Merge pull request #50 from rzaharia/master

Updated all readme files to contain links to the next step

00_Introduction/Readme.md

@@ -156,4 +156,4 @@ Finally, clone a copy of this Github repository.
 
 In the next part of our compiler writing journey, we will start with
 the code to scan our input file and find the *tokens* that are the
-lexical elements of our language.
+lexical elements of our language. [Next step](../01_Scanner/Readme.md)

01_Scanner/Readme.md

@@ -326,4 +326,4 @@ struct token {
 
 In the next part of our compiler writing journey, we will build
 a recursive descent parser to interpret the grammar of our input
-files, and calculate & print out the final value for each file.
+files, and calculate & print out the final value for each file. [Next step](../02_Parser/Readme.md)

02_Parser/Readme.md

@@ -537,4 +537,4 @@ calculate the correct value of the expressions.
 
 In the next part of our compiler writing journey, we will modify
 the parser so that it also does the semantic analysis of the
-expressions to get the right maths results.
+expressions to get the right maths results. [Next step](../03_Precedence/Readme.md)

03_Precedence/Readme.md

@@ -434,4 +434,4 @@ In the next part of our compiler writing journey, we will replace the
 interpreter. In its place, we will write a translator that generates
 x86-64 assembly code for each AST node that has a maths operator.
 We will also generate some assembly preamble and postamble to
-support the assembly code that the generator outputs.
+support the assembly code that the generator outputs. [Next step](../04_Assembly/Readme.md)

04_Assembly/Readme.md

@@ -371,4 +371,4 @@ two.
 
 In the next part of our compiler writing journey, we will add some
 statements to our language, so that it starts to resemble a proper
-computer language.
+computer language. [Next step](../05_Statements/Readme.md)

05_Statements/Readme.md

@@ -321,4 +321,4 @@ identifiers, and to more cleanly separate the generic code generator and
 the CPU-specific generator.
 
 In the next part of our compiler writing journey, we will add variables
-to the language. This will require a significant amount of work.
+to the language. This will require a significant amount of work. [Next step](../06_Variables/Readme.md)

06_Variables/Readme.md

@@ -512,4 +512,4 @@ errors (variable use without a declaration).
 
 In the next part of our compiler writing journey, we will
 add the six comparison operators to our language. That will
-allow us to start on the control structures in the part after that.
+allow us to start on the control structures in the part after that. [Next step](../07_Comparisons/Readme.md)

07_Comparisons/Readme.md

@@ -293,4 +293,4 @@ That was a nice and easy addition to the compiler. The next part of the
 journey will be much more complicated.
 
 In the next part of our compiler writing journey, we will add IF statements
-to the compiler and make use of the comparison operators that we just added.
+to the compiler and make use of the comparison operators that we just added. [Next step](../08_If_Statements/Readme.md)

08_If_Statements/Readme.md

@@ -641,4 +641,4 @@ I know that Nils Holm chose a different approach when he was implementing SubC,
 you should look at his code just to see this different solution to the same problem.
 
 In the next part of our compiler writing journey, we will
-add another control structure: the WHILE loop.
+add another control structure: the WHILE loop. [Next step](../09_While_Loops/Readme.md)

09_While_Loops/Readme.md

@@ -225,4 +225,4 @@ So we can stop now, our job is done! No, of course not. We are still
 working towards getting the compiler to compile itself.
 
 In the next part of our compiler writing journey, we will add FOR loops
-to the language.
+to the language. [Next step](../10_For_Loops/Readme.md)

10_For_Loops/Readme.md

@@ -313,4 +313,4 @@ tackle next? There are so many things we could look at:
 
 I've decided to look at functions. So,
 in the next part of our compiler writing journey, we will begin
-the first of several stages to add functions to our language.
+the first of several stages to add functions to our language. [Next step](../11_Functions_pt1/Readme.md)

11_Functions_pt1/Readme.md

@@ -202,4 +202,4 @@ We've made a good start at adding functions to our language. For now,
 it's a pretty simplistic function declaration only.
 
