fix: update storage backend

Author: cofin

.pre-commit-config.yaml

@@ -17,7 +17,7 @@ repos:
       - id: mixed-line-ending
       - id: trailing-whitespace
   - repo: https://github.com/charliermarsh/ruff-pre-commit
-    rev: "v0.12.10"
+    rev: "v0.12.11"
     hooks:
       - id: ruff
         args: ["--fix"]

pyproject.toml

@@ -83,6 +83,7 @@ doc = [
 ]
 extras = [
   "adbc_driver_manager",
+  "fsspec[s3]",
   "pgvector",
   "pyarrow",
   "polars",

sqlspec/base.py

@@ -64,7 +64,7 @@ def _cleanup_sync_pools(self) -> None:
                     config.close_pool()
                     cleaned_count += 1
                 except Exception as e:
-                    logger.warning("Failed to clean up sync pool for config %s: %s", config_type.__name__, e)
+                    logger.debug("Failed to clean up sync pool for config %s: %s", config_type.__name__, e)
 
         if cleaned_count > 0:
             logger.debug("Sync pool cleanup completed. Cleaned %d pools.", cleaned_count)
@@ -87,14 +87,14 @@ async def close_all_pools(self) -> None:
                     else:
                         sync_configs.append((config_type, config))
                 except Exception as e:
-                    logger.warning("Failed to prepare cleanup for config %s: %s", config_type.__name__, e)
+                    logger.debug("Failed to prepare cleanup for config %s: %s", config_type.__name__, e)
 
         if cleanup_tasks:
             try:
                 await asyncio.gather(*cleanup_tasks, return_exceptions=True)
                 logger.debug("Async pool cleanup completed. Cleaned %d pools.", len(cleanup_tasks))
             except Exception as e:
-                logger.warning("Failed to complete async pool cleanup: %s", e)
+                logger.debug("Failed to complete async pool cleanup: %s", e)
 
         for _config_type, config in sync_configs:
             config.close_pool()
@@ -129,7 +129,7 @@ def add_config(self, config: "Union[SyncConfigT, AsyncConfigT]") -> "type[Union[
         """
         config_type = type(config)
         if config_type in self._configs:
-            logger.warning("Configuration for %s already exists. Overwriting.", config_type.__name__)
+            logger.debug("Configuration for %s already exists. Overwriting.", config_type.__name__)
         self._configs[config_type] = config
         return config_type
 

sqlspec/loader.py

@@ -13,15 +13,11 @@
 
 from sqlspec.core.cache import CacheKey, get_cache_config, get_default_cache
 from sqlspec.core.statement import SQL
-from sqlspec.exceptions import (
-    MissingDependencyError,
-    SQLFileNotFoundError,
-    SQLFileParseError,
-    StorageOperationFailedError,
-)
+from sqlspec.exceptions import SQLFileNotFoundError, SQLFileParseError, StorageOperationFailedError
 from sqlspec.storage.registry import storage_registry as default_storage_registry
 from sqlspec.utils.correlation import CorrelationContext
 from sqlspec.utils.logging import get_logger
+from sqlspec.utils.text import slugify
 
 if TYPE_CHECKING:
     from sqlspec.storage.registry import StorageRegistry
@@ -60,7 +56,7 @@ def _normalize_query_name(name: str) -> str:
     Returns:
         Normalized query name suitable as Python identifier.
     """
-    return TRIM_SPECIAL_CHARS.sub("", name).replace("-", "_")
+    return slugify(name, separator="_")
 
 
 def _normalize_dialect(dialect: str) -> str:
@@ -76,19 +72,6 @@ def _normalize_dialect(dialect: str) -> str:
     return DIALECT_ALIASES.get(normalized, normalized)
 
 
-def _normalize_dialect_for_sqlglot(dialect: str) -> str:
-    """Normalize dialect name for SQLGlot compatibility.
-
-    Args:
-        dialect: Dialect name from SQL file or parameter.
-
-    Returns:
-        SQLGlot-compatible dialect name.
-    """
-    normalized = dialect.lower().strip()
-    return DIALECT_ALIASES.get(normalized, normalized)
-
-
 class NamedStatement:
     """Represents a parsed SQL statement with metadata.
 
@@ -218,8 +201,7 @@ def _calculate_file_checksum(self, path: Union[str, Path]) -> str:
             SQLFileParseError: If file cannot be read.
         """
         try:
-            content = self._read_file_content(path)
-            return hashlib.md5(content.encode(), usedforsecurity=False).hexdigest()
+            return hashlib.md5(self._read_file_content(path).encode(), usedforsecurity=False).hexdigest()
         except Exception as e:
             raise SQLFileParseError(str(path), str(path), e) from e
 
@@ -253,19 +235,13 @@ def _read_file_content(self, path: Union[str, Path]) -> str:
             SQLFileNotFoundError: If file does not exist.
             SQLFileParseError: If file cannot be read or parsed.
         """
-
         path_str = str(path)
 
         try:
             backend = self.storage_registry.get(path)
             return backend.read_text(path_str, encoding=self.encoding)
         except KeyError as e:
             raise SQLFileNotFoundError(path_str) from e
-        except MissingDependencyError:
-            try:
-                return path.read_text(encoding=self.encoding)  # type: ignore[union-attr]
-            except FileNotFoundError as e:
-                raise SQLFileNotFoundError(path_str) from e
         except StorageOperationFailedError as e:
             if "not found" in str(e).lower() or "no such file" in str(e).lower():
                 raise SQLFileNotFoundError(path_str) from e
@@ -419,8 +395,7 @@ def _load_directory(self, dir_path: Path) -> int:
         for file_path in sql_files:
             relative_path = file_path.relative_to(dir_path)
             namespace_parts = relative_path.parent.parts
-            namespace = ".".join(namespace_parts) if namespace_parts else None
-            self._load_single_file(file_path, namespace)
+            self._load_single_file(file_path, ".".join(namespace_parts) if namespace_parts else None)
         return len(sql_files)
 
     def _load_single_file(self, file_path: Union[str, Path], namespace: Optional[str]) -> None:
@@ -533,44 +508,6 @@ def add_named_sql(self, name: str, sql: str, dialect: "Optional[str]" = None) ->
         self._queries[normalized_name] = statement
         self._query_to_file[normalized_name] = "<directly added>"
 
