Skip to main content

Storyboard data model and exporters — lacing-native panels along a timeline

Project description

artful

Storyboard data model and exporters — lacing-native panels along a timeline.

A storyboard is a sequence of panels along a timeline (a song, a video, a podcast clip). Each panel pins an interval of the master asset, optionally points at a project shot, and carries one or more images plus directorial annotations (caption, framing, camera, transition, notes). Panels are persisted as lacing annotations, so a storyboard is queryable, exportable, and round-trippable through every adapter lacing already supports (TextGrid, EAF, JAMS, OTIO, WebVTT, Web Annotation, …) without artful having to reinvent any of it.

artful is not the renderer. It's the panels-along-a-timeline plan that drives a renderer: an LLM authors panels from a script, a human reviews the contact sheet, the same panel data feeds downstream image / video generation.

Install

pip install artful

Requires pydantic>=2.6 and lacing>=0.0.13.

Quick start

from artful import (
    PanelBody, PanelImage, Storyboard,
    panel_intervals_from_panels, save_storyboard, load_storyboard,
)
from lacing import MemoryStore

sb = Storyboard(
    title="The Bells — v1",
    asset_id="song-asset-id-abc",
    style="noir, candlelight",
    panels=(
        PanelBody(
            panel_id="p1",
            caption="Thor at the piano",
            framing="medium",
            images=(PanelImage(path="composite.png", role="seed"),),
        ),
        PanelBody(
            panel_id="p2",
            caption="Bells over winter sky",
            framing="wide",
            camera="slow push-in",
        ),
    ),
)

# Pin each panel to a time interval (seconds on the master timeline).
intervals = panel_intervals_from_panels([
    ("p1", 0.0, 4.0),
    ("p2", 4.0, 8.0),
])

store = MemoryStore()
save_storyboard(sb, store, panel_intervals=intervals)

# ... later, possibly in a different process / backend ...
loaded = load_storyboard(store, asset_id="song-asset-id-abc")

The store can be any lacing.IntervalAnnotationStore (MemoryStore, SQLiteStore, PostgresStore, …); the persistence layer is lacing's.

Authoring with LLMs (Markdown)

Markdown is artful's canonical "give an LLM a storyboard to read or write" format. It round-trips losslessly through to_markdown / from_markdown.

from artful import to_markdown, from_markdown

md = to_markdown(loaded, intervals)
# Show `md` to an LLM, let it edit panels, then…
edited_sb, edited_intervals = from_markdown(edited_md)
save_storyboard(edited_sb, store, panel_intervals=edited_intervals)

The Markdown shape:

# The Bells — v1

- asset_id: `song-asset-id-abc`
- style: noir, candlelight
- aspect: 16:9

## panel p1 [0.00..4.00]s

- framing: medium

Thor at the piano

![seed](composite.png)

## panel p2 [4.00..8.00]s

- framing: wide
- camera: slow push-in

Bells over winter sky

Reviewing as a contact sheet (HTML)

from artful import to_html

with open("storyboard.html", "w") as f:
    f.write(to_html(loaded, intervals))

to_html produces a self-contained HTML page with embedded styles and <img> tags pointing at each panel's path / url. Open it in a browser to review.

Data model

Type What it is
Storyboard Title, asset_id (the timeline), panels, style hint, aspect ratio.
PanelBody One panel: panel_id, optional shot_id, images, caption, framing, camera, transition_in, notes. The interval lives on the lacing annotation's reference, not in the body.
PanelImage One image for a panel — artifact_id (sha-256 lacing artifact ref), url, or path, plus a role (thumbnail | seed | reference | alternate) and caption.

All three are frozen Pydantic models with extra="forbid". Importing artful registers PanelBody against lacing under the body-schema URI annot://schema/storyboard-panel/v1, so lacing.validate_body(...) works out of the box.

Why lacing-native?

Storyboards are an annotation problem: panels are intervals on a timeline with structured bodies. Rather than invent a new persistence layer, artful encodes each panel as a lacing.Annotation whose body_schema_uri is the panel schema. The benefits:

  • Allen interval algebra for free: overlaps, gaps, adjacencies.
  • Multiple storyboards over the same asset stay separate via the tier field on each annotation.
  • Provenance tracked across edits through lacing's Provenance model.
  • Format adapters — read or write panels as TextGrid tiers, JAMS annotations, OTIO clips, Web Annotation JSON-LD, etc., via lacing.

A small StoryboardMetaBody schema (URI annot://schema/storyboard-meta/v1) carries title / style / aspect as a single timeless annotation alongside the panels.

API surface

from artful import (
    # data model
    Storyboard, PanelBody, PanelImage,
    new_panel_id,

    # persistence (round-trip with any lacing.IntervalAnnotationStore)
    save_storyboard, load_storyboard,
    panel_intervals_from_panels,

    # exports
    to_markdown, from_markdown,
    to_html,

    # body-schema URIs
    PANEL_BODY_SCHEMA_URI,
    STORYBOARD_META_BODY_SCHEMA_URI,
    StoryboardMetaBody,
)

PDF export is deferred to the optional [pdf] extra (reportlab).

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

artful-0.0.7.tar.gz (22.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

artful-0.0.7-py3-none-any.whl (16.1 kB view details)

Uploaded Python 3

File details

Details for the file artful-0.0.7.tar.gz.

File metadata

  • Download URL: artful-0.0.7.tar.gz
  • Upload date:
  • Size: 22.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.11.21 {"installer":{"name":"uv","version":"0.11.21","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}

File hashes

Hashes for artful-0.0.7.tar.gz
Algorithm Hash digest
SHA256 1febce9f64dc306c743631b3ea314f52c5f14da4b028f24a446545fb3f92daed
MD5 88b71558361f274f9046087a76b2a22b
BLAKE2b-256 8898bd05abcdd474d9cd3fdf982af76b6decb073440cd801e3b64897b1636f9e

See more details on using hashes here.

File details

Details for the file artful-0.0.7-py3-none-any.whl.

File metadata

  • Download URL: artful-0.0.7-py3-none-any.whl
  • Upload date:
  • Size: 16.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.11.21 {"installer":{"name":"uv","version":"0.11.21","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}

File hashes

Hashes for artful-0.0.7-py3-none-any.whl
Algorithm Hash digest
SHA256 8a43ee0aa1cd7aa2c09829c84e17621dbdd101c0bd7e6cc4d61f242a9b9e8640
MD5 ddde7f9e0bf6994ffe2959e6a6ad51a2
BLAKE2b-256 7d85bd5cc3356829c90c0ab00b36456673fc34be9154ae76742c711b7f8e269b

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page