 In the next part of our compiler writing journey, we will begin
-the process to add types to our compiler.
+the process to add types to our compiler. [Next step](../12_Types_pt1/Readme.md)

12_Types_pt1/Readme.md

@@ -538,4 +538,4 @@ Still not the most elegant assembly code, but it does work. Also,
 
 In the next part of our compiler writing journey, we will
 add function calls with one argument, and returning a value
-from  a function.
+from  a function. [Next step](../13_Functions_pt2/Readme.md)

13_Functions_pt2/Readme.md

@@ -619,4 +619,4 @@ plus a new data type. As I expected, it wasn't trivial but I think the
 changes are mostly sensible.
 
 In the next part of our compiler writing journey, we will port our
-compiler to a new hardware platform, the ARM CPU on a Raspberry Pi.
+compiler to a new hardware platform, the ARM CPU on a Raspberry Pi. [Next step](../14_ARM_Platform/Readme.md)

14_ARM_Platform/Readme.md

@@ -285,4 +285,4 @@ As we go forward, I'll try to keep both `cg.c` and `cg_arm.c`
 functionally in sync.
 
 In the next part of our compiler writing journey, we will add the `char`
-pointer to the language, as well as the '*' and '&' unary operators.
+pointer to the language, as well as the '*' and '&' unary operators. [Next step](../15_Pointers_pt1/Readme.md)

15_Pointers_pt1/Readme.md

@@ -378,4 +378,4 @@ one to the address of `x`, which is incorrect. I'll
 have to work out how to fix this.
 
 In the next part of our compiler writing journey, we will
-try to fix this problem.
+try to fix this problem. [Next step](../16_Global_Vars/Readme.md)

16_Global_Vars/Readme.md

@@ -202,4 +202,4 @@ cc -o out out.s lib/printint.c
 ## Conclusion and What's Next
 
 In the next part of our compiler writing journey,
-I promise to tackle the issue of adding offsets to pointers.
+I promise to tackle the issue of adding offsets to pointers. [Next step](../17_Scaling_Offsets/Readme.md)

17_Scaling_Offsets/Readme.md

@@ -364,4 +364,4 @@ the operation. I eyeballed the output and it seems to be what I want.
 Only time will tell.
 
 In the next part of our compiler writing journey,
-I don't know what I will do!
+I don't know what I will do! [Next step](../18_Lvalues_Revisited/Readme.md)

18_Lvalues_Revisited/Readme.md

@@ -491,4 +491,4 @@ I'd like to tackle arrays. This will be a combination of dereferencing,
 lvalues and rvalues, and scaling the array indices by the size of the
 array's elements. We have all the semantic components in place, but we'll
 need to add tokens, parsing and the actual index functionality. It should
-be an interesting topic like this one was.
+be an interesting topic like this one was. [Next step](../19_Arrays_pt1/Readme.md)

19_Arrays_pt1/Readme.md

@@ -464,4 +464,4 @@ scale, add to the base address, and set as lvalue/rvalue. Once this
 was right, the existing code generator produces the right assembly output.
 
 In the next part of our compiler writing journey, we'll add character
-and string literals to our language and find a way to print them out.
+and string literals to our language and find a way to print them out. [Next step](../20_Char_Str_Literals/Readme.md)

20_Char_Str_Literals/Readme.md

@@ -304,4 +304,4 @@ characters inside them. But there was some work done on the code
 generator, too.
 
 In the next part of our compiler writing journey, we'll add some
-more binary operators to the language that the compiler recognises.
+more binary operators to the language that the compiler recognises. [Next step](../21_More_Operators/Readme.md)

21_More_Operators/Readme.md

@@ -589,4 +589,4 @@ revisit this later.
 In the next part of our compiler writing journey, I'd like to
 tackle local variables. Once we can get these to work, we can extend
 them to also include function parameters and arguments. This will
-take two or more steps.
+take two or more steps. [Next step](../22_Design_Locals/Readme.md)

22_Design_Locals/Readme.md

@@ -266,4 +266,4 @@ I think that's about all I want to do in terms of design before I start on
 the next parts of our compiler writing journey. I'll tackle local variables
 by themselves in the next part, and try to add in function calls and
 parameters in the following part. But it might take three or more steps
-to get all of the new proposed features implemented. We'll see.
+to get all of the new proposed features implemented. We'll see. [Next step](../23_Local_Variables/Readme.md)

23_Local_Variables/Readme.md

@@ -342,4 +342,4 @@ doing some thinking about the design of a solution, it turned out to be
 easier than I expected. Somehow I suspect the next step will be the tricky one.
 
