Skip to main content

Format Catalog

The iterable.catalog module exposes machine-readable metadata about supported data formats, merging the declarative format registry with runtime capability flags.

Functions​

list_formats()​

Return sorted canonical format ids.

from iterable.catalog import list_formats

print(list_formats()[:5])

describe_format(format_id, include_capabilities=True)​

Describe a format by canonical id or alias.

from iterable.catalog import describe_format

info = describe_format("xml")
print(info["example_args"])  # {'tagname': 'item'}
print(info["limitations"])
print(info["capabilities"])

export_catalog(format="dict", include_capabilities=True)​

Export the full catalog as a dict or JSON string.

from iterable.catalog import export_catalog

catalog = export_catalog(format="dict", include_capabilities=False)
json_text = export_catalog(format="json")

CI artifact​

The repository maintains dev/formats.json, generated by:

python dev/scripts/export_formats_json.py

CI tests verify this file matches the live catalog export when the registry changes.

LLM context utilities​

Use iterable.ai.context to build safe prompt samples:

from iterable.ai.context import sample_for_llm, redact_for_llm

rows = sample_for_llm("data.csv", max_rows=10, strategy="stratified")
safe_rows = redact_for_llm(rows)

See AI-Powered Documentation for documentation generation.

Agent index​

See root llms.txt for entry points and links.