Types: Add support for BLOB type, per base64 encoding

Author: amotl

src/sqlalchemy_cratedb/compiler.py

@@ -254,6 +254,9 @@ def visit_TIMESTAMP(self, type_, **kw):
         """
         return "TIMESTAMP %s" % ((type_.timezone and "WITH" or "WITHOUT") + " TIME ZONE",)
 
+    def visit_BLOB(self, type_, **kw):
+        return "STRING"
+
 
 class CrateCompiler(compiler.SQLCompiler):
     def visit_getitem_binary(self, binary, operator, **kw):

src/sqlalchemy_cratedb/dialect.py

@@ -34,6 +34,7 @@
 )
 from .sa_version import SA_1_4, SA_2_0, SA_VERSION
 from .type import FloatVector, ObjectArray, ObjectType
+from .type.binary import LargeBinary
 
 TYPES_MAP = {
     "boolean": sqltypes.Boolean,
@@ -158,6 +159,7 @@ def process(value):
     sqltypes.Date: Date,
     sqltypes.DateTime: DateTime,
     sqltypes.TIMESTAMP: DateTime,
+    sqltypes.LargeBinary: LargeBinary,
 }
 
 

src/sqlalchemy_cratedb/type/__init__.py

@@ -1,11 +1,13 @@
 from .array import ObjectArray
+from .binary import LargeBinary
 from .geo import Geopoint, Geoshape
 from .object import ObjectType
 from .vector import FloatVector, knn_match
 
 __all__ = [
     Geopoint,
     Geoshape,
+    LargeBinary,
     ObjectArray,
     ObjectType,
     FloatVector,

src/sqlalchemy_cratedb/type/binary.py

@@ -0,0 +1,44 @@
+import base64
+
+import sqlalchemy as sa
+
+
+class LargeBinary(sa.String):
+    """A type for large binary byte data.
+
+    The :class:`.LargeBinary` type corresponds to a large and/or unlengthed
+    binary type for the target platform, such as BLOB on MySQL and BYTEA for
+    PostgreSQL.  It also handles the necessary conversions for the DBAPI.
+
+    """
+
+    __visit_name__ = "large_binary"
+
+    def bind_processor(self, dialect):
+        if dialect.dbapi is None:
+            return None
+
+        # TODO: DBAPIBinary = dialect.dbapi.Binary
+
+        def process(value):
+            if value is not None:
+                # TODO: return DBAPIBinary(value)
+                return base64.b64encode(value).decode()
+            else:
+                return None
+
+        return process
+
+    # Python 3 has native bytes() type
+    # both sqlite3 and pg8000 seem to return it,
+    # psycopg2 as of 2.5 returns 'memoryview'
+    def result_processor(self, dialect, coltype):
+        if dialect.returns_native_bytes:
+            return None
+
+        def process(value):
+            if value is not None:
+                return base64.b64decode(value)
+            return value
+
+        return process