Forum Replies Created
-
Posts
-
Hi,
The answer to your first questions is no: BizTalk flat file schema wizard does not support the use of regular expressions in specifying the tag identifier.
The answer to your second question is yes: there are other approaches to work around this issue, see the following link to MSDN forum thread for one example:
Good luck.
Hi
I dont think the way you are trying to work out this problem will take you through. BizTalk would not be able to understand what kind of a record it is as the records 2,3 ,4 are unbounded. You need to have a tag identifier with each of the record in order to let BizTalk know what kind of record it is.
So if your record begins with 002, 003 and subsequently you have to go on the record node and specify the tag identifier. This will help BizTalk identify the record correctly. When you do this remember to reoder your records correctly as tag identifier will not be in the output xml for the flat file.
Puneet
Hi Chas
What I would suggest is not to use a custom pipeline in the receive location, you could subscribe the message into and orchestration and then call both the pipelines in that orchestration to do further processing. Else you can have two orchestrations that subscribe to this message and call the individual pipelines in each of the orchestration.
If you still are bent on using the pipelines on the receive locations, then there is another thing that you can do (I would not recommend it though). You can have two send ports that subscribe to this incoming message (which will of course use a passthrough pipeline) and have them output it to two different locations, from where you can get them into BizTalk with the help of two receive locations using two different pipelines that you wanted to use in the first place.
Puneet
But it is passively waiting for a website to send the xml message, right? How the Biztalk actively get the xml from the website? Thx
Hi,
Expose a wcf service in BizTalk receive port with the xml message you are expecting from website.
Where I can find the schema of the Error Message such that I can create in Orchestration catch exception block? Thanks.
For an orchestration, you can implement a Scope shape with one or more Catch Exception shapes along with it, and use a Send port to output the error message if the flow of the orchestration falls into one of the Catch Exception shapes.
Hi Robert,
You could consider using caching technique in your BizTalk solution, here's an article explaining that topic:
http://www.codeproject.com/…/Considerations-for-caching-in-BizTalk
Good luck.
Hi
Well i know it's a bit unconventional :). The data that i'm fetching can only be done with this assembly and while mapping the incoming to the outgoing message i realized that it would be convenient to call a method in that assembly to get the specific value i needed in the mapping (to complement the outgoing message with that data). This can be done with only one script functoid and a few other functoids.
The alternative is to do the traditional way e.g assign the data from assembly to a mesage, ad the schema to the map and it becomes a multimap and then do the mapping. I find using multimapps to be cumbersome to work with when you need to do changes to a schema or replace it and the mapping itself would become more complicated. The data in the assembly is stored in a XDocument (x number of records) and i query that document for the value i need (given certain parameters), achieving this in the map is what makes it "complicated".
i could change the assembly to only give one record as a result of a query to the data source but the query takes time(and i would need to do it many times in the map) and that is why i need to pre fetch it in the orchestration and store the records in a variable in the assembly to later fetch with a method that querys the XDocumet and returns the value i need.
i have come up with a middle solution but not as simple as what i was trying to achieve nor "complicated" as the traditional way. But if you have any idea on how to achieve this please share 🙂 (i suppose one could use some reflection code to find the instantiated object but feels like a bit forceful ).
Regards Robert
Why would you want to do that, unless you are doing something to that data in the orchestration, before trying to access it in the map? If you are just fetching that data in the map, you don't need to instantiate the object in the orchestration.
Hi,
If you just want to implement a scenario of content-based routing without orchestration, you can do what the following post suggests:
