SAP Community - Witalij Rudnicki

Re: SAP CAP, HDI and Datasphere

2025-06-17T14:46:50.060000+02:00

Do you plan to only read data from Datasphere, or do you plan to write anything back to Datasphere from the CAP app?

Re: SAP CodeJam Topics

2025-06-23T16:48:03.487000+02:00

@Saikrishna_Sriram thank you for your question. Any particular topics within SAC, Datasphere and/or BDC that you had in mind? As the "CodeJam" event name suggests it is primarily focused on code, not on products.

Re: SAP Community Events in Warsaw - 2025

2025-06-27T19:48:31.721000+02:00

Could you pls add the link to submit talk proposals for SIT?

Re: Cloning a Corporate GitHub Repository (Python) to VS Code on MAC

2025-07-15T19:42:00.652000+02:00

Thanks for sharing this, @SugandhaSachdeva

While I understand what you are trying to achieve, I think the level of details described in your post are not enough for most of the readers to understand the problem and the solution.

🇵🇹 Foundations of AI applications with the SAP HANA Cloud (Porto, PT)

2025-07-16T00:12:22.595000+02:00

This SAP CodeJam event is on the topic of SAP HANA Cloud for AI Applications: Machine Learning, Vectors, and Knowledge Graphs. You do not need to have any background in the topic, but a lot of curiosity and some basic programming skills!

This is a free in-person event and is planned for Friday, September 26th, in Porto, Portugal 🇵🇹 from 10:00 to 16:00 WEST (UTC+1). The language of the event and the content will be in English.

The event is being hosted by SAP Inside Track Porto at
Forte de Gaia, Autograph Collection
R. de Serpa Pinto 60, 4400-307 Vila Nova de Gaia, Portugal

Note that spaces are limited: we can only accommodate 30 attendees.

By the end of the session, you’ll have built examples that demonstrate how SAP HANA Cloud can serve as a versatile back-end for modern Artificial Intelligence applications:

Leverage the SAP HANA Vector Engine to ingest, index and query high‑dimensional embeddings—building a vector store within HANA db that supports semantic similarity search in text.
Implement Retrieval‑Augmented Generation (RAG) by exploring both HANA‑native and hybrid approaches, understanding SQL, PAL, and 3rd party embeddings to enrich generative outputs with precise, up‑to‑date data.
Model and Query a Knowledge Graph inside SAP HANA using RDF triples (subject‑predicate‑object), mastering SPARQL alongside SQL interoperability to extract semantic relationships.
Integrate HANA with LangChain to build AI apps that orchestrate prompt workflows, execute database statements, and navigate knowledge graphs—all directly from Python, in this case.

Re: Request for SAP CodeJam Event in the day before of the SAP Inside Track Porto 2025

2025-07-16T00:54:34.750000+02:00

https://community.sap.com/t5/sap-codejam/foundations-of-ai-applications-with-the-sap-hana-cloud-porto-pt/ev-p/14153367

Re: Request to host a CodeJam in September 2025 (s'Hertogenbosch)

2025-07-16T23:36:37.722000+02:00

Sorry for the delay in responding to you @cVi .

We were busy with all the events until last week. Is this request, month, and topics selection still valid for you?

Best regards,
--Vitaliy (aka Witalij)

Re: Dashboard Design Best-Practices

2025-07-17T12:33:29.002000+02:00

Hi @MarcZauritz . Currently the last link to Best Practices is broken (should be https://community.sap.com/t5/technology-blog-posts-by-sap/sap-analytics-cloud-best-practices/ba-p/14051467). Regards.

Using Python in SAP Business Application Studio – my notes

2025-07-18T13:55:15.744000+02:00

The Python Tools extension, which enhances the Python coding experience, was introduced in SAP Business Application Studio (referred to as "BAS" below) exactly two years ago. Here are my notes from using it.

This post is not a tutorial or a comprehensive guide to best practices. It’s a collection of personal notes, which I hope you find helpful when working with Python in BAS. The focus here is on running Python code in BAS, not on deploying to SAP BTP runtimes like Cloud Foundry (CF) or Kyma. This is when you are usually working on Machine Learning or AI projects in BAS, like during SAP CodeJams.

I assume you're not an absolute beginner with SAP Business Application Studio—or at least you're familiar with Visual Studio Code.

For the examples in this post, I’ll be using the SAP Business Application Studio available in the SAP BTP Trial environment.

