Creating Apps & Python

Many thanks everyone who joined us for this edition of the Python & Pizza night at our office in Rotterdam. We had a blast eating pizza and developing apps with you, and we loved seeing your amazing work! 

![Python & Pizza header optie 1.jpg](https://cms.viktor.ai/uploads/Python_and_Pizza_header_optie_1_2c83eac8d3.jpg)


We loved having so many developers from different companies (Boskalis, Dura Vermeer, Heijmans, Mocean Offshore, Omexom, Port of Rotterdam, Royal HaskoningDHV, RPS, Smit) and with different levels of experiencing developing (VIKTOR) web apps together. Whether you just finished the tutorial, build your first app, or where there to continue working on your current project, it was great seeing such enthusiasm. 
It was especially remarkable that the guys from [MO4](https://mo4.online/) managed to onboard themselves and build a working application to place wind turbines on a map in just 3 hours! 

![Python & Pizza wind turbine 2.jpg](https://cms.viktor.ai/uploads/Python_and_Pizza_wind_turbine_2_d4bd7d742e.jpg)


As you can see from the pictures, they were even so nice to give us a presentation of their application at the end of the evening. This was followed by a short sneak peek of upcoming platform updates by our Product Owner Stijn. 

We are already looking forward to a next edition! Will we see you there as well? 


large_python and pizza blog.jpg

medium_python and pizza blog.jpg

small_python and pizza blog.jpg

thumbnail_python and pizza blog.jpg

python and pizza blog.jpg

Looking back at a successful night developing web apps with Python and VIKTOR

Python & Pizzas: Looking back at a successful night developing web apps

### Content: 

1. An introduction to developing successful applications
2. Designing and planning your application
3. How to start coding
4. Gathering and processing feedback
5. Increasing the adoption of your app



How to create successful application that ensure adoption

medium_White Papers & guides - Building succesful applications.jpg

small_White Papers & guides - Building succesful applications.jpg

thumbnail_White Papers & guides - Building succesful applications.jpg

WP_pink

How to create a succesful that people will adopt

large_White Papers & Guides - Building Succesful applications

medium_White Papers & Guides - Building Succesful applications

small_White Papers & Guides - Building Succesful applications

thumbnail_White Papers & Guides - Building Succesful applications

WP_pink_2

Do you want to build a successful application? This guide is here to help you make apps that provide real value to your work and that other people will definitely want to use too!

This guide will provide actionable insights that are useful for anyone involved in creating custom software tools, such as:

- **People developing the tools** (developers, engineers, data scientists, etc..)
- **End-users**, who will use the tools when they are released
- **Project managers** who are interested in workflow automation/optimization
- **Program managers,** who oversee and stimulate digital transformation initiatives
- **Domain experts**, who provide their expertise to people creating software tools
- And **anyone else i in the process of creating awesome software tools**

Over the past years, we, the application developers at VIKTOR, have built many apps for many different industries. We have learned a lot from that and now want to share our experiences with you. 

The principles and processes we explain in this guide are the same as we apply daily. They have helped us build anything from simple apps designed to save time to advanced engineering applications that are used to automatically calculate underground tunnels, dikes, and other large structures. 





Learn how you (developer, engineer, end-user, domain expert, project manager, etc.) can contribute to the creation of apps that provide real value to your work.

Build successful applications

This guide provides actionable insights for anyone involved in the creation of apps, such as engineers, developers, end-users, project managers, and domain experts. Learn how you can contribute to creating apps that provide real value to your work!

Building Successful Applications

VIKTOR updates

## Faster apps!

Dear developers, we've heard your feedback and are happy to let you know that we have improved the speed of the platform! This goes for both the navigation, as well as in the editor itself. Maybe you already noticed this, for example when loading your Workspace's users page. 

Please let us know if there is anything else you would like to see improved by leaving a message in the [community forum](https://community.viktor.ai/?utm_campaign=Newsletter%20developers&utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-8PVlqkiKlNJFc1hzgFQMn7BTQbgS3EaoOPtQg2qMh-Q_iY13vLrsXhRA1LhdHc874b-pzm).

</br>


[![CTA_Join the community](https://cms.viktor.ai/uploads/CTA_Join_the_community_22d97b332e.svg)](https://community.viktor.ai/)

## Quick and easy visualization of lines

It is now possible to visualize 2D lines, arcs, and polylines in the (3D) GeometryView. This makes rendering a line a lot quicker and easier as you no longer have to create a 3D element with a very thin thickness to do so! 

Below, you can see the Python code needed to visualize a line, along with an example of how this looks in the editor. 

![python code line.png](https://cms.viktor.ai/uploads/python_code_line_0c515ac939.png)

</br>

![Blog_newsletter_Oct_2022](https://cms.viktor.ai/uploads/Blog_newsletter_Oct_2022_4e999a72db.png)


</br>

## Coming soon...

### The new 'editor' app type

Because some apps only require a single editor, we will soon introduce the 'editor' app type! This app type has no folder structure, meaning end-users immediately go to the editor after entering a workspace. This way, you can create dashboards or simple calculators in a more user-friendly manner!

![editor only app type.png](https://cms.viktor.ai/uploads/editor_only_app_type_3c8148f906.png)

</br>

### New 'Apps' page

Previously, you could already create your own Workspaces. Soon, it will also be possible to create and manage Apps! We are currently building a dedicated 'Apps' page where you can view all your organization's Apps and adjust your own. Additionally, admins will be able to assign more fine-grained publishing rights to developers to determine who is working on which App.

![Details Page.png](https://cms.viktor.ai/uploads/Details_Page_33e6495c30.png)

</br>


## Other changes

**SDK**

- SDK version v13.5.0 is now available
- Various classes now accept or can return a File object 
- Mock functions were added for all external analysis classes to facilitate easier testing

**WORKER**

- Worker v5.2.0 is now available
- SCIA installer has been simplified, major version is now retrieved from program directory
- More descriptive error message is returned when trying to parse non-existing files
- Excel worker now correctly evaluates automatically updating cells when running macros
- Added the possibility to add a description when creating a worker in the Administrator menu

**CLI**

- CLI v0.24.1 is available
- Improved auto-reload on Windows
- Added `--use-filesystem` flag to use the filesystem instead of git to gather files during publication of an app
- Improved structure of help by grouping commands
- Fixed bug where configure retrieved the wrong Python path 

**PLATFORM**

- Fixed wrong name used in Workspace invite email 
- Fixed incorrectly sending Developer Upgrade email to non-developers
- Fixed issue with updating the details of a user on SSO environments
- Fixed wrong translation in user creation modal
- Simplified Workspaces search bar text to 'Search'
- Fixed issue where OutputField doesn't take up the full width
- Fixed issue where the developer email notifications did not show correct setting







Blog_newsletter_October_2022

Discover the latest platform updates. Join our community and start building now! 

VIKTOR platform updates | October 2022

What's new in VIKTOR (October 2022)

## Easier debugging in production

When using your apps, users sometimes get a “something went wrong” message. However, it’s not always 100% clear what went wrong and how it can be solved. That is why from now on, users can report errors whenever a problem occurs. Consequently, a report including all relevant information (e.g. error stack, used parameters) is sent to the developer. This not only makes it easier to debug your apps in production but also ensures users that the issue will be worked on!

![Blog_newsletter_september_report_problem](https://cms.viktor.ai/uploads/Blog_newsletter_september_report_problem_edcdf1228c.png)

*The new error message, including 'report problem' option'* 

</br>

![Blog_newsletter_september_report_problem_information](https://cms.viktor.ai/uploads/Blog_newsletter_september_report_problem_information_3224ae3002.png)

*Example of a detailed overview of a reported problem*

</br>

## Add and remove developer accounts yourself

For some time now, we have been working on gradually giving admins more control over their environments. Previously, admins could already create workspaces and add users to their environment. Now, it’s also possible to add and remove developers and specify publishing rights!

This makes starting with development much quicker for new users after they’ve been added to an environment and assigned developer rights, and takes away any security risks in case of developer offboarding, as access and publishing rights can either be granted or revoked immediately.

![Blog_newsletter_september_developer_access](https://cms.viktor.ai/uploads/Blog_newsletter_september_developer_access_2f3fae6730.png)

</br>

## Coming soon...

### Input validation of steps

Soon, you can stop users from going to the next step in an application whenever they’ve entered invalid input. Additionally, you can annotate which fields are invalid to optimally guide users towards a solution and prevent any problems caused earlier on from occurring in later steps. Annotation of invalid fields can also be applied to views, leading to better guidance in case a view can not be calculated.

</br>

## Other changes

### SDK

- SDK v13.3.1 and v13.4.0 are now available


### PLATFORM

- Added the platform version information on all screens 
- Added the ability to select and deselect all options in a MultiFileField 
- Bugfix – data view items were not expanding when children were dynamically added to them 
- Bugfix – Some texts were not properly shown in the simple app type dashboard and notifications 
- Bugfix – Scrollbar wasn't reset to top on page changes in Step and Page editor 


### CLI 

- Added support for multiple accounts in configuration file 
- Updated demo app to refer to updated documentation structure 
- Improved descriptions of configure command 
- Fixed failing tests in demo template app 
- Fixed failing configure command if configuration file does not yet exist 
- Fixed configure command suggesting invalid Python installation path 

Blog_newsletter_September

Easier debugging in production and the possibility for admins to add and remove developer accounts from their environments.

The latest VIKTOR platform updates, September 2022

What's new in VIKTOR (September 2022)

Artificial Intelligence

## What are distributed algorithms?

To enable communication between various, interconnected processors (computers) that do not have a shared memory, distributed algorithms were designed back in the 1960s. Distributed algorithms are a specific type of algorithm that simultaneously runs its different parts on interconnected computers to perform a task or solve a calculation. For the computer software to run properly, it is important that these computers are interconnected and able to communicate with each other. We call this a ‘distributed system’. 

Examples of areas in which distributed systems and algorithms are used are telecommunications (telephone, television), scientific computing (weather prediction), or [leader election](https://en.wikipedia.org/wiki/Leader_election) (the procedure of designating a single process as the organizer of a task distributed among several computers). 

## Distributed algorithms vs. regular algorithms

Distributed algorithms differ from ‘traditional’ algorithms regarding the number of processors that are used to complete a task. While a distributed algorithm runs on multiple processors, a traditional computer algorithm runs on just one. With distributed algorithms, each processor – at the same time – takes care of a different part of the ‘big’ algorithm, meaning they can draw on the 
capabilities of other computing devices and processors. 

![Distributed algorithm](https://cms.viktor.ai/uploads/Distributed_algorithm_56e7f1a576.svg)

</br>

## Why a distributed system? 

Distributed systems are beneficial in the way that many processors can be used to solve a single problem, and that they can be located anywhere. This provides scalability and an ensures for an improved performance compared to using individual processors. 

An example of a distributed system are the processors used to create NASA’s space picture of a black hole. This picture was not taken by just one telescope, but through a network of many telescopes responsible tasked to look at a certain point in space for a set number of seconds at a set time. 


![NASA black hole distributed system](https://cms.viktor.ai/uploads/NASA_black_hole_d328edf909.png)


</br>

### A simple distributed system

A simple example of a distributed system can be a ring network. Every processor (normally called node) in a ring network can only communicate with its direct neighbor. If a processor wants to communicate with any other processor, it must do this in X number of steps, in which X implicates the number of intermediate processors. 

![Distributed system](https://cms.viktor.ai/uploads/Distributed_system_2395a1e454.svg)

</br>

### Priority in distributed systems

When working with distributed systems, we want to achieve truly distributed solutions. Sometimes, however, a certain system might require priority over other systems. For example, to prioritize sending certain specific messages over others in message systems. With a truly distributed system, this priority is unknown. Luckily, with a selection algorithm we can overcome this issue. For example, the [Hirschberg-Sinclair algorithm](https://en.wikipedia.org/wiki/Hirschberg%E2%80%93Sinclair_algorithm) designed for leader election problems in ring networks.

### Leader election

Leader election is the procedure of designating a single process as the organizer of some task that is distributed among several computers. Before starting the task, all computers involved are either unaware of or unable to communicate with the leader of this task. By running the algorithm (election), each computer will recognize a specific computer as the task leader. Which computer this will be is decided based on a shared identity characteristic of all computers, such as a ranking in numbers. The computer with the ‘highest’ identity (in this case: the highest number) is ultimately chosen the leader. 


![Distributed system priority](https://cms.viktor.ai/uploads/Distributed_system_with_priority_69ed3f4c4b.svg)

</br>

## Distributed algorithms in a web app

In this application, the Hirschberg-Sinclair election algorithm is used in combination with a bidirectional ring network. This means that each processor in the ring can only communicate directly with its two neighbors. 

Each processor is assigned a random number, indicating priority (higher number equals higher priority). The algorithm then works in phases trying to figure out if it has the highest priority in its neighborhood. 

Every phase, all processors ask their neighbors whether they are the highest in priority, to which the neighbor returns a message either confirming or denying this. This way, all processors – except for one – gradually disappear from the network. Each processor that disappears, does no longer participate in the ranking, but still forwards the information from its neighbors. This process repeats itself for several rounds, which are also kept track of, so the network knows which processors should disappear and which should still be included.

<div style="padding:56.25% 0 0 0;position:relative;"><iframe src="https://player.vimeo.com/video/742652816?h=6d22fd2c59&&autoplay=1&loop=1&autopause=0amp;badge=0&amp;autopause=0&amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;" title="Distributed algorithms - 100 nodes"></iframe></div><script src="https://player.vimeo.com/api/player.js"></script>

</br>

There are a 6 states that the algorithm can be in every phase:

- **Idle**: The processors do nothing
- **Send out**: … or forwards other outgoing messages if it is not the original sender.
- **Send in**: The processor either responds to the outgoing message or forwards another ingoing message.
- **Receive out**: The processor is either asked to compare itself with the original sender or forward the message.
- **Receive in**: The processor is asked to forward and incoming message or if it is the original sender they will know if it is larger or smaller than their neighbors
- **Selected**: Processors that were larger than their neighbors stay, and others disappear from the network

<div style="padding:56.25% 0 0 0;position:relative;"><iframe src="https://player.vimeo.com/video/742654151?h=8b115f6cef&autoplay=1&loop=1&autopause=0&amp;badge=0&amp;autopause=0&amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;" title="Distributed algorithms - 100 nodes messages"></iframe></div><script src="https://player.vimeo.com/api/player.js"></script>

</br>

In the application, a local network is used – made with Pyro4. The network closely resembles a real distributed system, except for without the delay in time. Because it is an asynchronized algorithm, it is also possible to turn on a delay to resemble real time passing (i.e., a delay of 30 seconds to send information from a processor theoretically being in Australia to another oner one theoretically being in the Netherlands). 

## Try it yourself!

Are you interested in using this application for your own projects? You can now download the full Python code from our [GitHub repository](https://github.com/viktor-platform/sample-distributed-algorithms) or sign up for a [free VIKTOR account]( https://www.viktor.ai/start-building-apps) to start building!



