is it possible to call mapping in a c# Program?

There are some utility tools for pipeline and FFDASM.exe, etc in SDK/utiliti
es. and also it is possible in visual Studio to test the map by right click.
I would like to right a program that uses the map without calling the BizTal
k orchestration. It is working if I just define send and receive ports with 
a filter, but I wanna use it in a C# code, that I can call it frequently.

Any ideas? thanks in advance,
metin

[ Post a follow-up to this message ]

A map is basically just an XSLT.  In which case, it would be quite easy to
use it inside C#.  If you use functiods inside the map, you might be able to
use debug process, to step thru the flow, and see what Interface or function
Biztalk uses to call it.  I have never tried this, so it is only a guess.

--
Todd Sussman
My opinion is neither copyrighted nor trademarked, and it's price
competitive. If you like, I'll trade for one of yours.

Remove The REMOVE! to reply.

"metin" <anonymous@discussions.microsoft.com> wrote in message
news:336815E9-D889-4E02-A8C4-416446C44227@microsoft.com...
> is it possible to call mapping in a c# Program?
>
> There are some utility tools for pipeline and FFDASM.exe, etc in
SDK/utilities. and also it is possible in visual Studio to test the map by
right click.
> I would like to right a program that uses the map without calling the
BizTalk orchestration. It is working if I just define send and receive ports
with a filter, but I wanna use it in a C# code, that I can call it
frequently.
>
> Any ideas? thanks in advance,
> metin

[ Post a follow-up to this message ]

thanks, I'll try that.
what I am trying to do is actually;
-reading a flat file,
-converting to XML,
-use map (xslt) to convert
- write output into a file again as a flat file.

The reason for that: There are many of different files like that and these f
iles are bigger than 10 GB. Instead of using my current orchestration, I wan
t to use only the FFDAsm, map and a C# program which controls the whole thin
g without going to the Data
base. This helps also developer who can only use Visual studio to create sch
ema and map instead of writing custom C# code for every mapping.

-metin

[ Post a follow-up to this message ]

Any XML file greater than about one megabyte is VERY hard to do XSLT on.
XSLT requires the XML to be loaded in a DOM.  This means that your gigabyte
file would need to be expanded exponentially growing in memory.  This is
physically impossible to do.  (BizTalk won't help you there either, this is
just the nature of those technologies...)

Any transformations you'd like to do should be done programmatically in a
streaming way.

Kind regards,
Christof


"metin" <anonymous@discussions.microsoft.com> wrote in message
news:5EDB1A98-E81A-44AB-A0B8-AD4FCF4B03DD@microsoft.com...
> thanks, I'll try that.
> what I am trying to do is actually;
> -reading a flat file,
> -converting to XML,
> -use map (xslt) to convert
> - write output into a file again as a flat file.
>
> The reason for that: There are many of different files like that and these
files are bigger than 10 GB. Instead of using my current orchestration, I
want to use only the FFDAsm, map and a C# program which controls the whole
thing without going to the Database. This helps also developer who can only
use Visual studio to create schema and map instead of writing custom C# code
for every mapping.
>
> -metin

[ Post a follow-up to this message ]

Actually, we are procesing right now 2 GB file in 2 hours, by splitting it i
nto small chunks. remember the question regarding costom properties, that wa
s for that problem too.

this is true, we also observed that files upto 5 MB are being processed very
 quickly in our machine (4 Processor server). It is also processing upto 100
 MB files in accepted time frame but the bigger files, no chance!

I am thinking just for the transformation of some of the large files files(1
0GB), I don't need the orchestration part. I can read the file as a file str
eam and build small chunks and let them be transformed and then write again 
into output file, read the
next chunk. In this way I will use the same mapping and schema utilities. Be
cause the pepole who are creating maping and defining the schemas are differ
ent than the ones who are developing. So if I create a custom devlopment the
n Maintenance will be not s
o easy, because then the people will need C# or other programming language e
xperiences, rather than just using the Visual studio for Mapping.

I wrote actually similiar program on top of the SDK\sample\ direct submit, w
hich is slower than file adapter.

If I would do it for myself I would write a PERL script and everything would
 be easy :-)))

Regards,
Metin

[ Post a follow-up to this message ]