Merge e1cf48e into cacc0f6

README.rst

@@ -87,6 +87,11 @@ This file should be named mappings.json. Here is a sample :
                 "myobject.subproperty":{
                     "dest":"subproperty",
                     "type":"TEXT"
+                },
+                "transformed_field": {
+                  "dest": "field",
+                  "type": "BOOLEAN",
+                  "transform": "package.module.transform_str_to_bool"
                 }
             }
         }
@@ -99,6 +104,14 @@ Please notice the following :
 - If the original document in mongodb has a embedded document, everything is flattened to be inserted in PostgreSQL
 - One can define indices in two different ways : Using the array ``indices`` and a SQL definition or autogenerate index
  by setting the ``index`` field to true
+- The ``transform`` (if set) points to a function used to transform the field from the Mongo document
+
+Example of transform function:
+
+.. code-block:: python
+
+   def transform_str_to_bool(mongo_val):
+       return (mongo_val in ['true', 'True', 'yes', 'Yes', '1'])
 
 The connector also supports arrays of documents. Let say your Mongo database stores the following documents :
 

mongo_connector/doc_managers/mappings.py

@@ -11,6 +11,11 @@
 from mongo_connector.doc_managers.mapping_schema import MAPPING_SCHEMA
 from mongo_connector.errors import InvalidConfiguration
 
+from importlib import import_module
+import logging
+
+
+LOG = logging.getLogger(__name__)
 
 _formatter = DocumentFlattener()
 
@@ -82,6 +87,65 @@ def get_primary_key(mappings, namespace):
     return mappings[db][collection]['pk']
 
 
+def get_transformed_value(mapped_field, mapped_document, key):
+    val = mapped_document[key]
+
+    if 'transform' in mapped_field:
+        transform = mapped_field['transform']
+        transform_path = transform.rsplit('.', 1)
+        module_path = 'mongo_connector.doc_managers.transforms'
+
+        if len(transform_path) == 2:
+            module_path, transform_path = transform_path
+
+        else:
+            transform_path = transform_path[0]
+
+        try:
+            module = import_module(module_path)
+            transform = getattr(module, transform_path)
+
+        except (ImportError, ValueError) as err:
+            LOG.error('Impossible to use transform function: {0}'.format(err))
+
+        else:
+            try:
+                new_val = transform(val)
+
+            except Exception as err:
+                LOG.error(
+                    'An error occured during field transformation: {0}'.format(
+                        err
+                    )
+                )
+
+            else:
+                val = new_val
+
+    return val
+
+
+def get_transformed_document(mappings, db, collection, mapped_document):
+    mapped_fields = {
+        mapping['dest']: mapping
+        for _, mapping in iteritems(mappings[db][collection])
+        if 'dest' in mapping and mapping['type'] not in (
+            ARRAY_TYPE,
+            ARRAY_OF_SCALARS_TYPE
+        )
+    }
+    keys = list(mapped_fields.keys())
+    keys.sort()
+
+    return {
+        key: get_transformed_value(
+            mapped_fields[key],
+            mapped_document, key
+        ) if key in mapped_fields else mapped_document[key]
+        for key in mapped_document
+    }
+
+
 def is_mapped(mappings, namespace, field_name=None):
     db, collection = db_and_collection(namespace)
     return db in mappings and collection in mappings[db] and \

mongo_connector/doc_managers/postgresql_manager.py

@@ -142,7 +142,13 @@ def _upsert(self, namespace, document, cursor, timestamp):
         mapped_document = get_mapped_document(self.mappings, document, namespace)
 
         if mapped_document:
-            sql_insert(cursor, collection, mapped_document, self.mappings[db][collection]['pk'])
+            sql_insert(
+                cursor,
+                collection,
+                mapped_document,
+                self.mappings,
+                db, collection
+            )
 
             self._upsert_array_fields(collection, cursor, db, document, mapped_document, namespace, timestamp)
             self.upsert_scalar_array_fields(collection, cursor, db, document, mapped_document, namespace, timestamp)

mongo_connector/doc_managers/sql.py

@@ -9,12 +9,26 @@
 from past.builtins import long, basestring
 from psycopg2._psycopg import AsIs
 
