search

deepset-ai / haystack

:mag: AI orchestration framework to build customizable, production-ready LLM applications. Connect components (models, vector DBs, file converters) to pipelines or agents that can interact with your data. With advanced retrieval methods, it's best suited for building RAG, question answering, semantic search or conversational agent chatbots.
https://haystack.deepset.ai
Apache License 2.0
17.57k stars 1.91k forks source link

Feature Request: Offline Rendering for pipeline.show() #7896

Open touhi99 opened 4 months ago

touhi99 commented 4 months ago

Is your feature request related to a problem? Please describe. The current implementation of pipeline.show() uses Mermaid for graph rendering Visualizing Pipelines , which requires an internet connection to function. This dependency makes it unsuitable for on-premises environments or scenarios where an internet connection is unavailable, limiting the ability to visualize pipelines offline.

Describe the solution you'd like Introduce an alternative rendering engine that does not require an internet connection, allowing users to visualize pipelines offline. This engine should offer similar functionality and quality to the current Mermaid-based rendering.

Describe alternatives you've considered One potential alternative could be developing a Haystack component that utilizes the open-source mermaid.ink service. This service could be integrated to handle rendering within an offline environment, maintaining compatibility with Mermaid's syntax and capabilities.

Additional context

anakin87 commented 4 months ago

For context, previously we were also supporting ghraphviz rendering engine. It was removed in #6961.