Downloading filename problem

[Ness Mountain]

I need to have php download a datafile to use locally, and using the
following code it will download it, but for some reason it ignores my
filename="ticket.csv", and it just insists on calling the file by the name
of the php program.


header("Content-Type: application/force-download");
header("Content-Type: text/comma-separated-values");
header("Content-Type: application/download");
header("Content-Disposition: attachment; filename=ticket.csv");
header("Content-Transfer-Encoding: binary");

I see from other postings that browsers are sometimes grouchy about this,
but I'm not getting "ticket.csv.php" -- I'm just getting the name of the
program file, box_office.php. I need to be able to set the default
filename.

Doesn't filename= work at all??? Thanks for your help!

Ness Blackbird

[Gary Petersen]

On Sat, 30 Aug 2003 02:38:24 -0500, in message
<Xns93E7724547F3nesswillowmountainco@129.250.170.8 4>, the AI program named
"Ness Mountain" <dontspamness@willowmountain.com> randomly printed:

> I need to have php download a datafile to use locally, and using the
> following code it will download it, but for some reason it ignores my
> filename="ticket.csv" [...]

According to the HTTP spec (RFC 2616 section 19), you
should only need the "Content-type: application/octet-stream"
header to cause the "Save response As..." dialog to
be displayed, and the rfc also seems to indicate that
the filename parameter is just a suggestion; I don't
think that browsers have to respect it.

[Ness Mountain]

Gary Petersen <garyp1492@remove.meearthlink.invalid> wrote in
news:pan.2003.09.01.05.48.17.29287.426@REMOVE.MEea rthlink.INVALID:

>> I need to have php download a datafile to use locally, and using the
>> following code it will download it, but for some reason it ignores my
>> filename="ticket.csv" [...]

>
> According to the HTTP spec (RFC 2616 section 19), you
> should only need the "Content-type: application/octet-stream"
> header to cause the "Save response As..." dialog to
> be displayed, and the rfc also seems to indicate that
> the filename parameter is just a suggestion; I don't
> think that browsers have to respect it.
>

Well, that's a bummer!!!