Slashdot Mirror
Refactoring: Improving the Design of Existing Code
kabz writes "Refactoring (as I'll refer to the book from here on in) is a heavy and beautifully produced 418 page hardback book. The author is a UK-based independent consultant who has worked on many large systems and has written several other books including UML-Distilled. Refactoring is a self-help book in the tradition of Code Complete by Steve McConnell. It defines its audience clearly as working programmers and provides a set of problems, a practical and easily followed set of remedies and a rationale for applying those techniques." Read below for the rest of Johnathan's review.
Refactoring: Improving the Design of Existing Code
author
Martin Fowler with Kent Beck, John Brant, William Opdyke and Don Roberts.
pages
431
publisher
Addison-Wesley
rating
9/10
reviewer
Jonathan Watmough
ISBN
0201485672
summary
expands and formalizes the idea of applying explicit refactorings
Code refactoring is the process of restructuring code in a controlled way to improve the structure and clarity of code, whilst maintaining the meaning of the code being restructured. Many maintenance problems stem from poorly written code that has become overly complex, where objects are overly familiar with each other, and where solutions implemented expeditiously contribute to the software being hard to understand and hard to add features to.
Typically refactorings are applied over a testable or local scope, with existing behavior being preserved. Refactoring as defined in this book is not about fixing bad designs, but instead should be applied at lower levels.
Testing a la Extreme Programming is emphasized as a control for ensuring that program meaning is not changed by refactoring. It is not over emphasized, and this is not a book about testing, but it is often mentioned and stays in the background through the book.
The refactorings presented in the book are not intended as a comprehensive solution for all problems, but they do offer a means to regain control of software that has been implemented poorly, or where maintenance has been shown to simply replace old bugs with newer ones.
The book is divided into two main sections, introductory material that introduces and discusses refactorings, and a lengthy taxonomy of refactorings that includes both examples and further discussion. The introductory material consists of a long worked example through simple Java code that implements printing a statement for a video store. Despite the simplicity of the code, Fowler shows in clear detail where improvements can be made, and how those improvements make the code both impressively easy to understand, and easy to maintain and add features.
Several key refactorings are demonstrated in the opening chapter including Extract Method, Move Method and Replace Conditional with Polymorphism. This is a book about programming in the object oriented paradigm, so as you might expect, the first two refactorings refer to extracting and moving object methods either into new methods, or between objects. The third example provides a means to replace special cased behavior in a single object type by deriving a sub type of the object and moving type specific code to the sub types. This is a fundamental technique in object oriented programming, and is discussed here in practical terms.
Now that several actual refactorings have been introduced, Fowler provides a solid and well thought-out discussion of the why's, when's and when not's of refactoring. For example, code can decay as features are added, and programmers special-case, or bodge additional functionality into existing objects. Fowler argues that the bitrot and decay makes software more unreliable, leads to bugs and can accelerate as the problem gets worse. Faced with these problems, refactoring should be used to improve local design and clean up and improve code, leading to better software, that is easier to maintain, easier to debug, and easier to improve with new features as requirements change.
However, there is a caveat, in that since software functionality should remain unchanged during refactoring, the process of refactoring consumes resources, but provides no easily measurable value. Fowler confronts this issue in a section that discusses how to communicate with managers, that you are performing refactoring work. He denies being subversive, but his conclusion is that refactoring should essentially be folded in with normal work as it improves the overall result.
This is a bit like goofing off on the basis that you'll think better after 20 minutes of fooseball. I'd definitely subscribe to that theory, but many others may not.
Kent Beck guests in Chapter Three for a review of the issues in typical software that suggest a refactoring may be needed. This chapter is entitled Bad Smells in Code, and most of the smells presented will be familiar to any reasonably experienced programmer, and they will be a great learning experience for less experienced programmers. I got the same feeling reading this chapter as I did when I first read Code Complete. Here was someone writing down names and describing problems that I had a vague unease about, but was too inexperienced to really articulate or do something about. Typically the refactorings address the same kind of issues that a code review with a skilled experienced programmer would address. Long parameter lists, too long methods, objects delving about in each others private variables, case statements, related code spread across different objects etc. None of these problems are debilitating in themselves, but added up, they lead to software that can be prone to error and difficult to maintain.
Most of the remaining substance of the book, 209 pages, is given over to a taxonomy of refactorings. These 72 refactorings are covered in detail with comprehensive simple examples presented in Java. Each refactorings is given a clear name, a number and a line or two of descriptive text. The motivation for the refactoring is then discussed, often including caveats and cautions. The mechanics of implementing the refactoring are then listed, with 1 or more (and often more) examples of implementing the refactoring. Refactorings range from the very simple to more complex examples such as Convert Procedural Design to Objects.