Interactive visualization distributed algorithm

A web-based application showcasing an interactive distributed algorithm, implemented with nothing but Python

Interactive visualization of distributed algorithms with Python

1. An introduction to Artificial Intelligence
2. Artificial Intelligence vs Machine Learning
3. AI and ML in organizations
4. Building a Machine Learning application
5. About VIKTOR

artificial intelligence AI machine learning ML engineering construction

medium_A4 - White Paper COVERwp ai cover.png

small_A4 - White Paper COVERwp ai cover.png

thumbnail_A4 - White Paper COVERwp ai cover.png

WP_green

artificial intelligence machine learning engineering construction

large_artificial intelligence machine learning engineering construction

medium_artificial intelligence machine learning engineering construction

small_artificial intelligence machine learning engineering construction

thumbnail_artificial intelligence machine learning engineering construction

Blog_AI_and_Machine_learning_engineering_construction_1

Have you ever wondered how you can use AI to automate engineering processes? This white paper is for all engineers, managers, or directors that want to know more about applying AI-methods, such as machine learning, in their organization. 

Find out more about the different types of AI and how you can use them to improve and automate (engineering) processes within your organization. You will also see some examples of real AI apps used by organizations in the industry and learn what it takes to build a machine learning application yourself. 

Keep scrolling to download the white paper.

Find how you can use AI automate engineering processes in your organization.

Download the White Paper and get INSPIRED

Have you ever wondered how you can use AI to automate engineering processes? This white paper is for all engineers, managers, or directors that want to know more about applying AI-methods, such as machine learning, in their organization. 

Artificial Intelligence: Engineering and Construction

## The traveling salesman problem

The traveling salesman problem (TSP) is a popular [NP-hard problem](https://en.wikipedia.org/wiki/NP-hardness) that tries to answer the following question: “Given a list of cities and the distances between each pair of cities, what is the shortest possible route that visits each city exactly one time before returning to the city of origin?”. 

Being first introduced in 1930, the TSP has been an intensively studied optimization problem ever since. Even though the TSP is a computationally difficult problem, by now there are many exact algorithms known that can be used to solve instances with up to tens of thousands of cities completely. Even problems with millions of cities can be approximated with a small fraction of 1%.

Today, the TSP is applied in instances such as planning, logistics, and the manufacture of microchips. 

## TSP in a VIKTOR app

The goal of this application is to showcase some of the approaches to the TSP in the VIKTOR environment. While no new approach is developed in this application, the app gives you the opportunity to play around with different algorithms to try to get the best solution for a generated topology. 

### Two topologies

In the application, you can choose between two different topologies: 

1.	**Circle:** This topology chooses X number of points on a circle with a radius of Y based on your definition of X and Y. We have added this option so you can check whether the algorithm is able to determine the shortest possible route, which is always a circle.

<div style="padding:56.25% 0 0 0;position:relative;"><iframe src="https://player.vimeo.com/video/733567691?h=ae69e6b6a9&amp;badge=0&amp;autopause=0&amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;" title="Traveling salesman problem | Generic algorithm | Circle topology"></iframe></div><script src="https://player.vimeo.com/api/player.js"></script>

</br>

2.	**Random:** This topology has a grid of 0-1 and chooses pseudorandom X and Y locations. It is also possible to enter a Z seed to initialize the pseudorandom number generator. This allows you to recreate the same random topology, which you, for example, can use for different algorithms so you can compare them. 

<div style="padding:56.25% 0 0 0;position:relative;"><iframe src="https://player.vimeo.com/video/733569835?h=0e51409f2f&amp;badge=0&amp;autopause=0&amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;" title="Traveling salesman problem | Two-opt algorithm | Random topology"></iframe></div><script src="https://player.vimeo.com/api/player.js"></script>

</br>

For each topology that you generate, storage is produced. This is done by creating a hash with the coordinates of all the cities. In this storage, the shortest known distance over all your runs is saved as the topology high score. For each topology that is run, you can compare the determined distance with the high score.  

### Three algorithms

The application contains three different algorithms, some well-known, some lesser known, that have previously been used to solve the TSP. If you switch between the algorithms, you will see a different explanatory text appear that provides information on the specifics of the corresponding algorithm. Additionally, all the input variables contain a text box with a description matching the algorithms.

