A fantastic book along these lines is Allen Holub's "Compiler Design in C". It's old (1990) and out of print, but you can get the PDF for free from Holub's site [1].
Over the course of the book, a C compiler is developed. To handle lexical analysis and parsing, tools similar to lex and yacc are developed.
Here's an excerpt from the preface to give an idea of the approach:
> This book presents the subject of Compiler Design in a way that's understandable to a programmer, rather than a mathematician. My basic premise is that the best way to learn how to write a compiler is to look at one in depth; the best way to understand the theory is to build tools that use that theory for practical ends. So, this book is built around working code that provides immediate practical examples of how given theories are applied. I have deliberately avoided mathematical notation, foreign to many programmers, in favor of English descriptions of the theory and using the code itself to explain a process. If a theoretical discussion isn't clear, you can look at the code that implements the theory. I make no claims that the code presented here is the only (or the best) implementation of the concepts presented. I've found, however, that looking at an implementation-at any implementation--can be a very useful adjunct to understanding the theory, and the reader is well able to adapt the concepts presented here to alternate implementations.
> The disadvantage of my approach is that there is, by necessity, a tremendous amount of low-level detail in this book. It is my belief, however, that this detail is both critically important to understanding how to actually build a real compiler, and is missing from virtually every other book on the subject. Similarly, a lot of the low-level details are more related to program implementation in general than to compilers in particular. One of the secondary reasons for learning how to build a compiler, however, is to learn how to put together a large and complex program, and presenting complete programs, rather than just the directly compiler-related portions of those programs, furthers this end. I've resolved the too-many-details problem, to some extent, by isolating the theoretical materials into their own sections, all marked with asterisks in the table of contents and in the header on the top of the page. If you aren't interested in the nuts and bolts, you can just skip over the sections that discuss code.
...
> In a sense, this book is really an in-depth presentation of several, very well documented programs: the complete sources for three compiler-generation tools are presented, as is a complete C compiler. (A lexical-analyzer generator modeled after the UNIX lex utility is presented along with two yacc-like compiler compilers.) As such, it is more of a compiler-engineering book than are most texts-a strong emphasis is placed on teaching you how to write a real compiler. On the other hand, a lot of theory is covered on the way to understanding the practice, and this theory is central to the discussion. Though I've presented complete implementations of the programs as an aid to understanding, the implementation details aren't nearly as important as the processes that are used by those programs to do what they do. It's important that you be able to apply these processes to your own programs.
Over the course of the book, a C compiler is developed. To handle lexical analysis and parsing, tools similar to lex and yacc are developed.
Here's an excerpt from the preface to give an idea of the approach:
> This book presents the subject of Compiler Design in a way that's understandable to a programmer, rather than a mathematician. My basic premise is that the best way to learn how to write a compiler is to look at one in depth; the best way to understand the theory is to build tools that use that theory for practical ends. So, this book is built around working code that provides immediate practical examples of how given theories are applied. I have deliberately avoided mathematical notation, foreign to many programmers, in favor of English descriptions of the theory and using the code itself to explain a process. If a theoretical discussion isn't clear, you can look at the code that implements the theory. I make no claims that the code presented here is the only (or the best) implementation of the concepts presented. I've found, however, that looking at an implementation-at any implementation--can be a very useful adjunct to understanding the theory, and the reader is well able to adapt the concepts presented here to alternate implementations.
> The disadvantage of my approach is that there is, by necessity, a tremendous amount of low-level detail in this book. It is my belief, however, that this detail is both critically important to understanding how to actually build a real compiler, and is missing from virtually every other book on the subject. Similarly, a lot of the low-level details are more related to program implementation in general than to compilers in particular. One of the secondary reasons for learning how to build a compiler, however, is to learn how to put together a large and complex program, and presenting complete programs, rather than just the directly compiler-related portions of those programs, furthers this end. I've resolved the too-many-details problem, to some extent, by isolating the theoretical materials into their own sections, all marked with asterisks in the table of contents and in the header on the top of the page. If you aren't interested in the nuts and bolts, you can just skip over the sections that discuss code.
...
> In a sense, this book is really an in-depth presentation of several, very well documented programs: the complete sources for three compiler-generation tools are presented, as is a complete C compiler. (A lexical-analyzer generator modeled after the UNIX lex utility is presented along with two yacc-like compiler compilers.) As such, it is more of a compiler-engineering book than are most texts-a strong emphasis is placed on teaching you how to write a real compiler. On the other hand, a lot of theory is covered on the way to understanding the practice, and this theory is central to the discussion. Though I've presented complete implementations of the programs as an aid to understanding, the implementation details aren't nearly as important as the processes that are used by those programs to do what they do. It's important that you be able to apply these processes to your own programs.
[1] https://holub.com/compiler/