-from mongo_connector.doc_managers.mappings import get_mapped_document
-from mongo_connector.doc_managers.utils import extract_creation_date, get_array_fields, db_and_collection, \
-    get_array_of_scalar_fields, ARRAY_OF_SCALARS_TYPE, ARRAY_TYPE, get_nested_field_from_document
+from mongo_connector.doc_managers.mappings import (
+    get_mapped_document,
+    get_transformed_value,
+    get_transformed_document
+)
+
+from mongo_connector.doc_managers.utils import (
+    extract_creation_date,
+    get_array_fields,
+    db_and_collection,
+    get_array_of_scalar_fields,
+    ARRAY_OF_SCALARS_TYPE,
+    ARRAY_TYPE,
+    get_nested_field_from_document
+)
+
 
 LOG = logging.getLogger(__name__)
 
+
 all_chars = (chr(i) for i in range(0x10000))
 control_chars = ''.join(c for c in all_chars if unicodedata.category(c) == 'Cc')
 control_char_re = re.compile('[%s]' % re.escape(control_chars))
@@ -60,25 +74,41 @@ def sql_bulk_insert(cursor, mappings, namespace, documents):
     db, collection = db_and_collection(namespace)
 
     primary_key = mappings[db][collection]['pk']
-    keys = [
-        v['dest'] for k, v in iteritems(mappings[db][collection])
-        if 'dest' in v and v['type'] != ARRAY_TYPE
-        and v['type'] != ARRAY_OF_SCALARS_TYPE
-        ]
-    keys.sort()
 
+    mapped_fields = {
+        mapping['dest']: mapping
+        for _, mapping in iteritems(mappings[db][collection])
+        if 'dest' in mapping and mapping['type'] not in (
+            ARRAY_TYPE,
+            ARRAY_OF_SCALARS_TYPE
+        )
+    }
+    keys = list(mapped_fields.keys())
+    keys.sort()
     values = []
 
     for document in documents:
         mapped_document = get_mapped_document(mappings, document, namespace)
-        document_values = [to_sql_value(extract_creation_date(mapped_document, mappings[db][collection]['pk']))]
+        document_values = [
+            to_sql_value(
+                extract_creation_date(
+                    mapped_document,
+                    mappings[db][collection]['pk']
+                )
+            )
+        ]
 
         if not mapped_document:
             break
 
         for key in keys:
             if key in mapped_document:
-                document_values.append(to_sql_value(mapped_document[key]))
+                val = get_transformed_value(
+                    mapped_fields[key],
+                    mapped_document, key
+                )
+                document_values.append(to_sql_value(val))
+
             else:
                 document_values.append(to_sql_value(None))
         values.append(u"({0})".format(u','.join(document_values)))
@@ -128,7 +158,9 @@ def get_document_keys(document):
     return keys
 
 
-def sql_insert(cursor, tableName, document, primary_key):
+def sql_insert(cursor, tableName, document, mappings, db, collection):
+    primary_key = mappings[db][collection]['pk']
+
     creationDate = extract_creation_date(document, primary_key)
     if creationDate is not None:
         document['_creationDate'] = creationDate
@@ -152,7 +184,10 @@ def sql_insert(cursor, tableName, document, primary_key):
         )
 
     try:
-        cursor.execute(sql, document)
+        cursor.execute(
+            sql,
+            get_transformed_document(mappings, db, collection, document)
+        )
     except Exception as e:
         LOG.error(u"Impossible to upsert the following document %s : %s", document, e)
 

tests/test_mappings.py

@@ -5,6 +5,10 @@
 from mongo_connector.doc_managers import mappings
 
 
+def parseInt(val):
+    return int(val)
+
+
 class TestPostgreSQLMappings(TestCase):
     def test_clean_and_flatten_doc(self):
         mapping = {
@@ -506,6 +510,62 @@ def test_valid_mapping_array_of_scalar(self):
 
         mappings.validate_mapping(mapping)
 
