753177cde93
maint: Move non-dynamically linked functions from DLD-FUNCTIONS/ to corefcn/ directory
Rik <rik@octave.org>
parents:
14501
diff
changeset
|
37 #include "defun.h" |
10022 | 38 #include "unwind-prot.h" |
39 #include "gripes.h" | |
40 #include "utils.h" | |
41 | |
17787
175b392e91fe
Use GNU style coding conventions for code in libinterp/
Rik <rik@octave.org>
parents:
17744
diff
changeset
|
42 // This allows safe indexing with char. |
175b392e91fe
Use GNU style coding conventions for code in libinterp/
Rik <rik@octave.org>
parents:
17744
diff
changeset
|
43 // In C++, char may be (and often is) signed! |
10022 | 44 #define ORD(ch) static_cast<unsigned char>(ch) |
15215
9020dddc925a
use std::numeric_limits for integer max and min values
John W. Eaton <jwe@octave.org>
parents:
15195
diff
changeset
|
45 #define TABSIZE (std::numeric_limits<unsigned char>::max () + 1) |
10022 | 46 |
47 // This is the quick search algorithm, as described at | |
48 // http://www-igm.univ-mlv.fr/~lecroq/string/node19.html | |
11586
12df7854fa7c
strip trailing whitespace from source files
John W. Eaton <jwe@octave.org>
parents:
11572
diff
changeset
|
49 static void |
10022 | 50 qs_preprocess (const Array<char>& needle, |
15215
9020dddc925a
use std::numeric_limits for integer max and min values
John W. Eaton <jwe@octave.org>
parents:
15195
diff
changeset
|
51 octave_idx_type *table) |
10022 | 52 { |
53 const char *x = needle.data (); | |
54 octave_idx_type m = needle.numel (); | |
55 | |
17787
175b392e91fe
Use GNU style coding conventions for code in libinterp/
Rik <rik@octave.org>
parents:
17744
diff
changeset
|
56 for (octave_idx_type i = 0; i < TABSIZE; i++) |
175b392e91fe
Use GNU style coding conventions for code in libinterp/
Rik <rik@octave.org>
parents:
17744
diff
changeset
|
57 table[i] = m + 1; |
175b392e91fe
Use GNU style coding conventions for code in libinterp/
Rik <rik@octave.org>
parents:
17744
diff
changeset
|
58 for (octave_idx_type i = 0; i < m; i++) |
175b392e91fe
Use GNU style coding conventions for code in libinterp/
Rik <rik@octave.org>
parents:
17744
diff
changeset
|
59 table[ORD(x[i])] = m - i; |
10022 | 60 } |
61 | |
62 | |
11586
12df7854fa7c
strip trailing whitespace from source files
John W. Eaton <jwe@octave.org>
parents:
11572
diff
changeset
|
63 static Array<octave_idx_type> |
10022 | 64 qs_search (const Array<char>& needle, |
65 const Array<char>& haystack, | |
15215
9020dddc925a
use std::numeric_limits for integer max and min values
John W. Eaton <jwe@octave.org>
parents:
15195
diff
changeset
|
66 const octave_idx_type *table, |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
67 bool overlaps = true) |
10022 | 68 { |
69 const char *x = needle.data (); | |
70 octave_idx_type m = needle.numel (); | |
71 const char *y = haystack.data (); | |
72 octave_idx_type n = haystack.numel (); | |
73 | |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
74 // We'll use deque because it typically has the most favorable properties for |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
75 // the operation we need. |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
76 std::deque<octave_idx_type> accum; |
10133
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
77 if (m == 1) |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
78 { |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
79 // Looking for a single character. |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
80 for (octave_idx_type i = 0; i < n; i++) |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
81 { |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
82 if (y[i] == x[0]) |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
83 accum.push_back (i); |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
84 } |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
85 } |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
86 else if (m == 2) |
10022 | 87 { |
10133
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
88 // Two characters. |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
89 if (overlaps) |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
90 { |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
91 for (octave_idx_type i = 0; i < n-1; i++) |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
92 { |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
93 if (y[i] == x[0] && y[i+1] == x[1]) |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
94 accum.push_back (i); |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
95 } |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
96 } |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
97 else |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
98 { |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
99 for (octave_idx_type i = 0; i < n-1; i++) |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
100 { |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
101 if (y[i] == x[0] && y[i+1] == x[1]) |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
102 accum.push_back (i++); |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
103 } |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
