bin/transport

@@ -14,19 +14,27 @@ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLI
 
 
 Usage :
-	transport --config <path-to-file.json> --procs <number-procs>
-@TODO: Create tables if they don't exist for relational databases
-example of configuration :
+	transport help -- will print this page
 
-1. Move data from a folder to a data-store
-	transport [--folder <path> ] --config <config.json>	 #-- assuming the configuration doesn't have folder 
-	transport --folder <path> --provider <postgresql|mongo|sqlite> --<database|db> <name> --table|doc <document_name>
-In this case the configuration should look like :
-	{folder:..., target:{}}
-2. Move data from one source to another
-	transport --config <file.json>
-	{source:{..},target:{..}} or [{source:{..},target:{..}},{source:{..},target:{..}}]
+    transport move <path> [index]
+        <path>  path to the configuration file
+        <index> optional index within the configuration file
 
+e.g: configuration file (JSON formatted)
+    - single source to a single target
+
+    {"source":{"provider":"http","url":"https://cdn.wsform.com/wp-content/uploads/2020/06/agreement.csv"}
+        "target":{"provider":"sqlite3","path":"transport-demo.sqlite","table":"agreement"}
+    }
+
+    - single source to multiple targets
+    {
+        "source":{"provider":"http","url":"https://cdn.wsform.com/wp-content/uploads/2020/06/agreement.csv"},
+        "target":[
+            {"provider":"sqlite3","path":"transport-demo.sqlite","table":"agreement},
+            {"provider":"mongodb","db":"transport-demo","collection":"agreement"}
+        ]
+    }
 
 """
 import pandas as pd
@@ -36,51 +44,123 @@ import sys
 import transport
 import time
 from multiprocessing import Process
-SYS_ARGS = {}
-if len(sys.argv) > 1:
+import typer
+import os
+import transport
+from transport import etl
+from transport import providers
 
-    N = len(sys.argv)
-    for i in range(1,N):
-        value = None
-        if sys.argv[i].startswith('--'):
-            key = sys.argv[i][2:] #.replace('-','')
-            SYS_ARGS[key] = 1			
-            if i + 1 < N:
-                value = sys.argv[i + 1] = sys.argv[i+1].strip()
-            if key and value and not value.startswith('--'):
-                SYS_ARGS[key] = value
+# SYS_ARGS = {}
+# if len(sys.argv) > 1:
     
+#     N = len(sys.argv)
+#     for i in range(1,N):
+#         value = None
+#         if sys.argv[i].startswith('--'):
+#             key = sys.argv[i][2:] #.replace('-','')
+#             SYS_ARGS[key] = 1			
+#             if i + 1 < N:
+#                 value = sys.argv[i + 1] = sys.argv[i+1].strip()
+#             if key and value and not value.startswith('--'):
+#                 SYS_ARGS[key] = value
                 
-        i += 2
         
-if __name__ == '__main__' :
-	#
-	# Load information from the file ...
-	if 'help' in SYS_ARGS :
+#         i += 2
+
+app = typer.Typer()
+
+# @app.command()
+def help() :     
 	print (__doc__)
-	else:
-		try:
-			_info = json.loads(open(SYS_ARGS['config']).read())
-			if 'index' in SYS_ARGS :
-				_index = int(SYS_ARGS['index'])
-				_info = [_item for _item in _info if _info.index(_item) == _index]
-				pass
-			elif 'id' in SYS_ARGS :
-				_info = [_item for _item in _info if 'id' in _item and _item['id'] == SYS_ARGS['id']]
-			
-			procs = 1 if 'procs' not in SYS_ARGS else int(SYS_ARGS['procs'])
-			jobs = transport.factory.instance(provider='etl',info=_info,procs=procs)
-			print ([len(jobs),' Jobs are running'])
-			N = len(jobs)
+def wait(jobs):
     while jobs :
-				x = len(jobs)
-				jobs = [_job for _job in jobs if _job.is_alive()]
-				if x != len(jobs) :
-					print ([len(jobs),'... jobs still running'])
+        jobs = [thread for thread in jobs if thread.is_alive()]
         time.sleep(1)
-			print ([N,' Finished running'])
-		except Exception as e:
 
