News and blogs on our products, services, and events
Recently active
In my previous blog I showed you how you can create a neural network to predict the hardenability of steel. This was done with the help of R and Keras. I also showed you how R functionality can be integrated in SQL templates. This time I will cover the subject of forecasting and how this can be done using R. Since integrating R code in SQL templates always follows the same procedure, I will not be covering that subject again in this blog. The main goal here is to show you some of the possibilities R has to offer regarding forecasting. I will show you how to prepare a dataset, fit different types of models to your data, using those models to forecast, visualize your results, and calculate evaluation metrics. ForecastingForecasting is a method of predicting the future based on past data. For example, company X has kept track of all sales orders of product Y for the last 2 years and wants to predict next month's demand for Y. A simple method that company X can use is to sum up demand p
SQL Server Management Studio has been the backbone for writing queries and managing SQL Server databases for a long time now. I’ve been using SSMS for well over a decade.A new kid on the block - Azure Data Studio - is gaining popularity. This IDE has been around for a couple of years now and is a close relative to Visual Studio Code. And with the possibility of plugging in extensions such as GitHub Copilot, I’ve recently completely switched to Azure Data Studio.In this blog, I’ll highlight some of the features of Azure Data Studio and suggest some key binding changes which I did make the transition from SSMS to Azure Data Studio easier. Familiarize yourself with F1Not unlike Visual Studio Code, virtually all commands, actions and settings can be done in Azure Data Studio via the F1 menu. If you want to do something and you don’t see a button - always press F1.The F1 menu of Azure Data Studio Tab display sizeThis may be a personal preference - I always ensure tabs are back to their orig
As of Windows Server 2019, Microsoft disabled the encryption protocols TLS 1.0 and 1.1 by default, and requires all connections to, for example, SQL Server instances on Windows Server 2019 to use TLS 1.2. This can cause problems with Crystal Reports documents that use the default SQL Server drivers included with Crystal Reports, instead of the Microsoft OLE DB Driver.To resolve this, you need to replace the drivers used by each report with the latest Microsoft OLE DB Driver for SQL Server, which you can download here. To update the driver in the report file:Open the Set Datasource Location dialog from the Database menu in the Crystal Reports designer. Select the connection to replace and click Make New Connection to create a connection using the Microsoft OLE DB Driver for SQL Server or select an existing MSOLEDBSQL connection. Do not use the Microsoft OLE DB Provider for SQL Server (SQLOLEDB). Select Update to replace the connection.Repeat this for all Crystal Reports files. An altern
When a developer creates a code template and assigns it to a program object, the order in which the code is executed is often neglected. This can lead to some serious issues, one of which we are discussing today. Deadlocks. A deadlock usually does not occur until a product is used in a production environment. Even though transactions are rolled back and no real harm is done, the developer will often have to solve the deadlock issues with great haste. This blog post will provide some guidelines on how to find and solve a deadlock. In this blog post, I’m assuming that the isolation level is the current default isolation level for products – Read Committed. How deadlocks occur A deadlock will occur when a transaction (A) is trying to access one or more rows which are in use by another transaction (B). Normally this is not a problem, the transaction (A) will wait until the rows are released. But the transaction (A) will wait indefinitely when the other transaction (B) is waiting
The Thinkwise Platform supports various types of report generation solutions. The choice for a certain type of reporting has functional implications. For instance, not all reporting solutions support the usage of sub-reports. Other reporting solutions may offer more features regarding the use of SVG images or HTML markup.There are also non-functional aspects to consider. License costs is often noted. Another important aspect is the portability of the reports. Not every type of report can run on both Windows and Linux environments.Quite some research has gone into the portability of the usage of Crystal Reports lately. We’d like to share some of our findings in this blog.Runtime installationAdministrators who have deployed applications using Crystal Reports know that it requires the installation of the Crystal Reports runtime. Either on the desktop clients when the Windows GUI is used or on the web server when the reports are generated by Indicium, the Application Tier, or the Web GUI.A
When I leave my house every morning, I pick up my car keys from the counter and put them in my pocket. To me, this is a very common way of interacting with an object. I grab the keys, move them into my pocket and let go of the keys. But when it comes to our software, this would happen very differently. I would select the Car keys record, put it in edit mode and change a field indicating the location from Counter to My pocket. I would then save the record and.. *Poof!* – The keys are now in my pocket. Very unnatural if you ask me. Editing the car keys? I sound like a locksmith! This new feature, however, will bring our applications one step closer to the real world: Drag-drop of records! Model-driven drag-dropAs with every feature, the definition of drag-drop is a part of the model. To keep things as easy and accessible as possible, no new logic concepts were introduced. The existing concepts of tasks and corresponding context logic are used. Besides that, variants play a bi
