irc::tokenstream is a token parser which using std::string and std::vector builds a list of tokens using irc parsing rules.

e.g.:
:arse PRIVMSG #chan :foo bar baz
becomes
a[0] = ":arse", a[1] = "PRIVMSG", a[2] = "#chan", a[3] = "foo bar baz".

*** SOMEONE *** needs to optimize this or at least verify its neatness (without making it total craq and unreadable). Feel free to mess with my code.
Any optimizations you make, run them against the COMMENTED OUT TEST SUITE at the top of main() in src/inspircd.cpp and ensure ALL output is the same
with no crashes. (note: if you comment out and build with the test suite, all inspircd will do is output test data and exit!)


git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@4396 e03df62e-2008-0410-955e-edbf42e46eb7

3 files changed, 117 insertions, 0 deletions

diff --git a/include/hashcomp.h b/include/hashcomp.h
index 36e5dd0c3..ad6a2133e 100644
--- a/include/hashcomp.h
+++ b/include/hashcomp.h
@@ -79,6 +79,19 @@ namespace irc
 	        bool operator()(const in_addr &s1, const in_addr &s2) const;
 	};
 
+	class tokenstream
+	{
+	 private:
+		std::string tokenbuffer;
+		std::vector<std::string> tokens;
+	 public:
+		tokenstream(std::string &source);
+		~tokenstream();
+
+		unsigned int GetNumTokens();
+		const std::string& GetToken(unsigned int index);
+	};
+
 
 	/** The irc_char_traits class is used for RFC-style comparison of strings.
 	 * This class is used to implement irc::string, a case-insensitive, RFC-
diff --git a/src/hashcomp.cpp b/src/hashcomp.cpp
index 3271e599c..08fbc8e2f 100644
--- a/src/hashcomp.cpp
+++ b/src/hashcomp.cpp
@@ -177,3 +177,47 @@ std::istream& operator>>(std::istream &is, irc::string &str)
 	str = tmp.c_str();
 	return is;
 }
+
+irc::tokenstream::tokenstream(std::string &source)
+{
+	std::string::iterator last_starting_position = source.begin();
+	bool last_pushed = false;
+
+	for (std::string::iterator n = source.begin(); n != source.end(); n++)
+	{
+		if ((last_pushed) && (*n == ':'))
+		{
+			/* If we find a token thats not the first and starts with :,
+			 * this is the last token on the line
+			 */
+			tokens.push_back(std::string(n+1, source.end()));
+			break;
+		}
+
+		last_pushed = false;
+
+		if ((*n == ' ') || (n+1 == source.end()))
+		{
+			/* If we find a space, or end of string, this is the end of a token.
+			 */
+			tokens.push_back(std::string(last_starting_position, n+1 == source.end() ? n+1  : n));
+			last_starting_position = n+1;
+			last_pushed = true;
+		}
+	}
+}
+
+irc::tokenstream::~tokenstream()
+{
+}
+
+unsigned int irc::tokenstream::GetNumTokens()
+{
+	return tokens.size();
+}
+
+const std::string& irc::tokenstream::GetToken(unsigned int index)
+{
+	return tokens[index];
+}
+
diff --git a/src/inspircd.cpp b/src/inspircd.cpp
index 23a232c46..9635f707a 100644
--- a/src/inspircd.cpp
+++ b/src/inspircd.cpp
@@ -944,6 +944,66 @@ int InspIRCd::Run()
 
 int main(int argc, char** argv)
 {
+	/* TEST SUITE FOR TOKENSTREAM
+	 *
+	 * Expected output:
+	 * 
+	 * String: 'PRIVMSG #test FOO BAR'
+	 * NumItems = 4
+	 * Token 0 = 'PRIVMSG'
+	 * Token 1 = '#test'
+	 * Token 2 = 'FOO'
+	 * Token 3 = 'BAR'
+	 * String: 'PRIVMSG #test :FOO BAR BAZ'
+	 * NumItems = 3
+	 * Token 0 = 'PRIVMSG'
+	 * Token 1 = '#test'
+	 * Token 2 = 'FOO BAR BAZ'
+	 * String: ':PRIVMSG #test :FOO BAR BAZ'
+	 * NumItems = 3
+	 * Token 0 = ':PRIVMSG'
+	 * String: 'AAAAAAA'
+	 * NumItems = 1
+	 * Token 0 = 'AAAAAAA'
+	 * String: ''
+	 * NumItems = 0
+	 *
+	std::string a = "PRIVMSG #test FOO BAR";
+	printf("String: '%s'\n",a.c_str());
+	irc::tokenstream test(a);
+	printf("NumItems = %d\n",test.GetNumTokens());
+	printf("Token 0 = '%s'\n",test.GetToken(0).c_str());
+	printf("Token 1 = '%s'\n",test.GetToken(1).c_str());
+	printf("Token 2 = '%s'\n",test.GetToken(2).c_str());
+	printf("Token 3 = '%s'\n",test.GetToken(3).c_str());
+
+	std::string b = "PRIVMSG #test :FOO BAR BAZ";
+	printf("String: '%s'\n",b.c_str());
+	irc::tokenstream test2(b);
+	printf("NumItems = %d\n",test2.GetNumTokens());
+	printf("Token 0 = '%s'\n",test2.GetToken(0).c_str());
+	printf("Token 1 = '%s'\n",test2.GetToken(1).c_str());
+	printf("Token 2 = '%s'\n",test2.GetToken(2).c_str());
+
+	std::string c = ":PRIVMSG #test :FOO BAR BAZ";
+	printf("String: '%s'\n",c.c_str());
+	irc::tokenstream test3(c);
+	printf("NumItems = %d\n",test3.GetNumTokens());
+	printf("Token 0 = '%s'\n",test3.GetToken(0).c_str());
+
+	c = "AAAAAAA";
+	printf("String: '%s'\n",c.c_str());
+	irc::tokenstream test4(c);
+	printf("NumItems = %d\n",test4.GetNumTokens());
+	printf("Token 0 = '%s'\n",test4.GetToken(0).c_str());
+
+	c = "";
+	printf("String: '%s'\n",c.c_str());
+	irc::tokenstream test5(c);
+	printf("NumItems = %d\n",test5.GetNumTokens());
+
+	exit(0); */
+	
 	try
 	{
 		ServerInstance = new InspIRCd(argc, argv);