Re: cvs: php4(php_4_0_4) / configure.in /main php_version.h

At 10:31 4/12/2000, Thies C. Arntzen wrote:

On Sun, Dec 03, 2000 at 11:48:25PM +0200, Zeev Suraski wrote: At 23:45 3/12/2000, Sterling Hughes wrote:

I think that's why Sascha wanted the whitespace changes separated from the "actual" changes, so we could inspect the "actual" changes more carefully before they become a part of the release branch. It saves us from wading through the commit to see which changes are whitespace related and which changes are actually signifigant if the whitespace changes are removed.

Hmm, well, I think that it's fairly easy to ignore lines that are identical except for added spaces.

    i disagree. i bet that i can could you a ~100 line diff with
    mostly whitespace changes and even you will
    overlook something important.

I'll take that challenge :) I've read plenty of whitespace-intensive patches in the past, and reading through them was very easy. The reason the whitespace patch (which is a bad name, as there were other 'look and feel' changes there) was committed at the same time as the logic fix was that while patching in the logic fix, the mess in that module annoyed me and slowed me down (I'm weird, I know). I was half way through the logic fix already, and that's why they were combined. I think it's an ok guideline not to commit unrelated changes together (in general), even though it's often not very practical, and thus it should be a guideline, as opposed to a rule. Zeev -- Zeev Suraski <zeev@zend.com> CTO, Zend Technologies Ltd. http://www.zend.com/