There is a lot of demand from the Thinkwise Community for easy uploading of multiple files via the Universal GUI; see this idea for example.For the Web GUI, @Harm Horstman developed a great solution that leverages a custom upload page in a preview component, but unfortunately this solution is less suitable for the Universal GUI as it bypasses Indicium and requires a separate application to be installed on the web server.Fortunately, it is relatively easy to develop a similar component that uses the Indicium API to upload files and integrate it in the Universal GUI. In this blog post we'll show you how to do that. Click to enlargeConfiguring the preview componentTo display a custom component in the Universal GUI, we use the preview component as described here. In this preview component, we will display a URL pointing to our custom HTML page that contains the upload functionality. The first step is to model a screen containing the preview component. To do this, we create an upload table
The Thinkwise Platform has powerful functionality to connect to external applications using OAuth. Customers often use external applications, for example a payment system. In that case they often want to import the data to the Thinkwise database, as the external system is their single source of truth. We can use the Thinkwise Platform to synchronize this data. With the support of system flows we can even automate this to automatically collect the latest data.In this blog I will use Exact Online as the external application to synchronize the projects from Exact to a database.Creating the data modelI made a very simple model for demo purposes. As a developer you can extract a lot more data from the API.The data model contains two tables, the oauth_token_info table to save the OAuth access and refresh tokens. The project table contains the information we will extract from the Exact Online API.Data modelApp RegistrationTo call the API first we need to register an application in Exact Onlin
With the arrival of SQL Server 2016 Microsoft added the R Services add-on, which enables the execution of R scripts directly from T-SQL. Starting with SQL Server 2017, R Services was renamed to machine learning services to reflect the fact that Python is now supported as well. Both languages are highly popular in data science and machine learning. Python is an easy to learn general purpose language, while R was developed mainly for statisticians. Although there is a large amount of overlap between the languages, people with a background in statistics seem to prefer R over Python. On the other hand people new to programming or without a sufficient background in statistics prefer Python because of it's easy to learn syntax. Personally I think it is more a matter of choosing the right tool for the job. I would use Python for creating complex deep learning models and use R to analyze the results afterwards. In this blog I will show you that it is actually possible to do both using R
We often get questions about customizing the Universal GUI, particularly the login page. We plan to offer several model-options to customize it in the future, such as the logo, colors, title text and background image, and to display maintenance messages, for example. But with the latest release of the Universal GUI, it is already possible to make these adjustments manually, using custom CSS! SetupStarting with the latest (beta) release of the Universal GUI it is possible to inject your own custom CSS to change the styling of various parts of the user interface.To get started, simply rename the custom.sample.css file in the public folder of the Universal GUI to custom.css. This file will then automatically be loaded by the Universal GUI. By renaming the file, it will no longer be overwritten with Universal GUI updates.Next, add your own CSS to the custom.css file and you’re off! CSS classesThe Universal GUI provides static CSS classes for a selection of elements, mainly on the login pag
SQL server version 2016 has introduced the feature of setting up system-versioning to tables (also known as ‘system-versioned’ temporal tables). This means that SQL server stores a history of data changes and deletions which take place within a table. There are various methods of creating history tables, but with SQL server you can allow a history table to be made automatically for a table. To build a test scenario, let’s first create a table called article. code:create table article(article_number varchar(10) not null,article_name varchar(100) null,article_price numeric(9,2) null,constraint article_pk primary key clustered(article_number asc)) In order to create a history table, SQL first needs start_date_time and end_date_time columns on the table which is to be logged. code:alter table articleadd start_date_time datetime2 generated always as row start hidden not null default '1900-01-01 00:00:00.0000000', --Can be anything, but has to be in the past. end_date_time datetim
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
The Thinkwise Platform version 2021.3 has revised the way the Creation screen works, opening up more elegant ways to leverage tooling to automate this process.The generation screen consists of a number of steps:Generate definition Validate definition Generate source code Execute source code Run unit testsThis has remained unchanged between versions 2021.2 and 2021.3. However, the way these steps are performed has fundamentally changed.In the previous version, the steps in the generation screen were bound to a specific client. A developer would start Generate definition and the definition generation would be orchestrated start-to-end by the users client. This has served us well for many years, but there are some drawbacks that we’ve addressed in this release. The new Creation approachAll ‘jobs’ are no longer performed by a client, but are instead performed by Indicium. This means that generation, validation and such will continue even if the developer closes the screen or even the Softw
