Skip to main content

Inspect format transformations

Compare the authored mappings and representative outputs for each format stage.

How to use this explorer

Move between stages with the arrows or the numbered stage list. Each stage shows the input it receives on the left and the output it produces on the right, with the lines that changed marked. The configuration that makes the change sits below the comparison, and the final stage shows the complete pipeline.

The 4 stages in order

  1. JSON Input
    A representative JSON record before a schema-backed encoding is selected.
  2. Partial snippet: JSON → Avro preparation
    The installed Edge validator reports `Unknown component or field 'avro'` for the documented processor, so this validated stage prepares schema-compatible JSON but does not claim to emit Avro bytes.
  3. Avro → Parquet
    Configure a representative Parquet object output and review column types and partitioning.
  4. Auto-Detection
    Apply the two authored checks for JSON objects and an Avro prefix, then report an unknown format.

Interactive pipeline

FORMAT TRANSFORMATIONS

Curated encoding walkthrough

Stage 1 of 4

JSON Input

A representative JSON record before a schema-backed encoding is selected.

1HighlightsAuthored emphasis only—not a computed diff.
Copy & download

Input

{"sensor_id": "sensor-42","temperature_celsius": 23.5,"humidity_percent": 45.2,"timestamp": "2024-01-15T10:30:00Z"}

Output

Authored highlight: JSON review points:Field names and numeric types are explicitNo schema artifact is attached to this record
Partial snippet01-json-input.yaml
pipeline:
  processors: [] # JSON input; no format conversion yet