In migrating to CRM 2011, we're discovering different developers used different capitalization of custom entities and custom fields. This creates a headache for custom programming using the early bound methods. http://msdn.microsoft.com/en-us/library/gg327844.aspx. Is there a way to normalize the entity/field names before (or after) the migration?
1
votes
If renaming isn't possible via a "supported" method, perhaps its possible to generate the entity classes using only lowercase with CrmSvcUtil.exe? I see no parameters however. While CrmSvcUtil generates per schema, I observed the following SDK function (others too I assume) require using lowercase: RetrieveAttributeRequest.LogicalName
– roadsunknown
1 Answers
2
votes
As far as I'm aware. The only way to achieve the correct capitalisation you desire is to recreate the entities with the appropriate names.
Before or after the migration is mostly the upgrade of the CRM Server installation and modifying the database schema to reflect the upgrade, while still maintaining the current data and customisation data.
Thats as far as the "Supported" spiel goes.
As for an actual workaround. If your looking to upgrade anyway I'd be tempted to restore your current CRM 4 system to a test domain. Then look at how feasible it is to change the schema names in the actual "untouchable" crm database. I believe there is a MetaDataSchema.Entity table where this is centrally stored so i'd test this to see how usable this is and what impact it has on say the webservice.
So you face a similiar choice that I face on multiple occasions while working with Dynamics CRM. Go with the supported way, or a bit of "Yee-Ha" development. Sorry it's probably not what you want to hear!
Edit:
In regards to what to change I can't say for sure as I haven't got a CRM 4.0 system to hand only a 2011 at this moment in time. However as an example, there will be a MetadataSchema.Entity table in the [OrganisationName_MSCRM] Database. Of which certain columns will jump out. Name, PhysicalName and Logical Name.
Logical name is the one that CRM users which it defaults to lowercase no matter how you enter it.
I believe PhysicalName and Name are the ones you would be looking to change into lower case.
The actual "Name" of the entity, eg logical name is "account" whereas in CRM where it is displayed in a user friendly way is related through a table called MetadataSchema.LocalizedLabel through the foreign key "ObjectId" which in this case would be the "EntityId" field.
This is where I would be looking to do the changes as it shouldn't have an impact on the rest of the data due to the "logicalname" field being the one CRM probably uses.
As far as your generation for strongly types classes goes.
if you use latebound such as
relatedEntity.LogicalName = "new_related_account";
relatedEntity["relatedaccountid"] = entity["accountid"];
then all the properties and logical names need to be lower case. As this will use the "logicalname" property previously identified in the MetadataSchema table.
Howver using SVCUtil I can only assume it looks at the "Name" and "Physical Name" attributes to give a slightly more user friendly coding experience when it generates the file for use in custom applications.
Though if you are looking to use the early bound class generation it shouldn't be a problem as the definition file generated will provide intellisense on the correct capitalisation of attributes and properties, and if you are using late bound like the example previously it's all lower case. So it's more it will just be a bit untidy to look at than completely impractical =)