-			print (e)
+@app.command()
+def move (path,index=None):
+
+    _proxy = lambda _object: _object.write(_object.read())
+    if os.path.exists(path):
+        file = open(path)
+        _config = json.loads (file.read() )
+        file.close()
+        if index :
+            _config = _config[ int(index)]
+            etl.instance(**_config)
+        else:
+            etl.instance(config=_config)
+                 
+        #
+        # if type(_config) == dict :
+        #     _object = transport.etl.instance(**_config) 
+        #     _proxy(_object)
+        # else:
+        #     #
+        #     # here we are dealing with a list of objects (long ass etl job)
+        #     jobs = []
+        #     failed = []
+        #     for _args in _config :
+        #         if index and _config.index(_args) != index :
+        #             continue
+
+        #         _object=transport.etl.instance(**_args)
+        #         thread = Process(target=_proxy,args=(_object,))
+        #         thread.start()
+        #         jobs.append(thread())
+        #         if _config.index(_args) == 0 :
+        #             thread.join()
+            # wait(jobs)
+@app.command()
+def version():
+     print (transport.version.__version__)
+@app.command()
+def generate (path:str):
+	"""
+	This function will generate a configuration template to give a sense of how to create one
+	"""
+	_config = [
+          {
+               "source":{"provider":"http","url":"https://raw.githubusercontent.com/codeforamerica/ohana-api/master/data/sample-csv/addresses.csv"},
+             "target":
+            [{"provider":"file","path":"addresses.csv","delimiter":"csv"},{"provider":"sqlite","database":"sample.db3","table":"addresses"}]
+            }
+            ]
+	file = open(path,'w')
+	file.write(json.dumps(_config))
+	file.close()
+@app.command()
+def usage():
+     print (__doc__)	
+if __name__ == '__main__' :
+     app()
+# 	#
+# 	# Load information from the file ...
+# 	if 'help' in SYS_ARGS :
+# 		print (__doc__)
+# 	else:
+# 		try:
+# 			_info = json.loads(open(SYS_ARGS['config']).read())
+# 			if 'index' in SYS_ARGS :
+# 				_index = int(SYS_ARGS['index'])
+# 				_info = [_item for _item in _info if _info.index(_item) == _index]
+# 				pass
+# 			elif 'id' in SYS_ARGS :
+# 				_info = [_item for _item in _info if 'id' in _item and _item['id'] == SYS_ARGS['id']]
+			
+# 			procs = 1 if 'procs' not in SYS_ARGS else int(SYS_ARGS['procs'])
+# 			jobs = transport.factory.instance(provider='etl',info=_info,procs=procs)
+# 			print ([len(jobs),' Jobs are running'])
+# 			N = len(jobs)
+# 			while jobs :
+# 				x = len(jobs)
+# 				jobs = [_job for _job in jobs if _job.is_alive()]
+# 				if x != len(jobs) :
+# 					print ([len(jobs),'... jobs still running'])
+# 				time.sleep(1)
+# 			print ([N,' Finished running'])
+# 		except Exception as e:
+			
+# 			print (e)
 			
 	

setup.py

@@ -17,7 +17,7 @@ args    = {
     "license":"MIT",
     "packages":["transport"]}
 args["keywords"]=['mongodb','couchdb','rabbitmq','file','read','write','s3','sqlite']
-args["install_requires"] = ['pymongo','sqlalchemy<2.0.0','pandas','pandas-gbq','numpy','cloudant','pika','nzpy','boto3','boto','pyarrow','google-cloud-bigquery','google-cloud-bigquery-storage','flask-session','smart_open','botocore','psycopg2-binary','mysql-connector-python']
+args["install_requires"] = ['pyncclient','pymongo','sqlalchemy<2.0.0','pandas','typer','pandas-gbq','numpy','cloudant','pika','nzpy','boto3','boto','pyarrow','google-cloud-bigquery','google-cloud-bigquery-storage','flask-session','smart_open','botocore','psycopg2-binary','mysql-connector-python']
 args["url"] =   "https://healthcareio.the-phi.com/git/code/transport.git"
 args['scripts'] = ['bin/transport']
 if sys.version_info[0] == 2 :

transport/__init__.py

@@ -28,7 +28,7 @@ import importlib
 import sys 
 import sqlalchemy
 if sys.version_info[0] > 2 : 
-	from transport.common import Reader, Writer,Console #, factory
+	# from transport.common import Reader, Writer,Console #, factory
 	from transport import disk
 
 	from transport import s3 as s3
@@ -97,7 +97,7 @@ class factory :
 	TYPE = {"sql":{"providers":["postgresql","mysql","neteeza","bigquery","mariadb","redshift"]}}
 	PROVIDERS = {
 		"etl":{"class":{"read":etl.instance,"write":etl.instance}},
-		"console":{"class":{"write":Console,"read":Console}},
+		# "console":{"class":{"write":Console,"read":Console}},
 		"file":{"class":{"read":disk.DiskReader,"write":disk.DiskWriter}},
 		"sqlite":{"class":{"read":disk.SQLiteReader,"write":disk.SQLiteWriter}},
         "postgresql":{"port":5432,"host":"localhost","database":None,"driver":pg,"default":{"type":"VARCHAR"},"class":{"read":sql.SQLReader,"write":sql.SQLWriter}},
@@ -124,6 +124,9 @@ class factory :
 			#
 			# Legacy code being returned
 			return factory._instance(**_args);
+			
+			
+
 		else:
 			return instance(**_args)
 	@staticmethod
@@ -157,27 +160,49 @@ class factory :
 		return anObject
 
