Flowfile 0.9.4

Project combining flowfile core (backend) and flowfile_worker (compute offloader) and flowfile_frame (api)

Flowfile

Main Repository: Edwardvaneechoud/Flowfile
Documentation: Website - Core - Worker - Frontend - Technical Architecture

Flowfile is an open-source data platform that combines a visual pipeline builder, a data catalog with Delta Lake storage, scheduling, Kafka ingestion, sandboxed Python execution, and a Polars-compatible Python API — all in a single pip install.

Quick Start

pip install Flowfile
flowfile run ui

This starts the backend services and opens the visual ETL interface in your browser.

What You Get

• Visual pipeline builder with 30+ nodes for joins, filters, aggregations, fuzzy matching, pivots, and more
• Data catalog with Delta Lake storage, version history, and lineage tracking
• Scheduling — interval-based or triggered by catalog table updates
• Kafka/Redpanda ingestion as a canvas node with automatic schema inference
• Sandboxed Python execution in isolated Docker containers
• Code generation — export visual flows as standalone Python/Polars scripts
• Flow parameters — ${variable} substitution, configurable via UI or CLI
• Cloud storage — S3, Azure Data Lake Storage, Google Cloud Storage
• Database connectivity — PostgreSQL, MySQL, SQL Server, Oracle, DuckDB, and more
• Python API with Polars-like syntax and visual flow graph generation

Python API

import flowfile as ff
from flowfile import col, open_graph_in_editor

df = ff.from_dict({
    "id": [1, 2, 3, 4, 5],
    "category": ["A", "B", "A", "C", "B"],
    "value": [100, 200, 150, 300, 250]
})

result = df.filter(col("value") > 150).with_columns([
    (col("value") * 2).alias("double_value")
])

# Open the pipeline on the visual canvas
open_graph_in_editor(result.flow_graph)

Common Operations

import flowfile as ff
from flowfile import col, when, lit

# Read from various sources
df = ff.read_csv("data.csv")
df_pq = ff.read_parquet("data.parquet")

# Transform
filtered = df.filter(col("value") > 150)
with_status = df.with_columns([
    when(col("value") > 200).then(lit("High")).otherwise(lit("Low")).alias("status")
])

# Aggregate
by_category = df.group_by("category").agg([
    col("value").sum().alias("total"),
    col("value").mean().alias("average")
])

# Join
joined = df.join(other_df, left_on="id", right_on="product_id")

# Visualize any pipeline
ff.open_graph_in_editor(joined.flow_graph)

Code Generation

Export visual flows as standalone Python/Polars scripts:

Package Components

• Core Service (flowfile_core) — ETL engine, catalog, scheduler, auth
• Worker Service (flowfile_worker) — CPU-intensive data processing
• Web UI — Browser-based visual pipeline builder
• FlowFrame API (flowfile_frame) — Polars-compatible Python library
• Scheduler (flowfile_scheduler) — Interval and table-trigger scheduling

CLI

flowfile run ui                              # Start web UI
flowfile run core --host 0.0.0.0             # Start core service
flowfile run worker --host 0.0.0.0           # Start worker service
flowfile run flow pipeline.json              # Run a flow
flowfile run flow pipeline.json --param key=value  # Run with parameters

More Options

• Desktop App: Download from GitHub Releases
• Docker: docker compose up -d for self-hosted deployments
• Browser Demo: demo.flowfile.org (WASM, no server)

Resources

• Documentation: Comprehensive guides
• Main Repository: Latest code and examples
• Technical Architecture: Design overview