104 } |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
105 } |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
106 else if (n >= m) |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
107 { |
2e4fc7fdba15
optimize strfind with 1 or 2 characters
Jaroslav Hajek <highegg@gmail.com>
parents:
10086
diff
changeset
|
108 // General case. |
10022 | 109 octave_idx_type j = 0; |
110 | |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
111 if (overlaps) |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
112 { |
11586
12df7854fa7c
strip trailing whitespace from source files
John W. Eaton <jwe@octave.org>
parents:
11572
diff
changeset
|
113 while (j < n - m) |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
114 { |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
115 if (std::equal (x, x + m, y + j)) |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
116 accum.push_back (j); |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
117 j += table[ORD(y[j + m])]; |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
118 } |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
119 } |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
120 else |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
121 { |
11586
12df7854fa7c
strip trailing whitespace from source files
John W. Eaton <jwe@octave.org>
parents:
11572
diff
changeset
|
122 while (j < n - m) |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
123 { |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
124 if (std::equal (x, x + m, y + j)) |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
125 { |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
126 accum.push_back (j); |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
127 j += m; |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
128 } |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
129 else |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
130 j += table[ORD(y[j + m])]; |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
131 } |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
132 } |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
133 |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
134 if (j == n - m && std::equal (x, x + m, y + j)) |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
135 accum.push_back (j); |
10022 | 136 } |
137 | |
138 octave_idx_type nmatch = accum.size (); | |
10258 | 139 octave_idx_type one = 1; |
140 Array<octave_idx_type> result (dim_vector (std::min (one, nmatch), nmatch)); | |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
141 octave_idx_type k = 0; |
11586
12df7854fa7c
strip trailing whitespace from source files
John W. Eaton <jwe@octave.org>
parents:
11572
diff
changeset
|
142 for (std::deque<octave_idx_type>::const_iterator iter = accum.begin (); |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
143 iter != accum.end (); iter++) |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
144 { |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
145 result.xelem (k++) = *iter; |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
146 } |
10022 | 147 |
148 return result; | |
149 } | |
150 | |
15039
e753177cde93
maint: Move non-dynamically linked functions from DLD-FUNCTIONS/ to corefcn/ directory
Rik <rik@octave.org>
parents:
14501
diff
changeset
|
151 DEFUN (strfind, args, , |
17787
175b392e91fe
Use GNU style coding conventions for code in libinterp/
Rik <rik@octave.org>
parents:
17744
diff
changeset
|
152 "-*- texinfo -*-\n\ |
15039
e753177cde93
maint: Move non-dynamically linked functions from DLD-FUNCTIONS/ to corefcn/ directory
Rik <rik@octave.org>
parents:
14501
diff
changeset
|
153 @deftypefn {Built-in Function} {@var{idx} =} strfind (@var{str}, @var{pattern})\n\ |
e753177cde93
maint: Move non-dynamically linked functions from DLD-FUNCTIONS/ to corefcn/ directory
Rik <rik@octave.org>
parents:
14501
diff
changeset
|
154 @deftypefnx {Built-in Function} {@var{idx} =} strfind (@var{cellstr}, @var{pattern})\n\ |
17398
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
155 @deftypefnx {Built-in Function} {@var{idx} =} strfind (@dots{}, \"overlaps\", @var{val})\n\ |
10022 | 156 Search for @var{pattern} in the string @var{str} and return the\n\ |
157 starting index of every such occurrence in the vector @var{idx}.\n\ | |
17398
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
158 \n\ |
10022 | 159 If there is no such occurrence, or if @var{pattern} is longer\n\ |
19386
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
160 than @var{str}, or if @var{pattern} itself is empty, then @var{idx} is the\n\ |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
161 empty array @code{[]}.\n\ |
17398
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
162 The optional argument @qcode{\"overlaps\"} determines whether the pattern\n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
163 can match at every position in @var{str} (true), or only for unique\n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
164 occurrences of the complete pattern (false). The default is true.\n\ |
10022 | 165 \n\ |
11553
01f703952eff
Improve docstrings for functions in DLD-FUNCTIONS directory.