 import time
-def instance(**_args):
+def instance(**_pargs):
 	"""
 	creating an instance given the provider, we should have an idea of :class, :driver
 	:provider
 	:read|write = {connection to the database}
 	"""	
+	#
+	# @TODO: provide authentication file that will hold all the parameters, that will later on be used
+	#
+	_args = dict(_pargs,**{})
+	if 'auth_file' in _args :
+		path = _args['auth_file']
+		file = open(path)
+		_config = json.loads( file.read())
+		_args = dict(_args,**_config)
+		file.close()
+
 	_provider = _args['provider']
-	_group = None
-	
-	for _id in providers.CATEGORIES :
-		if _provider in providers.CATEGORIES[_id] :
-			_group = _id
-			break
-	if _group :
-		_classPointer = _getClassInstance(_group,**_args)
+	_context = list( set(['read','write','listen']) & set(_args.keys()) )
+	if _context :
+		_context = _context[0]
+	else:
+		_context = _args['context'] if 'context' in _args else 'read'
+	# _group = None
+	
+
+	# for _id in providers.CATEGORIES :
+	# 	if _provider in providers.CATEGORIES[_id] :
+	# 		_group = _id
+	# 		break
+	# if _group :
+	
+	if _provider in providers.PROVIDERS and _context in providers.PROVIDERS[_provider]:
+		
+		# _classPointer = _getClassInstance(_group,**_args)
+		_classPointer = providers.PROVIDERS[_provider][_context]
 		#
 		# Let us reformat the arguments
-		if 'read' in _args or 'write' in _args :
-			_args = _args['read'] if 'read' in _args else _args['write']
-			_args['provider'] = _provider
-		if _group == 'sql' :
+		# if 'read' in _args or 'write' in _args :
+		# 	_args = _args['read'] if 'read' in _args else _args['write']
+		# 	_args['provider'] = _provider
+		# if _group == 'sql' :
+		if _provider in providers.CATEGORIES['sql'] :
 			_info = _get_alchemyEngine(**_args)
 
 			_args = dict(_args,**_info)
@@ -202,12 +227,20 @@ def _get_alchemyEngine(**_args):
 	This function returns the SQLAlchemy engine associated with parameters, This is only applicable for SQL _items
 	:_args	arguments passed to the factory {provider and other}
 	"""
+	_provider = _args['provider']
+	_pargs = {}
+	if _provider == providers.SQLITE3 :
+		_path = _args['database'] if 'database' in _args else _args['path']
+		uri = ''.join([_provider,':///',_path])
+		
+	else:
+		
 		#@TODO: Enable authentication files (private_key)
 		_username = _args['username'] if 'username' in _args else ''
 		_password = _args['password'] if 'password' in _args else ''
 		_account = _args['account'] if 'account' in _args else ''
-	_database =  _args['database']	
-	_provider = _args['provider']
+		_database =  _args['database']	if 'database' in _args else _args['path']
+		
 		if _username != '':
 			_account = _username + ':'+_password+'@'
 		_host = _args['host'] if 'host' in _args else ''
@@ -229,31 +262,33 @@ def _get_alchemyEngine(**_args):
 	
 	
 		uri = ''.join([_provider,"://",_account,_fhost,'/',_database])
-	
+		_pargs = {'host':_host,'port':_port,'username':_username,'password':_password}
 	_engine =  sqlalchemy.create_engine (uri,future=True)
 	_out = {'sqlalchemy':_engine}
-	_pargs = {'host':_host,'port':_port,'username':_username,'password':_password}
+	
 	for key in _pargs :
 		if _pargs[key] != '' :
 			_out[key] = _pargs[key]
 	return _out
+@DeprecationWarning
 def _getClassInstance(_group,**_args):
 	"""
 	This function returns the class instance we are attempting to instanciate
 	:_group		items in providers.CATEGORIES.keys()
 	:_args		arguments passed to the factory class
 	"""		
-	if 'read' in _args or 'write' in _args :
-		_context = 'read' if 'read' in _args else _args['write']
-		_info = _args[_context]
-	else:
-		_context = _args['context'] if 'context' in _args else 'read'
-	_class = providers.READ[_group] if _context == 'read' else providers.WRITE[_group]
-	if type(_class) == dict and _args['provider'] in _class:
-		_class = _class[_args['provider']]
-
-	return _class
-
+	# if 'read' in _args or 'write' in _args :
+	# 	_context = 'read' if 'read' in _args else _args['write']
+	# 	_info = _args[_context]
+	# else:
+	# 	_context = _args['context'] if 'context' in _args else 'read'
+	# _class = providers.READ[_group] if _context == 'read' else providers.WRITE[_group]
+	# if type(_class) == dict and _args['provider'] in _class:
+	# 	_class = _class[_args['provider']]
+
+	# return _class
+
+@DeprecationWarning
 def __instance(**_args):
 	"""
 

