Thursday, January 17, 2008

Creating and Posting Inventory ProfitLoss journals in DAX using .NET Business Connector

Yesterday, I was helping a friend from Canada create a small solution that would create and post ProfitLoss journals in Microsoft Dynamics AX 4.0 using .NET Business Connector.

In the end, we created 2 solutions - one was entirely written in X++, and only simple class static method calls were made from C#. The other was completely written in C#, using various classes, available with .NET business connector.

I uploaded the solution to the following link, in case anyone would be interested to download it and play around with it or use in their projects.
Download ProfitLossPostingAppl

Also, here is the source code - it is a small console application, and I tried to add as many comments as possible, so that even complete X++ beginners would be able to easily read the code.

Notice that in the solution, there is a reference to the business connector dll.
You can find the Microsoft.Dynamics.BusinessConnectorNet.dll in the Client\Bin folder of your Dynamics AX installation. This dll is what provides you with access to the Dynamics AX application and the set of classes to use AX tables and classes.

using System;
using Microsoft.Dynamics.BusinessConnectorNet;

namespace ProfitLossPostingAppl
class AxProfitLossPostingEngine
static void Main(string[] args)
Axapta ax = new Axapta();
// company name, language, object server, configuration
// this uses Windows Authentication
ax.Logon("cmul", null, "localhost", null);

// Start a transaction

// AxaptaRecord is a class that allows to work with Tables in AX
AxaptaRecord header = ax.CreateAxaptaRecord("InventJournalTable");
AxaptaRecord journalName = ax.CreateAxaptaRecord("InventJournalName");
AxaptaRecord line = ax.CreateAxaptaRecord("InventJournalTrans");
AxaptaRecord inventTable = ax.CreateAxaptaRecord("InventTable");
AxaptaRecord warehouse = ax.CreateAxaptaRecord("InventDim");

// You can call static table methods using the following syntax
journalName = ax.CallStaticRecordMethod("InventJournalName", "find", "IPL") as AxaptaRecord;

// There is a set of predefined methods on the AxaptaRecord class, like the clear(), initValue, DML operations, etc.
// You can call table object methods as well, not only static
header.Call("initFromInventJournalName", journalName);

line.Call("initFromInventJournalTable", header);
// You cannot use table fields directly as in X++. Instead you have set/get methods
line.set_Field("itemId", "B-R14");

// Instead of using static table methods (like find) you can execute a direct SQL statement and receive the result in the AxaptaRecord object
inventTable.ExecuteStmt("select * from %1 where %1.ItemId == 'B-R14'");
// If you receive more that one record you can iterate through them using Next (as in AX)
line.Call("initFromInventTable", inventTable);

line.set_Field("Qty", 160.0);

warehouse.set_Field("InventLocationId", "MW");

warehouse = ax.CallStaticRecordMethod("InventDim", "findOrCreate", warehouse) as AxaptaRecord;

line.set_Field("InventDimId", warehouse.get_Field("inventDimId"));

// Notice AxaptaRecord is passed by reference here
ax.CallStaticRecordMethod("InventJournalTable", "initTotal", header);


// You can call static class methods the same way you call table static methods, but using a different method on Axapta class
// So in case you wrote the posting in X++, you would be able to call it, passing the JournalId as the argument
// int numOfLinesPosted = (int)ax.CallStaticClassMethod("DEV_ProfitLossEngine", "postProfitLossJournal", header.get_Field("JournalId"));

// Or, you can use the AxaptaObject class to accomplish the same from C#
// You can initialize a new class using the Axapta class method
// AxaptaObject journalCheckPost = ax.CreateAxaptaObject("InventJournalCheckPost");

// Or using a static method, if that suites your needs better
// Notice here that an object of type AxaptaRecord is passed into a method that expects InventJournalTable as the argument
AxaptaObject journalCheckPost = ax.CallStaticClassMethod("InventJournalCheckPost", "newPostJournal", header) as AxaptaObject;
// You can object methods the same way you would on a table
journalCheckPost.Call("parmShowInfoResult", false);
journalCheckPost.Call("parmThrowCheckFailed", true);
journalCheckPost.Call("parmTransferErrors", false);

int numOfLinesPosted = (int)journalCheckPost.Call("numOfPostedLines");

Console.WriteLine(String.Format("{0} line(s) have been successfully posted", numOfLinesPosted));
Console.WriteLine("JournalId is " + header.get_Field("JournalId"));
Console.WriteLine("Press any key to continue ...");

catch (Exception ex)



P.S. Of course, it would probably be a better idea to use classed InventJournalTableData, InventJournalTransData, etc.
But for the simplisity of the example, everything is done directly with tables.

Wednesday, January 16, 2008

Microsoft Dynamics AX 2009 - some of the new development features

Alexei Eremenko from Microsoft Russia has posted a number of articles about the new features that will be available in DAX 2009.
The original link, which is in Russian (but the main thing to look at is the code, which is universal):

I would like to make a brief review of the features he talked about for the English-speaking population and another feature I liked that Alexei did not mention.

Let's start with the support for union in SQL statements (but only when using Query* classes).

query = new Query();
query.queryType(QueryType::Union); // The other value of QueryType is "Join"

Another Exception type has been introduced, which now actually allows to catch the DuplicateKey exception:

Table t;

while select forupdate t
test.Field1 = ‘xyz’;
catch ( Exception::DuplicateKeyException, t )
infolog(‘Record already exists - ‘ + t.Field1 );

The bulk DML statements now allow using inner/outer joins, and you can access the result of the update_recordset operation to get the number of rows that were updated:

update_recordset batchJob setting
Status = BatchStatus::Canceled,
EndDateTime = thisDate,
Finishing = 1
where batchJob.Status == BatchStatus::Cancelling
notexists join batch
where (
(batch.Status == BatchStatus::Ready ||
batch.Status == BatchStatus::Executing ||
batch.Status == BatchStatus::Hold ||
batch.Status == BatchStatus::Cancelling)
&& batch.BatchJobId == batchJob.RecId

rowsUpdated = (batchJob.RowCount() > 0); // get the number of updated rows with rowCount()

And, the feature I enjoyed, is that we now have crossCompany support in X++. Meaning you can access data from tables from a number of companies in one query.
Here are code snippets to explain what I mean:

static void DataBaseAccess_CrossCompany(Args _args)
InventTable inventTable;
container companyContainer = ['IN1', 'QMS'];
while select crossCompany : companyContainer inventTable
where inventTable.ItemId == "B-R14"
print inventTable.ItemId, " -- ", inventTable.dataAreaId;

This code will print ItemId from 2 companies, even though InventTable has the property SaveDataPerCompany set to Yes.

The same functionality is available with the Query classes:

static void DataBaseAccess_CrossCompany_Query(Args _args)
Query query = new Query();
QueryBuildDataSource qbds = query.addDataSource(tableNum(InventTable));
QueryRun queryRun;
InventTable inventTable;
qbds.addRange(fieldNum(InventTable, ItemId)).value(queryValue("B-R14"));


queryRun = new QueryRun(query);
while (
inventTable = queryRun.get(tableNum(InventTable));
print inventTable.ItemId, " -- ", inventTable.dataAreaId;

If you don't add any specific company ranges, the data will be retrieved from all companies you have access to.

Microsoft Dynamics AX 2009 has a lot more to offer, of course. But enough for today.