[Catalyst] Re: Content-Disposition filename and delayed download

[Bill Moseley]

Very sorry for the duplicate -- was meant for another list.  Crap, now you
have to delete this one, too....

On Fri, Dec 3, 2010 at 5:49 AM, Bill Moseley <moseley at hank.org> wrote:

> Hi,
>
> What's the recommended approach to the filename used in downloads?  I wou=
ld
> like to allow users to specify the filename (or, say, use an existing
> user-provided name or title) which means I'd likely need to encode in utf=
-8.
>
> But, I'm not sure what conclusion to draw from looking at
> http://greenbytes.de/tech/tc2231/
>
> RFC2183 says the values should be 78 characters and limited to ASCII.  21=
84
> and then 2231 discusses encoding but it seems not well supported if I'm
> reading those test correctly.
>
> See: http://www.ietf.org/rfc/rfc2231.txt
>
>
> Now, say I have to dynamically generate the download and it might take
> longer than a normal web request.  In the past I've opened another window
> and used a meta refresh to poll the server until the download is ready.
>  Seems a bit old-school.
>
> Can anyone point me to an example of a better approach?  I've seem
> suggestions of using iframes or polling with AJAX.
>
> I've also see suggestions that it's important to not set no-cache headers
> for IE when sending the download.
>
>
> I'm not quite clear if I want to send Connection: close when polling.  The
> idea is I don't want to hold up a web server process when the client is
> polling for the download, but the load balancer should allow for keep-ali=
ve
> connection on the client side but not tie up a single web server process.
>
>
>
>
>
> --
> Bill Moseley
> moseley at hank.org
>



-- =

Bill Moseley
moseley at hank.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.scsys.co.uk/pipermail/catalyst/attachments/20101203/d9df7=
dae/attachment.htm