transport/bricks.py

@@ -0,0 +1,111 @@
+"""
+This file implements databricks handling, This functionality will rely on databricks-sql-connector
+LICENSE (MIT)
+Copyright 2016-2020, The Phi Technology LLC
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
+The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+
+@TODO:
+    - Migrate SQLite to SQL hierarchy
+    - Include Write in Chunks from pandas
+"""
+import os
+import sqlalchemy
+from transport.common import Reader,Writer
+import pandas as pd
+
+
+class Bricks:
+    """
+    :host
+    :token
+    :database
+    :cluster_path
+    :table
+    """
+    def __init__(self,**_args):
+        _host = _args['host']
+        _token= _args['token']
+        _cluster_path = _args['cluster_path']
+        self._schema = _args['schema'] if 'schema' in _args else _args['database']
+        _catalog = _args['catalog']
+        self._table = _args['table'] if 'table' in _args else None
+
+        #
+        # @TODO:
+        # Sometimes when the cluster isn't up and running it takes a while, the user should be alerted of this
+        #
+
+        _uri = f'''databricks://token:{_token}@{_host}?http_path={_cluster_path}&catalog={_catalog}&schema={self._schema}'''
+        self._engine = sqlalchemy.create_engine (_uri)
+        pass
+    def meta(self,**_args):
+        table = _args['table'] if 'table' in _args else self._table
+        if not table :
+            return []
+        else:
+            if sqlalchemy.__version__.startswith('1.') :
+                _m = sqlalchemy.MetaData(bind=self._engine)
+                _m.reflect(only=[table])
+            else:
+                _m = sqlalchemy.MetaData()
+                _m.reflect(bind=self._engine)
+            #
+            # Let's retrieve te information associated with a table
+            # 
+            return [{'name':_attr.name,'type':_attr.type} for _attr in _m.tables[table].columns]  
+
+    def has(self,**_args):
+        return self.meta(**_args)
+    def apply(self,_sql):
+        try:
+            if _sql.lower().startswith('select') :
+                return pd.read_sql(_sql,self._engine)
+        except Exception as e:
+            pass
+
+class BricksReader(Bricks,Reader):
+    """
+    This class is designed for reads and will execute reads against a table name or a select SQL statement
+    """
+    def __init__(self,**_args):
+        super().__init__(**_args)
+    def read(self,**_args):
+        limit = None if 'limit' not in _args else str(_args['limit'])
+        
+        if 'sql' in _args :
+            sql = _args['sql']
+        elif 'table' in _args  :
+            table = _args['table']
+            sql = f'SELECT * FROM {table}'
+        if limit :
+            sql = sql + f' LIMIT {limit}'
+
+        if 'sql' in _args or 'table' in _args :            
+            return self.apply(sql)
+        else:
+            return pd.DataFrame() 
+    pass
+class BricksWriter(Bricks,Writer):
+    def __init__(self,**_args):
+        super().__init__(**_args)
+    def write(self,_data,**_args):  
+        """
+        This data will write data to data-bricks against a given table. If the table is not specified upon initiazation, it can be specified here
+        _data:  data frame to push to databricks
+        _args:  chunks, table, schema
+        """    
+        _schema = self._schema if 'schema' not in _args else _args['schema']
+        _table = self._table if 'table' not in _args else _args['table']
+        _df = _data if type(_data) == pd.DataFrame else _data
+        if type(_df) == dict :
+            _df = [_df]
+        if type(_df) == list :
+            _df = pd.DataFrame(_df)
+        _df.to_sql(
+            name=_table,schema=_schema,
+            con=self._engine,if_exists='append',index=False);
+        pass

transport/common.py

@@ -93,29 +93,29 @@ class ReadWriter(Reader,Writer) :
 	This class implements the read/write functions aggregated
 	"""
 	pass
-class Console(Writer):
-	lock = RLock()
-	def __init__(self,**_args):
-		self.lock = _args['lock'] if 'lock' in _args else False
-		self.info = self.write
-		self.debug = self.write
-		self.log = self.write
-		pass
-	def write (self,logs=None,**_args):
-		if self.lock :
-			Console.lock.acquire()
-		try:
-			_params = _args if logs is None and _args else  logs
-			if type(_params) == list:
-				for row in _params :
-					print (row)
-			else:
-				print (_params)
-		except Exception as e :
-			print (e)
-		finally:
-			if self.lock :
-				Console.lock.release()
+# class Console(Writer):
+# 	lock = RLock()
+# 	def __init__(self,**_args):
+# 		self.lock = _args['lock'] if 'lock' in _args else False
+# 		self.info = self.write
+# 		self.debug = self.write
+# 		self.log = self.write
+# 		pass
+# 	def write (self,logs=None,**_args):
+# 		if self.lock :
+# 			Console.lock.acquire()
+# 		try:
+# 			_params = _args if logs is None and _args else  logs
+# 			if type(_params) == list:
+# 				for row in _params :
+# 					print (row)
+# 			else:
+# 				print (_params)
+# 		except Exception as e :
+# 			print (e)
+# 		finally:
+# 			if self.lock :
+# 				Console.lock.release()
 