Due to the difficulties of reproducing large and complex sections of code, Fowler sticks with relatively simple examples. These seem to grate on him more than the reader, and he can come across as somewhat embarrassed when we look at the employee, programmer, manager pay example for the tenth time. I certainly didn't have a problem with it though.
This is a very well written and fun to read book. I personally feel that much of the material is implied by from Code Complete, but Fowler does a fantastic job of expanding and formalizing the idea of applying explicit refactorings. Much like Code Complete gave a motivation for keeping code well commented and laid out, this book presents the case for care and feeding of how to structure software. To fight bitrot and technical debt, poorly structured and unclear code should be targeted and refactored to improve structure and clarity. This gives a very real payback in terms of less required maintenance, and ease in adding features later on down the line.
Despite the fact that all the examples are in Java, the ideas are easily transferable to C++ or any procedural object oriented language. I highly recommend this book.
You can purchase Refactoring: Improving the Design of Existing Code from amazon.com. Slashdot welcomes readers' book reviews -- to see your own review here, read the book review guidelines, then visit the submission page.
Typically refactorings are applied over a testable or local scope, with existing behavior being preserved. Refactoring as defined in this book is not about fixing bad designs, but instead should be applied at lower levels.
Testing a la Extreme Programming is emphasized as a control for ensuring that program meaning is not changed by refactoring. It is not over emphasized, and this is not a book about testing, but it is often mentioned and stays in the background through the book.
The refactorings presented in the book are not intended as a comprehensive solution for all problems, but they do offer a means to regain control of software that has been implemented poorly, or where maintenance has been shown to simply replace old bugs with newer ones.
The book is divided into two main sections, introductory material that introduces and discusses refactorings, and a lengthy taxonomy of refactorings that includes both examples and further discussion. The introductory material consists of a long worked example through simple Java code that implements printing a statement for a video store. Despite the simplicity of the code, Fowler shows in clear detail where improvements can be made, and how those improvements make the code both impressively easy to understand, and easy to maintain and add features.
Several key refactorings are demonstrated in the opening chapter including Extract Method, Move Method and Replace Conditional with Polymorphism. This is a book about programming in the object oriented paradigm, so as you might expect, the first two refactorings refer to extracting and moving object methods either into new methods, or between objects. The third example provides a means to replace special cased behavior in a single object type by deriving a sub type of the object and moving type specific code to the sub types. This is a fundamental technique in object oriented programming, and is discussed here in practical terms.
Now that several actual refactorings have been introduced, Fowler provides a solid and well thought-out discussion of the why's, when's and when not's of refactoring. For example, code can decay as features are added, and programmers special-case, or bodge additional functionality into existing objects. Fowler argues that the bitrot and decay makes software more unreliable, leads to bugs and can accelerate as the problem gets worse. Faced with these problems, refactoring should be used to improve local design and clean up and improve code, leading to better software, that is easier to maintain, easier to debug, and easier to improve with new features as requirements change.
However, there is a caveat, in that since software functionality should remain unchanged during refactoring, the process of refactoring consumes resources, but provides no easily measurable value. Fowler confronts this issue in a section that discusses how to communicate with managers, that you are performing refactoring work. He denies being subversive, but his conclusion is that refactoring should essentially be folded in with normal work as it improves the overall result.
This is a bit like goofing off on the basis that you'll think better after 20 minutes of fooseball. I'd definitely subscribe to that theory, but many others may not.
Kent Beck guests in Chapter Three for a review of the issues in typical software that suggest a refactoring may be needed. This chapter is entitled Bad Smells in Code, and most of the smells presented will be familiar to any reasonably experienced programmer, and they will be a great learning experience for less experienced programmers. I got the same feeling reading this chapter as I did when I first read Code Complete. Here was someone writing down names and describing problems that I had a vague unease about, but was too inexperienced to really articulate or do something about. Typically the refactorings address the same kind of issues that a code review with a skilled experienced programmer would address. Long parameter lists, too long methods, objects delving about in each others private variables, case statements, related code spread across different objects etc. None of these problems are debilitating in themselves, but added up, they lead to software that can be prone to error and difficult to maintain.