Rik <octave@nomad.inbox5.com>
parents:
11523
diff
changeset
|
166 If a cell array of strings @var{cellstr} is specified\n\ |
17398
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
167 then @var{idx} is a cell array of vectors, as specified above.\n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
168 \n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
169 Examples:\n\ |
10022 | 170 \n\ |
171 @example\n\ | |
172 @group\n\ | |
173 strfind (\"abababa\", \"aba\")\n\ | |
174 @result{} [1, 3, 5]\n\ | |
175 \n\ | |
17398
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
176 strfind (\"abababa\", \"aba\", \"overlaps\", false)\n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
177 @result{} [1, 5]\n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
178 \n\ |
10022 | 179 strfind (@{\"abababa\", \"bebebe\", \"ab\"@}, \"aba\")\n\ |
14360
97883071e8e4
doc: Correct off-by-1 spacings in all .cc docstrings
Rik <octave@nomad.inbox5.com>
parents:
14138
diff
changeset
|
180 @result{}\n\ |
10022 | 181 @{\n\ |
182 [1,1] =\n\ | |
183 \n\ | |
184 1 3 5\n\ | |
185 \n\ | |
186 [1,2] = [](1x0)\n\ | |
187 [1,3] = [](1x0)\n\ | |
188 @}\n\ | |
189 @end group\n\ | |
190 @end example\n\ | |
19062
f9959972949a
Remove references to deprecated strmatch function.
Ben Abbott <bpabbott@mac.com>
parents:
18763
diff
changeset
|
191 @seealso{findstr, regexp, regexpi, find}\n\ |
10022 | 192 @end deftypefn") |
193 { | |
194 octave_value retval; | |
195 int nargin = args.length (); | |
10677
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
196 bool overlaps = true; |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
197 |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
198 if (nargin == 4 && args(2).is_string () && args(3).is_scalar_type ()) |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
199 { |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
200 std::string opt = args(2).string_value (); |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
201 if (opt == "overlaps") |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
202 { |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
203 overlaps = args(3).bool_value (); |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
204 nargin = 2; |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
205 } |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
206 else |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
207 { |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
208 error ("strfind: unknown option: %s", opt.c_str ()); |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
209 return retval; |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
210 } |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
211 } |
10022 | 212 |
213 if (nargin == 2) | |
214 { | |
18100
6a71e5030df5
Follow coding convention of defining and initializing only 1 variable per line in liboctinterp.
Rik <rik@octave.org>
parents:
17787
diff
changeset
|
215 octave_value argstr = args(0); |
6a71e5030df5
Follow coding convention of defining and initializing only 1 variable per line in liboctinterp.
Rik <rik@octave.org>
parents:
17787
diff
changeset
|
216 octave_value argpat = args(1); |
10086
76df75b10c80
allow more cells in strfind/strrep + supply a general mechanism
Jaroslav Hajek <highegg@gmail.com>
parents:
10074
diff
changeset
|
217 if (argpat.is_string ()) |
10022 | 218 { |
10086
76df75b10c80
allow more cells in strfind/strrep + supply a general mechanism
Jaroslav Hajek <highegg@gmail.com>
parents:
10074
diff
changeset
|
219 Array<char> needle = argpat.char_array_value (); |
12786
4e70c4e61c24
strfind.cc: Fix infinite loop when pattern contains 255 (Bug #33670).
Rik <octave@nomad.inbox5.com>
parents:
12483
diff
changeset
|
220 OCTAVE_LOCAL_BUFFER (octave_idx_type, table, TABSIZE); |
10022 | 221 qs_preprocess (needle, table); |
222 | |
223 if (argstr.is_string ()) | |
19386
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
224 if (argpat.is_empty ()) |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
225 // Return a null matrix for null pattern for MW compatibility |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
226 retval = Matrix (); |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
227 else |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
228 retval = octave_value (qs_search (needle, |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
229 argstr.char_array_value (), |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
230 table, overlaps), |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
231 true, true); |
10022 | 232 else if (argstr.is_cell ()) |
233 { | |
234 const Cell argsc = argstr.cell_value (); | |
235 Cell retc (argsc.dims ()); | |
236 octave_idx_type ns = argsc.numel (); | |
237 | |
238 for (octave_idx_type i = 0; i < ns; i++) | |
239 { | |
240 octave_value argse = argsc(i); | |
241 if (argse.is_string ()) | |
19386
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
242 { |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
243 if (argpat.is_empty ()) |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
244 retc(i) = Matrix (); |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
245 else |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
246 retc(i) = octave_value (qs_search (needle, |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
247 argse.char_array_value (), |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
248 table, overlaps), |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
249 true, true); |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
250 } |
10022 | 251 else |
252 { | |
11553
01f703952eff
Improve docstrings for functions in DLD-FUNCTIONS directory.