 
 """

transport/disk.py

@@ -62,34 +62,25 @@ class DiskWriter(Writer):
 	"""
 	THREAD_LOCK = Lock()
 	def __init__(self,**params):
-		Writer.__init__(self)
-		self.cache['meta'] = {'cols':0,'rows':0,'delimiter':None}
-		if 'path' in params:
-			self.path = params['path']
-		else:
-			self.path = 'data-transport.log'
-		self.delimiter = params['delimiter'] if 'delimiter' in params else None
-		# if 'name' in params:
-		# 	self.name = params['name'];
-		# else:
-		# 	self.name = 'data-transport.log'
-		# if os.path.exists(self.path) == False:
-		# 	os.mkdir(self.path)
-	def meta(self):
-		return self.cache['meta']
-	def isready(self):
-		"""
-			This function determines if the class is ready for execution or not
-			i.e it determines if the preconditions of met prior execution
-		"""
-		return True
-		# p =  self.path is not None and os.path.exists(self.path)
-		# q = self.name is not None 
-		# return p and q
-	def format (self,row):
-		self.cache['meta']['cols'] += len(row) if isinstance(row,list) else len(row.keys())
-		self.cache['meta']['rows'] += 1
-		return (self.delimiter.join(row) if self.delimiter else json.dumps(row))+"\n"
+		super().__init__()
+		self._path = params['path']
+		self._delimiter = params['delimiter'] if 'delimiter' in params else None
+		self._mode = 'w' if 'mode' not in params else params['mode']
+	# def meta(self):
+	# 	return self.cache['meta']
+	# def isready(self):
+	# 	"""
+	# 		This function determines if the class is ready for execution or not
+	# 		i.e it determines if the preconditions of met prior execution
+	# 	"""
+	# 	return True
+	# 	# p =  self.path is not None and os.path.exists(self.path)
+	# 	# q = self.name is not None 
+	# 	# return p and q
+	# def format (self,row):
+	# 	self.cache['meta']['cols'] += len(row) if isinstance(row,list) else len(row.keys())
+	# 	self.cache['meta']['rows'] += 1
+	# 	return (self.delimiter.join(row) if self.delimiter else json.dumps(row))+"\n"
 	def write(self,info,**_args):
 		"""
 			This function writes a record to a designated file
@@ -97,21 +88,15 @@ class DiskWriter(Writer):
 			@param	row	row to be written
 		"""
 		try:
-			_mode = 'a' if 'overwrite' not in _args else 'w'
+			
+			
 			DiskWriter.THREAD_LOCK.acquire()		
-			f = open(self.path,_mode)
-			if self.delimiter :
-				if type(info) == list :
-					for row in info :
-						f.write(self.format(row))
-				else:
-					f.write(self.format(info))
-			else:
-				if not type(info) == str :
-					f.write(json.dumps(info)+"\n")
-				else:
-					f.write(info)
-			f.close()
+			
+			_delim = self._delimiter if 'delimiter' not in _args else _args['delimiter']
+			_path = self._path if 'path' not  in _args else _args['path']
+			_mode = self._mode if 'mode' not in _args else _args['mode']
+			info.to_csv(_path,index=False,sep=_delim)
+			pass
 		except Exception as e:
 			#
 			# Not sure what should be done here ...
@@ -220,17 +205,23 @@ class SQLiteWriter(SQLite,DiskWriter) :
 		#
 		# If the table doesn't exist we should create it
 		#
-	def write(self,info):
+	def write(self,info,**_args):
 		"""
 		"""
 		
-		if not self.fields :
-			self.init(list(info.keys()))
+		#if not self.fields :
+		#	#if type(info) == pd.DataFrame :
+		#	#	_columns = list(info.columns) 
+        #   #self.init(list(info.keys()))
 		
 		if type(info) == dict :
 			info = [info]
 		elif type(info) == pd.DataFrame :
+			info = info.fillna('')
 			info = info.to_dict(orient='records')
+        if not self.fields :
+            _rec = info[0]
+            self.init(list(_rec.keys()))
 
