ID:1484728
 
Applies to:Website
Status: Open

Issue hasn't been assigned a status value.
The following post shows an interesting example of downloading binary data using world.Export():

http://www.byond.com/forum/?post=1484404

The DM reference is not completely clear on this usage, so my request is to have it updated.

Alternatively it may be a good idea to put the reference in a wiki format somewhere so that others can contribute to it. I'll develop my own BYOND wiki software if it can be added as an official resource.

Thanks!
We had a wiki style reference that we removed due to, I think, performance problems, but it might be trivial to reinstate that. We'll take a look.
http://www.byond.com/members/?command=reference

This, Tom? It's been broken for a while.
I have a working implementation of MediaWiki that uses BYOND keys, but on my current connection I'm unable to upload it for testing, but once I manage to get it uploaded it can definitely be used for purposes of dynamic reference material.


There was also the Bwicki2 and Bwicki DMCGI programs that never really gained any steam.
I have been messing around with this method to download binary data, and unlike what I have been told, it seems pretty stable and reliable, I haven't tested with large files yet, so I can't tell if it would corrupt the file somewhere alongside the process.

Tom, may you bring in some clarification on this matter?
It would be limited to GET sizes, which are 65535 bytes including the various headers and whatnot.
If we're talking about the linked use, I don't think that applies since the DMB/RSC aren't actually in the GET headers, but rather are the content requested - unless world.Export() limits what it will recieve to 65535 bytes, you mean.
world.Export() to a web-server uses the GET protocol, which is limited to 65535 bytes on what it can send and receive, I don't think it would support anything bigger than that properly but I could be wrong on how the incoming data is handled as opposed to the outgoing stuff (which is definitely limited on size). Easy enough to test it though.
I only mention it since it seems that nobody posted in that thread about failing to download anything, and it seems unlikely that it never, ever, got tested with files over 65k.
It seems there are cases where it fails to download the file.
I couldn't be able to retrieve an .xml file, neither a .zip.

Instead it seems it is successfully download .dmb and .rsc files, as I have been experimentin with and I can tell it -almost- works fine.

I wish a better implementation of this method would be possible in future v500 updates, if possible.

I guess i am not the only one that is interested on this.