🚀 go-pugleaf

"Alim Akyol" <akyolalim@yahoo.com> wrote in message
news:xn0esw68cdluat000@forums.borland.com...

> Is there any component for this purpose?

There are several.  Indy (http://www.indyproject.org), ICS
(http://www.overbyte.be), and Synapse (http://www.ararat.cz/synapse) all
have HTTP client components available.  Or you can use Microsoft's WinInet
or WinHTTP APIs (Internet Explorer uses these internally).

> Can I use IHTMLDocument2 interface to load pages without
> using TWebBrowser?

You can try creating a standalone instance of the IHTMLDocument2 object and
then load HTML into the document as needed:

    IHTMLDocument2 *pDocument = NULL;

    if( SUCCEEDED(CoCreateInstance(CLSID_HTMLDocument, NULL,
CLSCTX_INPROC_SERVER, IID_IHTMLDocument2, (LPVOID*)&pDocument)) )
    {
        // use pDocument as needed ...
        pDocument->Release();
    }

However, without a valid browser to host the document, you will not be able
to directly navigate the document anywhere.  You would have to either
download the HTML manually and then load it into the IHTMLDocument2 via its
IPersistStreamInit interface, or else use CreateURLMoniker() to create a
moniker for a URL and then load it into the IHTMLDocument2 via its
IPersistMoniker interface.

> I could use TIdHTTP but I have a doubt about it. I don't want to
> parse pages, and handle frames, cookies, SSL pages, etc.

You will have to, using any of the above options, because they are all
one-resource-at-a-time downloaders, so you have to parse the HTML to figure
out which additional requests to make.  TWebBrowser (and thus IE) have to do
the same thing internally.


Gambit