if(x=5)
-
Zac Howland wrote:
nterestingly enough, I've run into opposition to it from supervisors when coding standards are discussed. I've never quite understood why they wouldn't want to avoid possible bugs, but sometimes they are almost religious about it.
I don't especially like this style but it's not the kind of thing I would forbid in a coding standard. It should concentrate on more important things, e.g., in C/C++, uninitialised variables.
Kevin
[coding standards]
Kevin McFarlane wrote:
It should concentrate on more important things
Like which ways brackets should be placed! :mad: Hating coding style guides and the people that are content in contriving them
"We trained hard, but it seemed that every time we were beginning to form up into teams we would be reorganised. I was to learn later in life that we tend to meet any new situation by reorganising: and a wonderful method it can be for creating the illusion of progress, while producing confusion, inefficiency and demoralisation." -- Caius Petronius, Roman Consul, 66 A.D.
-
Ravi beat me to it, but this is the first thing that comes to mind. I'd been coding C++ for about 8 months, paid employment for about 2. I spent a day looking for this bug, every time I saw the line if (x=5) my brain just turned it in to if (x==5) which is obviously what I meant. I got into the habit of writing if (5 == x ) after that, and every now and again, I'd get the compiler error that told me I'd made the same typo again :-)
Christian Graus - Microsoft MVP - C++ Metal Musings - Rex and my new metal blog
Christian Graus wrote:
if (5 == x )
Actually, there is quite some resistence to this. Their justification is that you don't avoid bugs by tricks or writing unreadable code. You happen to have such a trick for this case, but many other bugs cannot be avoided by tricks. Sticking to a good discipline is the key. The good thing is when you think of that trick, you are avoiding the bug already, regardless of which way you actually code. :)
Best, Jun
-
Kevin McFarlane wrote:
It should concentrate on more important things, e.g., in C/C++, uninitialised variables.
Agreed. However, it is interesting what topics actually get fought over most when discussing coding standard guidelines. I've been in meetings with 2 different development teams at 2 different companies and had similar discussions over what I consider to be rather trivial issues.
If you decide to become a software engineer, you are signing up to have a 1/2" piece of silicon tell you exactly how stupid you really are for 8 hours a day, 5 days a week Zac
Zac Howland wrote:
I've been in meetings with 2 different development teams at 2 different companies and had similar discussions over what I consider to be rather trivial issues.
Guidelines I've come across usually split into recommended and mandatory. Then the things which people disagree over are put in recommendations. But really guidelines, e.g., for C++, should concentrate on the kind of stuff in Scott Meyers's Effective C++ books.
Kevin
-
[coding standards]
Kevin McFarlane wrote:
It should concentrate on more important things
Like which ways brackets should be placed! :mad: Hating coding style guides and the people that are content in contriving them
"We trained hard, but it seemed that every time we were beginning to form up into teams we would be reorganised. I was to learn later in life that we tend to meet any new situation by reorganising: and a wonderful method it can be for creating the illusion of progress, while producing confusion, inefficiency and demoralisation." -- Caius Petronius, Roman Consul, 66 A.D.
jhwurmbach wrote:
Hating coding style guides and the people that are content in contriving them
Well, they should be about Standards rather than style. That's the key. I have my preferences over curly bracket placement but they wouldn't be enforced in a standards document I would write.
Kevin
-
Zac Howland wrote:
I've been in meetings with 2 different development teams at 2 different companies and had similar discussions over what I consider to be rather trivial issues.
Guidelines I've come across usually split into recommended and mandatory. Then the things which people disagree over are put in recommendations. But really guidelines, e.g., for C++, should concentrate on the kind of stuff in Scott Meyers's Effective C++ books.
Kevin
Kevin McFarlane wrote:
Guidelines I've come across usually split into recommended and mandatory. Then the things which people disagree over are put in recommendations. But really guidelines, e.g., for C++, should concentrate on the kind of stuff in Scott Meyers's Effective C++ books.
I don't disagree with that one bit. Its just been my experience that it hasn't worked out that way (and it should be noted that in the first case, I was the only one who was even vaguely familiar with Meyer's books ... much less Robbins, Sutter, etc). I've even had objections arise when I tried to make a case for always forcing code to compile on the highest warning level with 0 errors/0 warnings before being checked in to version control (which is something I always thought should be a given).
If you decide to become a software engineer, you are signing up to have a 1/2" piece of silicon tell you exactly how stupid you really are for 8 hours a day, 5 days a week Zac
-
[coding standards]
Kevin McFarlane wrote:
It should concentrate on more important things
Like which ways brackets should be placed! :mad: Hating coding style guides and the people that are content in contriving them
"We trained hard, but it seemed that every time we were beginning to form up into teams we would be reorganised. I was to learn later in life that we tend to meet any new situation by reorganising: and a wonderful method it can be for creating the illusion of progress, while producing confusion, inefficiency and demoralisation." -- Caius Petronius, Roman Consul, 66 A.D.
jhwurmbach wrote:
Like which ways brackets should be placed!