+    def test_get_transform_value(self):
+        mapped_field = {
+            'type': 'INT',
+            'dest': 'str_to_int',
+            'transform': 'tests.test_mappings.parseInt'
+        }
+        doc = {
+            'str_to_int': '42'
+        }
+
+        got = mappings.get_transformed_value(mapped_field, doc, 'str_to_int')
+        self.assertEqual(got, 42)
+
+        mapped_field['transform'] = 'missing.no'
+        got = mappings.get_transformed_value(mapped_field, doc, 'str_to_int')
+        self.assertEqual(got, '42')
+
+        mapped_field['transform'] = 'tests.test_mappings.parseInt'
+        doc = {
+            'str_to_int': '42a'
+        }
+        got = mappings.get_transformed_value(mapped_field, doc, 'str_to_int')
+        self.assertEqual(got, '42a')
+
+    def test_get_transform_document(self):
+        mapping = {
+            'db': {
+                'col': {
+                    'str_to_int': {
+                        'type': 'INT',
+                        'dest': 'str_to_int',
+                        'transform': 'tests.test_mappings.parseInt'
+                    }
+                }
+            }
+        }
+        doc = {
+            'str_to_int': '42'
+        }
+
+        got = mappings.get_transformed_document(mapping, 'db', 'col', doc)
+        self.assertEqual(got, {'str_to_int': 42})
+
+        doc = {
+            'str_to_int': '42a'
+        }
+        got = mappings.get_transformed_document(mapping, 'db', 'col', doc)
+        self.assertEqual(got, {'str_to_int': '42a'})
+
+        mapping['db']['col']['str_to_int']['transform'] = 'missing.no'
+        doc = {
+            'str_to_int': '42'
+        }
+        got = mappings.get_transformed_document(mapping, 'db', 'col', doc)
+        self.assertEqual(got, {'str_to_int': '42'})
+
 
 if __name__ == '__main__':
     main()

tests/test_sql.py

@@ -80,6 +80,7 @@ def test_sql_bulk_insert(self):
         }
 
         sql.sql_bulk_insert(cursor, mapping, 'db.col', [doc])
+
         cursor.execute.assert_called_with(
             "INSERT INTO col (_creationDate,field1,field2_subfield) VALUES (NULL,'val',NULL)"
         )
@@ -161,6 +162,21 @@ def test_sql_bulk_insert_array(self):
         ])
 
     def test_sql_insert(self):
+        mapping = {
+            'db': {
+                'col': {
+                    'pk': '_id',
+                    '_id': {
+                        'type': 'INT',
+                        'dest': '_id'
+                    },
+                    'field': {
+                        'type': 'TEXT',
+                        'dest': 'field'
+                    }
+                }
+            }
+        }
         cursor = MagicMock()
         now = datetime.now()
 
@@ -169,7 +185,7 @@ def test_sql_insert(self):
         doc['_id'] = ObjectId.from_datetime(now)
         doc['field'] = 'val'
 
-        sql.sql_insert(cursor, 'table', doc, '_id')
+        sql.sql_insert(cursor, 'table', doc, mapping, 'db', 'col')
 
         doc['_creationDate'] = utils.extract_creation_date(doc, '_id')
 
@@ -182,7 +198,7 @@ def test_sql_insert(self):
             'field': 'val'
         }
 
-        sql.sql_insert(cursor, 'table', doc, '_id')
+        sql.sql_insert(cursor, 'table', doc, mapping, 'db', 'col')
 
         cursor.execute.assert_called_with(
             'INSERT INTO table  (field)  VALUES  (%(field)s) ',