[xvalovic]

Summary

Two features for the Legacy → Cloud migration:

1. Computed Attributes (CA). Legacy computed attributes are now migrated with the metrics step. Before this PR, any metric, insight or dashboard that used a CA failed to migrate.
2. Creator/modifier of metrics and insights. dashboard-permissions can now set createdBy/modifiedBy on metrics and insights too, not only on dashboards.

1. Computed attributes (c5cd9c9)

Legacy CA = attr.comp.X + label.comp.X. Each one becomes a Cloud computedAttribute entity with id label.comp.X:

SELECT CASE WHEN SUM({metric/<id>}) <= 10 THEN "0-10" ... ELSE "..." END BY {label/<base>} ALL OTHER

• Fetch and translate (metrics/computed_attribute.py, new): rel definitions are parsed with a strict grammar: one metric, numeric thresholds, literal bucket labels. A CA that does not match is logged as cannot be migrated and skipped, and the run continues.
• Deploy order (workflows/migrate_metrics.py): plain metrics → computed attributes → metrics that reference a CA. CA creation uses the same create / --overwrite-existing semantics as other objects (ComputedAttributeCreationStrategy).
• References. All references now use computed_attribute/… / "type": "computedAttribute" instead of label:
  • metric MAQL (content_tree_modifier.py)
  • insight attribute buckets, filters, and COUNT(attr.comp.X) measures (cloud_insight.py)
  • dashboard filter contexts (filter_context.py) and drills (drill_converter.py)
• Id mapping. IdMappings resolves attr.comp.* / label.comp.* even without a mapping row. The metrics step also writes computed_attribute_mappings.csv.
• Outputs: cloud_failed_computed_attributes.json and cloud_skipped_computed_attributes.json. With --dump-legacy / --dump-cloud, also legacy_computed_attributes.json and cloud_computed_attributes.json.

Requirements and limitations (also in the README)

• Cloud CA is Beta. It needs the enableComputedAttributes feature flag on the parent and on every child workspace where the migration runs; the flag is not inherited.
• Without the flag, GET …/computedAttributes returns 400 ("Feature 'enableComputedAttributes' is not enabled"), so the metrics step fails for that workspace, even when it has no CA content.
• Cloud CAs do not support exports, date dimensions or period-over-period.
• The translation wraps the metric in SUM(). This is exact for additive metrics. For metrics that are already averages (e.g. Avg Wait Time, Avg Agent Handle Time), bucket values may differ from Legacy and should be checked visually.

2. Creator/modifier of metrics and insights (e6991ec)

• dashboard-permissions gets --include-metrics and --include-insights. Legacy author → Cloud createdBy, Legacy contributor → Cloud modifiedBy, written through the workspace layout. Metrics and insights have no permissions, so only these two fields change. The objects are read from metric_mappings.csv / insight_mappings.csv (--metric-mapping-file / --insight-mapping-file).
• Behaviour change: dashboards now take modifiedBy from the Legacy contributor instead of author.

Testing

• Unit tests: pytest, 119 passed. New tests:
  • tests/test_computed_attributes.py: CA translation, unsupported definitions, mappings
  • CA cases for metrics and insights (with_computed_attribute_*)
  • tests/test_layout_manager.py: createdBy / modifiedBy for metrics and insights

[coderabbitai]

Important

• 🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

⚙️ Run configuration

• Configuration used: Organization UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: fa16bc0a-3cf3-4990-8618-953b845a00a6

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

_{Comment @coderabbitai help to get the list of available commands.}

[marmil-cz]

LGTM