Rik <octave@nomad.inbox5.com>
parents:
11523
diff
changeset
|
253 error ("strfind: each element of CELLSTR must be a string"); |
10022 | 254 break; |
255 } | |
256 } | |
257 | |
258 retval = retc; | |
259 } | |
260 else | |
261 error ("strfind: first argument must be a string or cell array of strings"); | |
262 } | |
10086
76df75b10c80
allow more cells in strfind/strrep + supply a general mechanism
Jaroslav Hajek <highegg@gmail.com>
parents:
10074
diff
changeset
|
263 else if (argpat.is_cell ()) |
76df75b10c80
allow more cells in strfind/strrep + supply a general mechanism
Jaroslav Hajek <highegg@gmail.com>
parents:
10074
diff
changeset
|
264 retval = do_simple_cellfun (Fstrfind, "strfind", args); |
10022 | 265 else |
11553
01f703952eff
Improve docstrings for functions in DLD-FUNCTIONS directory.
Rik <octave@nomad.inbox5.com>
parents:
11523
diff
changeset
|
266 error ("strfind: PATTERN must be a string or cell array of strings"); |
10022 | 267 } |
268 else | |
269 print_usage (); | |
270 | |
271 return retval; | |
272 } | |
273 | |
274 /* | |
14501
60e5cf354d80
Update %!tests in DLD-FUNCTIONS/ directory with Octave coding conventions.
Rik <octave@nomad.inbox5.com>
parents:
14360
diff
changeset
|
275 %!assert (strfind ("abababa", "aba"), [1, 3, 5]) |
60e5cf354d80
Update %!tests in DLD-FUNCTIONS/ directory with Octave coding conventions.
Rik <octave@nomad.inbox5.com>
parents:
14360
diff
changeset
|
276 %!assert (strfind ("abababa", "aba", "overlaps", false), [1, 5]) |
60e5cf354d80
Update %!tests in DLD-FUNCTIONS/ directory with Octave coding conventions.
Rik <octave@nomad.inbox5.com>
parents:
14360
diff
changeset
|
277 %!assert (strfind ({"abababa", "bla", "bla"}, "a"), {[1, 3, 5, 7], 3, 3}) |
60e5cf354d80
Update %!tests in DLD-FUNCTIONS/ directory with Octave coding conventions.
Rik <octave@nomad.inbox5.com>
parents:
14360
diff
changeset
|
278 %!assert (strfind ("Linux _is_ user-friendly. It just isn't ignorant-friendly or idiot-friendly.", "friendly"), [17, 50, 68]) |
19386
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
279 %!assert (strfind ("abc", ""), []) |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
280 %!assert (strfind ("abc", {"", "b", ""}), {[], 2, []}) |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
281 %!assert (strfind ({"abc", "def"}, ""), {[], []}) |
10022 | 282 |
14501
60e5cf354d80
Update %!tests in DLD-FUNCTIONS/ directory with Octave coding conventions.
Rik <octave@nomad.inbox5.com>
parents:
14360
diff
changeset
|
283 %!error strfind () |
60e5cf354d80
Update %!tests in DLD-FUNCTIONS/ directory with Octave coding conventions.
Rik <octave@nomad.inbox5.com>
parents:
14360
diff
changeset
|
284 %!error strfind ("foo", "bar", 1) |
19386
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
285 %!error <unknown option: foobar> strfind ("foo", 100, "foobar", 1) |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
286 %!error <each element of CELLSTR must be a string> strfind ({"A", 1}, "foo") |
f054112912df
Change strfind to return empty array for empty pattern for compatibility (bug #43649).
Rik <rik@octave.org>
parents:
19269
diff
changeset
|
287 %!error <first argument must be a string> strfind (100, "foo") |
14501
60e5cf354d80
Update %!tests in DLD-FUNCTIONS/ directory with Octave coding conventions.