Most of the remaining substance of the book, 209 pages, is given over to a taxonomy of refactorings. These 72 refactorings are covered in detail with comprehensive simple examples presented in Java. Each refactorings is given a clear name, a number and a line or two of descriptive text. The motivation for the refactoring is then discussed, often including caveats and cautions. The mechanics of implementing the refactoring are then listed, with 1 or more (and often more) examples of implementing the refactoring. Refactorings range from the very simple to more complex examples such as Convert Procedural Design to Objects.
Due to the difficulties of reproducing large and complex sections of code, Fowler sticks with relatively simple examples. These seem to grate on him more than the reader, and he can come across as somewhat embarrassed when we look at the employee, programmer, manager pay example for the tenth time. I certainly didn't have a problem with it though.
This is a very well written and fun to read book. I personally feel that much of the material is implied by from Code Complete, but Fowler does a fantastic job of expanding and formalizing the idea of applying explicit refactorings. Much like Code Complete gave a motivation for keeping code well commented and laid out, this book presents the case for care and feeding of how to structure software. To fight bitrot and technical debt, poorly structured and unclear code should be targeted and refactored to improve structure and clarity. This gives a very real payback in terms of less required maintenance, and ease in adding features later on down the line.
Despite the fact that all the examples are in Java, the ideas are easily transferable to C++ or any procedural object oriented language. I highly recommend this book.
You can purchase Refactoring: Improving the Design of Existing Code from amazon.com. Slashdot welcomes readers' book reviews -- to see your own review here, read the book review guidelines, then visit the submission page.
This is a bit like goofing off on the basis that you'll think better after 20 minutes of fooseball. I'd definitely subscribe to that theory, but many others may not.
It's not at all like that. If you really need an analogy to understand this very simple concept, it's like seeing that your desk is overflowing with paperwork and spending some time filing everything properly. A little time invested can make it a lot quicker to find what you are looking for and help you deal with one thing at a time.
Reviewed back in 1999 on Slashdot.
http://books.slashdot.org/article.pl?sid=99/09/16/1333202
I don't mean to be a wet blanket, but the book has been out for quite some time -- checking Amazon, July of 1999. It is pretty great, and I would recommend it to those who somehow managed to miss it up to this point, but a review almost nine years later? Slow news day much?
On the other hand, maybe periodically prodding towards the direction of higher internal quality (to be distinguished from external quality, that which is perceived in a black-box fashion by your customers, the relationship between these two qualities is of course a matter of much friction and debate between the managing and laboring classes) isn't a bad call. Lord knows any extra ammo to convince people that this is worthwhile is appreciated. As much as we like to think of ourselves as poets, I sometimes think the traditional profession most software development resembles is "butcher" or "janitor", just one messy hack-job and sweep under the rug after another after four or five decades of which you can retire and grumble on the beach about putting cyanide in the guacamole, fondly reminiscing about your red swingline or asr-33 or what have you.
News for Geeks in Austin, TX
Does your book cover that?
And bury them deep.
"new" code is a capital investment and gets very high tax benefits.
"refactored" code is a pure cost.
I do a lot of refactoring but i always have to sneak it in under the cover of a "new" project instead of a "bug fix" project.
She was like chocolate when she drank... semi-sweet at first and then increasingly bitter.
THe Pragmatics Programmer: From Journeyman to Master" was published as well in 1999 and is written by Andrew Hunt and David Thomas. I just got it for Christmas and have been enthralled with it. They spend some time dealing with the refactoring of code as well as wonderul insight into a wealth of other areas.
This is yet another book that I wish I had read years ago. Working a few years in industry really makes you realize how much you can learn from other people. But alas, the problem of youth is that you always think you're the exception.
Either give it away or get top dollar, but never sell yourself cheap.
At several places where I worked management was always happy to allow cycles to be spent on the process of 'refactoring' the code.
Unfortunately, in my experience, the process of 'refactoring' involved making code more complex by adding to it. In one case, I saw the product of the 'refactoring' process wrap two pieces of functionality into two separate EJBs (with a whole 'dto-pojo' conversion scheme for data "isolation"). In another, I saw some functionality wrapped into a collection of beans--which was later wrapped in another layer of beans, and so forth, until 20 lines of code which set up a call into the javax.xml.translate package (for performing an XSLT transformation) into something like 8 bean layers. The 20 lines of code was at the heart of an 8-layer onion, each layer added by someone else's "refactoring" operation.
