1 ## Copyright (C) 1996-2012 Kurt Hornik
3 ## This file is part of Octave.
5 ## Octave is free software; you can redistribute it and/or modify it
6 ## under the terms of the GNU General Public License as published by
7 ## the Free Software Foundation; either version 3 of the License, or (at
8 ## your option) any later version.
10 ## Octave is distributed in the hope that it will be useful, but
11 ## WITHOUT ANY WARRANTY; without even the implied warranty of
12 ## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
13 ## General Public License for more details.
15 ## You should have received a copy of the GNU General Public License
16 ## along with Octave; see the file COPYING. If not, see
17 ## <http://www.gnu.org/licenses/>.
20 ## @deftypefn {Function File} {} findstr (@var{s}, @var{t})
21 ## @deftypefnx {Function File} {} findstr (@var{s}, @var{t}, @var{overlap})
22 ## Return the vector of all positions in the longer of the two strings
23 ## @var{s} and @var{t} where an occurrence of the shorter of the two starts.
24 ## If the optional argument @var{overlap} is true, the returned vector
25 ## can include overlapping positions (this is the default). For example:
29 ## findstr ("ababab", "a")
30 ## @result{} [1, 3, 5];
31 ## findstr ("abababa", "aba", 0)
36 ## @strong{Caution:} @code{findstr} is scheduled for deprecation. Use
37 ## @code{strfind} in all new code.
38 ## @seealso{strfind, strmatch, strcmp, strncmp, strcmpi, strncmpi, find}
41 ## Note that this implementation swaps the strings if second one is longer
42 ## than the first, so try to put the longer one first.
44 ## Author: Kurt Hornik <Kurt.Hornik@wu-wien.ac.at>
47 function v = findstr (s, t, overlap = true)
49 if (nargin < 2 || nargin > 3)
53 if (all (size (s) > 1) || all (size (t) > 1))
54 error ("findstr: arguments must have only one non-singleton dimension");
57 ## Make S be the longer string.
58 if (length (s) < length (t))
66 ## zero length target: return empty set
70 ## length one target: simple find
74 ## length two target: find first at i and second at i+1
75 v = find (s(1:l_s-1) == t(1) & s(2:l_s) == t(2));
78 ## length three or more: match the first three by find then go through
79 ## the much smaller list to determine which of them are real matches
80 limit = l_s - l_t + 1;
81 v = find (s(1:limit) == t(1)
82 & s(2:limit+1) == t(2)
83 & s (3:limit+2) == t(3));
86 ## Need to search the index vector if our find was too short
87 ## (target length > 3), or if we don't allow overlaps. Note though
88 ## that there cannot be any overlaps if the first character in the
89 ## target is different from the remaining characters in the target,
90 ## so a single character, two different characters, or first character
91 ## different from the second two don't need to be searched.
92 if (l_t >= 3 || (! overlap && l_t > 1 && any (t(1) == t(2:l_t))))
93 ## force strings to be both row vectors or both column vectors
94 if (all (size (s) != size (t)))
98 ## determine which ones to keep
99 keep = zeros (size (v));
102 for idx = 1:length (v)
103 keep(idx) = all (s(v(idx) + ind) == t);
106 ## First possible position for next non-overlapping match.
108 for idx = 1:length (v)
109 if (v(idx) >= next && s(v(idx) + ind) == t)
111 ## Skip to the next possible match position.
127 ## Always return a row vector, because that's what the old one did.
135 %!assert (findstr ("abababa", "a"), [1, 3, 5, 7])
136 %!assert (findstr ("abababa", "aba"), [1, 3, 5]);
137 %!assert (findstr ("aba", "abababa", 0), [1, 5]);
139 %% Test input validation
141 %!error findstr ("foo", "bar", 3, 4);
142 %!error findstr (["AB" ; "CD"], "C");