Sunday, April 26, 2009

Lookup methods tutorial + custom list lookup implementation

One of the great features available in Dynamics AX is the support of lookup forms that provide a user-friendly interface for selecting a value for a field from a list, and are highly customizable, allowing the developer great flexibility in meeting user needs.

Recently, I was posed a question of how to present a user with a list of custom-defined values in a lookup form.

The simple (and suggested) approach here is to create a new BaseEnum, containing all the specified values, and add a ComboBox control for this enumeration to the form. If the values are to be displayed conditionally (only a subset is shown each time), you can build a superset of all values in AOT, and use SysFormEnumComboBox class to limit the actual output at runtime.

But, this would not work in my case, because the values in the list were dependent on other data in the form plus the current date. Using a temporary table for this scenario seemed like an overkill (but possible). So i decided to investigate the system class SysLookup for hints on how this can be done with minimum effort. The implementation of this method is provided below.

As part of posting this implementation, I decided to also briefly describe some of the other lookup method options available to an AX developer, turning this post into a small tutorial on lookup methods.

Tutorial_LookupMethods_Screenshot

1. The approach used most in the application is the implicit lookup based on table fields. As you can see from the control properties, DataSource and DataField specify which table field the control is based on. In my example, it is based on SalesTable.CustAccount, which uses the ExtendedDataType CustAccount, which has a relation to CustTable.AccountNum. Therefore, any time you add this field to a form, it will automatically provide a lookup button to select one of the customers.

Relations being specified on EDTs is at times confusing to people, who are used to seeing the relations between tables on the tables themselves. Technically speaking, such definitions on EDTs are incorrect. But no worries, AX supports the “correct” scenario out of the box as well. The relation to CustTable could have been specified on the table just as well.

2. 2nd most used approach is to specify the ExtendedDataType on the control directly. This is used to allow the user to specify a value to filter the records on directly from the form, for example (without the need to go to Extended Query form). It is also indirectly used on all RunBase dialogs in the system (when creating a DialogField, we specify the EDT to be used, which is transfered to the ExtendedDataType property on the corresponding control). In the tutorial, it is again the CustAccount EDT that is specified in the properties.

3. The 2 above examples both used metadata to define what lookup is to be displayed. The 3rd most used approach is relying on SysTableLookup class and builds the lookup at runtime. The code is relatively straightforward, and is described in more detail on MSDN. Using this approach, the developer can specify the query to filter the data being displayed in the lookup form. Note, that SysTableLookup only allows to have data from one table in the lookup form (+ display methods on this table). SysMultiTableLookup is an extension I have created a while ago, that adds this and other useful functionality to SysTableLookup class.

4. The remaining 2 standard approaches are rarely used in the application. Lookup based on ExtendedDataType is very similar to approach described under #2, but is executed from code at runtime. This way, you can change the lookup data displayed dynamically, based on some conditions (For example, on Ledger Journal lines, the offset account lookup shows vendors, customers, banks, etc. based on the offset account type specified on the line).

5. BaseEnum values are usually represented as a ComboBox (and, in special cases, CheckBox or RadioButton) control on forms in Dynamics AX. The lookup is provided by the kernel automatically for this type of control. But, sometimes, it is required to show the values of an enumeration in a string control – most common scenario is providing filter capabilities based on enums, where multiple values can be specified at once, similar to extended query form filters. In fact, this approach is actually used on SysQueryForm for enum fields. As you can see from the code, the lookup call is also very simple in this case.

6. This approach does not currently exist in the application, and the goals for its implementation were already described above. The lookup method code in this case looks rather straightforward, here it is:

public void lookup()
{
Counter yearCount;
List valueList = new List(Types::String);

for (yearCount = 0; yearCount < 5; yearCount++)
{
valueList.addEnd(strFmt("Year %1", year(SystemDateGet()) - yearCount));
}

SysLookup::lookupList(this, valueList, "List of years");
}

As you can see, the data displayed in this lookup depends on current system state (date) and other data (yearCount). It uses the class List to hold the values to be displayed.

You can download the project (Compatible with Axapta 3.0 – Dynamics AX 2009) with the tutorial and custom list lookup implementation by following the link below:

Note: The xpo contains changes to SysLookup class/form (only usr layer has been exported, for your convenience). Be careful when importing those, and don’t import them into the production environment.

Thursday, April 23, 2009

Be careful with join clauses when writing complex queries

Join clause evaluation is dependent on its position in the sql statement. In SQL Management Studio, if you try to specify the conditions incorrectly, you will receive the following error message:

SQLstatement

In X++, you will not receive any compilation errors, nor any runtime errors. In complex scenarios with a lot of queries, this might go unnoticed, and will be extremely hard to weed out at a later stage, when data inconsistencies crawl in.

Take, as an example, the following job. At first glance, the 2 methods look exactly the same, and it seems as if they should work just fine. But in reality, the second method will return incorrect results, because inventSum.InventDimId will be treated as a constant (empty string as the default value) and not as a table field used in the same select statement.

QueryBuild classes have a major advantage in this situation, as you cannot easily add join clauses (links) unless adding to the child (joined) queryBuildDataSource.

static void JoinClauseWarningJob(Args _args)
{
#define.ItemId("ESB-005")

void testCorrectJoin()
{
InventTable inventTable;
InventSum inventSum;
InventDim inventDim;

select inventTable
where inventTable.ItemId == #ItemId
join inventSum
where inventSum.ItemId == inventTable.ItemId
join inventDim
where inventDim.inventDimId == inventSum.InventDimId;

info(strfmt("Correct join where clause: %1", inventSum.AvailPhysical));
}

void testIncorrectJoin()
{
InventTable inventTable;
InventSum inventSum;
InventDim inventDim;

select inventTable
where inventTable.ItemId == #ItemId
join inventDim
where inventDim.inventDimId == inventSum.InventDimId
join inventSum
where inventSum.ItemId == inventTable.ItemId;

info(strfmt("Incorrect join where clause: %1", inventSum.AvailPhysical));
}
// Actually execute the code
testCorrectJoin();
testIncorrectJoin();
}

So, the suggestion is simple: Use QueryBuild classes (or AOT queries) whenever possible, and pay attention to the order of tables and join clauses in the select statements that you write.