Product updates
Ideas
Groups
E-learning
Documentation
Events

Community
Search

Topics started by Robert Jan de Nie

24 Topics

Most views

Recently active Most replies

Robert Jan de NieThinkwise blogger

published in News & Blogs

Manage branches with ThinkwiseBlog

Modern businesses require a short time to market to keep their competitive advantage. This means the software development process needs to be agile. In order to support maximum agility and especially when working in (multiple) teams it is smart to work in branches rather than work on the main model.In this article we are going to focus on working with feature branches whilst still being able to go through the complete DTAP flow and have the ability to apply hotfixes if the need arises.As a quick recap, a branch is a separate version of your application model, created from a specific point in time, that allows developers to make independent changes, fixes, and updates without affecting the main version. Branches are used to isolate new features, bug fixes, and other development work, and later merge them back into the main version when they are ready to be released. This allows multiple developers to work on different aspects of the application simultaneously, promoting collaboration an

Robert Jan de NieThinkwise blogger

published in News & Blogs

Check constraints using the Software FactoryBlog

In any application it is important to validate a user’s input.Is the value of the order smaller than the maximum order value? Is the start-date earlier than the end-date? Does this action lead to a stock amount smaller than zero?There are a few options how you, as a developer, can implement these checks:Domain (using min/max value) Default Trigger Check constraintA closer lookLet’s take the following example and work out if and how we can implement its validation:Is the start-date earlier than the end-date?Domain Validation through a domain is not possible in this case since we need to compare two values Default Clear out the end-date field and send a warning to the user informing that the end-date should occur after the start date Advantage: The user is being informed while the data is being entered. Risk: Business rule is only being enforced on the screens where the default has been implemented. When this data is created through other means, for example a task or external interfa

Robert Jan de NieThinkwise blogger

published in News & Blogs

Using query store to analyse performanceBlog

Have you ever had a system slow down and everybody waiting for you to fix it? Or wondered what happened in the database an hour ago? Or been curious what impact a change in your query had to the speed of execution? Or been confused about how to best monitor database performance? Let's talk about a way to answer the above questions using the Query Store that is being provided in SQL Server since SQL Server 2016. This is not going to be an in-depth view of all the features of the query store but merely to make you aware it exists, the way I use it and some of the key benefits. If you are looking for a detailed analysis of the complete feature set, please refer to the Microsoft documentation. First thing we need to do is enable the query store, fortunately, this is very easy. Navigate to database properties Select ‘query store’ on the left hand panel Set the operation mode of the query store to 'Read Write' Click OK Hit refresh on your database in the Object Explorer Verify there is an ex

Robert Jan de NieThinkwise blogger

Filtered indexesCompleted

Since a recent version of the Software Factory it is possible to create unique indexes that exclude NULL values. This is a useful feature, however, I think it would be great if we can take it one step further.I’d like to propose the possibility to create filtered indexes, where the developer him/herself can supply a where clause that will be applied to the index.For myself, I have this working using some dynamic model fiddling, but I think it would be pretty straight-forward to implement as a default feature.This is an example from the index script that is being created using my adjustment:/* Create index 'ix_un_def_dimensions_len' on table 'article_default_dimension'. */*snip*create unique nonclustered index ix_un_def_dimensions_len on "article_default_dimension"( article_id, default_length)where default_width is nullgo/* Create index 'ix_un_def_dimensions_len_width' on table 'article_default_dimension'. */*snip*create unique nonclustered index ix_un_def_dimensions_len_width on "arti

Robert Jan de NieThinkwise blogger

Support for sequencesCompleted

All supported databases support the concept of sequences. To clarify what a sequence is, I'll quote Microsoft: A sequence is a user-defined schema bound object that generates a sequence of numeric values according to the specification with which the sequence was created. The sequence of numeric values is generated in an ascending or descending order at a defined interval and can be configured to restart (cycle) when exhausted. Sequences, unlike identity columns, are not associated with specific tables. Applications refer to a sequence object to retrieve its next value. The relationship between sequences and tables is controlled by the application. User applications can reference a sequence object and coordinate the values across multiple rows and tables. Sequences are quite useful in various scenario's, for example to assign (unique) batch-numbers the products that have been produced and are being stored in a stock table. I think it would be useful if we could make use of thes

Robert Jan de NieThinkwise blogger

Check procedures - Control procedure has been takenOpen

Within applications developped with the Software Factory, there are several area's where data or a formula needs to be checked. Imagine a task that will confirm an order. Usually, you want to check the current status of the order, so that for example, current orders cannot be confirmed twice. Also, you want to help the user, have the right task buttons available, so you apply the same logic to hide the task button, based on current status. Unfortunately, these checks need to be put into two seperate control procedures, once for the task, once for the context procedure. In the worst case, the developer decides to write these checks twice. If he is a bit smarter he writes a specific function in order to test this case and handle the situation accordingly. This would be, for now, also the advice I'd give to people; write a function to test and use this function in both task and context procedure. Bad thing about this approach, is that a developer can forget to do one o

