`jellycell.config`¶

Pydantic schemas for jellycell.toml and TOML I/O helpers.

The schema here is the source of truth for the project-level config. PEP-723 [tool.jellycell] overrides at file scope are merged on top of this at runtime (handled by the format layer, not here).

Module Contents¶

Classes¶

`ProjectInfo`	The `[project]` table.
`PathsConfig`	The `[paths]` table. All paths are relative to the project root.
`RunConfig`	The `[run]` table.
`ViewerConfig`	The `[viewer]` table.
`LintConfig`	The `[lint]` table.
`JournalConfig`	The `[journal]` table — analysis-trajectory log.
`ArtifactsConfig`	The `[artifacts]` table — how jellycell picks default artifact paths.
`Config`	Full jellycell.toml schema.

Functions¶

default_config

Produce a default jellycell.toml for a new project.

Data¶

ArtifactLayout

API¶

class jellycell.config.ProjectInfo(/, **data: Any)¶

Bases: pydantic.BaseModel

The [project] table.

Initialization

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

model_config¶: ‘ConfigDict(…)’

name: str¶: None

class jellycell.config.PathsConfig(/, **data: Any)¶

Bases: pydantic.BaseModel

The [paths] table. All paths are relative to the project root.

Initialization

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

model_config¶: ‘ConfigDict(…)’

notebooks: str¶: ‘notebooks’

data: str¶: ‘data’

artifacts: str¶: ‘artifacts’

site: str¶

‘site’

Static HTML catalogue — where jellycell render + jellycell export publish their outputs. Distinct from manuscripts/ (prose

tearsheets, markdown, GitHub-native). The live viewer reads and serves from here; projects may git-ignore it if they don’t need a checked-in static site.

manuscripts: str¶: ‘manuscripts’

cache: str¶: ‘.jellycell/cache’

class jellycell.config.RunConfig(/, **data: Any)¶

Bases: pydantic.BaseModel

The [run] table.

Initialization

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

model_config¶: ‘ConfigDict(…)’

kernel: str¶: ‘python3’

subprocess: bool¶: True

timeout_seconds: int¶: 600

class jellycell.config.ViewerConfig(/, **data: Any)¶

Bases: pydantic.BaseModel

The [viewer] table.

Initialization

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

model_config¶: ‘ConfigDict(…)’

host: str¶: ‘127.0.0.1’

port: int¶: 5179

watch: list[str]¶: ‘Field(…)’

class jellycell.config.LintConfig(/, **data: Any)¶

Bases: pydantic.BaseModel

The [lint] table.

Initialization

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

model_config¶: ‘ConfigDict(…)’

enforce_artifact_paths: bool¶: True

enforce_declared_deps: bool¶: False

warn_on_large_cell_output: str¶: ‘10MB’

jellycell.config.ArtifactLayout¶: None

class jellycell.config.JournalConfig(/, **data: Any)¶

Bases: pydantic.BaseModel

The [journal] table — analysis-trajectory log.

When enabled, jellycell run appends a one-section entry to manuscripts/journal.md per invocation: timestamp, notebook, cell-change summary, any new/invalidated artifacts, optional --message commentary. Opt-out-by-default because the audit trail is usually more valuable than clean.

Initialization

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

model_config¶: ‘ConfigDict(…)’

enabled: bool¶

True

Write to <manuscripts_dir>/journal.md on every jellycell run.

Turn off (enabled = false) for transient exploration projects that don’t want the trail, or when the journal lives outside this project.

path: str¶

‘journal.md’

Relative path under manuscripts/ to write to. Defaults to journal.md; override (e.g. "log/runs.md") if you prefer another location.

class jellycell.config.ArtifactsConfig(/, **data: Any)¶

Bases: pydantic.BaseModel

The [artifacts] table — how jellycell picks default artifact paths.

Only affects path-less jc.figure() / jc.table() calls where jellycell chooses the location. Explicit paths (jc.save(x, "artifacts/foo.json")) always win unchanged.

Initialization

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

model_config¶: ‘ConfigDict(…)’

layout: jellycell.config.ArtifactLayout¶

‘flat’

Default artifact layout:

flat (default): artifacts/<name>.<ext>. Simplest; fine for single-notebook projects.
by_notebook: artifacts/<notebook-stem>/<name>.<ext>. Good when one project has many notebooks producing similarly-named artifacts.
by_cell: artifacts/<notebook-stem>/<cell-name>/<name>.<ext>. Every artifact’s path makes its producer obvious to agents and humans without opening the manifest.