Everyone knows they should always be placed on the same line! All your code should look like this:
if (isTrue()) { for (int i = 0; i < SomeNumber; i++) { int q = i; DosomethingWithQ(q); } PrintSomething(); };P
If you decide to become a software engineer, you are signing up to have a 1/2" piece of silicon tell you exactly how stupid you really are for 8 hours a day, 5 days a week Zac
-
Kevin McFarlane wrote:
Guidelines I've come across usually split into recommended and mandatory. Then the things which people disagree over are put in recommendations. But really guidelines, e.g., for C++, should concentrate on the kind of stuff in Scott Meyers's Effective C++ books.
I don't disagree with that one bit. Its just been my experience that it hasn't worked out that way (and it should be noted that in the first case, I was the only one who was even vaguely familiar with Meyer's books ... much less Robbins, Sutter, etc). I've even had objections arise when I tried to make a case for always forcing code to compile on the highest warning level with 0 errors/0 warnings before being checked in to version control (which is something I always thought should be a given).
If you decide to become a software engineer, you are signing up to have a 1/2" piece of silicon tell you exactly how stupid you really are for 8 hours a day, 5 days a week Zac
Zac Howland wrote:
Its just been my experience that it hasn't worked out that way (and it should be noted that in the first case, I was the only one who was even vaguely familiar with Meyer's books
According to Francis Glassborow of the Association of C and C++ Users (ACCU), most programmers don't read books on programming, so I'm not surprised. He told me this in an email after he'd wriiteen that he'd lent a colleague Effective C++ and it had been returned unopened.
Kevin
-
Henize wrote:
In C# if statments can only evaluate booleans so if(x=5) would not give it a boolean and cause a compiler error.
Which it "borrowed" from Java ...
Henize wrote:
C# is the way of the future.
Just like you don't use a hammer when you need a screwdriver, you don't use C# (or Java or VB) when you need C++.
If you decide to become a software engineer, you are signing up to have a 1/2" piece of silicon tell you exactly how stupid you really are for 8 hours a day, 5 days a week Zac
Check out the Microsoft research project called Singularity. Its an operating system written in C#. Around %99 of it(including the kernel) is written in C#.
Zac Howland wrote:
Which it "borrowed" from Java ...
Yes, Java was a good start and had the right idea. C# can do more. .NET allows any language to be compiled to the .NET platform.
static int Sqrt(int x) { if (x<0) throw new ArgumentOutOfRangeException(); int temp, y=0, b=0x8000, bshft=15, v=x; do { if (v>=(temp=(y<<1)+b<>=1)>0); return y; :omg:
-
In C# if statments can only evaluate booleans so if(x=5) would not give it a boolean and cause a compiler error. C# is the way of the future.
static int Sqrt(int x) { if (x<0) throw new ArgumentOutOfRangeException(); int temp, y=0, b=0x8000, bshft=15, v=x; do { if (v>=(temp=(y<<1)+b<>=1)>0); return y; :omg:
Henize wrote:
static int Sqrt(int x) { if (x<0) throw new ArgumentOutOfRangeException(); int temp, y=0, b=0x8000, bshft=15, v=x; do { if (v>=(temp=(y<<1)+b<>=1)>0); return y;
Talking about errorprone statements ;P btw, I have never thought of that, not even after I had this error happen five times in a 500k lines application. Good tip, will keep that in mind and ask the quality manager to add it to the code convention.
WM.
Yaaarrrr What about weapons of mass-construction? -
jhwurmbach wrote:
Like which ways brackets should be placed!
Everyone knows they should always be placed on the same line! All your code should look like this:
if (isTrue()) { for (int i = 0; i < SomeNumber; i++) { int q = i; DosomethingWithQ(q); } PrintSomething(); };P
If you decide to become a software engineer, you are signing up to have a 1/2" piece of silicon tell you exactly how stupid you really are for 8 hours a day, 5 days a week Zac
Zac Howland wrote:
[brackets] Everyone knows they should always be placed on the same line!
You her a multitude of voices from all directions shouting: "For that you will burn on the stake, heretic!" ;P Thats what is fundamental about coding styles: When you have three programmers together, there are at least four stlyes assembled.
"We trained hard, but it seemed that every time we were beginning to form up into teams we would be reorganised. I was to learn later in life that we tend to meet any new situation by reorganising: and a wonderful method it can be for creating the illusion of progress, while producing confusion, inefficiency and demoralisation." -- Caius Petronius, Roman Consul, 66 A.D.
