Import Workflows

The CLI can import common local data formats into TaiDB. Imports are useful for building local search indexes, migrating small datasets, or preparing examples.

Import JSONL

Use JSON Lines when each line is an object. TaiDB infers keys and values from the importer implementation.

taidb import ./app.taidb ./data.jsonl --format jsonl

Add a key prefix:

taidb import ./app.taidb ./data.jsonl --format jsonl --prefix docs:

Import CSV

taidb import ./app.taidb ./data.csv --format csv

Use a batch size that fits your memory and durability requirements:

taidb import ./app.taidb ./data.csv --format csv --batch-size 8192

Import a directory

Directory import is useful for local document indexes.

taidb import ./docs.taidb ./notes --format dir --prefix file:

Auto-detect format

taidb import ./app.taidb ./data.jsonl --format auto

Auto-detection is convenient, but explicit formats are better in scripts.

Import from another TaiDB database

taidb import-db ./target.taidb ./source.taidb --prefix old:

If the source database is encrypted:

taidb import-db ./target.taidb ./source.taidb \
  --source-key-file ./source.key \
  --prefix migrated:

Verify after import

taidb stats ./app.taidb --json
taidb verify ./app.taidb

For larger imports, record:

• source file count or row count
• batch size
• compression settings
• encryption settings
• total import time
• final stats --json output

This makes import performance easier to compare across releases.