[Info-vax] Streaming a File on OpenVMS with Caché

[David Froble]

Bill Gunshannon wrote:
> In article <bd6e7824-f4c1-4856-b4e8-0d60e6e7af3d at googlegroups.com>,
> 	Steven Schweda <sms.antinode at gmail.com> writes:
>>> Seems to me there is.  The recieving end has a program
>>> that, obviously, expects a particular input.  Tell the
>>> sending side(s) the format that the data needs to be in and
>>> don't accept anything else.
>>    I love your customer orientation, 
> 
> I am very customer oriented (you should try working with the datacenter
> here!!) but reality is reality.
> 
>>                                      but please explain how a
>> remote user on a non-VMS system can determine the record
>> format of a file when it arrives at (or is reconstituted on)
>> the VMS system.
> 
> If you tell him the format it needs to be in and how to transfer it so that
> it will not get changed along the way you won't have to "reconstitute" it.
> 
> Now, we really haven't been told just what the data is so this is really
> all just academic.  But in cases like this the best bet is to make the
> data an human readable, ASCII text file and the use ASCII mode in FTP
> to transfer it.  That will even work on VMS. :-)  
> 
>>> [...] the only way [...]
>>    In my experience, statements like that are practically
>> always wrong.
> 
> You must not work with real users much.  I used to get classlists from
> the faculty for the creation of accounts.  Supposedly, they all used
> the same function from a web page to get their classlists.  And yet,
> with 10 faculty members I got classlists in 7 totally different formats
> in many cases not even containing the same fields.  As you might imagine,
> this made automating the process of adding new users somewhat problematic.
> The solution was, as stated, to get the list of new students all in one
> format (in my case CSV) in a single file, in a specified layout.  Has
> worked find ever since!!
> 
> bill
> 
> 

I do agree that having a set of formats that are acceptable is reasonable.

However, I can imagine some cases where the entity receiving the data 
wants the data more than the sender wants to send the data.  In such a 
case, the receiver just might have to take what is sent, and feel lucky 
to get it.

Since the OP >>STILL<< (as far as I've seen) hasn't described much of 
what's actually going on, everything is speculation, and I'm not sure 
why the topic is still active.