Avro migration replay

What it does

Replay sample records from a writer schema into a reader schema and inspect defaults, aliases, promotions and failures locally.

How to use it

1. Paste writer and reader schemas plus a JSON array of sample records, then replay the migration.
2. Inspect reader-visible values and a per-record change trace. Export the report or a reusable fixture bundle.
3. Processing stays in a local worker. No Kafka, schema registry, upload or account connection.

Example

The example renames a field through a reader alias, promotes an int and supplies a reader default while dropping an old field.

Methodology

A bounded interpreter validates sample data against the writer schema and applies Avro reader-schema resolution. Per-record outputs and traces distinguish defaults, aliases, promotions, omitted fields and failures.

Limitations

128 KiB per schema, 1 MiB samples, 100 values, depth 48, 1,500 schema nodes, 200,000 steps, 2 MiB report and a 15-second worker deadline. Avro JSON union tags and byte strings are required; longs use a decimal-string extension for lossless JSON. Logical types use underlying values; decimal annotations must agree. No binary/OCF codec test, Kafka connection, registry lookup or runtime compatibility certification.

Automation

This tool runs in your browser. No upload API or MCP endpoint is available.