-    def get_sql(self, name: str) -> "SQL":
-        """Get a SQL object by statement name.
-
-        Args:
-            name: Name of the statement (from -- name: in SQL file).
-                  Hyphens in names are converted to underscores.
-
-        Returns:
-            SQL object ready for execution.
-
-        Raises:
-            SQLFileNotFoundError: If statement name not found.
-        """
-        correlation_id = CorrelationContext.get()
-
-        safe_name = _normalize_query_name(name)
-
-        if safe_name not in self._queries:
-            available = ", ".join(sorted(self._queries.keys())) if self._queries else "none"
-            logger.error(
-                "Statement not found: %s",
-                name,
-                extra={
-                    "statement_name": name,
-                    "safe_name": safe_name,
-                    "available_statements": len(self._queries),
-                    "correlation_id": correlation_id,
-                },
-            )
-            raise SQLFileNotFoundError(name, path=f"Statement '{name}' not found. Available statements: {available}")
-
-        parsed_statement = self._queries[safe_name]
-        sqlglot_dialect = None
-        if parsed_statement.dialect:
-            sqlglot_dialect = _normalize_dialect_for_sqlglot(parsed_statement.dialect)
-
-        return SQL(parsed_statement.sql, dialect=sqlglot_dialect)
-
     def get_file(self, path: Union[str, Path]) -> "Optional[SQLFile]":
         """Get a loaded SQLFile object by path.
 
@@ -659,3 +596,41 @@ def get_query_text(self, name: str) -> str:
         if safe_name not in self._queries:
             raise SQLFileNotFoundError(name)
         return self._queries[safe_name].sql
+
+    def get_sql(self, name: str) -> "SQL":
+        """Get a SQL object by statement name.
+
+        Args:
+            name: Name of the statement (from -- name: in SQL file).
+                  Hyphens in names are converted to underscores.
+
+        Returns:
+            SQL object ready for execution.
+
+        Raises:
+            SQLFileNotFoundError: If statement name not found.
+        """
+        correlation_id = CorrelationContext.get()
+
+        safe_name = _normalize_query_name(name)
+
+        if safe_name not in self._queries:
+            available = ", ".join(sorted(self._queries.keys())) if self._queries else "none"
+            logger.error(
+                "Statement not found: %s",
+                name,
+                extra={
+                    "statement_name": name,
+                    "safe_name": safe_name,
+                    "available_statements": len(self._queries),
+                    "correlation_id": correlation_id,
+                },
+            )
+            raise SQLFileNotFoundError(name, path=f"Statement '{name}' not found. Available statements: {available}")
+
+        parsed_statement = self._queries[safe_name]
+        sqlglot_dialect = None
+        if parsed_statement.dialect:
+            sqlglot_dialect = _normalize_dialect(parsed_statement.dialect)
+
+        return SQL(parsed_statement.sql, dialect=sqlglot_dialect)

sqlspec/protocols.py

@@ -4,7 +4,7 @@
 and runtime isinstance() checks.
 """
 
-from typing import TYPE_CHECKING, Any, ClassVar, Optional, Protocol, Union, runtime_checkable
+from typing import TYPE_CHECKING, Any, Optional, Protocol, Union, runtime_checkable
 
 from typing_extensions import Self
 
@@ -14,7 +14,6 @@
 
     from sqlglot import exp
 
-    from sqlspec.storage.capabilities import StorageCapabilities
     from sqlspec.typing import ArrowRecordBatch, ArrowTable
 
 __all__ = (
@@ -194,9 +193,8 @@ class ObjectStoreItemProtocol(Protocol):
 class ObjectStoreProtocol(Protocol):
     """Protocol for object storage operations."""
 
-    capabilities: ClassVar["StorageCapabilities"]
-
     protocol: str
+    backend_type: str
 
     def __init__(self, uri: str, **kwargs: Any) -> None:
         return
@@ -330,7 +328,7 @@ async def write_arrow_async(self, path: "Union[str, Path]", table: "ArrowTable",
         msg = "Async arrow writing not implemented"
         raise NotImplementedError(msg)
 
-    async def stream_arrow_async(self, pattern: str, **kwargs: Any) -> "AsyncIterator[ArrowRecordBatch]":
+    def stream_arrow_async(self, pattern: str, **kwargs: Any) -> "AsyncIterator[ArrowRecordBatch]":
         """Async stream Arrow record batches from matching objects."""
         msg = "Async arrow streaming not implemented"
         raise NotImplementedError(msg)

sqlspec/storage/__init__.py

@@ -8,16 +8,6 @@
 - Capability-based backend selection
 """
 
-from sqlspec.protocols import ObjectStoreProtocol
-from sqlspec.storage.capabilities import HasStorageCapabilities, StorageCapabilities
-from sqlspec.storage.registry import StorageRegistry
+from sqlspec.storage.registry import StorageRegistry, storage_registry
 
-storage_registry = StorageRegistry()
-
-__all__ = (
-    "HasStorageCapabilities",
-    "ObjectStoreProtocol",
-    "StorageCapabilities",
-    "StorageRegistry",
-    "storage_registry",
-)
+__all__ = ("StorageRegistry", "storage_registry")