goto... Who uses it?
-
In SQL - fairly often to jump to the error handler at the end of our sprocs. I'll admit there's no good reason we do this, since it's easy enough for us to avoid this with if statements, but it's a pattern used in our original code and so for consistency we stuck with it:
Create Procedure MyProc as
Begin Tran -- Do stuff... if @@error <> 0 goto errorHandler Commit Tran Return 0errorHandler:
Rollback Tran
Return 1cheers, Chris Maunder The Code Project | Co-founder Microsoft C++ MVP
The problem with that view ("for consistency") is that you will forever be in the same programming mode. The form of the code should be
if @@error <> 0
BEGIN
Rollback Tran
Return 1
END
else
BEGIN
Commit Tran
Return 0
ENDIn my world there are no exceptions! In C# I do accept the break statement within loops to allow an early escape. But no GOTOs. I programmed during the structured programming discussions in the 1970s. The amount of time required to comprehend GOTO filled code was enormous. And so easily fixed.
Gus Gustafson
-
DanielSheets wrote:
You are convinced that goto's and ternary operators are examples of bad programming.
Good grief. I didn't say ternary operators were bad; I said that since they are no longer needed in your specific code, they can be removed. What I learned was that the code
text == "" ? "0" : ...
does not throw an exception in C# if text is null. My view that gotos are bad is based on years of experience of finding bugs caused by their use (even in assembly where JMPs are a given and cause all sorts of problems when you aren't careful.) Condescension is in the eye of the beholder. In this case, you had a code review, extremely valid suggestions were made on how to improve it. You concede that much of what you are doing isn't needed, but insist on keeping it that way. That's fine, but don't go around blasting everyone for being condescending, a snob or not seeing the full picture.
DanielSheets wrote:
I don't care what you think
Apparently you do else you wouldn't reply. :)
I did have valid suggestions. And they were good suggestions. Was this part of your code review?
Joe Woodbury wrote:
"its" is spelled "it's" in this context, but it should probably read "it was".
-
"correct use" depends on your definition of "correct". However, the main point is that even if at first it makes sense, or even simplifies code, it suffers from the lack of long-tem maintainability: Unlike other control statements, goto lacks an associated block of control, and thus makes it considerably harder to detect or verify the effect of any change you make to code that may or may not be executed, once or repeatedly, depending on goto statements in potentially several entirely different place(s). Goto would be much less of a problem if jump labels weren't self-declaring and global: if you could localize the use of a label like variable names, that would greatly restrict the potential (ab)use of goto commands, and thus ease the effort to understand the code and verify changes.
Stefan_Lang wrote:
"correct use" depends on your definition of "correct". However, the main point is that even if at first it makes sense, or even simplifies code, it suffers from the lack of long-tem maintainability:
Just to make it clear I have 40+ years of programming experience and I started before OO existed and used languages where one had to use goto along with using languages where goto didn't exist. And a number in between. I understand not only how goto can be misused but how any number of constructs can be used incorrectly. I also understand how absolutes are never that. (Where the former is a technical point and the latter part is not.) And that is the context in which I made my response.
-
Eddy Vluggen wrote:
Readable code leads to maintainable projects. The better a project can be maintained, the more chances it will survive in the long run.
.... which is completely offset by his use of 'goto'.
If it's not broken, fix it until it is
Kevin Marois wrote:
.... which is completely offset by his use of 'goto'.
Nope. The code as presented was readable and was maintainable. Given that maintenance is a proven and known expense then I would much rather have the code that was presented versus some esoteric trick that is technically correct but neither readable nor maintainable.
-
This isn't a programming question. Anyway... I find it useful in very few situations. It can make for cleaner code if used correctly. Of course, it can also be over used.
-
Nemanja Trifunovic wrote:
assembly, -C- and FORTRAN
FTFY - rarely used in C either. (Edit - C with strikethrough looked too much like a Euro symbol)
Rob Grainger wrote:
rarely used in C either.
Take a look at any non-trivial C code base (Linux kernel, for instance) and you'll see tons of goto statements.
-
Been writing code since '85. Never used a goto in production code and I would have serious issue[s] with anyone who did.
Yes, quite. So you would have serious issue[s] with the people who wrote: jpeglib libsoap lua runtime mongoose gifencod etc. (just the examples I have at hand). Never had much of a problem maintaining/debugging a flow control that uses a label. Switch statements have probably caused more problems. But then I'm just being objective.
-
I can't ever remember needing it in c#. That said, it's been so long I doubt I'd spot a place where it might make things better.
Sort of a cross between Lawrence of Arabia and Dilbert.[^]
-Or-
A Dead ringer for Kate Winslett[^]This is when you would use a goto in C#. Basically a fall through in a switch statement.
-
I've not used it in a long time. There are a few of them in our legacy code, but no new code have them; they are mostly used for quick exit of a function to do cleanup.
Nihil obstat
In C, or C++, the only situation I can recall using it for was to break out of multiple nested loops.
for (int i=0; i
Of course, in Java, Fortran (and presumably C#), one just uses a labelled break statement (exit statement in Fortran) for this purpose - main difference being that the label goes at the beginning of the loop being broken out of, rather than the end.
-
Mauro Leggieri wrote:
For the other side, as an assembly programmer... I see JMPs everywhere hehehe.
..simply because there's little alternative in assembly. Assumed we weren't talking about assembly, but higher-level languages - there are few people still working professionally with assembly.
Bastard Programmer from Hell :suss: If you can't read my code, try converting it here[^]
Days ago started to program a driver and remembered me that, altough in some places you can see try/leave/except blocks is modern code, because is programmed in plain C, "goto" is still used a lot mainly for cleanup purposes.
-
I strongly disagree. This calls for splitting the function into two, if not more, pieces and/or rethinking the algorithm. I've seen code like this cause way too many bugs when something is introduced in the middle of BigFunction which doesn't get cleaned up at the end.
You have the right to disagree, I'm saying this is a case where I see goto have it's usefulness. Yes you can split into multiple function but sometimes there is 10+ parameters involved and creating subfunction doesn't make it clearer. I respect your opinion and your strict mode but don't get me wrong... all languages [more or less] have the goto keyword or an equivalent.