Robert Jan de NieThinkwise blogger

Mock data set as queryCompleted

Currently to create mock data you can use either the mock data sets or create a preparation query (per unit test) in order to create some mock data this way.Especially with larger mock datasets it can be tedious to enter all the mock data manually.I’d like to be able to use the power of SQL to create mock data sets. Instead of creating a preparation query per unit test to create some mock data I’d like to be able to write a query that will create me some mock data. This query should be run before the current preparation query of the associated unit test.This way you will be able to re-use your mock data preparation query instead of writing them per unit test and have a more powerful option to create some mock data.

Robert Jan de NieThinkwise blogger

Insert your own code into CATCH part of 'atomic transaction'Completed

For a while now, there is a checkbox that lets your code run in an atomic transaction. This is a great improvement allowing you to easily let your code run inside a transaction.I would like to propose an additional improvement, by letting the user put in a template within the catch-part of the atomic transaction. This way, you can add your own error handling in the catch part of the code as well.Below an example of what this could look like:CREATE or ALTER procedure demoasbegin --PART 1 of default transaction code, generated by default by Software Factory begin try begin transaction; --Force an error select 1/0; --PART 2 of default transaction code, generated by default by Software Factory commit transaction; end try begin catch if @@trancount > 0 begin rollback transaction; end; --USER ENTERED additional error handling --for example, writing some error in a log. insert into error_log (error_log_message)

Robert Jan de NieThinkwise blogger

Automatically create unique index for lookup columnOpen

There is a nice validation called 'Reference source table has possible double look-up values’. It even tells us the solution to resolve it: 'Create a unique index on the source table of the reference, to ensure the look-up value is always unique. The unique index should have the following fields:- All primary key columns which are also a foreign key column.- The lookup display column.’Why can't the Software Factory create these indexes automatically? It would save developers a lot of time and would improve the quality of the data quite a bit.

Robert Jan de NieThinkwise blogger

Retry on deadlockOpen

When a user currently encounters a deadlock and she is chosen as the deadlock-victim, the user gets a message asking to rerun the transaction.Solving (frequent) deadlocks is something a developer should do, but they still can happen. Often, when a user retries to run the transaction, the problem has gone away.I’d like to propose a client side (GUI/Indicium) solution that will re-try the transaction without the user intervening. This way, the user is not seeing a technical message regarding the deadlock and the action the user wanted to make still pulls trough.I would implement this as follows (my suggestion)Store a setting in SF specifying the amount of deadlock retries (default = 5) GUI/Indicium: Retry the transaction as many times as specified in SF (with an appropriate wait-time) When a deadlock occurs, store the deadlock graph XML in a deadlock-log table in IAM Write to logfile a deadlock has occured (warning) The advantage to this solution is two-fold:The user can (mostly) conti

Robert Jan de NieThinkwise blogger

Generate code group when generating definition and codegroup has not been generatedCompleted

Pretty self-explanatory, when there are no code groups generated for a specific code-group and you hit generate definition on the result tab, nothing happens and you have to go back and generate the code-group first. This is annoying and can easily be done for you.Please? :)

Robert Jan de NieThinkwise blogger

Show more information during 'Generate source code' stepCompleted

However the source code generation step has never failed me, I sometimes wonder if it is doing something. I’d like to have a more verbose output of what it is actually doing and at what step the generation process is.Currently we only see this and just have to wait:It would be nice if we are able to see the smaller steps that are happening inside a specific step. The advantage is two-fold. It keeps you up to date and makes it visible if there are any steps that are running slowly.

Robert Jan de NieThinkwise blogger

Excluding multiple NULL values for indexes by defaultCompleted

In SQL Server, a unique index does not allow multiple NULLs by default. There is a way around this by adding a where clause to the index. (where unique_field is not null). However, this can not be modelled in the SF.

Robert Jan de NieThinkwise blogger

Fixed output of a function call for a unit testOpen

Code that calls a function can be difficult to test if that function performs significant logic. I want to isolate the code we are testing from the logic buried in the functions that it calls. To create independent tests, you should be able to fix the output of some functions.For example when you have a task that confirms a sales order, you don’t want to unit-test the function that checks for a credit limit. The credit-limit check should just return ‘ok’.Another big advantage is that this will reduce the amount of mock data required to perform the unit test. In some cases you have to fill half a database in order to get a function to return a ‘true’ whilst not even testing that function.