In Java, because modern IDEs allow you to write code without thinking, the problem with code is not that there isn't enough code (to prevent incestuous classes from being overly familiar with each other), but that there is too much code as programmers unfamiliar with the problem decided to add beans and interprocess communication and multiple threads without properly sizing the problem. (Right now I'm looking at an internal system which may need to process 1 transaction a second, tops, built in an inter-cooperating network of 8 EJBs, which someone thought would help improve transactional performance. Eight? A second system essentially replicates an in-memory SQL system rolled in-house: the system has been buggy because the reverse index processing had a race condition. Um, why wasn't that built in a dozen classes on top of MySQL instead of built with a couple of hundred classes that reinvent the wheel poorly?)
While I'm glad someone wrote a book on refactoring methodology, in my experience what we need is a book which describes how to write "simple" code: code that is just as complicated as it needs to be, and no more. And a book which also describes how to simplify overly-complicated code, how to pick simpler techniques, and how to manage programmers who have an "itch" so they go scratch it somewhere else--I think that would be a much more useful book.
It's easy to add complexity. It's hard to simplify.
And if anyone whines about how old code needs to be rewritten, point them at this
http://www.joelonsoftware.com/articles/fog0000000027.html Old code doesn't rust, it gets better, as bugs are fixed. Lou Montulli again: "I laughed heartily as I got questions from one of my former employees about FTP code the he was rewriting. It had taken 3 years of tuning to get code that could read the 60 different types of FTP servers, those 5000 lines of code may have looked ugly, but at least they worked." That's just in a PC application. Try refactoring the 'ugliness' out of an embedded system and see how long your employer still has customers, and how long you still have a job. And it's interesting that evolution, an unconscious process that far outperforms human 'intelligent' designers doesn't have any concept of ugliness at all. Maybe that concept is just an artifact of your limited ability to deal with complexity.
echo -e 'global _start\n _start:\n mov eax, 2\n int 80h\n jmp _start' > a.asm; nasm a.asm -f elf; ld a.o -o a;
If the application is "done" and no major components are to be added, then by all means leave it alone.
However, most systems aren't like that. They need to be changed due to changes in requirements, new functionality, business rules, etc., and that's where refactoring helps you - it helps isolate functionality and help the system evolve. (The app should have been designed with that in mind from the beginning, but you and I both know that it never is.)
The main problem I have with refactoring (both the book and the concept) is that it's way too easy to go overboard with it because it presents a Right Way of writing code. Sometimes methods need to be 20 lines, sometimes classes need to do more than one thing, sometimes inheritance just gets in the way. Fowler respects these ideas, but I've known people who read this book and take refactoring to its logical extreme, which results in overly fragmented code.
I also despise the XP directive to not comment code, which Fowler promotes.
In all it's a good book but it's best to read it when after you've had a few years of real world experience and you can tell what should and shouldn't be taken seriously.
It presented examples in Smalltalk and C++.
Stumbling in the dark
I hear slavering of jaws
Eaten by a grue.
The fallacy you're doing there is shifting the scales in that comparison. One side is learning Jazz at all, the other is being a _star_ in programming.
You'll find that being a star in _any_ discipline, Jazz included, isn't just a matter of being given a crash course in music notation. There are many who can learn Jazz, but there are few which are stars at it.
And there are a ton of people who just aren't any good at Jazz, no matter how much you teach them. There are those who are born tone deaf, or lack the coordination, or those who just aren't interested in a musical career, or those who find that learning uber-boring, or a dozen other cases which you just can't turn into _stars_ in any kind of music.
I'll further go and say that in Jazz (or any other kind of music) it's a relatively low pay career and with very few oportunity to become a super-star. So I'd wager that any music school gets an already filtered set of candidates. It's already those who love that genre, are sure that that's what they want to do, etc. It's already the people passionate and motivated.
In programming, we get a ton of drooling burger-flippers who think they can just get a quick training in Java and earn the big bucks. Not because they like it, not because it fits their personality type, not because they showed any kind of aptitude or inclination, just because they think they can fake their way to the big bucks. And it shows.
If you want a more apt comparison, compare them to the gang wanting to be a rock star in high school and college. The prospect of glamour and big bucks is there, and every other high-school boy wants to be in a rock band. The problem is that most suck, and will _never_ be even an acceptable player with any instrument. And aren't particularly motivated to train hard either. How many actually become a _star_? I'll say that's lost in the decimals.
So basically to sum up the comparison to Jazz:
1. Yes, you can teach Jazz. You can teach CS too. That's why we have that kind of colleges, you know.
2. You can't just take any guy off the street and turn him into a _star_ musician. If you went and took random people laid off by McDonald and tried to turn them into musicians, very few would even make an acceptable musician, and _extremely_ few will ever be considerable a _star_. Same as programming.