1.	**2-opt:** The first and easiest algorithm to be implemented in this app is the 2-opt local search algorithm. This algorithm detects if the route crosses over itself and, if it does, it tries to reorder itself so that it does not. The improvement threshold of this algorithm checks how much your route has improved. If it can no longer be improved, it quits trying. 

<div style="padding:56.25% 0 0 0;position:relative;"><iframe src="https://player.vimeo.com/video/733569411?h=e6c8a73680&amp;badge=0&amp;autopause=0&amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;" title="Traveling salesman problem | Two-opt algorithm | Circle topology"></iframe></div><script src="https://player.vimeo.com/api/player.js"></script>

</br>

2.	**Genetical algorithm:** The Genetical Algorithm (GA) is an evolutionary algorithm, based on Charles Darwin's theory of natural evolution. After initializing the population (which is defined as different possible routes), we select the fittest individuals and combine them with each other to get new, and hopefully better, solutions. We can also implement random mutations in the population. This is helpful as otherwise there is a chance that the solution will never improve because we keep having the same individuals when breeding.

<div style="padding:56.25% 0 0 0;position:relative;"><iframe src="https://player.vimeo.com/video/733568236?h=d9aca7601d&amp;badge=0&amp;autopause=0&amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;" title="Traveling salesman problem | Generic algorithm | Random topology"></iframe></div><script src="https://player.vimeo.com/api/player.js"></script>

 </br>

3.	**Self-organizing maps:** The self-organizing maps algorithm implements a neural network that is closely related to the topology we generate. The purpose of the technique is to represent the model with a lower number of dimensions while maintaining the relations of similarity of the nodes contained in it. The algorithm used in this app generates a circular array of neurons, behaving as an elastic ring. During the execution of the algorithm, the neurons are searching for cities closest to their neighborhood. The ring then warps around the cities. To ensure convergence, we can add a learning rate to the algorithm. The higher the rate, the more the neurons will move. However, we might want to make the algorithm less aggressive the longer it runs. Because of this, we can also add decay to the learning rate.

<div style="padding:56.25% 0 0 0;position:relative;"><iframe src="https://player.vimeo.com/video/733568419?h=6863c4ecdc&amp;badge=0&amp;autopause=0&amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;" title="Traveling salesman problem | Self-organizing maps"></iframe></div><script src="https://player.vimeo.com/api/player.js"></script>

</br>

Here, you can see a small part of the Python code for the self-organizing maps: 

```python
def self_organizing_maps(
    cities: np.ndarray, iterations: int, learning_rate: float = 0.8, popSize: int = 100, decay: float = 0.0003
) -> Union[list, list, list]:
    """Solve the TSP using a Self-Organizing Map."""
    problem = _read_tsp(cities)

    # Obtain the normalized set of cities (w/ coord in [0,1])
    cities = problem.copy()

    cities[["x", "y"]] = _normalize(cities[["x", "y"]])

    # The population size
    n = popSize

    # Generate an adequate network of neurons:
    network = _generate_network(n)

    routes = [_get_route(cities, network).values.tolist()]
    distances = [path_distance(problem[["x", "y"]].to_numpy())]

    for i in range(iterations):
        # Choose a random city
        city = cities.sample(1)[["x", "y"]].values
        winner_idx = _select_closest(network, city)
        # Generate a filter that applies changes to the winner's gaussian
        gaussian = _get_neighborhood(winner_idx, n // 10, network.shape[0])
        # Update the network's weights (closer to the city)
        network += gaussian[:, np.newaxis] * learning_rate * (city - network)
        # Decay the variables
        learning_rate = learning_rate * (1 - decay)
        n = n * (1 - decay)

        # Adding data for the frames
        route = _get_route(cities, network)
        problem = problem.reindex(route)
        distances.append(path_distance(problem[["x", "y"]].to_numpy()))
        route = route.values.tolist()
        routes.append(route)

    return routes, np.arange(0, iterations + 1, 1).tolist(), distances
```

## If you want to use the application...

You can now download the full Python code from our [GitHub repository](https://github.com/viktor-platform/sample-travelling-salesman) or sign up for a [free VIKTOR account](https://www.viktor.ai/start-building-apps) to use it in your own applications! 



