add make_sql_cell function (#140)

* add make_sql_cell function

* break tests into unit and e2e, add sql cell test

* Ruff snuck in an extra dupe line!

* changelog

Author: Matt Kafonek

CHANGELOG.md

@@ -7,6 +7,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 For pre-1.0 releases, see [0.0.35 Changelog](https://github.com/noteable-io/origami/blob/0.0.35/CHANGELOG.md)
 
 ## [Unreleased]
+### Added
+- `origami.models.notebook.make_sql_cell` convenience function, returns a `CodeCell` with appropriate metadata
 
 ## [1.0.0-alpha.2] - 2023-07-26
 ### Changed

origami/models/notebook.py

@@ -8,6 +8,8 @@
 Devs: as usual with Pydantic modeling, the top-level model (Notebook) is at the bottom of this file,
 read from bottom up for most clarity.
 """
+import random
+import string
 import uuid
 from typing import Any, Dict, List, Literal, Optional, Union
 
@@ -97,6 +99,34 @@ def output_collection_id(self) -> Optional[Union[str, uuid.UUID]]:
         return self.metadata.get("noteable", {}).get("output_collection_id")
 
 
+def make_sql_cell(
+    cell_id: Optional[str] = None,
+    source: str = '',
+    db_connection: str = '@noteable',
+    assign_results_to: Optional[str] = None,
+) -> CodeCell:
+    cell_id = cell_id or str(uuid.uuid4())
+    # Remove first line of source if it starts with %%sql. That is the right syntax for regular
+    # code cells with sql magic support, but Noteable SQL cells should have just the sql source
+    if source.startswith('%%sql'):
+        lines = source.splitlines()
+        source = '\n'.join(lines[1:])
+
+    if not assign_results_to:
+        name_suffix = "".join(random.choices(string.ascii_lowercase, k=4))
+        assign_results_to = 'df_' + name_suffix
+    metadata = {
+        'language': 'sql',
+        'type': 'code',
+        'noteable': {
+            'cell_type': 'sql',
+            'db_connection': db_connection,
+            'assign_results_to': assign_results_to,
+        },
+    }
+    return CodeCell(cell_id=cell_id, source=source, metadata=metadata)
+
+
 class MarkdownCell(CellBase):
     cell_type: Literal["markdown"] = "markdown"