From c95e1281fc5ecfe3f65a59fa426f16a54b2a6a50 Mon Sep 17 00:00:00 2001 From: Nathan Binkert Date: Wed, 22 Jun 2005 09:52:02 -0400 Subject: [PATCH] fix tokenize base/str.cc: Fix tokenize so that it doesn't behave incorrectly when there are empty strings. test/tokentest.cc: Clean up the test function so it's easier to see what's going on --HG-- extra : convert_revision : c7a3db7bc516d3575b1cc4ab7afbd0f1fbe1ec6f --- base/str.cc | 23 +++++++++-------- test/tokentest.cc | 64 ++++++++++++++++++++++++++++------------------- 2 files changed, 50 insertions(+), 37 deletions(-) diff --git a/base/str.cc b/base/str.cc index 3a11bb17d..15f44dad2 100644 --- a/base/str.cc +++ b/base/str.cc @@ -26,11 +26,10 @@ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ -#include - -#include #include +#include +#include #include #include @@ -75,15 +74,17 @@ tokenize(vector& v, const string &s, char token, bool ignore) string::size_type first = 0; string::size_type last = s.find_first_of(token); - if (ignore) { - if (last == first) { - while (last == first) - last = s.find_first_of(token, ++first); + if (s.empty()) + return; - if (last == string::npos) { - v.push_back(s); - return; - } + if (ignore && last == first) { + while (last == first) + last = s.find_first_of(token, ++first); + + if (last == string::npos) { + if (first != s.size()) + v.push_back(s.substr(first)); + return; } } diff --git a/test/tokentest.cc b/test/tokentest.cc index cd2182141..7f27d58fe 100644 --- a/test/tokentest.cc +++ b/test/tokentest.cc @@ -26,7 +26,7 @@ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ -#include +#include #include #include @@ -35,35 +35,47 @@ int main(int argc, char *argv[]) { - if (argc != 3) { - cout << "Usage: " << argv[0] << " \n"; - exit(1); - } + using namespace std; - int i; - string test = argv[1]; - vector tokens1; - vector tokens2; - char token = argv[2][0]; + if (argc != 3) { + cout << "Usage: " << argv[0] << " \n"; + exit(1); + } - cout << "string = \"" << test << "\", token = \'" << token << "\'\n"; - cout << "testing without ignore\n"; - tokenize(tokens1, test, token, false); + int i; + string test = argv[1]; + vector tokens1; + vector tokens2; + char token = argv[2][0]; - if (tokens1.size()) { - for (i = 0; i < tokens1.size() - 1; i++) - cout << tokens1[i] << "(" << tokens1[i].size() << "), "; - cout << tokens1[i] << "(" << tokens1[i].size() << ")\n"; - } + cout << "string = \"" << test << "\", token = \'" << token << "\'\n"; + cout << "testing without ignore\n"; + tokenize(tokens1, test, token, false); - cout << "testing with ignore\n"; - tokenize(tokens2, test, token, true); + if (tokens1.size()) { + int size = tokens1.size(); + cout << "size = " << size << "\n"; + for (i = 0; i < size; i++) { + cout << "'" << tokens1[i] << "' (" << tokens1[i].size() + << ")" << ((i == size - 1) ? "\n" : ", "); + } + } else { + cout << "no tokens" << endl; + } - if (tokens2.size()) { - for (i = 0; i < tokens2.size() - 1; i++) - cout << tokens2[i] << "(" << tokens2[i].size() << "), "; - cout << tokens2[i] << "(" << tokens2[i].size() << ")\n"; - } + cout << "testing with ignore\n"; + tokenize(tokens2, test, token, true); - return 0; + if (tokens2.size()) { + int size = tokens2.size(); + cout << "size = " << size << "\n"; + for (i = 0; i < size; i++) { + cout << "'" << tokens2[i] << "' (" << tokens2[i].size() + << ")" << ((i == size - 1) ? "\n" : ", "); + } + } else { + cout << "no tokens" << endl; + } + + return 0; } -- 2.30.2