Audience:
Customers/Partners who want to customize the CRM Connector for GP to
support update/delete of new entities (entities which are not supported
out-of-box). It is expected that the user performing these steps
is aware of the map creation process in BizTalk Server 2004.
Scenario:
You wants to integrate new entities (example: CRM Lead and GP Prospect)
and support create/update/delete operations asynchronously.
A “Create” map is comparatively easier but for a "Update" or "Delete"
map the primary key of the destination system will be required. The primary key in the destination application is required to determine which entity instance needs to be updated or deteled. (For
example: to update the CRM Lead on update of a GP Prospect, the CRM
leadid (primary key of CRM Lead) will be required to identify which CRM
Lead needs to be updated)
This piece of information can be easily queried from the CRM Connector
for GP database, if you know how links are managed by the CRM Connector
for GP.
Example:
Consider you need to create an "Update" map between CRM Lead and GP
Prospect. PROSPID in GP and leadid in CRM are the primary keys for
these entities. When an update on a GP Prospect occurs an xml for that
prospect reaches the CRM Connector for GP which has the PROSPID
populated in it. In the update map, you need to keep an advanced
functoid between PROSPID and leadid. This functoid will query the CRM
Connector for GP database to determine which leadid is linked with the
PROSPID. I suggest you create your own dll which has a
public method which can query the link. Then in the advanced script
functoid you link this dll and the method.
Link management in CRM Connector for GP database
Intentityinstance(Table
in integration database): Primary key is instanceid. Each row will
correspond to one entity instance of an application.
“appentitykey” will contain the information about the primary key of
that entity instance. So all the integrated entity instances of either
CRM or GP will have entries in this table.
Intentityinstancelink(Table
in integration database): Instanceid1 and instanceid2 of this table are
corresponding to instanceid of Intentityinstance table. Linkid is the
primary key. So this table is actually just linking two instances of
Intentityinstance table.
So
in our example, to find out leadid in CRM, first query
Intentityinstance table where appentitykey contains PROSPID and the
value of PROSPID. So you have the instanceid of the prospect instance.
Now
query intentityinstancelink table where either instanceid1 equals to
“the instanceid of propect instance” or instanceid2 equals to “the
instanceid of propect instance”. If the link exists you will find one
link in this table. If instanceid1 matches then you take instanceid2 or
vice-versa. So you have “instanceid of lead instance”.
Now
query intentityinstance table again to query the lead instance.
Appentitykey will contain the leadid guid. Parse and extract the value
and copy this to the leadid node of crm schema in the map.
I
am attaching a code snippet(in C#.Net) which will give the feel of
how you can query the integration database. This code can’t be
used directly(as variables and references are not specified), I suggest
you understand the logic and modify the code to suit your purpose.
