L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Design et création · SKILL.md

visualization-export

visualization-export

Construct and export Lux Clause, Vis, and VisList visualizations.

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentation de la Skill

GitHub

name: visualization-export description: "Construct and export Lux Clause, Vis, and VisList visualizations." disable-model-invocation: true metadata: disco-role: operating license: Apache 2.0

Lux visualization export router

Use this sub-skill when the user needs to construct Lux Clause, Vis, or VisList objects directly; enumerate wildcard visualization collections; refresh a visualization against a new dataframe; or export chart code/specifications from an individual Vis.

Route here for

  • Creating a single visualization with lux.vis.Vis.Vis and intent strings or Clause objects.
  • Creating a collection with lux.vis.VisList.VisList, including wildcard "?", list-valued attributes, and manually composed lists of Vis objects.
  • Exporting a single Vis with to_altair(standalone=False), to_matplotlib(), to_vegalite(prettyOutput=True), or to_code(language=...).
  • Handling special-character columns, long labels, and the difference between df.exported[0] as a Vis and multiple exported widget selections as a VisList.
  • Explaining VisList.sort, showK, normalize_score, refresh_source, map, get, and the present-but-not-implemented set method.

Route elsewhere

  • Pandas import-hook basics, dataframe intent, recommendations, df.recommendation, and df.current_vis: use pandas-intent-recommendations.
  • Global plotting backend, plot style, chart scale, label length, top-k, sort, widget setup, and custom recommendation actions: use configuration-actions.
  • Semantic data types, datetime/geographic/id behavior, groupby/index views, and type overrides: use special-data-types.
  • PostgreSQL-backed LuxSQLTable or SQL export workflows: use sql-backend.

Required references

Read these bundled references before giving detailed guidance:

  1. references/api-reference.md for verified Lux 0.5.1 signatures and method behavior.
  2. references/workflows.md for construction, wildcard, refresh, export, standalone-data, and df.exported workflows.
  3. references/troubleshooting.md for multi-visualization errors, export failures, special-character columns, long labels, stale sources, and unsupported code-export languages.

A safe offline smoke script is bundled at scripts/vis_export_smoke.py. Use it to verify that the installed Lux package can construct Clause, Vis, and VisList objects and emit Altair, Matplotlib, and Vega-Lite export tokens on a tiny local dataframe.

Skills enregistrées