--===============1878693146==
Content-Type: multipart/signed;
  boundary="nextPart1498174.Y9K0EDNxZn";
  protocol="application/pgp-signature";
  micalg=pgp-sha1
Content-Transfer-Encoding: 7bit

--nextPart1498174.Y9K0EDNxZn
Content-Type: text/plain;
  charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

Am Dienstag, 7. Juni 2005 13:09 schrieb Frans Englich:
> On the topic of parsing PHP and so forth; one way is to let the builder(if
> you use KHTML, you could inherit from KHTMLDocumentBuilder such that your
> builder can play nice with KHTML), and let it build specialize processing
> instructions, "QuantaPHPProcessingInstructionImpl", for PIs with name
> "php", and register each such node in a list in the "QuantaDocumentImpl".
> In this way you can quickly loop over all php instructions(without having
> to walk the tree), and access the PHP code as strings, which you can invo=
ke
> a PHP parser on(and attach any data structures created from the PHP parsi=
ng
> on the QuantaPHPProcessingInstructionImpl, perhaps).
>
> This is a difficult and confusing topic!

This is even more difficult. As you can do sick things with php like

<?
function foo() {
  echo 'href=3D"'.fancyUrl().'">'.fancyDescription()."</a>"
}
?>
=2E..
go to <a<? foo() ?> and bar it ...

you'll get a valid document in the delivered output, but you'll never get a=
=20
valid xml/html document on the server side. Hence, you cannot represent PHP=
=20
processing instructions like static nodes within the DOM tree.

mfg
	Leo

--nextPart1498174.Y9K0EDNxZn
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)

iD8DBQBCp2Hxj5jssenUYTsRAg2pAJ0WKuloXPk0e4Fbg37cDqTSQE+QiQCcC73K
Ir4+sQOTa/C8NDYExqTWanI=
=k3v7
-----END PGP SIGNATURE-----

--nextPart1498174.Y9K0EDNxZn--

--===============1878693146==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
quanta-devel mailing list
quanta-devel@kde.org
https://mail.kde.org/mailman/listinfo/quanta-devel

--===============1878693146==--