Rik <octave@nomad.inbox5.com>
parents:
14360
diff
changeset
|
288 %!error <PATTERN must be a string> strfind ("foo", 100) |
10022 | 289 */ |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
290 |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
291 static Array<char> |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
292 qs_replace (const Array<char>& str, const Array<char>& pat, |
11586
12df7854fa7c
strip trailing whitespace from source files
John W. Eaton <jwe@octave.org>
parents:
11572
diff
changeset
|
293 const Array<char>& rep, |
15215
9020dddc925a
use std::numeric_limits for integer max and min values
John W. Eaton <jwe@octave.org>
parents:
15195
diff
changeset
|
294 const octave_idx_type *table, |
10677
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
295 bool overlaps = true) |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
296 { |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
297 Array<char> ret = str; |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
298 |
18100
6a71e5030df5
Follow coding convention of defining and initializing only 1 variable per line in liboctinterp.
Rik <rik@octave.org>
parents:
17787
diff
changeset
|
299 octave_idx_type siz = str.numel (); |
6a71e5030df5
Follow coding convention of defining and initializing only 1 variable per line in liboctinterp.
Rik <rik@octave.org>
parents:
17787
diff
changeset
|
300 octave_idx_type psiz = pat.numel (); |
6a71e5030df5
Follow coding convention of defining and initializing only 1 variable per line in liboctinterp.
Rik <rik@octave.org>
parents:
17787
diff
changeset
|
301 octave_idx_type rsiz = rep.numel (); |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
302 |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
303 if (psiz != 0) |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
304 { |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
305 // Look up matches, without overlaps. |
10677
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
306 const Array<octave_idx_type> idx = qs_search (pat, str, table, overlaps); |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
307 octave_idx_type nidx = idx.numel (); |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
308 |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
309 if (nidx) |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
310 { |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
311 // Compute result size. |
10677
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
312 octave_idx_type retsiz; |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
313 if (overlaps) |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
314 { |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
315 retsiz = 0; |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
316 // OMG. Is this the "right answer" MW always looks for, or |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
317 // someone was just lazy? |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
318 octave_idx_type k = 0; |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
319 for (octave_idx_type i = 0; i < nidx; i++) |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
320 { |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
321 octave_idx_type j = idx(i); |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
322 if (j >= k) |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
323 retsiz += j - k; |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
324 retsiz += rsiz; |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
325 k = j + psiz; |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
326 } |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
327 |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
328 retsiz += siz - k; |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
329 } |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
330 else |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
331 retsiz = siz + nidx * (rsiz - psiz); |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
332 |
18763
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
333 if (retsiz == 0) |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
334 ret.clear (dim_vector (0, 0)); |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
335 else |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
336 { |
18763
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
337 ret.clear (dim_vector (1, retsiz)); |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
338 const char *src = str.data (); |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
339 const char *reps = rep.data (); |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
340 char *dest = ret.fortran_vec (); |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
341 |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
342 octave_idx_type k = 0; |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
343 for (octave_idx_type i = 0; i < nidx; i++) |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
344 { |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
345 octave_idx_type j = idx(i); |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
346 if (j >= k) |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
347 dest = std::copy (src + k, src + j, dest); |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
348 dest = std::copy (reps, reps + rsiz, dest); |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
349 k = j + psiz; |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
350 } |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
351 |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
352 std::copy (src + k, src + siz, dest); |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
353 } |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
354 } |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
355 } |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
356 |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
357 return ret; |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
358 } |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
359 |
15039
e753177cde93
maint: Move non-dynamically linked functions from DLD-FUNCTIONS/ to corefcn/ directory
Rik <rik@octave.org>
parents:
14501
diff
changeset
|
360 DEFUN (strrep, args, , |
17787
175b392e91fe
Use GNU style coding conventions for code in libinterp/
Rik <rik@octave.org>
parents:
17744
diff
changeset
|
361 "-*- texinfo -*-\n\ |
17398
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
362 @deftypefn {Built-in Function} {@var{newstr} =} strrep (@var{str}, @var{ptn}, @var{rep})\n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
363 @deftypefnx {Built-in Function} {@var{newstr} =} strrep (@var{cellstr}, @var{ptn}, @var{rep})\n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
364 @deftypefnx {Built-in Function} {@var{newstr} =} strrep (@dots{}, \"overlaps\", @var{val})\n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
365 Replace all occurrences of the pattern @var{ptn} in the string @var{str}\n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
366 with the string @var{rep} and return the result.\n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
367 \n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
368 The optional argument @qcode{\"overlaps\"} determines whether the pattern\n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
369 can match at every position in @var{str} (true), or only for unique\n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
370 occurrences of the complete pattern (false). The default is true.\n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
371 \n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
372 @var{s} may also be a cell array of strings, in which case the replacement is\n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
373 done for each element and a cell array is returned.\n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
374 \n\ |
67c6fc2b9f63
doc: Add info about "overlaps" option to strfind, strrep.
