Re: sscanf() and fscanf() ready for testing.

At 10:24 PM 6/2/00 -0400, clayton collie wrote:

i have finally gotten sscanf() working (fscanf() is 99% the same code, so it should be ok). preliminary testing is going well, but i'd like to open the code up to wider exposure before actually submitting it. (BTW how do you get cvs access ?)

You fill in the CVS submission form at http://www.php.net/cvs-php3.php

    any brave souls willing to have a look ? i'm afraid im not too

imaginative with test cases.

Just from your description I'm not sure your approach was the right one. sscanf() should not be a copy of 99% of the code in fscanf() but if possible the code should be generalized so that sscanf() and fscanf() can both use the same code. How difficult is this? (I haven't actually looked at the code myself). It would make maintenance much easier and if someone fixes a bug or feature in fscanf() (or vice versa) it would automatically fix sscanf().

i have full source, user-level docs, instructions on integrating the source, a test php script and a (small) file with a set of test cases. i can attach to the NG, email, or stick a copy on a web site somewhere.

Where is the fscanf() implementation? I don't see it in the source tree (I must be not looking well enough). Andi --- Andi Gutmans <andi@zend.com> http://www.zend.com/