feat(wren): add context init --from-mdl to import MDL JSON as YAML project

[goldmedal]

Summary

• Extends wren context init with --from-mdl <path> to convert a legacy camelCase MDL JSON into a v2 YAML project in one command
• Adds --force flag to allow overwriting an existing project
• Pure file-conversion — no database connection required, consistent with context build

Changes

wren/src/wren/context.py

• _camel_to_snake() — known mapping table + generic regex fallback
• _convert_keys_to_snake() — recursive key conversion (inverse of _convert_keys)
• ProjectFile dataclass — relative_path + content
• convert_mdl_to_project() — MDL JSON → list of ProjectFile (pure function, testable)
• write_project_files() — writes files to disk with --force guard

wren/src/wren/context_cli.py

• init extended: --from-mdl (optional path) and --force (bool)
• Without --from-mdl: existing scaffold behavior unchanged
• With --from-mdl: loads JSON, converts, writes, prints next-steps hint

wren/tests/unit/test_convert_mdl.py (new, 25 tests)

• _camel_to_snake parametric tests + round-trip guarantee for all known keys
• Full integration test (models, views, relationships, instructions)
• write_project_files guard tests (refuse overwrite, force overwrite)
• convert → build round-trip test
• Edge cases: empty MDL, missing dataSource, unknown camelCase keys, empty relationships
• CLI tests via typer.testing.CliRunner

Test plan

• uv run pytest tests/unit/test_convert_mdl.py — 25/25 passed
• just lint — clean

🤖 Generated with Claude Code

Summary by CodeRabbit

• New Features
  • Import MDL JSON files into Wren projects using wren context init --from-mdl
  • Generate complete project structure with YAML configurations and SQL files from MDL source
  • Use --force flag to overwrite existing project configurations during initialization

[coderabbitai]

Important

Review skipped

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

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: defaults

Review profile: CHILL

Plan: Pro

Run ID: 08eeaf7f-2a7d-44a0-b551-6d8c7721c3a5

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:

• ✅ Review completed - (🔄 Check again to review again)

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feat/init-from-mdl

---

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.}

[coderabbitai]

🧹 Nitpick comments (3)

wren/src/wren/context_cli.py (1)

59-65: Add error handling for invalid JSON.

If the MDL file contains malformed JSON, json.loads raises JSONDecodeError with a raw traceback. Wrapping this in a try-except would provide a cleaner user experience.

🛡️ Proposed fix to handle JSON parse errors

-        mdl_json = json.loads(mdl_path.read_text())
+        try:
+            mdl_json = json.loads(mdl_path.read_text())
+        except json.JSONDecodeError as e:
+            typer.echo(f"Error: Invalid JSON in {mdl_path}: {e}", err=True)
+            raise typer.Exit(1)
         files = convert_mdl_to_project(mdl_json)

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@wren/src/wren/context_cli.py` around lines 59 - 65, Wrap the JSON parsing
call so malformed MDL yields a friendly error: surround the
json.loads(mdl_path.read_text()) with a try/except that catches
json.JSONDecodeError, call typer.echo with a clear message including the
mdl_path (and optionally the JSON error message) to stderr, and then raise
typer.Exit(1) instead of letting the raw traceback propagate; keep the
subsequent convert_mdl_to_project and write_project_files(project_path,
force=force) flow unchanged so only parsing errors are handled here.

wren/src/wren/context.py (1)

112-114: Consider defensive handling for missing model name.

If a model in the input MDL lacks a name field, line 114 raises an unhandled KeyError. Since this is a user-facing import tool, a clearer error message would improve UX.

