Add support for database URLs

README.md

@@ -16,7 +16,7 @@ pip install tap-mongodb
 
 ## Set up Config file
 Create json file called `config.json`, with the following contents:
-```
+```json
 {
   "password": "<password>",
   "user": "<username>",
@@ -25,15 +25,29 @@ Create json file called `config.json`, with the following contents:
   "database": "<database name>"
 }
 ```
-The folowing parameters are optional for your config file:
+
+All of the above attributes are required by the tap to connect to your mongo instance. 
+
+
+Alternatively, you can use a database URL to connect in which case the above settings will be ignored and are thus optional.
+Note that you have to include settings like `replica_set` or `ssl` directly in the URL instead of
+providing them separately if you use a database URL to establish the connection.
+
+For example, to connect to a database that uses DNS SRV records:
+```json
+{
+  "database_url": "mongodb+srv://user:[email protected]/test?w=majority&tls=true"
+}
+```
+
+The following parameters are optional for your config file:
 
 | Name | Type | Description |
 | -----|------|------------ |
 | `replica_set` | string | name of replica set |
 |`ssl` | Boolean | can be set to true to connect using ssl |
 | `include_schema_in_destination_stream_name` | Boolean | forces the stream names to take the form `<database_name>_<collection_name>` instead of `<collection_name>`|
 
-All of the above attributes are required by the tap to connect to your mongo instance. 
 
 ## Run in discovery mode
 Run the following command and redirect the output into the catalog file

setup.py

@@ -14,6 +14,7 @@
           'pymongo==3.8.0',
           'tzlocal==2.0.0',
           'terminaltables==3.1.0',
+          'dnspython==2.1.0'
       ],
       extras_require={
           'dev': [

tap_mongodb/__init__.py

@@ -8,6 +8,7 @@
 from bson import timestamp
 
 import singer
+from urllib.parse import urlparse
 from singer import metadata, metrics, utils
 
 import tap_mongodb.sync_strategies.common as common
@@ -18,7 +19,7 @@
 
 LOGGER = singer.get_logger()
 
-REQUIRED_CONFIG_KEYS = [
+REQUIRED_CONFIG_KEYS_IF_NO_DATABASE_URL_PROVIDED = [
     'host',
     'port',
     'user',
@@ -350,27 +351,38 @@ def do_sync(client, catalog, state):
 
 
 def main_impl():
-    args = utils.parse_args(REQUIRED_CONFIG_KEYS)
+    args = utils.parse_args([])
     config = args.config
 
-    # Default SSL verify mode to true, give option to disable
-    verify_mode = config.get('verify_mode', 'true') == 'true'
-    use_ssl = config.get('ssl') == 'true'
+    if not config.get('database_url', None):
+        args = utils.parse_args(REQUIRED_CONFIG_KEYS_IF_NO_DATABASE_URL_PROVIDED)
+        config = args.config
+        connection_params = {"host": config['host'],
+                             "port": int(config['port']),
+                             "username": config.get('user', None),
+                             "password": config.get('password', None),
+                             "authSource": config['database'],
+                             "ssl": use_ssl,
+                             "replicaset": config.get('replica_set', None),
+                             "readPreference": 'secondaryPreferred'}
 
-    connection_params = {"host": config['host'],
-                         "port": int(config['port']),
-                         "username": config.get('user', None),
-                         "password": config.get('password', None),
-                         "authSource": config['database'],
-                         "ssl": use_ssl,
-                         "replicaset": config.get('replica_set', None),
-                         "readPreference": 'secondaryPreferred'}
+        # Default SSL verify mode to true, give option to disable
+        verify_mode = config.get('verify_mode', 'true') == 'true'
+        use_ssl = config.get('ssl') == 'true'
 
-    # NB: "ssl_cert_reqs" must ONLY be supplied if `SSL` is true.
-    if not verify_mode and use_ssl:
-        connection_params["ssl_cert_reqs"] = ssl.CERT_NONE
+        # NB: "ssl_cert_reqs" must ONLY be supplied if `SSL` is true.
+        if not verify_mode and use_ssl:
+            connection_params["ssl_cert_reqs"] = ssl.CERT_NONE
 
-    client = pymongo.MongoClient(**connection_params)
+        client = pymongo.MongoClient(**connection_params)
+
+    else:
+        url = config.get('database_url')
+        url_parsed = urlparse(url)
+        config['user'] = url_parsed.username
+        config['database'] = url_parsed.path[1:]
+        config['host'] = url_parsed.netloc
+        client = pymongo.MongoClient(url)
 
     LOGGER.info('Connected to MongoDB host: %s, version: %s',
                 config['host'],