Python runtime

In SAP Business Application Studio, the dev space includes a system-level Python 3 binary at /bin/python3, which is primarily intended for OS-level scripts and tooling. This version is tied to the base container image and is only updated when the image itself is refreshed.

Managing Python versions

For application development, developers typically use user-managed Python runtimes. Since January 2024, you can select the Python version with which you want to work.

SAP Business Application Studio uses asdf to allow you to select which runtime versions to install and use for developing your application. You can check this in the BAS terminal with asdf current python.

By default, SAP Business Application Studio provides only one officially supported Python version.

If you need another version of Python, you can install it with asdf install python <version>. If you want the latest minor version, then use latest:, like asdf install python latest:3.12 in the case of the 3.12 version of Python.

At the time of writing this post, the asdf version used in BAS is 0.12.0. This version uses the asdf global and asdf local commands to set the actual runtime version. If you refer to the asdf documentation, these commands have been replaced with asdf set, so don't get confused.

Runtime management from the command palette

You can also install different versions from the command palette. Select > Runtime: Install, then Python, and the version you want to install. To set the default version of Python for execution, use the command > Runtime: Set Default.

These commands are provided by the built-in BAS Framework extension.

You can now run your Python programs in BAS...

...using a version of Python accordingly to your requirements.

Design-time

Being a fork of Visual Studio Code - Open Source ("Code - OSS")—SAP Business Application Studio provides basic support for Python files and Jupyter notebooks out of the box.

BAS additional extension "Python Tools"

To improve your Python coding experience, SAP Business Application Studio provides an additional extension called "Python Tools". This extension includes IntelliSense, formatting, linting, and debugging support for Python files and Jupyter notebooks.

To add "Python Tools" to your BAS dev space, go to the configuration of a stopped or a newly created dev space, and select Python Tools from Additional SAP Extensions.

Upon activation, you should see additional extensions listed now as built-in in BAS.

Note that BAS installs its extensions from Open VSX, an open-source registry for VS Code extensions. VS Code's Python extension is available at: https://open-vsx.org/extension/ms-python/python.

You can see Python extensions in the file system:

ls -lad /extbin/local/openvscode-server/extensions/ms-py*

These are extensions selected and integrated by the BAS product team at SAP.

Additional notes:

You can install other versions of the same extensions directly from https://open-vsx.org/, but there is no guarantee that they have been tested and will properly work with BAS. First of all, the extension version installed directly from the marketplace should be compatible with the version of SAP Business Application Studio (or the "Code—OSS" to be more precise).

For example, at the time of writing this post, the version of the Python extension at the VSX marketplace is 2025.04...

which is compatible with version 1.94.0 of Code-OSS and therefore BAS: https://github.com/microsoft/vscode-python/blob/v2025.4.0/package.json#L50

Usage of the Python extension in BAS

Upon installation of the Python extension in BAS, you can:

1/ Execute Python extension's commands from the palette:

2/ Trigger an execution of a program from the editor:

...which you can configure by editing the python.terminal settings:

3/ modify its @ext:ms-python.python settings:

Python virtual environments

To run your code, you usually need additional Python packages, such as the Python machine learning client for SAP HANA (hana-ml) or SAP Cloud SDK for AI (Python).

If you have experience with Python, then you already know all the advantages of using virtual environments. Although you can think of BAS devspace as an isolated project development environment, there is a good reason to use Python virtual environments there, as I already explained in Persisting Python environment in SAP Business Application Studio.

An additional reason to use a virtual environment is if you plan to test or run your project using different versions of Python managed by asdf. Each of the installed versions of Python will use their own location to store packages:

You should be ready to work with Python in SAP Business Application Studio now!

Please share your tips in the comments!

In my other blog posts, I focused on:
🪐 Using Jupyter in SAP Business Application Studio, and
🔨 Using conda-forge in SAP Business Application Studio.

------

-Vitaliy, aka @Sygyzmundovych

Re: 🇵🇱 Build AI services using the SAP Cloud Application Programming Model and SAP generative AI H

2025-07-18T13:59:07.628000+02:00

Sorry, I won't be able to meet and greet you in Wrocław at that time @KevinR

Enjoy it in Wrocław!

Re: 🇩🇪 Getting started with Generative AI Hub on SAP AI Core

2025-07-22T15:00:43.722000+02:00

We will be waving to you from Porto 🇵🇹 on that day from HANA AI CodeJam: https://community.sap.com/t5/sap-codejam/foundations-of-ai-applications-with-the-sap-hana-cloud-porto-pt/ev-p/14153367 👋

Re: 🇩🇪 Build AI services using the SAP Cloud Application Programming Model and Generative AI Hub

2025-07-22T15:01:19.303000+02:00

Re: Mio Yasutake

2025-07-22T15:26:51.245000+02:00

Great to see you among SAP Champions @MioYasutake 👏🏼

And it is somewhat difficult to answer your question, "What blog post or answer have you written that you're most proud of—and why?", cause I wrote a lot of them. But I think my fav is my series about understanding containers: https://community.sap.com/t5/tag/UnderstandContainers/tg-p/board-id/technology-blog-sap

Why? Because the best blog posts of yours are those you return to quite often as your own notes 🤓

Re: Marla Alschweiki

2025-07-22T15:45:50.057000+02:00

Congrats on joining SAP Mentors @Marla-Schwaiki 👏🏼

"Collaboration and continuous learning are key." I love it 🤓

Re: How to use Gemini 2.5 Flash Explicit Context Cache Capability with GenAI Hub SDK

2025-07-24T18:32:29.794000+02:00

If you are using Google native integration https://help.sap.com/doc/generative-ai-hub-sdk/CLOUD/en-US/_reference/gen_ai_hub.html#google-vertex-ai, then I can see it is based on Google Vertex AI, not Gemini API, so you'd rather need to reference this doc: https://cloud.google.com/vertex-ai/generative-ai/docs/context-cache/context-cache-create

But then I see caches are configured at the Google client level, not model: https://googleapis.github.io/python-genai/genai.html#genai.caches.Caches.create

So, by looking just at the code, I cannot see how you can program a cache with the Python Genaihub SDK. But it is just me quickly looking into the code without the definitive answer.

Maybe someone from the genAI Hub dev team can share more.

Regards,
--Witalij

Re: CodeJam Request Lisbon - September 2025

2025-07-30T15:37:31.118000+02:00

Hi @Jorge_Herrero_Blanco

I will be in Porto that week doing the SAP CodeJam on the topic of AI: https://community.sap.com/t5/sap-codejam/foundations-of-ai-applications-with-the-sap-hana-cloud-porto-pt/ev-p/14153367

If you are interested in that topic or in the "Getting started with Generative AI Hub on SAP AI Core" as somewhat more introductory into AI, then I could do these in Lisbon, but on Monday, September 29th.

Regards,
--Witalij

Re: SAP HANA

2025-07-31T18:46:10.219000+02:00

There are several options listed on the page; https://help.sap.com/docs/SAP_ANALYTICS_CLOUD/00f68c2e08b941f081002fd3691d86a7/bdf055159cbb4f36b26c93ccb0c56066.html?locale=en-US&version=LATEST

Is that what you were looking for?

Using Jupyter in SAP Business Application Studio – my notes

2025-08-01T17:29:38.575000+02:00

In my previous blog post Using Python in SAP Business Application Studio – my notes, I focused on Python.

This blog post focuses on running Jupyter notebooks in SAP Business Application Studio (referred to as "BAS" below). I assume you're not an absolute beginner with SAP Business Application Studio—or at least you're familiar with Visual Studio Code. If you need a general introduction to navigating, coding, and executing code in notebooks using the Jupyter extension, then check: https://code.visualstudio.com/docs/datascience/jupyter-notebooks#_create-or-open-a-jupyter-notebook

For the examples in this post, I’ll be using the SAP Business Application Studio, which is available in the SAP BTP Trial environment.

Creating a virtual environment with the Python extension

As already discussed in Persisting Python environment when using Jupyter notebooks in SAP Business Application Studio, one of the first things you might want to do, when working with the Python code in BAS, is to create a virtual environment. There, we discussed how to use the `python -m venv` command to create it. Now, let's use the extension commands.

🤓 Pro tip: if you want to understand what is happening when extension commands are executed, then:

Open the Output panel (eg, View -> Output from the menu).
Choose the output for the Python extension, and then in Settings, set the level to Info and Set As Default:
Alternatively, if the Python extension has not yet been loaded, then open the command palette and run the command >Python: Show Output:

Go to the command palette and run the command `Python: Create Environment...`...

