fix(ddl): CHECK sanitizer no longer deletes columns named CHECKOUT_*

The chunk-level CHECK-constraint detector used prefix matching
(upper.startswith("CHECK")), so any column whose name begins with CHECK —
CHECKOUT_INITIATED, CHECKOUT_ATTEMPTS, CHECKOUT_CONVERSION_RATE,
CHECKOUT_ABANDONMENT_RATE — was silently dropped from CREATE TABLE while the
generated dataset still carried its values. The INSERT then failed on columns
the table didn't have. Root cause of the "random" schema/data-mismatch build
failures (Cloudscape, Atlas Freight partial, Tixr run 02ec3e7baba6) — it fired
whenever the LLM blueprint author named a column CHECKOUT_*, which
checkout-funnel briefs (Tixr) invite.

Detector now requires the CHECK keyword followed by a parenthesized
expression, and the CONSTRAINT branch uses a word-boundary match so an FK
named CHECKOUT_FK survives. Inline CHECK removal already used \bCHECK\b and
was safe. 4 regression tests added.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

cdw_connector.py

@@ -258,7 +258,15 @@ class SnowflakeDeployer:
                 continue
 
             upper = stripped.upper()
-            is_table_check = upper.startswith("CHECK") or ("CHECK" in upper and upper.startswith("CONSTRAINT"))
+            # A table-level CHECK constraint is the keyword CHECK followed by a
+            # parenthesized expression — NOT any chunk that merely starts with the
+            # letters CHECK. A column named CHECKOUT_* (e.g. CHECKOUT_ATTEMPTS
+            # NUMBER(38,2)) must survive; prefix matching silently deleted such
+            # columns from CREATE TABLE while the dataset still carried their
+            # values (the FACT_PLATFORM_ENGAGEMENT / CHECKOUT_INITIATED bug).
+            is_table_check = bool(re.match(r"CHECK\s*\(", upper)) or (
+                upper.startswith("CONSTRAINT") and re.search(r"\bCHECK\s*\(", upper)
+            )
             if is_table_check:
                 removed_any = True
                 continue

tests/test_cdw_ddl_sanitizer.py

@@ -0,0 +1,70 @@
+"""Regression tests for the CREATE TABLE CHECK-constraint sanitizer.
+
+The bug: the chunk-level CHECK detector used prefix matching
+(upper.startswith("CHECK")), so any COLUMN whose name begins with CHECK —
+CHECKOUT_INITIATED, CHECKOUT_ATTEMPTS, CHECKOUT_CONVERSION_RATE... — was
+silently deleted from the CREATE TABLE while the generated dataset still
+carried its values. The INSERT then referenced columns the table didn't have
+(the FACT_PLATFORM_ENGAGEMENT failure caught by the dataset_writer guard).
+"""
+
+from cdw_connector import SnowflakeDeployer
+
+
+def _sanitize(statement: str) -> str:
+    return SnowflakeDeployer()._sanitize_create_table_statement(statement)
+
+
+def test_checkout_columns_survive_and_table_check_is_removed():
+    stmt = (
+        "CREATE TABLE FACT_PLATFORM_ENGAGEMENT (\n"
+        "    ENGAGEMENT_KEY NUMBER NOT NULL,\n"
+        "    CHECKOUT_ATTEMPTS NUMBER(38,0) NOT NULL,\n"
+        "    CHECKOUT_CONVERSION_RATE NUMBER(38,2) NOT NULL,\n"
+        "    CHECKOUT_ABANDONMENT_RATE NUMBER(38,2) NOT NULL,\n"
+        "    CHECK (CHECKOUT_ATTEMPTS >= 0)\n"
+        ")"
+    )
+    out = _sanitize(stmt)
+    assert "CHECKOUT_ATTEMPTS NUMBER(38,0)" in out
+    assert "CHECKOUT_CONVERSION_RATE" in out
+    assert "CHECKOUT_ABANDONMENT_RATE" in out
+    assert "CHECK (" not in out
+
+
+def test_named_constraint_starting_with_check_is_only_removed_when_it_is_a_check():
+    stmt = (
+        "CREATE TABLE T (\n"
+        "    ID NUMBER NOT NULL,\n"
+        "    CHECKOUT_KEY NUMBER NOT NULL,\n"
+        "    CONSTRAINT CHECKOUT_FK FOREIGN KEY (CHECKOUT_KEY) REFERENCES DIM_CHECKOUT(CHECKOUT_KEY),\n"
+        "    CONSTRAINT POSITIVE_ID CHECK (ID > 0)\n"
+        ")"
+    )
+    out = _sanitize(stmt)
+    # FK constraint whose NAME contains CHECK must survive.
+    assert "FOREIGN KEY (CHECKOUT_KEY)" in out
+    # Real named CHECK constraint must go.
+    assert "POSITIVE_ID" not in out
+
+
+def test_inline_check_clause_removed_but_column_kept():
+    stmt = (
+        "CREATE TABLE T (\n"
+        "    ID NUMBER NOT NULL,\n"
+        "    QTY NUMBER CHECK (QTY >= 0) NOT NULL\n"
+        ")"
+    )
+    out = _sanitize(stmt)
+    assert "QTY NUMBER" in out
+    assert "CHECK (" not in out
+
+
+def test_statement_without_checks_is_untouched():
+    stmt = (
+        "CREATE TABLE DIM_CHECKOUT (\n"
+        "    CHECKOUT_KEY NUMBER NOT NULL,\n"
+        "    CHECKOUT_CHANNEL VARCHAR NOT NULL\n"
+        ")"
+    )
+    assert _sanitize(stmt) == stmt