 In the next part of our compiler writing journey, I will attempt to
-add function arguments and parameters to our compiler. Wish me luck!
+add function arguments and parameters to our compiler. Wish me luck! [Next step](../24_Function_Params/Readme.md)

24_Function_Params/Readme.md

@@ -494,4 +494,4 @@ code.
 
 In the next part of our compiler writing journey, I'll modify the compiler
 to make function calls with an arbirary number of arguments. Then I can
-move `input27a.c` and `input27b.c` into the `tests/` directory.
+move `input27a.c` and `input27b.c` into the `tests/` directory. [Next step](../25_Function_Arguments/Readme.md)

25_Function_Arguments/Readme.md

@@ -361,4 +361,4 @@ add the ability to write function prototypes. This will allow us to
 link in outside functions. I'm thinking of those original Unix functions
 and system calls which are `int` and `char *` based such as `open()`,
 `read()`, `write()`, `strcpy()` etc. It will be nice to compile
-some useful programs with our compiler.
+some useful programs with our compiler. [Next step](../26_Prototypes/Readme.md)

26_Prototypes/Readme.md

@@ -373,4 +373,4 @@ can compile! That's progress.
 ## Conclusion and What's Next
 
 In the next part of our compiler writing journey, I'll follow up on
-a comment above and improve the testing of our compiler's functionality.
+a comment above and improve the testing of our compiler's functionality. [Next step](../27_Testing_Errors/Readme.md)

27_Testing_Errors/Readme.md

@@ -289,4 +289,4 @@ try to:
  + add support for an external pre-processor
  + allow the compiler to compile multiple files named on the command line
  + add the `-o`, `-c` and `-S` flags to the compiler to make it feel
-   more like a "normal" C compiler
+more like a "normal" C compiler [Next step](../28_Runtime_Flags/Readme.md)

28_Runtime_Flags/Readme.md

@@ -358,4 +358,4 @@ int main()
 In the next part of our compiler writing journey, we will
 look at adding support for structs to our compiler. I think we
 might have to do another design step first before we get to
-implementing the changes.
+implementing the changes. [Next step](../29_Refactoring/Readme.md)

29_Refactoring/Readme.md

@@ -259,4 +259,4 @@ probably happen in the part where we implement enumerated values.
 
 In the next part of our compiler writing journey, I'll get back to
 what I wanted to cover in this part: the design side of implementing
-structs, unions and enums in our compiler.
+structs, unions and enums in our compiler. [Next step](../30_Design_Composites/Readme.md)

30_Design_Composites/Readme.md

@@ -600,4 +600,4 @@ In the next part of our compiler writing journey, I'll probably implement
 the declaration of struct types, but not actually write the code for them to
 be used. I'll do that in the following part. With both of these done, I'll
 hopefully be able to implement unions in a third part. Then, enums in the
-fourth part. We'll see!
+fourth part. We'll see! [Next step](../31_Struct_Declarations/Readme.md)

31_Struct_Declarations/Readme.md

@@ -519,4 +519,4 @@ will be complicated.
 
 In the next part of our compiler writing journey, I will try to 
 add the code to parse the '.' token so that we can access members
-in a struct variable.
+in a struct variable. [Next step](../32_Struct_Access_pt1/Readme.md)

32_Struct_Access_pt1/Readme.md

@@ -244,4 +244,4 @@ features will have to be made more general. So there will be a "mop up" stage
 in this journey.
 
 Now that we have structs mostly working, in the next part of our compiler writing
-journey, I will try to add unions.
+journey, I will try to add unions. [Next step](../33_Unions/Readme.md)

33_Unions/Readme.md

@@ -163,4 +163,4 @@ that pointer access into a union also works.
 ## Conclusion and What's Next
 
 This was another nice and easy part of our compiler writing journey.
-In the next part of our compiler writing journey, we will add enums.
+In the next part of our compiler writing journey, we will add enums. [Next step](../34_Enums_and_Typedefs/Readme.md)

34_Enums_and_Typedefs/Readme.md

@@ -408,4 +408,4 @@ bring in the C pre-processor. Now that we have structs, unions, enums
 and typedefs, we should be able to write a bunch of *header files*
 with definitions of some of the common Unix/Linux library functions.
 Then we will be able to include them in our source files and write
-some really useful programs.
+some really useful programs. [Next step](../35_Preprocessor/Readme.md)

35_Preprocessor/Readme.md

@@ -521,4 +521,4 @@ lines of code we'll need to make the compiler self-compiling, but I'm going
 to hazard a guess of between 7,000 to 9,000 lines. We'll see!
 
