Merge branch 'release/0.2.3'

CHANGES.txt

@@ -1,3 +1,8 @@
+Version 0.2.3 (2017-05-22)
+--------------------------
+Add extract_schema(uri) function (#26)
+Use standard regular expression for schema URIs (#24)
+
 Version 0.2.2 (2017-05-05)
 --------------------------
 Mark run ids archived to Glacier as processed (#23)

setup.py

@@ -18,7 +18,7 @@
 
 setup(
     name='snowplow_analytics_sdk',
-    version='0.2.2',
+    version='0.2.3',
     description='Snowplow Analytics Python SDK',
     author='Fred Blundun',
     url='https://github.com/snowplow/snowplow-python-analytics-sdk',

snowplow_analytics_sdk/json_shredder.py

@@ -16,27 +16,57 @@
 
 import re
 import json
+
 from snowplow_analytics_sdk.snowplow_event_transformation_exception import SnowplowEventTransformationException
 
+
+# TODO: remove in 0.3.0
+# See: https://github.com/snowplow/snowplow-python-analytics-sdk/issues/27
 SCHEMA_PATTERN = re.compile(""".+:([a-zA-Z0-9_\.]+)/([a-zA-Z0-9_]+)/[^/]+/(.*)""")
 
+SCHEMA_URI = ("^iglu:"                        # Protocol
+              "([a-zA-Z0-9-_.]+)/"            # Vendor
+              "([a-zA-Z0-9-_]+)/"             # Name
+              "([a-zA-Z0-9-_]+)/"             # Format
+              "([1-9][0-9]*"                  # MODEL (cannot start with 0)
+              "(?:-(?:0|[1-9][0-9]*)){2})$")  # REVISION and ADDITION
 
-def fix_schema(prefix, schema):
+SCHEMA_URI_REGEX = re.compile(SCHEMA_URI)
+
+
+def extract_schema(uri):
     """
-    Create an Elasticsearch field name from a schema string
+    Extracts Schema information from Iglu URI
+
+    >>> extract_schema("iglu:com.acme-corporation_underscore/event_name-dash/jsonschema/1-10-1")['vendor']
+    'com.acme-corporation_underscore'
     """
-    match = re.match(SCHEMA_PATTERN, schema)
+    match = re.match(SCHEMA_URI_REGEX, uri)
     if match:
-        snake_case_organization = match.group(1).replace('.', '_').lower()
-        snake_case_name = re.sub('([^A-Z_])([A-Z])', '\g<1>_\g<2>', match.group(2)).lower()
-        model = match.group(3).split('-')[0]
-        return "{}_{}_{}_{}".format(prefix, snake_case_organization, snake_case_name, model)
+        return {
+            'vendor': match.group(1),
+            'name': match.group(2),
+            'format': match.group(3),
+            'version': match.group(4)
+
+        }
     else:
         raise SnowplowEventTransformationException([
-            "Schema {} does not conform to regular expression {}".format(schema, SCHEMA_PATTERN)
+            "Schema {} does not conform to regular expression {}".format(uri, SCHEMA_URI)
         ])
 
 
+def fix_schema(prefix, schema):
+    """
+    Create an Elasticsearch field name from a schema string
+    """
+    schema_dict = extract_schema(schema)
+    snake_case_organization = schema_dict['vendor'].replace('.', '_').lower()
+    snake_case_name = re.sub('([^A-Z_])([A-Z])', '\g<1>_\g<2>', schema_dict['name']).lower()
+    model = schema_dict['version'].split('-')[0]
+    return "{}_{}_{}_{}".format(prefix, snake_case_organization, snake_case_name, model)
+
+
 def parse_contexts(contexts):
     """
     Convert a contexts JSON to an Elasticsearch-compatible list of key-value pairs