-
Zac Howland wrote:
Its just been my experience that it hasn't worked out that way (and it should be noted that in the first case, I was the only one who was even vaguely familiar with Meyer's books
According to Francis Glassborow of the Association of C and C++ Users (ACCU), most programmers don't read books on programming, so I'm not surprised. He told me this in an email after he'd wriiteen that he'd lent a colleague Effective C++ and it had been returned unopened.
Kevin
Which explains why there are so many horrible programmers running around getting paid to do their worst ...
If you decide to become a software engineer, you are signing up to have a 1/2" piece of silicon tell you exactly how stupid you really are for 8 hours a day, 5 days a week Zac
-
Check out the Microsoft research project called Singularity. Its an operating system written in C#. Around %99 of it(including the kernel) is written in C#.
Zac Howland wrote:
Which it "borrowed" from Java ...
Yes, Java was a good start and had the right idea. C# can do more. .NET allows any language to be compiled to the .NET platform.
static int Sqrt(int x) { if (x<0) throw new ArgumentOutOfRangeException(); int temp, y=0, b=0x8000, bshft=15, v=x; do { if (v>=(temp=(y<<1)+b<>=1)>0); return y; :omg:
Henize wrote:
Check out the Microsoft research project called Singularity. Its an operating system written in C#. Around %99 of it(including the kernel) is written in C#.
Sun tried to do this with Java a few years ago. It failed and for good reason.
Henize wrote:
Yes, Java was a good start and had the right idea. C# can do more. .NET allows any language to be compiled to the .NET platform.
Don't get me wrong, there are some nifty features in the .Net framework. That said, managed languages are very limited in their scope of applications. Garbage collection, no matter how efficient the algorithm is, will always be slower than optimizing your memory usage based on how you are using it. I haven't done a whole lot with C#, but I do a lot in Java. C++ is like having a whole general purpose toolbox while Java/C# is like having a nice screwdriver set.
If you decide to become a software engineer, you are signing up to have a 1/2" piece of silicon tell you exactly how stupid you really are for 8 hours a day, 5 days a week Zac
-
Zac Howland wrote:
[brackets] Everyone knows they should always be placed on the same line!
You her a multitude of voices from all directions shouting: "For that you will burn on the stake, heretic!" ;P Thats what is fundamental about coding styles: When you have three programmers together, there are at least four stlyes assembled.
"We trained hard, but it seemed that every time we were beginning to form up into teams we would be reorganised. I was to learn later in life that we tend to meet any new situation by reorganising: and a wonderful method it can be for creating the illusion of progress, while producing confusion, inefficiency and demoralisation." -- Caius Petronius, Roman Consul, 66 A.D.
:laugh:;Phehe ... sorry, just couldn't resist
If you decide to become a software engineer, you are signing up to have a 1/2" piece of silicon tell you exactly how stupid you really are for 8 hours a day, 5 days a week Zac
-
Henize wrote:
Check out the Microsoft research project called Singularity. Its an operating system written in C#. Around %99 of it(including the kernel) is written in C#.
Sun tried to do this with Java a few years ago. It failed and for good reason.
Henize wrote:
Yes, Java was a good start and had the right idea. C# can do more. .NET allows any language to be compiled to the .NET platform.
Don't get me wrong, there are some nifty features in the .Net framework. That said, managed languages are very limited in their scope of applications. Garbage collection, no matter how efficient the algorithm is, will always be slower than optimizing your memory usage based on how you are using it. I haven't done a whole lot with C#, but I do a lot in Java. C++ is like having a whole general purpose toolbox while Java/C# is like having a nice screwdriver set.
If you decide to become a software engineer, you are signing up to have a 1/2" piece of silicon tell you exactly how stupid you really are for 8 hours a day, 5 days a week Zac
-
The Singularity project is going quite well. It is very interesting, check it out.
static int Sqrt(int x) { if (x<0) throw new ArgumentOutOfRangeException(); int temp, y=0, b=0x8000, bshft=15, v=x; do { if (v>=(temp=(y<<1)+b<>=1)>0); return y; :omg:
Henize wrote:
The Singularity project is going quite well. It is very interesting, check it out.
When they produce an OS that is both stable and efficient enough to compete with a native OS, then I'll look more into it. As it stands right now, it is just another in a line of attempts to use "lazy languages" to do things they weren't designed for. I've seen this fail twice in the past (with both Sun and Microsoft).
If you decide to become a software engineer, you are signing up to have a 1/2" piece of silicon tell you exactly how stupid you really are for 8 hours a day, 5 days a week Zac