Giters

sabberworm / PHP-CSS-Parser

A Parser for CSS Files written in PHP. Allows extraction of CSS files into a data structure, manipulation of said structure and output as (optimized) CSS

Home Page:http://www.sabberworm.com/blog/2010/6/10/php-css-parser

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Auto calculating box models

jarodium opened this issue · comments

commented

Is it possible to calculate automatic widths and heights of the elements?

commented

Could you give me an example of what you mean? I don’t think this (if I understand you correctly) is possible solely with knowledge of the CSS without knowing the DOM.

commented

I think you have it right.
I was thinking of forcing the user to have a width definition on a root element. That way, the parser should assume 100% is the root element.

Example:

<div style="width:1005px">
       <div id="sub1">

       </div>
</div>

Let us make some assumptions based on a "reset.css". Margins and Paddings are defaulted at 0, but this first div must have a width. Now, assuming that there are no widths, all elements below must have 100% width.

If the root element has a 1px border, then 100% for the child divs would be 1003px.

I'm trying to find the best possible solution for converting a fully styled webpage to PDF, but the converters out there lack in CSS parsing so I was thinking of mashing your library against a PDF generator, with PHP.

Hopefully I make sense.

Regards

commented

So you could use PHP’s DOMDocument::loadHTMLFile to construct a DOM from your HTML. Unfortunately, there is no CSSOM in PHP. And you’re missing the glue to combine that DOM with a PHP-CSS-Parser Document object. For this to work there are two important parts missing:

  • A way of matching DeclarationBlock objects to DOM elements using their selectors. Selectors currently aren’t in any way parsed. So you’d need to be able to adopt the parser to understand that it’s parsing an element selector, class selector, id selector, and how the selectors are nested. You’d then need to be able to check whether a specific DOM nodes matches a selector or not.
  • The ability to apply cascading rules to DOM elements to have all calculated styles on every element. This includes converting relative units like em and % to their absolute counterparts by taking the parent element‘s calculated style value.

As a result of this, you’d have a tree of DOM nodes where each node has a complete set of calculated styles which you then could use to draw your PDF. You could choose to only apply certain styles like width, height, font-size and so forth or go for a complete implementation.

If you decide to go forward with this plan, I’d be very interested to incorporate the results of the selector parsing back into my code as that has been on my to do list for a long time. I wish you the best of luck with this endeavour.

commented

There is actually a DOM query library that takes CSS selectors. Maybe you could use that somehow and skip the step where you improve PHP-CSS-Parser’s selector parsing.