Robert Jan de NieThinkwise blogger

Overview of requirements containing an open 'discussion'Completed

When you are working on the requirements, there is a tab to put in discussion points regarding this requirement. It is however very hard to find out what requirements still have an open discussion item on them. Can we have an overview list of the requirements that contain a discussion, so it is easier to find and respond to these items?

Robert Jan de NieThinkwise blogger

Default add parameters for custom protocol process flowsOpen

Now you have to add them manually. Why not add them by default?

Robert Jan de NieThinkwise blogger

Add translation objects for RolesOpen

IAM is available in a few languages, but users of IAM need to assign roles to user groups based on the name of the role and the description. With internationalisation in mind and not knowing what language the IAM user understands, we need to be able to provide the names and descriptions of the roles in each language that IAM is available in. Not just the language the application is made in.

Robert Jan de NieThinkwise blogger

Generated checkbox -> generated by control procedureMerged

At the moment several tables in the Software Factory have the ‘generated’ checkbox, indicating a record has been generated rather than created manually by the user. I run into the problem that there are quitte a lot of dynamic model control procedures (power to the SF!) and sometimes have a hard time figuring out what control procedure has generated an item. Can we please have an extra (optional) field where you can put in some additional information regarding the generation of the record, i.e. generated by what CP for instance?

Robert Jan de NieThinkwise blogger

See what program objects are linked to a control procedureOpen

I’d like to be able to see within the control procedure screen what program objects are linked to it. This could just be one field with an aggregate of the assigned control procedures. This way you can easily see/find what control procedure is linked to a specific program object.Nice quick win, I’d say…

Robert Jan de NieThinkwise blogger

No lookup on generated tablesOn the backlog

It is possible to create a look-up on a generated table, for instance a task-lookup. This should not be possible since generated objects get dropped when generating.

Robert Jan de NieThinkwise blogger

Show technical info of current cursor locationOpen

I’d like to propose an addition to the developer tab on the ribbon menu, helping developers to find some additional information about the record and field they have currently selected.This should be done based on cursor location (i.e. clicking into a certain field). For example in the screenshot below, I am in the Sales order line screen (table) and have selected the customer column.I’d like to see some basic information straight away, rather than going into the debug menu. It could be something like this:

Robert Jan de NieThinkwise blogger

Change order(number) of requirementsClosed

A quick idea, when writing up requirements, it is sometimes useful to have them in a certain order. You can do this by changing the ‘sequence no’ field, but you can only re-arrange so much. I’d like to have options available similar to the ones to re-arrange the column order number of a table. So move up, move down and re-number. Be aware this request is not related to this one: https://community.thinkwisesoftware.com/ideas-18/renumber-requirements-task-230 Thanks!

Robert Jan de NieThinkwise blogger

Format queries in debuggerOpen

Sometimes select-queries become very hard to read, especially when you use display expressions. A lot of nesting takes place.In order to make them readable, I use the ApexSQL formatting plugin, but I think it would help a lot of people if the GUI did the propper formatting for you. Especially when you need to debug or tune something.This is an example from a current project:select top 1000 t1.[sales_order_id], t1.[administration_id], t1.[customer_id], t1.[sales_order_number], t1.[sales_order_type], t1.[based_on_complaint_id], t1.[payment_term_id], t1.[delivery_term_id], t1.[customer_reference], t1.[delivery_address_id], t1.[type_delivery_date], t1.[delivery_date], t1.[delivery_year], t1.[delivery_month], t1.[delivery_week], rtrim(t1.[currency_code]) as [currency_code], t1.[logistics_remark], t1.[invoice_organisation_id], t1.[invoice_recipient_person_id], t1.[avg_invoice_recipient_person_id], t1.[invoice_address_id], t1.[regular_confirmation_or_pro_forma], t1.[certificate_recipient_pers

Robert Jan de NieThinkwise blogger

Move running tasks to ribbonOpen

The running tasks button is currently very hard to find. It is small and potentially in the same colour as the background of the ‘application header bar’ where the icon is being held.I propose moving this button to a more visible location and also showing how many tasks are running/outstanding.Below a quick paint-sketch outlining my idea. (Please note, I am not a UX designer)

Badges

HJ van Dalfsenhas earned the badge Transcriber
C.G. van der Knaaphas earned the badge Geek
Mark_Plaggenborghas earned the badge Transcriber
Daan Heemskerkhas earned the badge Eye-opener
Dave Bieleveld Starcodehas earned the badge Preparator

Show all badges

Powered by Gainsight

Terms & Conditions Cookie settings

Sign up

Already have an account? Login

Login

Login to the Thinkwise Software Community

Login

Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.

Username or e-mail

Back to overview

Scanning file for viruses.

Sorry, we're still checking this file's contents to make sure it's safe to download. Please try again in a few minutes.

OK

This file cannot be downloaded

Sorry, our virus scanner detected that this file isn't safe to download.

OK