fix!: correctly identify keys to make deterministic

Author: eakmanrq

sqlmesh/core/model/common.py

@@ -153,14 +153,16 @@ def _add_variables_to_python_env(
 
     variables = {k: v for k, v in (variables or {}).items() if k in used_variables}
     if variables:
-        python_env[c.SQLMESH_VARS] = Executable.value(variables)
+        python_env[c.SQLMESH_VARS] = Executable.value(variables, use_deterministic_repr=True)
 
     if blueprint_variables:
         blueprint_variables = {
             k: SqlValue(sql=v.sql(dialect=dialect)) if isinstance(v, exp.Expression) else v
             for k, v in blueprint_variables.items()
         }
-        python_env[c.SQLMESH_BLUEPRINT_VARS] = Executable.value(blueprint_variables)
+        python_env[c.SQLMESH_BLUEPRINT_VARS] = Executable.value(
+            blueprint_variables, use_deterministic_repr=True
+        )
 
     return python_env
 

sqlmesh/migrations/v0085_deterministic_repr.py

@@ -12,6 +12,9 @@
 from sqlmesh.utils.migration import index_text_type, blob_text_type
 
 
+KEYS_TO_MAKE_DETERMINISTIC = ["__sqlmesh__vars__", "__sqlmesh__blueprint__vars__"]
+
+
 # Make sure `SqlValue` is defined so it can be used by `eval` call in the migration
 @dataclass
 class SqlValue:
@@ -82,6 +85,8 @@ def migrate(state_sync, **kwargs):  # type: ignore
 
         if python_env:
             for key, executable in python_env.items():
+                if key not in KEYS_TO_MAKE_DETERMINISTIC:
+                    continue
                 if isinstance(executable, dict) and executable.get("kind") == "value":
                     old_payload = executable["payload"]
                     try:

sqlmesh/migrations/v0086_check_deterministic_bug.py

@@ -0,0 +1,74 @@
+import json
+from sqlglot import exp
+
+from sqlmesh.core.console import get_console
+
+
+KEYS_TO_MAKE_DETERMINISTIC = ["__sqlmesh__vars__", "__sqlmesh__blueprint__vars__"]
+
+
+def migrate(state_sync, **kwargs):  # type: ignore
+    engine_adapter = state_sync.engine_adapter
+    schema = state_sync.schema
+    snapshots_table = "_snapshots"
+    versions_table = "_versions"
+    if schema:
+        snapshots_table = f"{schema}.{snapshots_table}"
+        versions_table = f"{schema}.{versions_table}"
+
+    result = engine_adapter.fetchone(
+        exp.select("schema_version").from_(versions_table), quote_identifiers=True
+    )
+    if not result:
+        # This must be the first migration, so we can skip the check since the project was not exposed to 85 migration bug
+        return
+    schema_version = result[0]
+    if schema_version < 85:
+        # The project was not exposed to the bugged 85 migration, so we can skip it.
+        return
+
+    warning = (
+        "SQLMesh detected that it may not be able to fully migrate the state database. This should not impact "
+        "the migration process, but may result in unexpected changes being reported by the next `sqlmesh plan` "
+        "command. Please run `sqlmesh diff prod` after the migration has completed, before making any new "
+        "changes. If any unexpected changes are reported, consider running a forward-only plan to apply these "
+        "changes and avoid unnecessary backfills: sqlmesh plan prod --forward-only. "
+        "See https://sqlmesh.readthedocs.io/en/stable/concepts/plans/#forward-only-plans for more details.\n"
+    )
+
+    for (
+        name,
+        identifier,
+        version,
+        snapshot,
+        kind_name,
+        updated_ts,
+        unpaused_ts,
+        ttl_ms,
+        unrestorable,
+    ) in engine_adapter.fetchall(
+        exp.select(
+            "name",
+            "identifier",
+            "version",
+            "snapshot",
+            "kind_name",
+            "updated_ts",
+            "unpaused_ts",
+            "ttl_ms",
+            "unrestorable",
+        ).from_(snapshots_table),
+        quote_identifiers=True,
+    ):
+        parsed_snapshot = json.loads(snapshot)
+        python_env = parsed_snapshot["node"].get("python_env")
+
+        if python_env:
+            for key, executable in python_env.items():
+                if (
+                    key not in KEYS_TO_MAKE_DETERMINISTIC
+                    and isinstance(executable, dict)
+                    and executable.get("kind") == "value"
+                ):
+                    get_console().log_warning(warning)
+                    return

sqlmesh/utils/metaprogramming.py

@@ -424,10 +424,11 @@ def is_value(self) -> bool:
         return self.kind == ExecutableKind.VALUE
 
     @classmethod
-    def value(cls, v: t.Any, is_metadata: t.Optional[bool] = None) -> Executable:
-        return Executable(
-            payload=_deterministic_repr(v), kind=ExecutableKind.VALUE, is_metadata=is_metadata
-        )
+    def value(
+        cls, v: t.Any, is_metadata: t.Optional[bool] = None, use_deterministic_repr: bool = False
+    ) -> Executable:
+        payload = _deterministic_repr(v) if use_deterministic_repr else repr(v)
+        return Executable(payload=payload, kind=ExecutableKind.VALUE, is_metadata=is_metadata)
 
 
 def serialize_env(env: t.Dict[str, t.Any], path: Path) -> t.Dict[str, Executable]: