Re: Re: HTML_Table: thead, tfoot and tbody

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 8/3/2005 11:50 AM, Christian Weiske spake the following words: >>But this would mean that there needs a bigger change in the internal >>API. Currently (almost) table all data is stored in $_structure. Now we The internal API really doesn't matter. In fact, I think it can be optimized. This would probably be a good time to go at it. Maybe we can also plan working on an E_STRICT php5 HTML_Table2 release, taking advantage of any new features in php5 that seem useful. > Don't think so. All which had to be changed is the > setHeaderCellContents() or how its called - the setcellcontents can stay > the same, setfootcell... is totally new. So we need only the switch in > setheadercontents to store in the body ($_structure) or the new head table. Let's call it an array :) Yes, I think that should be doable. And if setHeaderCellContents (or whatever) defaults to behave properly when the switch is set in the constructor, that sounds like a great idea to me. We did something similar in HTML_CSS when we decided to allow duplicates. Without the header switch, everything worked as before. With it, the API changed only slightly, mostly being a matter of the toArray() output. The internals were, of course, completely changed. So I'm convinced this is a doable solution probably the best way to solve this problem. Switching rows between the 3 tables isn't a good idea, imho, so let's not implement that :) >>need either three arrays or an extension of $_structure that allows row > > +1 for three arrays. I'm +1 for arrays, too. > And the output function has to be changed, but that's obvious. That nead not bother us that much :) Klaus -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (MingW32) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFC8JappSl7k8JGkr4RAshPAJ9o6n4cTLCT3wzU+eCLIXJbXmlOOACgrxmH XX7R1ZTka0ZrZNqjvz07fgs= =zWkX -----END PGP SIGNATURE-----