feat(annotations): enhance deserialization to handle list format and flatten position objects

[palewire]

• Update deserialize_model() to accept both dict and list input formats
• Add position object flattening to extract x0/x1/y0/y1 (range) and x/y (text) fields
• Improve handling of annotations from different sources (API vs serialization)
• Add comprehensive test coverage for new deserialization scenarios

This change makes the annotation models more flexible by supporting multiple input formats and automatically flattening nested position objects into their component coordinate fields, improving compatibility across different data sources.

[Copilot]

Pull Request Overview

This PR enhances annotation model deserialization to handle multiple input formats (dict with UUID keys from API and list format from serialization) and automatically flattens nested position objects into top-level coordinate fields (x/y for text annotations, x0/x1/y0/y1 for range annotations). This improves compatibility across different data sources.

• Updated deserialize_model() methods to accept both dict[str, dict] and list[dict] input formats
• Added position object flattening to extract coordinate fields directly to the annotation dictionary
• Added comprehensive integration tests demonstrating real-world usage with scatter plot annotations

Reviewed Changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
tests/unit/models/test_range_annotation_permissive.py	Updated test assertions to reflect flattened position fields at top level
tests/samples/scatter/policy-poll.json	Added sample chart configuration with text annotations for integration testing
tests/samples/scatter/policy-poll.csv	Added sample CSV data corresponding to policy-poll.json
tests/integration/test_scatter_annotations.py	Added comprehensive integration tests for scatter plot annotation deserialization
datawrapper/charts/multiple_column.py	Updated deserialization to handle both dict and list formats for MultipleColumn annotations
datawrapper/charts/models/text_annotations.py	Updated deserialization with list format support and position flattening
datawrapper/charts/models/range_annotations.py	Updated deserialization with list format support and position flattening

Comments suppressed due to low confidence (2)

datawrapper/charts/multiple_column.py:138

• The 'position' object is extracted from anno_data but never removed from anno_dict. This leaves the nested position object in the result alongside the flattened x/y fields, which is inconsistent with the behavior in text_annotations.py (line 329) and range_annotations.py (line 191) where position is removed with pop(). Add 'anno_dict.pop("position", None)' after line 125 to maintain consistency.

        for anno_id, anno_data in items_to_process:
            # Extract position data
            position = anno_data.get("position", {})
            x = position.get("x") if isinstance(position, dict) else None
            y = position.get("y") if isinstance(position, dict) else None
            plot = position.get("plot") if isinstance(position, dict) else None

            # Extract showInAllPlots (defaults to False for text annotations)
            show_in_all = anno_data.get("showInAllPlots", False)

            # Build annotation dict with id
            anno_dict = {**anno_data, "id": anno_id}

            # Add position fields
            if x is not None:
                anno_dict["x"] = x
            if y is not None:
                anno_dict["y"] = y

            # Add MultipleColumnChart-specific fields
            if plot is not None:
                anno_dict["plot"] = plot
            anno_dict["show_in_all_plots"] = show_in_all

            result.append(anno_dict)

datawrapper/charts/multiple_column.py:138

• Missing connector line handling that exists in the parent TextAnnotation.deserialize_model() (lines 334-341 in text_annotations.py). MultipleColumnTextAnnotation should handle connectorLine deserialization by checking if enabled is False and setting it to None, consistent with the parent class behavior.

        for anno_id, anno_data in items_to_process:
            # Extract position data
            position = anno_data.get("position", {})
            x = position.get("x") if isinstance(position, dict) else None
            y = position.get("y") if isinstance(position, dict) else None
            plot = position.get("plot") if isinstance(position, dict) else None

            # Extract showInAllPlots (defaults to False for text annotations)
            show_in_all = anno_data.get("showInAllPlots", False)

            # Build annotation dict with id
            anno_dict = {**anno_data, "id": anno_id}

            # Add position fields
            if x is not None:
                anno_dict["x"] = x
            if y is not None:
                anno_dict["y"] = y

            # Add MultipleColumnChart-specific fields
            if plot is not None:
                anno_dict["plot"] = plot
            anno_dict["show_in_all_plots"] = show_in_all

            result.append(anno_dict)

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Similar to MultipleColumnTextAnnotation, the 'position' object should be removed from anno_dict after extracting fields to maintain consistency with range_annotations.py. The position object remains in the result dictionary alongside the flattened fields.

[palewire]

@copilot open a new pull request to apply changes based on this feedback