Blog_traveling_salesman_problem

Use one of three algorithms for solving the traveling salesman problem in a free web-based application on the VIKTOR platform.

Traveling salesman problem: Three algorithms in a web application

Three algorithms to solve the traveling salesman problem in a web app

## Run apps on new versions of Python!

It is now possible to run your published apps on Python versions 3.8, 3.9, and 3.10. Use the config file to specify your preferred Python version to benefit from all the improvements in functionality and performance that these versions offer!

Visit the [community topic](https://community.viktor.ai/t/support-for-python-3-7-3-10/546/3?utm_campaign=Newsletter+developers&utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz--mrBIZAoTC_p-HmPbbHDPZFd6NqYerI0h2mlVq54bIZeRk2Z1f3hOV7Fbmh2GnnCwpi0OW) if you have any questions about this feature.

</br>

![python.png](https://cms.viktor.ai/uploads/python_bcc57b8e1d.png)

</br>

## Easier unit testing for views, file uploads, workers, and more

Testing your code has never been easier! We have added decorators that you can use to mock views, file uploads, workers, and more in VIKTOR - no more need to write these mocks yourself. 

More information about [mocking](https://docs.viktor.ai/docs/create-apps/development-tools-and-tips/automated-testing?_highlight=unit&_highlight=testing#mocking) can be found in the documentation.

</br>

![Newsletter_Aug_unittesting](https://cms.viktor.ai/uploads/Newsletter_Aug_unittesting_e0d963e059.png)

</br>

## Improved search functionality in the docs

In the developer survey, many of you told us you would like to see the search functionality improved in the documentation. So, we started working on it! The new search bar is now up and running. In the meantime, we will continue to improve the documentation further. 

Please visit the [documentation](https://docs.viktor.ai/docs/welcome/?utm_campaign=Newsletter%20developers&utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz--mrBIZAoTC_p-HmPbbHDPZFd6NqYerI0h2mlVq54bIZeRk2Z1f3hOV7Fbmh2GnnCwpi0OW) to check out the new search bar and let us know what you think of it in the [community forum](https://community.viktor.ai/?utm_campaign=Newsletter%20developers&utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz--mrBIZAoTC_p-HmPbbHDPZFd6NqYerI0h2mlVq54bIZeRk2Z1f3hOV7Fbmh2GnnCwpi0OW)!

</br>

![Newsletter_Aug_searchbar](https://cms.viktor.ai/uploads/Newsletter_Aug_searchbar_0a470ed0bf.png)

</br>

## Coming soon...

### Self-servicing of VIKTOR development accounts

Soon, admins will be able to assign developer rights and create a development workspace for new users when adding them to an environment. Existing users can also be given developer rights by going to users > edit user and ticking the 'developer' box!

</br>

![Newsletter_Aug_selfservicing_dev_acc](https://cms.viktor.ai/uploads/Newsletter_Aug_selfservicing_dev_acc_d50474d0fe.png)

</br>

## Other changes

### SDK

- SDK v13.3.0 has been released

</br>

### PLATFORM

- Bugfix - Development Workspace was not deleted after developer was removed
- Bugfix - TableField scrollbar was overlapping with number cell text
- Bugfix - Create workspace button now only shown to environment administrators
- Bugfix - Privileged API requests were not working on Development Workspaces
- Bugfix - Assignment filter on workers page to work properly work with Development Workspaces
- Bugfix - Granting adminstrator access to a user would revoke their Development Workspace access, showing it to the other Admins
- Simplified workspaces search bar text to 'Search'
 
</br>

### CLI

- Updated `configure` command to only ask for Python path if isolation mode is venv 
- Added support to publish with Python version defined in viktor.config.toml 
- Fixed bug where `configure` command fails to prompt for Python path if Python not in PATH 
- Added support for simple app type 

Newsletter img (2).svg

Support for running and publishing apps on Python 3.8, 3.9, and 3.10, easier unit testing by using mocks, and an improved search bar for the documentation!