 		SQLiteWriter.LOCK.acquire()
 		try:

transport/etl.py

@@ -35,6 +35,9 @@ import json
 import sys
 import transport
 import time
+import os
+
+
 from multiprocessing import Process
 SYS_ARGS = {}
 if len(sys.argv) > 1:
@@ -52,199 +55,303 @@ if len(sys.argv) > 1:
                 
         
         i += 2
-
-class Post(Process):
-    def __init__(self,**args):
+class Transporter(Process):
+    """
+    The transporter (Jason Stathem) moves data from one persistant store to another
+    - callback functions
+        :onFinish   callback function when finished
+        :onError    callback function when an error occurs
+    :source source data specification
+    :target destination(s) to move the data to
+    """
+    def __init__(self,**_args):
         super().__init__()
-        self.store = args['target']
-        if 'provider' not in args['target'] :
-            pass
-            self.PROVIDER = args['target']['type']
-            # self.writer = 	transport.factory.instance(**args['target'])
-        else:
-            self.PROVIDER = args['target']['provider']
-            self.store['context'] = 'write'
-            # self.store = args['target']
-            self.store['lock'] = True
-            # self.writer = transport.instance(**args['target'])
+        # self.onfinish = _args['onFinish'] 
+        # self._onerror = _args['onError']
+        self._source = _args['source']
+        self._target = _args['target']
+        
         #
-        # If the table doesn't exists maybe create it ?
+        # Let's insure we can support multiple targets
+        self._target = [self._target] if type(self._target) != list else self._target
+       
+        pass
+    def read(self,**_args):
+        """
+        This function
+        """
+        _reader = transport.factory.instance(**self._source)
         #
-        self.rows = args['rows']
-        # self.rows 	=	 args['rows'].fillna('')
+        # If arguments are provided then a query is to be executed (not just a table dump)
+        return _reader.read() if 'args' not in self._source else _reader.read(**self._source['args'])
         
-    def log(self,**_args) :
-        if ETL.logger :
-            ETL.logger.info(**_args)
+    def _delegate_write(self,_data,**_args):
+        """
+        This function will write a data-frame to a designated data-store, The function is built around a delegation design pattern
+        :data   data-frame or object to be written
+        """
+        if _data.shape[0] > 0 :
+            for _target in self._target :
+                if 'write' not in _target :
+                    _target['context'] = 'write'
+                    # _target['lock'] = True
+                else:
+                    # _target['write']['lock'] = True
+                    pass
+                _writer = transport.factory.instance(**_target)
+                _writer.write(_data,**_args)
+                if hasattr(_writer,'close') :
+                    _writer.close()
         
-    def run(self):
-        _info = {"values":self.rows} if 'couch' in self.PROVIDER else self.rows	
+    def write(self,_df,**_args):
+        """
+        """
+        SEGMENT_COUNT = 6
+        MAX_ROWS = 1000000
+        # _df = self.read()
+        _segments = np.array_split(np.range(_df.shape[0]),SEGMENT_COUNT) if _df.shape[0] > MAX_ROWS else np.array( [np.arange(_df.shape[0])])
+        # _index = 0
         
-        writer = transport.factory.instance(**self.store)
-        writer.write(_info)
-        writer.close()
         
+        for _indexes in _segments :
+            _fwd_args = {} if not _args else _args
             
-class ETL (Process):
-    logger = None
-    def __init__(self,**_args):
-        super().__init__()
+            self._delegate_write(_df.iloc[_indexes],**_fwd_args)
+            #
+            # @TODO: Perhaps consider writing up each segment in a thread/process (speeds things up?)
+            pass
+
+def instance(**_args):
+    _proxy = lambda _agent: _agent.write(_agent.read())
+    if 'source' in _args and 'target' in _args :
+        
+        _agent =  Transporter(**_args)  
+        _proxy(_agent)
+        
+    else:
+        _config = _args['config']
+        _items =  [Transporter(**_item) for _item in _config ]
+        _MAX_JOBS = 5
+        _items = np.array_split(_items,_MAX_JOBS)
+        for _batch in _items :
+            jobs = []
+            for _item in _batch :
+                thread = Process(target=_proxy,args = (_item,))
+                thread.start()
+                jobs.append(thread)
+            while jobs :
+                jobs = [thread for thread in jobs if thread.is_alive()]
+                time.sleep(1)
     