 In the next part of our compiler writing journey, we will add the `break`
-and `continue` keywords to our loop constructs.
+and `continue` keywords to our loop constructs. [Next step](../36_Break_Continue/Readme.md)

36_Break_Continue/Readme.md

@@ -362,4 +362,4 @@ in the process of implementing them.
 Now that we have the `break` keyword in the language, I will attempt to
 add `switch` statements in the next part of our compiler writing journey.
 This is going to require the addition of switch jump tables, and I know
-this is going to be complicated. So get ready for an interesting next step.
+this is going to be complicated. So get ready for an interesting next step. [Next step](../37_Switch/Readme.md)

37_Switch/Readme.md

@@ -667,4 +667,4 @@ I'm starting to get annoyed with the compulsory '{' ... '}'
 around all of our compound statements.
 So, in the next part of our compiler writing journey, 
 I will bite the bullet and attempt to solve the "dangling else"
-problem.
+problem. [Next step](../38_Dangling_Else/Readme.md)

38_Dangling_Else/Readme.md

@@ -401,4 +401,4 @@ If you do, I hope that you found all the parts of the journey so far useful.
 
 In the next part of our compiler writing journey, I will pick something that
 currently doesn't work but we need to work to self-compile our compiler, and
-fix it.
+fix it. [Next step](../39_Var_Initialisation_pt1/Readme.md)

39_Var_Initialisation_pt1/Readme.md

@@ -212,4 +212,4 @@ put comments into the code where I will have to write the assignment
 functionality to go with declarations.
 
 In the next part of our compiler writing journey, we will try to add
-global variable declarations with assignments to our compiler.
+global variable declarations with assignments to our compiler. [Next step](../40_Var_Initialisation_pt2/Readme.md)

40_Var_Initialisation_pt2/Readme.md

@@ -563,4 +563,4 @@ they say. I'm happy, though, because the changes to the symbol table
 make much more sense than what I had before.
 
 In the next part of our compiler writing journey, we will try to
-add local variable initialisation to the compiler.
+add local variable initialisation to the compiler. [Next step](../41_Local_Var_Init/Readme.md)

41_Local_Var_Init/Readme.md

@@ -151,4 +151,4 @@ Right now I'm guessing about 60 parts and an 75% chance of completing by year's
 But we still have a bunch of small, but possibly difficult, features to add to the
 compiler.
 
-In the next part of our compiler writing journey, I will add cast parsing to the compiler.
+In the next part of our compiler writing journey, I will add cast parsing to the compiler. [Next step](../42_Casting/Readme.md)

42_Casting/Readme.md

@@ -383,4 +383,4 @@ covered most bases here but not all of them, so expect to see
 some `void *` edge cases that I haven't spotted yet.
 
 In the next part of our compiler writing journey, we'll add some missing
-operators.
+operators. [Next step](../43_More_Operators/Readme.md)

43_More_Operators/Readme.md

@@ -361,4 +361,4 @@ right-associativity. After that, we could reuse some of the code generation
 code in `genAST()` to make our lives a bit easier.
 
 In the next part of our compiler writing journey, it looks like I'll be adding
-constant folding to the compiler.
+constant folding to the compiler. [Next step](../44_Fold_Optimisation/Readme.md)

44_Fold_Optimisation/Readme.md

@@ -298,4 +298,4 @@ it's good to see one type of optimisation now.
 
 In the next part of our compiler writing journey, we will replace our
 current global declaration parser with code that evaluates expressions
-using `binexpr()` and this new constant folding code.
+using `binexpr()` and this new constant folding code. [Next step](../45_Globals_Again/Readme.md)

45_Globals_Again/Readme.md

@@ -221,4 +221,4 @@ int main() {
 
 In the next part of our compiler writing journey, I will probably feed
 more of the compiler source to itself and see what we still have to
-implement.
+implement. [Next step](../46_Void_Functions/Readme.md)

46_Void_Functions/Readme.md

@@ -270,4 +270,4 @@ Two big things that we will need to tackle are static functions and
 variables, and the `sizeof()` operator. 
 
