Is there anything wrong with doing the following in C?

[SunnyD]

sprintf(stringptr, "%s", stringptr);

I know it appears redundant, however I'm using it in the context of a reparser/string builder... so there's a method to my madness. It "works", however I'm worried that under certain contexts it might blow up. I suppose I would really need to know how sprintf really works internally to answer that.

 

[Markbnj]

I don't think you'll have any problems. I can imagine scenarios where that might cause an issue, but I can't imagine sprintf behaving in a way that makes any of those scenarios possible. The easiest one to envision would be if the formatting takes place in a loop that reads and then writes one character at a time, and an input character generates >1 output characters (i.e. the write pointer gets ahead of the read pointer). Don't see it happening here.

 

[SunnyD]

Originally posted by: Markbnj
I don't think you'll have any problems. I can imagine scenarios where that might cause an issue, but I can't imagine sprintf behaving in a way that makes any of those scenarios possible. The easiest one to envision would be if the formatting takes place in a loop that reads and then writes one character at a time, and an input character generates >1 output characters (i.e. the write pointer gets ahead of the read pointer). Don't see it happening here.

In my implementation, that doesn't happen. I have a source string, and the destination string is always at least as big as the source string. What's happening is that in the loop that scans each character out of the source, if it encounters a character that needs to be escaped it then goes and reallocates a wider buffer for the destination string (wide enough for the original string + the number of characters that are being added from the escaping).

I finished it up, knocked out the bugs, and everything looks like it's working. I just wanted to make sure I wasn't going to have any nasty surprises from sprintf in this case.