💡 Optional: Add validation for missing name

     for model in mdl_json.get("models", []):
         model_snake = _convert_keys_to_snake(model)
-        name = model_snake["name"]
+        name = model_snake.get("name")
+        if not name:
+            raise ValueError(f"Model missing required 'name' field: {model_snake}")
         dir_path = f"models/{name}"

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@wren/src/wren/context.py` around lines 112 - 114, The loop over
mdl_json.get("models", []) accesses model_snake["name"] directly which raises
KeyError for missing model names; in the block that iterates models (where
mdl_json, _convert_keys_to_snake and model_snake are used) add a defensive check
for the "name" key (or use model_snake.get("name")) and if missing raise a
clear, user-facing error (e.g., ValueError) or log a descriptive message
including the model payload or index and skip processing, so import failures
surface with an actionable message rather than a raw KeyError.

wren/tests/unit/test_convert_mdl.py (1)

309-323: Consider extracting CliRunner setup to a fixture.

The CliRunner import and instantiation is repeated in each CLI test. A shared fixture would reduce duplication.

♻️ Optional: Extract runner fixture

Add a fixture at the module level:

`@pytest.fixture`
def runner():
    from typer.testing import CliRunner
    return CliRunner()

Then use it in tests:

-def test_cli_init_from_mdl(tmp_path: Path, sample_mdl_file: Path):
-    from typer.testing import CliRunner
-
-    from wren.cli import app
-
-    runner = CliRunner()
+def test_cli_init_from_mdl(tmp_path: Path, sample_mdl_file: Path, runner):
+    from wren.cli import app
+
     project_dir = tmp_path / "project"
     result = runner.invoke(

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@wren/tests/unit/test_convert_mdl.py` around lines 309 - 323, Extract the
repeated CliRunner setup into a pytest fixture and update tests to use it:
create a module-level fixture named e.g. runner that imports and returns
typer.testing.CliRunner(), then update test functions like
test_cli_init_from_mdl to accept the runner fixture instead of creating a new
CliRunner instance inside the test; ensure references in the test use the
injected runner when invoking app (from wren.cli import app) and remove the
in-test import/instantiation to eliminate duplication.

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Nitpick comments:
In `@wren/src/wren/context_cli.py`:
- Around line 59-65: Wrap the JSON parsing call so malformed MDL yields a
friendly error: surround the json.loads(mdl_path.read_text()) with a try/except
that catches json.JSONDecodeError, call typer.echo with a clear message
including the mdl_path (and optionally the JSON error message) to stderr, and
then raise typer.Exit(1) instead of letting the raw traceback propagate; keep
the subsequent convert_mdl_to_project and write_project_files(project_path,
force=force) flow unchanged so only parsing errors are handled here.

In `@wren/src/wren/context.py`:
- Around line 112-114: The loop over mdl_json.get("models", []) accesses
model_snake["name"] directly which raises KeyError for missing model names; in
the block that iterates models (where mdl_json, _convert_keys_to_snake and
model_snake are used) add a defensive check for the "name" key (or use
model_snake.get("name")) and if missing raise a clear, user-facing error (e.g.,
ValueError) or log a descriptive message including the model payload or index
and skip processing, so import failures surface with an actionable message
rather than a raw KeyError.

In `@wren/tests/unit/test_convert_mdl.py`:
- Around line 309-323: Extract the repeated CliRunner setup into a pytest
fixture and update tests to use it: create a module-level fixture named e.g.
runner that imports and returns typer.testing.CliRunner(), then update test
functions like test_cli_init_from_mdl to accept the runner fixture instead of
creating a new CliRunner instance inside the test; ensure references in the test
use the injected runner when invoking app (from wren.cli import app) and remove
the in-test import/instantiation to eliminate duplication.

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: f366aa19-5fc6-4e45-aff0-fb26d1d3304a

📥 Commits

Reviewing files that changed from the base of the PR and between 6a8ea38 and 4dc3b27.

⛔ Files ignored due to path filters (1)

• wren/uv.lock is excluded by !**/*.lock

📒 Files selected for processing (3)

• wren/src/wren/context.py
• wren/src/wren/context_cli.py
• wren/tests/unit/test_convert_mdl.py