Simplify template component handling to use migrations instead of schema placement

[SaxonF]

Summary

This PR refactors the template component system to simplify SQL handling by removing the complex schema placement logic and consolidating all SQL components into a single "migration" component type. This eliminates hundreds of lines of SQL parsing and merging code while providing a clearer, more predictable behavior for template authors.

Key Changes

• Removed SQL merging logic: Deleted ~750 lines of complex SQL statement parsing, merging, and deduplication code that attempted to intelligently merge ALTER statements into CREATE statements. This logic was error-prone and difficult to maintain.

• Unified component types: Replaced multiple schema-based component types (tables, functions, triggers, policies, extensions, etc.) with a single explicit migration component type. Unknown component types now return an error instead of falling back to SQL handling.

• Simplified migration file creation: Migration files are now created directly in the migrations directory with timestamp-based naming (YYYYMMDDHHMMSS_name.sql) rather than being distributed across schema directories based on component type and schema placement configuration.

• Removed schema placement configuration: Eliminated the schema_placement configuration system and related code that allowed customizing where different SQL component types were stored. All migrations now go to a single migrations directory.

• Removed schema_paths configuration: Deleted the logic that automatically added schema_paths to the config when SQL components were added. Migrations are now the primary way to manage database schema changes.

• Deterministic migration timestamps: Added a migrationTimestamp function that accepts a sequence counter to ensure unique timestamps when multiple migrations are added in a single operation, preventing filename collisions.

Implementation Details

• The executeMigrationComponent function replaces the previous executeSQLComponent function with a much simpler implementation that directly writes SQL content to a timestamped migration file.

• Removed dependencies on multigres parser and custom SQL parsing utilities that were used for statement analysis.

• Updated tests to reflect the new behavior: migrations are created with timestamp prefixes in the migrations directory, and unsupported component types now properly return errors instead of silently falling back to SQL handling.

• The runtimeState struct now tracks migrationSeq (sequence counter) instead of addedSql (list of added files).

https://claude.ai/code/session_0161GS2RCbLFW2XqrwtjRYfp

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

🗂️ Base branches to auto review (1)

• develop

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Central YAML (base), Organization UI (inherited)

Review profile: CHILL

Plan: Pro

Run ID: d47881b3-717d-4946-ad12-1d9aed6300c3

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

Tip

Try Coding Plans. Let us write the prompt for your AI agent so you can ship faster (with fewer bugs).
Share your feedback on Discord.

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

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