Rik <rik@octave.org>
parents:
15215
diff
changeset
|
375 Example:\n\ |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
376 \n\ |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
377 @example\n\ |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
378 @group\n\ |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
379 strrep (\"This is a test string\", \"is\", \"&%$\")\n\ |
14360
97883071e8e4
doc: Correct off-by-1 spacings in all .cc docstrings
Rik <octave@nomad.inbox5.com>
parents:
14138
diff
changeset
|
380 @result{} \"Th&%$ &%$ a test string\"\n\ |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
381 @end group\n\ |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
382 @end example\n\ |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
383 \n\ |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
384 @seealso{regexprep, strfind, findstr}\n\ |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
385 @end deftypefn") |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
386 { |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
387 octave_value retval; |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
388 int nargin = args.length (); |
10677
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
389 bool overlaps = true; |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
390 |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
391 if (nargin == 5 && args(3).is_string () && args(4).is_scalar_type ()) |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
392 { |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
393 std::string opt = args(3).string_value (); |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
394 if (opt == "overlaps") |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
395 { |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
396 overlaps = args(4).bool_value (); |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
397 nargin = 3; |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
398 } |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
399 else |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
400 { |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
401 error ("strrep: unknown option: %s", opt.c_str ()); |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
402 return retval; |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
403 } |
21defab4207c
make strrep not skipping overlaps by default, allow an option in both strfind/strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10521
diff
changeset
|
404 } |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
405 |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
406 if (nargin == 3) |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
407 { |
18100
6a71e5030df5
Follow coding convention of defining and initializing only 1 variable per line in liboctinterp.
Rik <rik@octave.org>
parents:
17787
diff
changeset
|
408 octave_value argstr = args(0); |
6a71e5030df5
Follow coding convention of defining and initializing only 1 variable per line in liboctinterp.
Rik <rik@octave.org>
parents:
17787
diff
changeset
|
409 octave_value argpat = args(1); |
6a71e5030df5
Follow coding convention of defining and initializing only 1 variable per line in liboctinterp.
Rik <rik@octave.org>
parents:
17787
diff
changeset
|
410 octave_value argrep = args(2); |
10086
76df75b10c80
allow more cells in strfind/strrep + supply a general mechanism
Jaroslav Hajek <highegg@gmail.com>
parents:
10074
diff
changeset
|
411 if (argpat.is_string () && argrep.is_string ()) |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
412 { |
10086
76df75b10c80
allow more cells in strfind/strrep + supply a general mechanism
Jaroslav Hajek <highegg@gmail.com>
parents:
10074
diff
changeset
|
413 const Array<char> pat = argpat.char_array_value (); |
76df75b10c80
allow more cells in strfind/strrep + supply a general mechanism
Jaroslav Hajek <highegg@gmail.com>
parents:
10074
diff
changeset
|
414 const Array<char> rep = argrep.char_array_value (); |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
415 |
12786
4e70c4e61c24
strfind.cc: Fix infinite loop when pattern contains 255 (Bug #33670).
Rik <octave@nomad.inbox5.com>
parents:
12483
diff
changeset
|
416 OCTAVE_LOCAL_BUFFER (octave_idx_type, table, TABSIZE); |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
417 qs_preprocess (pat, table); |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
418 |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
419 if (argstr.is_string ()) |
17787
175b392e91fe
Use GNU style coding conventions for code in libinterp/
Rik <rik@octave.org>
parents:
17744
diff
changeset
|
420 retval = qs_replace (argstr.char_array_value (), pat, rep, |
175b392e91fe
Use GNU style coding conventions for code in libinterp/
Rik <rik@octave.org>
parents:
17744
diff
changeset
|
421 table, overlaps); |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
422 else if (argstr.is_cell ()) |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
423 { |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
424 const Cell argsc = argstr.cell_value (); |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
425 Cell retc (argsc.dims ()); |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
426 octave_idx_type ns = argsc.numel (); |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
427 |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
428 for (octave_idx_type i = 0; i < ns; i++) |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
429 { |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
430 octave_value argse = argsc(i); |
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
431 if (argse.is_string ()) |
17787
175b392e91fe
Use GNU style coding conventions for code in libinterp/
Rik <rik@octave.org>
parents:
17744
diff
changeset
|
432 retc(i) = qs_replace (argse.char_array_value (), pat, rep, |
175b392e91fe
Use GNU style coding conventions for code in libinterp/
Rik <rik@octave.org>
parents:
17744
diff
changeset
|
433 table, overlaps); |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
434 else |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
435 { |
11553
01f703952eff
Improve docstrings for functions in DLD-FUNCTIONS directory.