But, of course, that won't stop idiot PHBs from trying.
A polar bear is a cartesian bear after a coordinate transform.
I have seen refactoring in action aswell.
It decreased the size of the program code, made the code more modular, faster, made possible to write good unit tests for it and decrease bug count per LOC, all in all rewarding the developer and user with a more stable, smaller, faster and more maintainable codebase.
It takes a man to suffer ignorance and smile
Be yourself no matter what they say
Two great reasons to refactor:
(1) Your requirements have changed.
(2) You are dealing with a piece of buggy code that is a nightmare to debug and test.
There's an interesting book you should read. I forget the name, but it's about how to improve the design of existing code.
how to invest, a novice's guide
First - you do realize the 2nd edition of Code Complete has a whole chapter on Refactoring, right? AND it recommends the other book for more details?
Have you actually read the book or are you just going from the review? Most of the methods are designed to work in a very small scale.
Examples - have you...
- Moved some code in the middle of a method to a separate method to make your code clear or re-usable?
- Renamed a variable with an unclear name?
- Added an intermediate variable for a complex calculation?
- Moved a complex boolean expression into a well-named boolean function?
- Added/removed a parameter to/from an existing method?
These are all refactorings, although some of the simplest. You might say that is just common sense, but that is kind of the point of it, and the particular steps are designed so that you won't screw up the code with common mistakes. Many refactoring tools exist too, and they are quite useful - for example, when adding a parameter to a method, one of my refactoring tools show me where that was used and I can edit any with a click (and also note if they all need it, or if I should give it a default value). If I change the name of a method, it can replace everywhere it is used (a search and replace can do the same, except if another object has a method with the same name it will screw things up - so you have to watch each replace carefully).
You might also note that many refactorings match other concepts in Code Complete.
None of these are supposed to "fix" a broken design by itself. They are meant to slowly improve the code so that it is easier to keep working on it.
The other forms of documentation are meaningful names. If you want to document a sub-expression, which does not have a name, give it a name, by assigning it into a variable before using it. If a piece of code does not have a name, then give it a name by putting it in a separate function.
Variable and function names are much less likely to go out of sync with the meaning of the code than comments are, are more concise and less redundant (so they avoid violating the DRY principle).
Remember, bad/wrong documentation is worse than no documentation.
There's lots of posts about how Refactoring is a waste of time, broke something or leads to bad code. I would put it to you that taking a saw and a hammer, one can do a lot to damage a perfectly good house. On the other hand, they can be used to build a nice new deck. I imagine that since I have no experience with a hammer that I would do a lot of damage. Why does everyone with no experience at Refactoring, and who tries it with poor results, assume that Refactoring is bad?
What to about this? Some options:
Refactoring (properly) leads to improved productivity and contentment, but learning any tool requires more than reading a book.
I'm surprised that the companion web site for the book hasn't been posted here yet. It's the catalog that will be of most interest.
What if you're not the one who does it "The first time around"? You mention that most code is generated by idiots... Don't you think that there's a good chance that many programmers might find themselves in a new position, being asked to implement new features into an old system that your predecessor (one of the idiots you mention) managed to somehow bludgeon together?
Another scenario (this time it's personal experience): A co-worker and I were asked if we could design an in house resource management system. Management wanted to keep this very low budget, meaning we had to use the tools available (VERY LIMITED) and nothing which would take more resources than a couple of co-op students knocking the idea about in their spare time (i.e. no resources at all). Although both of us have coding experience, neither of us had used the language available to us, nor do we do any sort of programming in our daily jobs.
We started discussed a couple of different approaches and we started knocking out bits of code that were really only meant for testing ideas about the language and putting together a rough sketch of the program. Unfortunately, one of the bosses saw a nearly-almost-halfway functional version out of this, showed it to his boss, who got excited, and then handed the code off to another employee to finish (actual resource allocation now that upper management approved). He unfortunately was a sloppy programmer and did a mediocre job at finishing the project. The system that resulted from that is still in use today, although over the last few months has started to show aberrant behaviour, reacting slowly, and any time a new feature is added it gets worse. "Tutorial" code we originally would have scrapped became a "functional" system, and it was made worse by a sloppy programmer who didn't understand the need for commenting (He actually deleted most of my comments out, because they "distracted" him).
In either of these situations, does it not sound like refactoring may do this code some good?
Write tests!
The only way you can make sure while refactoring an application that at the end of the process it will still behave the same way as it did before the refactoring is to write unit tests. Never (heavily) refactor untested code.