improved readmes

README.md

@@ -2,18 +2,5 @@
 
 TTK-91 is a assembly-like language made for a Computer organization course in Helsinki University. 
 The course used this language to teach how programming languages work, how code is compiled into machine language from a basic assembly-like language.
-This language is now retiring and as a tribute, I thought I would write a proper CLI compiler for the language in Haskell.
 
-
-## Todo
-* Implement
-	* registers
-	* memory
-	* LOAD, STORE
-	* arithmetic operations
-	* comparison
-	* jumps
-	* Stack + operators
-	* System calls
-	* instruction translation to binary and hex
-	* compiler configuration (stdin etc.)
+This is my take on writing the compiler. The main implementation is in C. The other ones are previous attempts when I hadn't read up on compiler design.

c/README.md

@@ -0,0 +1,38 @@
+# TTK-91 Compiler
+This is my implementation of the TTK-91 Compiler in C. 
+
+## Overview
+First the soucre file is read and then fed into `scanner.c` produced by `scanner.flex` and 
+lexing is done. After lexing the tokens are parsed by `parser.c` given by `parser.bison`. 
+The parser will output an AST (Abstract Syntax Tree) which will be then transformed into an IR
+(Intermediate Representation).
+
+
+## Scanner
+The scanner reads through the whole file and splits it up into tokens according to the rules
+defined in `scanner.flex`. The rules are made up of mostly regular expressions. Few notable
+mentions are the `comment` token which matches everything after `;` until the EOF (End Of File)
+and `label` token which matches all words made up of letters in the Finnish alphabet + numbers + `_`
+starting with a letter.
+
+## Parser
+The parser matches the tokens given by `scanner.c` to rules given in `parser.bison`. The left side
+represents a single non-terminal token and the right side represents all decompositions the
+non-terminal can be. For example if we had rule `expr: value TOKEN_PLUS value;`, and 
+`value: TOKEN_NUM;` we could match all strings (assuming the scanner is normal) that have the from
+`x + y`. We then could say that `expr: value TOKEN_PLUS value { /* do C things */}` and
+feed the tokens to a function that creates a C representation of the same expression.
+In the C block `$n` means the n-th tokens of the rule and $$ means the left side of the rule.
+Example: `value: TOKEN_NUM { $$ = $1; }` means that that rule is equal to the first token of the 
+rule. Adding to that `expr: value TOKEN_PLUS value { $$ = $1 + $3; }` means that expr is equal to
+both of the numbers added together. Read more about this from:
+`Introduction to Compilers and Language Design, Prof. Douglas Thain, Second Edition`.
+
+## Types
+TODO
+
+### Expr
+`expr` represents basically a line. It can contain a normal instruction, compiler definition or
+a comment. 
+
+### Value