-        self.name 	= _args['id'] if 'id' in _args else 'UNREGISTERED'
-        # if 'provider' not in _args['source'] :
-        #     #@deprecate
-        #     self.reader = transport.factory.instance(**_args['source'])
+    pass
+# class Post(Process):
+#     def __init__(self,**args):
+#         super().__init__()
+#         self.store = args['target']
+#         if 'provider' not in args['target'] :
+#             pass
+#             self.PROVIDER = args['target']['type']
+#             # self.writer = 	transport.factory.instance(**args['target'])
 #         else:
+#             self.PROVIDER = args['target']['provider']
+#             self.store['context'] = 'write'
+#             # self.store = args['target']
+#             self.store['lock'] = True
+#             # self.writer = transport.instance(**args['target'])
+#         #
+#         # If the table doesn't exists maybe create it ?
 #         #
-        #     # This is the new interface
-        #     _args['source']['context'] = 'read'	
+#         self.rows = args['rows']
+#         # self.rows 	=	 args['rows'].fillna('')
 		
-        #     self.reader = transport.instance(**_args['source'])
+#     def log(self,**_args) :
+#         if ETL.logger :
+#             ETL.logger.info(**_args)
 		
-        #
-        # do we have an sql query provided or not ....
-        # self.sql = _args['source']['sql'] if 'sql' in _args['source'] else None
-        # self.cmd = _args['source']['cmd'] if 'cmd' in _args['source'] else None
-        # self._oargs = _args['target'] #transport.factory.instance(**_args['target'])
-        self._source = _args ['source']
-        self._target = _args['target']
-        self._source['context'] = 'read'
-        self._target['context'] = 'write'
-       
-        self.JOB_COUNT =  _args['jobs']
-        self.jobs = []
-		# self.logger = transport.factory.instance(**_args['logger'])
-    def log(self,**_args) :
-        if ETL.logger :
-            ETL.logger.info(**_args)
-        
-    def run(self):
-        # if self.cmd :
-        #     idf = self.reader.read(**self.cmd)
-        # else:
-        #     idf = self.reader.read() 
-        # idf = pd.DataFrame(idf)		
-        # # idf = idf.replace({np.nan: None}, inplace = True)
+#     def run(self):
+#         _info = {"values":self.rows} if 'couch' in self.PROVIDER else self.rows	
        
-        # idf.columns = [str(name).replace("b'",'').replace("'","").strip() for name in idf.columns.tolist()]
-        # self.log(rows=idf.shape[0],cols=idf.shape[1],jobs=self.JOB_COUNT)
+#         writer = transport.factory.instance(**self.store)
+#         writer.write(_info)
+#         writer.close()
 
-        #
-        # writing the data to a designated data source 
-        #
-        try:
 		
+# class ETL (Process):
+#     logger = None
+#     def __init__(self,**_args):
+#         super().__init__()
 
-            _log = {"name":self.name,"rows":{"input":0,"output":0}}
-            _reader = transport.factory.instance(**self._source)
-            if 'table' in self._source :
-                _df = _reader.read()
-            else:
-                _df = _reader.read(**self._source['cmd'])
-            _log['rows']['input'] = _df.shape[0]
-            #
-            # Let's write the input data-frame to the target ...
-            _writer  = transport.factory.instance(**self._target)
-            _writer.write(_df)
-            _log['rows']['output'] = _df.shape[0]
+#         self.name 	= _args['id'] if 'id' in _args else 'UNREGISTERED'
+#         # if 'provider' not in _args['source'] :
+#         #     #@deprecate
+#         #     self.reader = transport.factory.instance(**_args['source'])
+#         # else:
+#         #     #
+#         #     # This is the new interface
+#         #     _args['source']['context'] = 'read'	
             
-            # self.log(module='write',action='partitioning',jobs=self.JOB_COUNT)
-            # rows = np.array_split(np.arange(0,idf.shape[0]),self.JOB_COUNT)
+#         #     self.reader = transport.instance(**_args['source'])
         