..and select the Venv option to create a virtual environment `.venv` (the name cannot be modified, at least in the current version of the Python extension):

...selecting the required base Python run-time installation (like ~/.asdf-inst/shims/python😞

If you have set output on the Info level for the Python extension, then you should see (interesting) details on how the environment was created:

One additional note: the .gitignore file has been automatically created for the virtual environment:

Select the Jupyter kernel

To execute code in your Jupyter notebook, you need to select a Jupyter kernel that will be used to execute the code.

Click Select Kernel and then on Python Environments... category:

...which will show you the list of Python environments that the Python extension found in your BAS dev space. Click on your virtual environment created for your project, like .venv:

Execute code in a Jupyter notebook

During the first execution of your code, Jupyter will check if the IPython Kernel executable (the Python package ipykernel) is installed in the selected Python environment. If not, it will automatically install it, as can be seen in the Jupyter extension's output, and then start it:

Once started, the Kernel exposes 5 ports for communication with your notebook, which can be seen in the run-time specification in a file ~/.local/share/jupyter/runtime/kernel-*.json. While this information should not be required in most of the basic scenarios, you might find it overwhelming that BAS displays pop-up messages for each port every time the kernel runtime starts:

While information about open ports is relevant when developing front-end applications using SAP Business Application Studio, it might be useless and overwhelming when working with Jupyter notebooks. So, click on the settings icon and turn off notifications from the "Exposing router ports" extension or disable this extension completely:

Limiting listed Python environments

You might also find it quite overwhelming to deal with the list of all the global Python locations when setting the kernel for a notebook. To reduce this list, select the command Jupyter: Filter Kernels from the command palette:

...which opens the setting jupyter.kernels.excludePythonEnvironments. Switch to the Workspace tab and click Add Item to specify which Python environments to exclude from the list. You might want to add the following items to the list:

/bin/python3
/usr/bin/python3
~/.asdf-inst/shims/python3.13
~/.asdf-inst/shims/python3
~/.asdf-inst/shims/python

🤓Pro tip: these values can also be edited in the project's file .vscode/settings.json.

{
    "jupyter.kernels.excludePythonEnvironments": [
        "/bin/python3",
        "/usr/bin/python3",
        "~/.asdf-inst/shims/python3.13",
        "~/.asdf-inst/shims/python3",
        "~/.asdf-inst/shims/python"
    ]
}

Next time you need to pick the kernel for a Jupyter notebook, you should see only your project's virtual Python environment:

You should be ready to work with your Jupyter notebooks in SAP Business Application Studio!

Please share your tips in the comments!

------

-Vitaliy, aka @Sygyzmundovych

Using conda-forge in SAP Business Application Studio – my notes

2025-08-01T23:07:46.509000+02:00

In my previous blog posts, I focused on Using Python in SAP Business Application Studio and Using Jupyter in SAP Business Application Studio with Python's virtual environments. But SAP Business Application Studio (referred to as "BAS" below) also allows you to work with the Conda environments:

Usually, Conda is used when one needs binary package management, and installation with pip fails because of the build requirements that cannot be successfully completed with the tools and authorizations a developer has in their BAS dev space.

👉 At the time of writing this post, using Conda might require additional commercial licenses from Anaconda, Inc. if downloading their tools and/or packages from their `default` channel. This might be a valid case for you, but this blog post covers only the use of the community-driven `conda-forge` channel and the minimum tools required to use it. To the best of my knowledge, I discuss only the technical setup here, but you might need to check any license implications for the dependencies of a project you are working on.

Install miniforge

Miniforge is a minimal installer for Conda and Mamba with the conda-forge channel set as the default (and only) channel.

Follow the installation steps, for example from https://github.com/conda-forge/miniforge?tab=readme-ov-file#unix-like-platforms-macos-linux--wsl

Download and run the installation script.

curl --location --remote-name --output-dir ~/tmp "https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-$(uname)-$(uname -m).sh"

bash ~/tmp/Miniforge3-$(uname)-$(uname -m).sh -b

Configure conda

Activate conda and its base environment.

eval "$(/home/user/miniforge3/bin/conda shell.$(basename "${SHELL}") hook)"

Initialize conda (it will add the shell's integration).

conda init

To prevent the conda's base environment from being activated on startup (because it will be taken care of by the Python extension), run the following command:

conda config --set auto_activate_base false

For changes to take effect, close and re-open your current shell.

