Re: Any Gotchas on using FTPAPI on PubNet?

[Russ Khoury <rkhoury@xxxxxxxxxxxxxxxxxxxx>]

   Mike:
   One of my clients connects with PubNet, running on a PC. I haven't
   moved it to use FTPAPI yet.
   I believe the double login is for a proxy (there was a thread on
   proxy's recently, so that may help).
   For those not familiar with it, PubNet is the book publishing
   industry's EDI - for document exchange between booksellers and
   publishers. Use to be a proprietary format (BISAC), but now uses X.12
Russ Khoury

   On 7/20/2011 10:44 AM, Mike Wills wrote:

I was just asking hoping that someone has worked with it before.

My guess is that they wrote a custom FTP server that is compatible
only to the point where an FTP client can connect to it. There is a
double login and appears to only support a couple commands. Here is
the documentation they provide: [1]http://murl.me/1g2. Also the error
codes: [2]http://murl.me/1g3. I'll just have to wait and see if I can get
some test information so I can play.

--
Mike Wills
[3]http://mikewills.me



On Wed, Jul 20, 2011 at 8:44 AM, Scott Klement [4]<sk@xxxxxxxxxxxxxxxx> wrote:

I'm not familiar with PubNet, I'm only familiar with the FTP protocol.
Technically, there's no such thing in the FTP protocol as a 'get'
command. When you type 'get' into an FTP client, the client understands
that you want to retrieve a file.  From your command line, it extracts
the "local file name" (in your example, 850file.txt) and uses that to
open a file on it's end.  Then it sends the RETR command to the server,
followed by the parameter you specified as the remote path name. (* in
your example)

Here's a description from RFC 959 (the standard that governs FTP) for
the RETR command:

RETRIEVE (RETR)
This command causes the server-DTP to transfer a copy of the
file, specified in the pathname, to the server- or user-DTP
at the other end of the data connection.  The status and
contents of the file at the server site shall be unaffected.

So according to RFC 959, the parameter to the 'GET' command is a path
name to the file to retrieve on from the server.  Not a wild card!  And
only a single file can be returned.  Furthermore, the files on the
server side are never modified/changed by a GET request.

Not sure if that really helps you at all...   honestly, your question
isn't really for us, but is for the people at PubNet.  You want to know
what their software will do with the * character you sent -- and I can't
help you with that, I didn't write their software.


On 7/19/2011 11:15 PM, Mike Wills wrote:

I am starting work on automatically downloading EDI documents from
PubNet via FTP.

Their sample shows doing a "get * 850file.txt" does that join all
documents that are sent into one document for parsing? That is what I
am understanding it does from the documentation.

-----------------------------------------------------------------------
This is the FTPAPI mailing list.  To unsubscribe, please go to:
[5]http://www.scottklement.com/mailman/listinfo/ftpapi
-----------------------------------------------------------------------

-----------------------------------------------------------------------
This is the FTPAPI mailing list.  To unsubscribe, please go to:
[6]http://www.scottklement.com/mailman/listinfo/ftpapi
-----------------------------------------------------------------------

References

   1. http://murl.me/1g2
   2. http://murl.me/1g3
   3. http://mikewills.me/
   4. mailto:sk@xxxxxxxxxxxxxxxx
   5. http://www.scottklement.com/mailman/listinfo/ftpapi
   6. http://www.scottklement.com/mailman/listinfo/ftpapi
-----------------------------------------------------------------------
This is the FTPAPI mailing list.  To unsubscribe, please go to:
http://www.scottklement.com/mailman/listinfo/ftpapi
-----------------------------------------------------------------------