 In the next part of our compiler writing journey, I will probably
-work on the `sizeof()` operator because `static` still scares me a bit!
+work on the `sizeof()` operator because `static` still scares me a bit! [Next step](../47_Sizeof/Readme.md)

47_Sizeof/Readme.md

@@ -137,4 +137,4 @@ Well, `sizeof()` turned out to be simple, at least for the functionality that
 we need for our compiler. In reality, `sizeof()` is quite complicated for
 a full-blown production C compiler.
 
-In the next part of our compiler writing journey, I will tackle `static`.
+In the next part of our compiler writing journey, I will tackle `static`. [Next step](../48_Static/Readme.md)

48_Static/Readme.md

@@ -208,4 +208,4 @@ going through the code, finding all C_GLOBAL uses and ensuring that I added
 appropriate C_STATIC code as well.
 
 In the next part of our compiler writing journey, I think it's time that
-I tackle the [ternary operator](https://en.wikipedia.org/wiki/%3F:).
+I tackle the [ternary operator](https://en.wikipedia.org/wiki/%3F:). [Next step](../49_Ternary/Readme.md)

49_Ternary/Readme.md

@@ -323,4 +323,4 @@ to itself, look at the parse errors I get and choose one or more of them
 to fix.
 
 > P.S. We've reached 5,000 lines of code and 90,000 words in the Readme
-  files. We must be nearly there!
+files. We must be nearly there! [Next step](../50_Mop_up_pt1/Readme.md)

50_Mop_up_pt1/Readme.md

@@ -281,4 +281,4 @@ functionality works.
 
 So that's a few small things fixed up in this part of our journey.
 What I will do now is step back, rethink the array/pointer design and
-try to fix this up in the next part of our compiler writing journey.
+try to fix this up in the next part of our compiler writing journey. [Next step](../51_Arrays_pt2/Readme.md)

51_Arrays_pt2/Readme.md

@@ -138,4 +138,4 @@ but just needed some more tweaking to cover all the functionality that
 we needed.
 
 In the next part of our compiler writing journey, we will go back to
-mopping up.
+mopping up. [Next step](../52_Pointers_pt2/Readme.md)

52_Pointers_pt2/Readme.md

@@ -320,4 +320,4 @@ we only expect C pre-processor lines when we see a '#' in column 1.
 
 In the next part of our compiler writing journey, I'll go back to feeding
 the compiler source code to itself, see what errors pop and up choose one
-or more to fix.
+or more to fix. [Next step](../53_Mop_up_pt2/Readme.md)

53_Mop_up_pt2/Readme.md

@@ -267,4 +267,4 @@ them.
 In the next part of our compiler writing journey, I will have to work
 on implementing register spills. I've been delaying this, but now most of
 the fatal errors from the compiler (when compiling itself) are register
-issues. So now it's time to sort this out.
+issues. So now it's time to sort this out. [Next step](../54_Reg_Spills/Readme.md)

54_Reg_Spills/Readme.md

@@ -460,4 +460,4 @@ but there is substantial room for improvement.
 
 While doing the above, I also fixed the problem with `&&` and `||`. I've
 decided to write these changes up in the next part, even though the
-code here already has these changes.
+code here already has these changes. [Next step](../55_Lazy_Evaluation/Readme.md)

55_Lazy_Evaluation/Readme.md

@@ -176,4 +176,4 @@ the only thing the compiler can't parse (in its own source code) is the
 declaration and use of local arrays. So, guess what...
 
 In the next part of our compiler writing journey, I'll try to work out
-how to declare and use local arrays.
+how to declare and use local arrays. [Next step](../56_Local_Arrays/Readme.md)

56_Local_Arrays/Readme.md

@@ -115,4 +115,4 @@ elements.
 ## Conclusion and What's Next
 
 In the next part of our compiler writing journey, I'll return to mopping
-up duties.
+up duties. [Next step](../57_Mop_up_pt3/Readme.md)

57_Mop_up_pt3/Readme.md

@@ -215,4 +215,4 @@ That's enough mopping up for now. When I fixed up the `+=` and `-=` problem,
 it highlighted a big issue with the `++` and `--` operators (prefix and
 postfix) as applied to pointers.
 
-In the next part of our compiler writing journey, I will tackle this issue.
+In the next part of our compiler writing journey, I will tackle this issue. [Next step](../58_Ptr_Increments/Readme.md)