Glosses on Linus Torvalds Coding Style

A gloss is a marginal note (see picture in the Wikipedia entry for an example). While one can't very well annotate someone else's web page, I will add my voice to the chorus of those who have commented on Linus Torvalds coding style document.

I strongly agree with almost everything said in this document, and have been glad for the "backup" when I've had to argue a point with some of my fellow workers. There are, however, a couple of items where I diverge.

The first point that I would gently argue is that 8-space tabs are too deep. I haven't "been looking at [my] screen for 20 straight hours" for a couple of years, but at the very outset I worked with 4-space tabs, and I find them very readable. I haven't worked with anyone who has strongly defended 8-space tabs. I will admit that I have had to try to persuade people to switch from 2-space to 4-space tabs, but that was back in the Apple ][ Basic days when people were trying to keep their programs as small as possible (remember cassettes?). I don't know of any readability studies on this topic, and would be interested to see them.

My second argument (or shall I say "disagreement" to be less disagreeable...) is with his distaste for mixed-case names. I find mixed-case easier to read (maybe it's just me), but, more importantly, I find it a lot easier to type than underscores. I know I'm not the best touch-typist in the world, but I always find myself looking down to make sure that my ring-finger is indeed hitting the dash/underscore key and not one of its neighbors. Using camelCase keeps my fingers on the home row much more, makes my typing faster and reduces the number of errors.

These are just friendly disagreements, however, and don't diminish the fact that I'm glad this document exists, and is available to us all.

Pythonics Anonymous

I think I need a twelve-step program to wean myself from Python.

When I first looked at Python a few years ago I was put off by the fact that white space was significant. I also thought that having to do "self." to indicate object fields (vs. global variables or parameters) was just noise. I got tired, however, of always looking up Perl syntax, and I really didn't care for the Perl TMTOWTDI (I prefer one good way to do things, not many). I then came across Eric Raymond's article and decided to learn and use Python.

Now, after a few years of working with the language I appreciate the fact that white space is significant (no more bracket-placement wars!). I also appreciate the cross-platform-ability and the fact that (even though it is open-source) it isn't GPL, which has allowed us to use it at a company I used to work for. I like having the various supporting packages (although the minidom has a problem when pretty-printing XML that I had to work around).

However, I now find myself thinking that I'd better go back to Java. I was put off Java because the reality took a really long time to catch up with the hype. Remember "Write once, run anywhere?" (which was really "Write once, debug everywhere?") It also seemed too heavyweight to me. I liked being able to run the Python interpreter in command line mode and experiment without having to fire up an editor, create files, save them and run them. It was also satisfying to be able to be able to write a one-line "Hello World".

It seems now, however, that the world (at least the world of enterprise software) has decided that It's A Java World. I guess I better get back on board. I know I'll fall off the wagon occasionally, but it's time to put Python back in the corner and get back to business.

Data Always Goes In A Database, Right?

Not necessarily.

I have worked on a number of software products where data (usually metadata) is stored in text or binary files. The code opens the file, reads from the file and parses the text (if necessary) into the structures used internally by the code. Over the years, the question has been asked more than once: why don't you put this data into a database?

There are a number of reasons not to use a database. First, all the data needs to be in memory for performance, and there is no advantage to keeping parts of it out on the disk. Second, the data is usually not appropriate for being "table-ized." That is, the data is complex and tree-structured rather than row-structured. I know that one can always flatten data into multiple tables with the appropriate keys and joins, but parsers built with tools such as Yacc and Lex will always be faster than doing multiple database joins and reads.

A third reason is that upgrading the data in the database requires scripts to add or modify tables or columns (deletes not being done to reduce the potential for referential integrity problems). Parsers can recognize a version number and do an upgrade-on-read, which doesn't require another program to run to perform the upgrade.

Fourth, files are easier for end-users to handle. For example, if a user needs to send some of his data to tech support, it's much easier for him to package up one or more files, rather than having to run some database dump utility and send the dump file. There is also no issue with (once having that database dump) the possibility that tech support does not have the particular version of the database software that the user is running.

A fifth reason (for text files at least), is that in a pinch the file can be opened and edited in a text editor if there are problems with it. With a database you have to run the appropriate database client software, then tease out the table relationships to understand where changes need to be made.

