Parsing html document

TWebbrowser can be used, try something like this:

Start by including (importing from mshtml.dll) the MSHTML.pas file in uses.

Cast the WebBrowser.Document as IHTMLDocument2 or IHTMLDocument3.

var
 MyDocument: IHTMLDocument3;
 Tables: IHTMLElementsCollection;
 Table: IHTMLTable2;
 Row: IHTMLTableRow2;
 Cell: IHTMLTableCell;

MyDocument := (webbrowser.Document as IHTMLDocument3);

You can access all table elements within the document with:

Tables := (MyDocument.Body.All.tags( 'TABLE' ) as IHTMLElementCollection);

Then look for the element you like to find. This should be easy if it has a name, because you can use:

Table := (Elements.item('NameOfMyTable',EmptyParam) as IHTMLTable2);

Or you have to loop each one:

for I := 0 to Tables.length - 1 do
begin
 Table := Tables.item(EmptyParam,I) as IHTMLTable2;
 ...
end;

When you find the table, get hold of the Rows property and handle each row:

for J:=0 to Table.Rows.Length-1 do
begin
 Row := Table.Rows.Item(J,'');
 for K:=0 to Row.cells.Length-1 do
 begin
  Cell := Row( EmptyParam, K ) as IHTMLTableCell;
  StringGrid1.Cells[K,J] := (( Cell as IHTMLElement).innerText);
 end;
end;

Havent tried running this, but it should be about right...

There are also plenty of different parsers on the net.

If you want a parser that is not reliant on having MSHTML installed then I recommend Turbo Power Internet Professional it has a good parser (and browser component).  It's on sourceforge at http://sourceforge.net/projects/tpipro/

oops,

this:

StringGrid1.Cells[i+1, j+1]

should be :

StringGrid1.Cells[j+1, i+1]