From 6778e3b87cc9a6f5d195a2c80e5b499e3d94558b Mon Sep 17 00:00:00 2001
From: dec05eba <dec05eba@protonmail.com>
Date: Sun, 27 Jan 2019 02:09:50 +0100
Subject: Add binds to emoji parsing, refactor

---
 src/IncomingMessage.cpp | 160 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 160 insertions(+)
 create mode 100644 src/IncomingMessage.cpp

(limited to 'src/IncomingMessage.cpp')

diff --git a/src/IncomingMessage.cpp b/src/IncomingMessage.cpp
new file mode 100644
index 0000000..e003a22
--- /dev/null
+++ b/src/IncomingMessage.cpp
@@ -0,0 +1,160 @@
+#include "../include/dchat/IncomingMessage.hpp"
+#include <assert.h>
+#include <string.h>
+
+namespace dchat
+{
+    enum class Token
+    {
+        NONE,
+        END_OF_FILE,
+
+        TEXT,
+        TYPE
+    };
+
+    struct Tokenizer
+    {
+        Tokenizer(const char *_text, usize _size)
+        {
+            assert(_text);
+            text = _text;
+            size = _size;
+            index = 0;
+            identifierRange = { 0, 0 };
+            typeRange = { 0, 0 };
+            typeDataRange = { 0, 0 };
+        }
+
+        enum class EnclosedType
+        {
+            TEXT,
+            DATA
+        };
+        
+        EnclosedType parseEnclosedData(char endSymbol)
+        {
+            bool foundEndOfType = false;
+
+            while(index < size)
+            {
+                char c = getChar();
+                ++index;
+                if(c == endSymbol)
+                {
+                    foundEndOfType = true;
+                    break;
+                }
+            }
+
+            if(!foundEndOfType)
+                return EnclosedType::TEXT;
+
+            return EnclosedType::DATA;
+        }
+
+        Token next()
+        {
+            if(index >= size)
+                return Token::END_OF_FILE;
+
+            char c = getChar();
+            if(c == '[')
+            {
+                usize start = index;
+                ++index;
+                if(parseEnclosedData(']') == EnclosedType::TEXT)
+                {
+                    identifierRange.start = start;
+                    identifierRange.end = index;
+                    return Token::TEXT;
+                }
+
+                if(index == size || getChar() != '(')
+                {
+                    identifierRange.start = start;
+                    identifierRange.end = index;
+                    return Token::TEXT;
+                }
+
+                typeRange.start = start + 1;
+                typeRange.end = index - 1;
+                typeDataRange.start = index + 1;
+
+                ++index;
+                switch(parseEnclosedData(')'))
+                {
+                    case EnclosedType::TEXT:
+                    {
+                        identifierRange.start = start;
+                        identifierRange.end = index;
+                        return Token::TEXT;
+                    }
+                    case EnclosedType::DATA:
+                    {
+                        typeDataRange.end = index - 1;
+                        return Token::TYPE;
+                    }
+                }
+            }
+            else
+            {
+                identifierRange.start = index;
+                ++index;
+                while(index < size)
+                {
+                    c = getChar();
+                    if(c == '[')
+                        break;
+                    ++index;
+                }
+                identifierRange.end = index;
+                return Token::TEXT;
+            }
+
+            assert(false);
+            return Token::NONE;
+        }
+
+        char getChar() const
+        {
+            assert(index < size);
+            return text[index];
+        }
+
+        const char *text;
+        usize size;
+        usize index;
+
+        Range identifierRange;
+        Range typeRange;
+        Range typeDataRange;
+    };
+
+    void parseIncomingMessage(const char *text, usize size, std::function<void(IncomingMessagePart)> callbackFunc)
+    {
+        Tokenizer tokenizer(text, size);
+        Token token = tokenizer.next();
+        while(token != Token::END_OF_FILE)
+        {
+            if(token == Token::TEXT)
+            {
+                callbackFunc(IncomingMessagePart { IncomingMessagePart::Type::TEXT, tokenizer.identifierRange });
+                token = tokenizer.next();
+            }
+            else if(token == Token::TYPE)
+            {
+                if(tokenizer.typeRange.length() == 5 && memcmp(text + tokenizer.typeRange.start, "emoji", 5) == 0)
+                {
+                    callbackFunc(IncomingMessagePart { IncomingMessagePart::Type::EMOJI, tokenizer.typeDataRange });
+                }
+                else
+                {
+                    Range typeToTextRange = { tokenizer.typeRange.start - 1, tokenizer.typeDataRange.end + 1 };
+                    callbackFunc(IncomingMessagePart{ IncomingMessagePart::Type::TEXT, typeToTextRange });
+                }
+                token = tokenizer.next();
+            }
+        }
+    }
+}
\ No newline at end of file
-- 
cgit v1.2.3