Sterling, thanks for you help!
I tried the code you sent and I had to add /wait to the open/direct
or else it timed out just like everything else.
I also found out that it is still not receiving a page and then writing it out,
it still tries to grab the whole thing into memory and then writes out the
whole thing when it has it all, but at least it's not timing out.
Finally, it actually happens to be an .rm non-compressed realplayer
file that I am trying to download and it's several MB in size.
I haven't yet tried to hack it to parse it's header but I may have to
do that so I khohw what the size of the file is. Unless you know
another trick to querying the size from the server, then I guess
that will be necessary, because otherwise I guess it just never
knows when the file is done. Apparently there's not going
to be any signal on the stream signifiying its end,
unless maybe the http server just disconnects?
Well, anyway, do you have a recommendation for bringing
in some stuff? Do I definitely need the /direct? Is rebol
still doing some buffering? (which would be a good and
necessary thing I should think.)
May you live to be 100!
-galt
From: [EMAIL PROTECTED] on 06/12/2000 02:24 PM
Please respond to [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
cc: (bcc: Galt Barber/LearningTree/US)
Subject: [REBOL] downloading big files Re:
You need to use OPEN/DIRECT as READ will do the entire page read.
try a: open/direct http://whatever-you-like
while [data: copy a] [write/append %output-file data]
Sterling
> Again, if there is no ftp available
> but the file is there as http:/something
> and you want to page it out to the disk
> as it comes in because it is so large,
> what does the code look like for that?
> Do you still use read? Do you need use open?
> Do you need to open the port directly and
> make a Get request or something for that url?
>
> Thanks for your time!!
>
> -galt
>
>
>
