Minification of margin and padding properties about csso HOT 7 CLOSED

No, it cannot be minified like this because 'not set' does not necessarily mean '0'.

from csso.

OK, but may give an option to choose this minification metod?

from csso.

What do you mean? not set -> 0 is an invalid assumption, optional or not. Let's say you have two classes: .test { margin: 3px } and .test2 { margin-top: 4px; margin-right: 4px; margin-bottom: 4px }, you apply them both to an element: div class="test test2". In this case, test2 overrides all margin properties set in test, except for margin-left which is 3px. If test2 was minified like you suggested, we'd end up with the wrong value of 0.

All the minification done by CSSO strictly adheres to the CSS specification. Perhaps our docs are misleading when we label basic minifications as 'safe' and the more advanced ones as 'structural' - actually all of them are safe and must stay such. I'll update the docs.

from csso.

I'm agree with Leonya, it is very unsafe to minify margins this way.

In any case the matter of CSS compressing is to compress CSS, not to correct it or to insert some absent values, sorry.

from csso.

Leonya: at your case you're absolutely right and I totally agree with you. But my case is different from the general use. I do not need a safe minification. I know what CSSO will do because I'd allow to do that.

at *nix, you can remove files with force metod "rm -rf", that look like what I want. Advanced users may want too.

I forked the project. I'd add that option for me :) Thanks anyway, CSSO works perfectly.

from csso.

hmert: keep in mind that you will have to fork it once again in september, it is CSSO2 in active development now — completely another codebase.

Please, comment this issue with your option implementation link after you will implement it. We will try to rethink this issue at least. :) Thank you.

from csso.

hmert: well, feel free to implement it in your fork, but I still don't get your reasoning. If you are going to use it for your own HTML/CSS which is "immune" to this problem, then why don't you write the CSS in the shorthand form in the first place?

from csso.