VIKTOR platform updates | August 2022

What's new in VIKTOR (August 2022)

[![CTA_Join the community](https://cms.viktor.ai/uploads/CTA_Join_the_community_22d97b332e.svg)](https://community.viktor.ai/)

</br>

## Create apps with only an editor

We have created a new app type called *editor* that allows you to easily create applications with only one page! Implementing the *editor* app type will be the same as for the recently introduced *[simple](https://docs.viktor.ai/docs/create-apps/app-types/#simple-type-app)* app type. However, with the *editor* app type users will directly enter the editor after opening a workspace, making it easier to navigate. This is especially useful if you don't require any data saving, for example when you are creating a dashboard or calculator app. 
 
Find more information about the different app types in the [documentation](https://docs.viktor.ai/docs/create-apps/app-types/). 

<div style="padding:49.84% 0 0 0;position:relative;"><iframe src="https://player.vimeo.com/video/774246715?h=51e3072c9c&badge=0&autopause=0&player_id=0&app_id=58479/embed" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen frameborder="0" style="position:absolute;top:0;left:0;width:100%;height:100%;"></iframe></div>

</br>

## Dedicated page for apps 

We have added a new page to the environment where developers and admins can view, create, and update all their (organization’s) apps. When opening an app from here, you will find information such as a description of the app, its maintainers, and details about the most recent and previously published versions (date, developer). The new Apps page gives us a lot of opportunities to add additional features, such as fine-grained control over the maintainers of apps. Keep on reading in the next highlight for more details. 
 
For more information about the Apps page, please visit the [documentation](https://docs.viktor.ai/docs/manage-apps/workspaces_and_apps/#apps-in-the-interface).

![Blog_newsletter_nov22 Dedicated page for apps](https://cms.viktor.ai/uploads/Blog_newsletter_nov22_Dedicated_page_for_apps_c5fd3c374f.png)


</br>

### Apps vs. Workspaces

The difference between an app and a workspace can be difficult. To make sure all is clear, we will quickly explain both concepts. 

A normal user mainly interacts with workspaces: this is where calculations can be made. Developers and admins however also interact with apps. Apps contain all a programmer's logic and code: how the app is structured, how input is organized, what the output looks like, and what calculations should be performed. In short, a workspace is an app plus data (from users and calculations). 

You can find more information on the apps versus workspaces [here](https://docs.viktor.ai/docs/manage-apps/workspaces_and_apps/#workspace-vs-app). 


</br>

## Assign developers to apps

You can now assign maintainers to your apps. Maintainers are the only developers that have the rights to publish new versions of an app. They also receive error reports on workspaces using this app. This way, access to sensitive app logic can be limited.  

![Blog_newsletter_nov22_maintainers](https://cms.viktor.ai/uploads/Blog_newsletter_nov22_maintainers_9a32c3234a.png)

</br>

## Other changes


**SDK**

- SDK version 13.6.0 is now available


**PLATFORM**

- Fixed issue with editor locking by the same developer 
- Changed icons in entity tree to be more intuitive 
- Fix that sometimes the wrong browser language is selected in the interface 
- Fix null value of generic entity field not present in initial revision 
- Fix inactive app maintainers got error reports for apps 
- VIKTOR_APP_SECRET is now also evaluated during the installation of packages 
- Improved platform stability 


**WORKER**

- Worker version 5.3.0 is now available
- Support for IDEA RCS v22


**CLI**

- CLI version 0.25.0 is now available
- Better error messages when using CLI in CI/CD pipelines
- Support for editor app type



VIKTOR Blog

What's new in VIKTOR (November 2022)