feat(pkg-py): Add ggsql visualization integration #201
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
cpsievert
force-pushed
the
feat/ggsql-integration
branch
from
c5c8f6e to
34a41da
Compare
Add core support for LLM-generated visualizations using ggsql syntax: - Add ggsql and altair dependencies to pyproject.toml - Create _ggsql.py with helpers for parsing and rendering visualizations - Extend AppState with visualization state fields (filter_viz_*, query_viz_*) - Implement visualize_dashboard and visualize_query tools in tools.py - Add prompt templates for visualization tools with ggsql syntax reference - Update system prompt with ggsql grammar documentation - Add visualization accessor methods to QueryChatBase - Export visualization data types (VisualizeDashboardData, VisualizeQueryData) The ggsql DSL allows the LLM to generate chart specifications that are rendered to Altair/Vega-Lite charts, supporting bar, line, point, area, and boxplot marks with various encodings. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Integrate ggsql visualization capabilities into all supported frameworks: Shiny: - Add visualization state to ServerValues dataclass - Implement ggvis(), ggsql(), ggtitle() reactive accessors - Add filter visualization re-rendering on data change - Create tabbed UI with Data/Filter Plot/Query Plot tabs Streamlit: - Add ggvis(), ggsql(), ggtitle() methods reading from session state - Create tabbed app layout with visualization tabs - Render Altair charts with expandable ggsql specs Gradio: - Add ggvis(), ggsql(), ggtitle() methods taking state dict - Create tabbed Blocks layout with visualization displays - Wire state changes to update all visualization outputs Dash: - Add visualization callbacks and state management - Create tabbed layout with dcc.Graph for Altair charts - Add ggsql spec display in collapsible sections All frameworks enable visualization tools by default and support both filter (dashboard) and query visualizations. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add comprehensive test coverage for ggsql visualization features: - test_ggsql.py: Unit tests for ggsql parsing and rendering - test_ggsql_integration.py: Integration tests for end-to-end visualization - test_viz_tools.py: Tests for visualize_dashboard and visualize_query tools - test_visualization_tabs.py: Playwright tests for UI tab interactions - Update test_state.py with visualization state field tests - Update test_tools.py and test_base.py for new tool configurations Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
cpsievert
force-pushed
the
feat/ggsql-integration
branch
from
34a41da to
a21712c
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
This PR integrates ggsql with querychat to enable LLM-generated SQL visualization queries that render as Altair charts. Two new tools (
visualize_dashboard,visualize_query) extend the existing tool system across all framework integrations.New Features
Two visualization tools:
visualize_dashboard: Creates persistent charts that update when filters changevisualize_query: Creates one-off charts from specific SQL queries with VISUALISE clauseNew accessor methods across all frameworks:
ggvis(source)- Returns Altair Chart objectggsql(source)- Returns ggsql specification stringggtitle(source)- Returns visualization titleTabbed UI: All framework
app()methods now display Data / Filter Plot / Query Plot tabsAuto re-rendering: Filter visualization automatically updates when filtered data changes (Shiny)
System prompt: ggsql syntax reference included when visualization tools are enabled
Implementation Details
qc.ggvis()(reactive)output_widgetqc.ggvis()(session state)st.altair_chart()qc.ggvis(state)(state dict)gr.Plot()qc.ggvis(state)(state dict)html.IframewithsrcDocFiles Changed (Key)
pkg-py/src/querychat/_ggsql.py- New ggsql helper modulepkg-py/src/querychat/tools.py- Visualization toolspkg-py/src/querychat/prompts/tool-visualize-*.md- Tool documentationpkg-py/src/querychat/prompts/prompt.md- System prompt with ggsql syntaxpkg-py/src/querychat/_querychat_base.py- Base class accessor methodspkg-py/src/querychat/_querychat_core.py- State management with viz fieldspkg-py/src/querychat/_shiny*.py- Shiny implementationpkg-py/src/querychat/_streamlit.py- Streamlit implementationpkg-py/src/querychat/_gradio.py- Gradio implementationpkg-py/src/querychat/_dash*.py- Dash implementationTest Coverage
test_ggsql.py- ggsql helper unit teststest_viz_tools.py- Tool function teststest_ggsql_integration.py- Integration tests1. ggsql Dependency
The
ggsqlpackage is currently referenced as a local path dependency:"ggsql @ {root:uri}/../ggsql/ggsql-python"Action needed: Update to PyPI reference once ggsql is published, or determine alternative distribution strategy.
2. Manual Testing
3. Documentation
4. Review Considerations
🤖 Generated with Claude Code