When all you know how to use is a hammer, everything looks like a nail.

Fail Early (But Not Often)

Civic-minded citizens have long been reminded to "Vote Early". A corrupt Chicago politician extended it to "Vote Early And Often." Failures should be handled as early as possible, but shouldn't (under most circumstances) happen very often.

My primary principle is "The earlier the failure, the less the consequences." For example, if I didn't check the return code on a file open, the code would merrily continue on and try to write to the bogus file descriptor returned from the "file open". Most current operating systems give you a bad handle error, but earlier ones would just return junk at best and might do something grotesque to the file system at worst. Also, the code has probably already gone past the point where it can put up a coherent error message that includes the filename and system error code. This the leads to my second principle about failure handling, which is: "The better the error message, the less tech support hates my guts." One annoying error message that I've had to deal with (as an end user) is "The [someLibrary.DLL] is in use by another program." Which program? Do I have to go through my Task Manager, or use "ps" on *nix, trying to kill one program at a time (not a good idea anyway!) to find out which is the offender?

Most contemporary programming languages have some sort of exception mechanism to help me write code to handle failures. In a language like C++ where I can throw just about anything, I create a superclass designed to handle failures (subclassed as appropriate) combined with a macro that catches all possible types that any code in my executable might throw. As long as I put my catch blocks in the right places (e.g., at the top of a thread or in top-level event handlers) I can return errors without bad side effects. Java, on the other hand, will only allow me to throw descendants of Throwable, which removes the need for the "glue" class and macro that I described for C++, but adds the complexity of having to deal with multiple subclasses of Throwable. There's no such thing as a free lunch.

What if I am working in a language that supports exceptions, but I am using some API that doesn't? For example, suppose I needed to use fopen (the low level 'C' routine) while working in C++. In this case I always check the return code from that call (as well as all other file system calls). If an error is returned from any of those calls, I convert it into a coherent exception and throw that exception.

To move farther back in the evolutionary ladder of languages, what happens when I am working in a language (such as shell script) where there is no exception mechanism at all? Here I check for errors after every possible line of shell script that can possibly fail ("assume the worst"). I keep the code readable by putting the status check off to the right of the "meaningful" code. This means that when I read the code I can concentrate on what it should be doing in most cases, rather than the edge cases of failure. For example, this shell script contains a shell function that displays an error message, the name of the script and how to get instructions for running the script.

Note that the error might not be caused by inappropriate usage, but this at least gives the person who ran the script something to see before reading the code in the script.

In the spirit of continuous improvement, I've tried moving the status test down into the shell function, but I discovered (at least on Cygwin) that making the function call changes the global status variable ($?). I even briefly considered doing something like this where the _chk_ function runs lines of script and then checks the error code.

_chk_( 'first line of script' )
_chk_( 'second line of script' )

I rejected this, however, because (a) the noise level is too high, and (b) I had serious concerns about quoting problems within each line of shell script code.

Sometimes you have to know when to stop refactoring.

The Two Commandments of Blogger

Learned from bitter experience...

1. Don't copy and paste from anywhere else, unless you have no special formatting such as lists or fonts. If you do, you will end up retyping your post anyway because Blogger incorrectly converts the formatting. I have tried this from a number of different sources including Google Documents and somehow the text after lists always ends up with the wrong linespacing.
2. Don't trust the preview. In scenario above, the preview looks fine. When I post, however, the linespacing is messed up and I have to "Save as Draft" to pull the post and re-edit it. This happens in both Firefox and Internet Explorer.

Thanks to my son for his help and suggestions.

The Costumer's Handshake

What is it about code formatting standards? After all, the important thing is that the code compiles, links and passes its tests, right?

No. All psychological and behavioral arguments aside, there is something attractive about well-formatted code. Admittedly, well-formatted code does not guarantee functional correctness, adequate performance and maintainability, but it is clear (at least to me and the folks that I prefer to work with) that it helps readability and therefore maintainability. And it certainly makes a better first impression than a slap-dash bunch of statements thrown together.

What does this have to do with costumers and handshakes?