Rik <octave@nomad.inbox5.com>
parents:
11523
diff
changeset
|
436 error ("strrep: each element of S must be a string"); |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
437 break; |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
438 } |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
439 } |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
440 |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
441 retval = retc; |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
442 } |
10074
5e2b4b7967cc
allow array of strings in strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10053
diff
changeset
|
443 else |
12483
7a5aacf65f81
Rewrite error strings in src/ to use variables named in documentation.
Rik <octave@nomad.inbox5.com>
parents:
11586
diff
changeset
|
444 error ("strrep: S must be a string or cell array of strings"); |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
445 } |
10086
76df75b10c80
allow more cells in strfind/strrep + supply a general mechanism
Jaroslav Hajek <highegg@gmail.com>
parents:
10074
diff
changeset
|
446 else if (argpat.is_cell () || argrep.is_cell ()) |
76df75b10c80
allow more cells in strfind/strrep + supply a general mechanism
Jaroslav Hajek <highegg@gmail.com>
parents:
10074
diff
changeset
|
447 retval = do_simple_cellfun (Fstrrep, "strrep", args); |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
448 else |
12483
7a5aacf65f81
Rewrite error strings in src/ to use variables named in documentation.
Rik <octave@nomad.inbox5.com>
parents:
11586
diff
changeset
|
449 error ("strrep: PTN and REP arguments must be strings or cell arrays of strings"); |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
450 } |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
451 else |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
452 print_usage (); |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
453 |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
454 return retval; |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
455 } |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
456 |
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
457 /* |
14501
60e5cf354d80
Update %!tests in DLD-FUNCTIONS/ directory with Octave coding conventions.
Rik <octave@nomad.inbox5.com>
parents:
14360
diff
changeset
|
458 %!assert (strrep ("This is a test string", "is", "&%$"), |
60e5cf354d80
Update %!tests in DLD-FUNCTIONS/ directory with Octave coding conventions.
Rik <octave@nomad.inbox5.com>
parents:
14360
diff
changeset
|
459 %! "Th&%$ &%$ a test string") |
60e5cf354d80
Update %!tests in DLD-FUNCTIONS/ directory with Octave coding conventions.
Rik <octave@nomad.inbox5.com>
parents:
14360
diff
changeset
|
460 %!assert (strrep ("abababc", "abab", "xyz"), "xyzxyzc") |
60e5cf354d80
Update %!tests in DLD-FUNCTIONS/ directory with Octave coding conventions.
Rik <octave@nomad.inbox5.com>
parents:
14360
diff
changeset
|
461 %!assert (strrep ("abababc", "abab", "xyz", "overlaps", false), "xyzabc") |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
462 |
18763
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
463 %!assert (size (strrep ("a", "a", "")), [0 0]) |
86818f2c4a92
Return 0x0 empty matrix from strrep for Matlab compatibility (bug #42341).
Rik <rik@octave.org>
parents:
18100
diff
changeset
|
464 |
14501
60e5cf354d80
Update %!tests in DLD-FUNCTIONS/ directory with Octave coding conventions.
Rik <octave@nomad.inbox5.com>
parents:
14360
diff
changeset
|
465 %!error strrep () |
60e5cf354d80
Update %!tests in DLD-FUNCTIONS/ directory with Octave coding conventions.
Rik <octave@nomad.inbox5.com>
parents:
14360
diff
changeset
|
466 %!error strrep ("foo", "bar", 3, 4) |
10053
830986c43dee
implement compiled strrep
Jaroslav Hajek <highegg@gmail.com>
parents:
10022
diff
changeset
|
467 */ |