commit 19d6f35133274e5eeb7520f96449d53ecf3bcdd5
Author: Eiji Miyamoto <eiji.miyamoto@student.manchester.ac.uk>
Date:   Wed Jul 12 12:29:19 2023 +0100

    improve cpp file as it did not tokenize non-blancket text and text after blancket

diff --git a/buffer_mecab.cpp b/buffer_mecab.cpp
index 42b9672..45c1491 100644
--- a/buffer_mecab.cpp
+++ b/buffer_mecab.cpp
@@ -9,26 +9,33 @@ void process_text(std::istream& sin, std::ostream& sout) {
     
     MeCab::Tagger* tagger = MeCab::createTagger("-Owakati");
     std::stringstream buffer;
-    //std::ostringstream buffer;
     std::string tokenized;
     bool in_bracket = false;
     for (char i : text) {
         buffer << i;
+        if (i == text.back()){
+            std::string token = tagger->parse(buffer.str().c_str());
+            token.erase(token.find_last_not_of(" \n\r\t") + 1);
+            tokenized += token;
+        }
         if (i == '[') {
+            std::string temp = buffer.str();
+            temp.pop_back();
+            buffer.str(temp);
             std::string parsed = tagger->parse(buffer.str().c_str());
             parsed.erase(parsed.find_last_not_of(" \n\r\t") + 1);
             tokenized += parsed;
-            // buffer = tagger->parse(buffer.str().c_str());
-            // tokenized += buffer;
             buffer.str("");
             buffer.clear();
             in_bracket = true;
         }
-        else if (in_bracket) {
+        if (in_bracket) {
             tokenized += i;
         }
-        else if (i == ']') {
+        if (i == ']') {
             in_bracket = false;
+            buffer.str("");
+            buffer.clear();
         }
     }