Well, I got it.
Is possible using a combination of these : r.MatchPos and r.MatchLen
for I := 1 to r.SubExprMatchCount do
begin
S := Format('line %d pos:%d len:%d',[I, r.MatchPos[I], r.MatchLen[I]]);
Memo1.Append(S);
end;
I can use StringReplace function to replace these values.
--
Marcos Douglas
On Sat, Sep 6, 2014 at 9:37 PM, Marcos Douglas <m...@delfire.net> wrote:
> Hi,
>
> I'm using RegExpr to parse some parts, just SOME parts, of a HTML and
> works perfectly. But, in one case I need identify a "pattern" to reuse
> using new "values".
>
> Please, see the example (using Lazarus) bellow:
>
> procedure TForm1.Button1Click(Sender: TObject);
> var
> InputStr: string;
> r: TRegExpr;
> begin
> r:= TRegExpr.Create;
> try
> InputStr := '<a href="/3604669/trabalho-28-07-2014-pg-1" title="Página
> 1, pg 1, 28/07/2014"> 28/07/2014 </a>';
> r.Expression := '\/([0-9]+)\/.*?pg\-(\d+).*?"';
> r.Exec(InputStr);
> ShowMessage(r.Match[0]); // = /3604669/trabalho-28-07-2014-pg-1"
> ShowMessage(r.Match[1]); // = 3604669
> ShowMessage(r.Match[2]); // = 1
> finally
> r.Free;
> end;
> end;
>
> The results are OK.
> I need to replace the position #1 and #2 with something like:
> #p1# and #p2#
>
> So, in other words, I need a result like this:
> Pattern = /#p1#/trabalho-28-07-2014-pg-#p2#"
>
> I tried to use r.Replace() or r.Substitute() but I think these methods
> don't do I want.
>
> Is possible?
>
> --
> Marcos Douglas
>
_______________________________________________
fpc-pascal maillist - fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-pascal