My wife likes to sew costumes for my daughter, who is deeply involved with the San Francisco Dickens Faire and other such semi-historical pastimes. At one of the Dickens Faires my wife approached a woman and asked her about her costume. This is not an uncommon occurrence at these events, and most people are only too happy to talk about their work. After chatting for a few minutes and establishing a rapport, my wife reached out, looked closer at the garment, and turned the edge of the woman's bodice to examine the stitching and the lining. The woman chuckled and said, "Ah, the costumer's handshake!" and we both immediately knew what she meant.

People who care about the quality of their (and other's) work know to look inside and underneath. Costumers know the handshake. Woodworkers know to open the drawers of a cabinet to check out craftsmanship of the joints inside. They look to see if something is veneer (likely to come off in time), or solid wood, which will last longer. Printers know to hold up a page of a book to the light to check that the imposition of back-to-back pages is in alignment, and to check the quality of the paper.

I don't know if such quick evaluations can help judge the quality of a body of code. There are obvious things to look for such as error handling and formatting consistency, but problems such as duplicated code and bad naming are harder to spot without spending more time getting to know the code.

I haven't worked with many open source projects, but the one I've spent the most time with (Blender - see www.blender.org) certainly doesn't pass a lot of my tests. Does that mean it's not successful? If open-source success is defined as "usage and visibility" then Blender certainly qualifies as a success. A lot of the commercial closed-source code I work with doesn't pass my tests either. Given that success for commercial closed-source is defined as profitability for the company that produces it, I can't comment on that in public.

I do know, however, that (by the craftsmanship standard), much of the code that I work with might not stand up to having its "lining" examined. But that is one of the goals I aim for.

Paranoid Programming

I read and enjoy Joel on Software (and used his CityDesk product for my website). I have to respond, however, to some of his comments in this article.

I strongly agree with most of it. For example, I'm glad to see conventions such as "Don’t put closing braces more than one screen away from the matching opening brace." (And I wish I could enforce that standard at the place I work at!) Also, he writes: "I have to admit that I’m a little bit scared of language features that hide things". I have seen very confused code caused by "features" in C++ like the parentheses operator on iterators. I always prefer to see a real method name such as "Next()" (thanks to Nick Pouschine for putting into words the solution to my unease with that syntax). And I never have, and never will, overload the plus operator for any reason.

I don't, however, see any reason to write off exceptions. I depend very heavily on exceptions in all the C++ and Python code I write. Joel's example of:

dosomething();
cleanup();

just seems wrong. The cleanup code should be on the destructor (or some equivalent in non-object-oriented languages), which should always get called at some point. Note that I'm assuming you've adopted the convention of "every object is correctly owned" so that memory is always freed at the appropriate point in the code, regardless of whether errors occurred or not.

[Apologies in advance for the CAPITALS.] I don't EVER need to "investigate the entire call tree of dosomething()" to see if there’s anything problematic in there. Rather I ALWAYS assume the worst, which is that any line of can cause an error (divide by zero, anyone?). I'm not paranoid, they really are out to get me!

I therefore write all my code to handle errors anywhere. That way, if an error happens in my code or some third-party code that I'm using, it is handled. Once I "accepted" exceptions, then I wrote code that I found more readable. I also was more confident in its error handling capabilities.

For example, instead of having all that nesting in the code from Raymond Chen's article that Joel references, I could put the variables into fields in an object, all the cleanup code into a destructor and then "flatten out" the error checking (I'll try to put an example in here later...). I don't claim that there are fewer lines of code (there may be more), but the error handling code doesn't get in the way of my reading and understanding of the intent of the code. This is especially true when errors have to be propagated back up a long call stack, where there may be many intermediate functions that would have to return the error. With exceptions, I only care about where they are thrown and where they need to be caught, not all the intermediate points in between.

To me, having the "other channel" of exception handling removes noise from the code and makes it more readable. I agree with Raymond that exceptions are tricky and you have to know how to do the right thing (e.g., don't throw pointers to stack-based objects and make sure to understand threading issues). Once, however, I had those principles understood and enforced, I was able to keep the error handling code "out of the way" of the rest of the code. This allows me to focus on the intent of the code as I read it, and not have to mentally "filter out" all the error handling code.

I certainly don't think that I'm smarter than Joel or Raymond. I do know, however, that I prefer using exceptions rather than having every function return error codes, both for code readability and robustness. I know they make me a more productive programmer.