Update conda's base environment to the latest packages from the community-maintained conda-forge channel.

conda update -n base --all --yes

You can also use the faster mamba command instead of conda. Both are parts of the Miniforge installation.

Point Python extension to your Conda executable

In Terminal, check the location of the conda command, which should be something like /home/user/miniforge3/condabin/conda:

which conda

In Settings, open python.condaPath and switch to the Remote tab. Input the value.

Create a .conda environment in your project

Open your project folder in BAS.

From the command palette, start the command Python: Create Environment..., then Conda environment type...

...and select the required Python version from the list:

You should see the message that the environment is created and set as the active one for the project.

Open a new Terminal session, and you should see the .conda environment set.

You can check the list of environments with the command:

conda env list

Install IPyKernel in your .conda environment

Unlike in a "venv" virtual environment, the Python extension might not automatically install IPyKernel in a conda environment. You need to install it manually.

In Terminal, with the .conda environment set in the command line, run the command:

conda install ipykernel --yes

Check with the command

conda list ipykernel

Select the .conda env in your notebook

Now, you can open your project's Jupyter notebook, select the .conda environment, and execute the code 🤓

You should be ready to work with conda-forge in SAP Business Application Studio now!

Please share your tips in the comments!

In my other blog posts, I focused on:
🐍Using Python in SAP Business Application Studio, and
🪐 Using Jupyter in SAP Business Application Studio.

------

-Vitaliy, aka @Sygyzmundovych

Re: From Mechanical Engineering to SAP: My Journey, Chall...

2025-08-01T23:20:52.532000+02:00

Thanks for sharing your story here @RACHID_AITBOUALE

I think you should publish it as well in the https://community.sap.com/t5/welcome-corner/gh-p/welcome-corner

Best regards.

SAP Community - Witalij Rudnicki

Re: SAP CAP, HDI and Datasphere

Re: SAP CodeJam Topics

Re: SAP Community Events in Warsaw - 2025

Re: Cloning a Corporate GitHub Repository (Python) to VS Code on MAC

🇵🇹 Foundations of AI applications with the SAP HANA Cloud (Porto, PT)

Re: Request for SAP CodeJam Event in the day before of the SAP Inside Track Porto 2025

Re: Request to host a CodeJam in September 2025 (s'Hertogenbosch)

Re: Dashboard Design Best-Practices

Using Python in SAP Business Application Studio – my notes

Python runtime

Managing Python versions

Runtime management from the command palette

You can now run your Python programs in BAS...

Design-time

BAS additional extension "Python Tools"

Additional notes:

Usage of the Python extension in BAS

Python virtual environments

You should be ready to work with Python in SAP Business Application Studio now!

In my other blog posts, I focused on:🪐 Using Jupyter in SAP Business Application Studio, and🔨 Using conda-forge in SAP Business Application Studio.

Re: 🇵🇱 Build AI services using the SAP Cloud Application Programming Model and SAP generative AI H

Re: 🇩🇪 Getting started with Generative AI Hub on SAP AI Core

Re: 🇩🇪 Build AI services using the SAP Cloud Application Programming Model and Generative AI Hub

Re: Mio Yasutake

Re: Marla Alschweiki

Re: How to use Gemini 2.5 Flash Explicit Context Cache Capability with GenAI Hub SDK

Re: CodeJam Request Lisbon - September 2025

Re: SAP HANA

Using Jupyter in SAP Business Application Studio – my notes

Creating a virtual environment with the Python extension

Select the Jupyter kernel

Execute code in a Jupyter notebook

Limiting listed Python environments

You should be ready to work with your Jupyter notebooks in SAP Business Application Studio!

Using conda-forge in SAP Business Application Studio – my notes

Install miniforge

Configure conda

Point Python extension to your Conda executable

Create a .conda environment in your project

Install IPyKernel in your .conda environment

Select the .conda env in your notebook

You should be ready to work with conda-forge in SAP Business Application Studio now!

In my other blog posts, I focused on:🐍Using Python in SAP Business Application Studio, and🪐 Using Jupyter in SAP Business Application Studio.

Re: From Mechanical Engineering to SAP: My Journey, Chall...

In my other blog posts, I focused on:
🪐 Using Jupyter in SAP Business Application Studio, and
🔨 Using conda-forge in SAP Business Application Studio.

In my other blog posts, I focused on:
🐍Using Python in SAP Business Application Studio, and
🪐 Using Jupyter in SAP Business Application Studio.