 #         #
-            # # @TODO: locks
-            # for i in np.arange(self.JOB_COUNT) :
-            #     # _id = ' '.join([str(i),' table ',self.name])
-            #     indexes = rows[i]
-            #     segment = idf.loc[indexes,:].copy() #.to_dict(orient='records')
-            #     _name = "partition-"+str(i)
-            #     if segment.shape[0] == 0 :
-            #         continue
-                
-            #     proc = Post(target = self._oargs,rows = segment,name=_name)
-            #     self.jobs.append(proc)
-            #     proc.start()
-                
-            #     self.log(module='write',action='working',segment=str(self.name),table=self.name,rows=segment.shape[0])
-            # while self.jobs :
-            # 	jobs = [job for job in proc if job.is_alive()]
-            # 	time.sleep(1)
-        except Exception as e:
-            print (e)
-        self.log(**_log)
-    def is_done(self):
-        self.jobs = [proc for proc in self.jobs if proc.is_alive()]
-        return len(self.jobs) == 0
-def instance(**_args):
-    """
-    :path ,index, id
-    :param _info    list of objects with {source,target}`
-    :param logger
-    """
-    logger = _args['logger'] if 'logger' in _args else None
-    if 'path' in _args :
-        _info = json.loads((open(_args['path'])).read())
+#         # do we have an sql query provided or not ....
+#         # self.sql = _args['source']['sql'] if 'sql' in _args['source'] else None
+#         # self.cmd = _args['source']['cmd'] if 'cmd' in _args['source'] else None
+#         # self._oargs = _args['target'] #transport.factory.instance(**_args['target'])
+#         self._source = _args ['source']
+#         self._target = _args['target']
+#         self._source['context'] = 'read'
+#         self._target['context'] = 'write'
+       
+#         self.JOB_COUNT =  _args['jobs']
+#         self.jobs = []
+# 		# self.logger = transport.factory.instance(**_args['logger'])
+#     def log(self,**_args) :
+#         if ETL.logger :
+#             ETL.logger.info(**_args)
+        
+#     def run(self):
+#         # if self.cmd :
+#         #     idf = self.reader.read(**self.cmd)
+#         # else:
+#         #     idf = self.reader.read() 
+#         # idf = pd.DataFrame(idf)		
+#         # # idf = idf.replace({np.nan: None}, inplace = True)
+
+#         # idf.columns = [str(name).replace("b'",'').replace("'","").strip() for name in idf.columns.tolist()]
+#         # self.log(rows=idf.shape[0],cols=idf.shape[1],jobs=self.JOB_COUNT)
 
+#         #
+#         # writing the data to a designated data source 
+#         #
+#         try:
            
-        if 'index' in _args :
-            _index = int(_args['index'])
-            _info = _info[_index]
             
-        elif 'id' in _args :
-            _info = [_item for _item in _info if '_id' in _item and _item['id'] == _args['id']]
-            _info = _info[0] if _info else _info
-    else:
-        _info = _args['info']
-    
-    if logger and type(logger) != str:
-        ETL.logger = logger 
-    elif logger == 'console':
-        ETL.logger = transport.factory.instance(provider='console',context='write',lock=True)
-    if type(_info) in [list,dict] :
-        _info = _info if type(_info) != dict else [_info]
-        #
-        # The assumption here is that the objects within the list are {source,target} 
-        jobs = []
-        for _item in _info :
+#             _log = {"name":self.name,"rows":{"input":0,"output":0}}
+#             _reader = transport.factory.instance(**self._source)
+#             if 'table' in self._source :
+#                 _df = _reader.read()
+#             else:
+#                 _df = _reader.read(**self._source['cmd'])
+#             _log['rows']['input'] = _df.shape[0]
+#             #
+#             # Let's write the input data-frame to the target ...
+#             _writer  = transport.factory.instance(**self._target)
+#             _writer.write(_df)
+#             _log['rows']['output'] = _df.shape[0]
+
+#             # self.log(module='write',action='partitioning',jobs=self.JOB_COUNT)
+#             # rows = np.array_split(np.arange(0,idf.shape[0]),self.JOB_COUNT)
+            
+#             # #
+#             # # @TODO: locks
+#             # for i in np.arange(self.JOB_COUNT) :
+#             #     # _id = ' '.join([str(i),' table ',self.name])
+#             #     indexes = rows[i]
+#             #     segment = idf.loc[indexes,:].copy() #.to_dict(orient='records')
+#             #     _name = "partition-"+str(i)
+#             #     if segment.shape[0] == 0 :
+#             #         continue
+                
+#             #     proc = Post(target = self._oargs,rows = segment,name=_name)
+#             #     self.jobs.append(proc)
+#             #     proc.start()
+                
+#             #     self.log(module='write',action='working',segment=str(self.name),table=self.name,rows=segment.shape[0])
+#             # while self.jobs :
+#             # 	jobs = [job for job in proc if job.is_alive()]
+#             # 	time.sleep(1)
+#         except Exception as e:
+#             print (e)
+#         self.log(**_log)
+#     def is_done(self):
+#         self.jobs = [proc for proc in self.jobs if proc.is_alive()]
+#         return len(self.jobs) == 0
+
+
+# def instance (**_args):
+#     """
+#     path to configuration file
+#     """
+#     _path   = _args['path']
+#     _config = {}
+#     jobs = []
+#     if os.path.exists(_path) :
+#         file    =  open(_path)
+#         _config = json.loads(file.read())
+#         file.close()
+#     if _config and type
+    
+
+# def _instance(**_args):
+#     """
+#     :path ,index, id
+#     :param _info    list of objects with {source,target}`
+#     :param logger
+#     """
+#     logger = _args['logger'] if 'logger' in _args else None
+#     if 'path' in _args :
+#         _info = json.loads((open(_args['path'])).read())
+        
+        
+#         if 'index' in _args :
+#             _index = int(_args['index'])
+#             _info = _info[_index]
+            
+#         elif 'id' in _args :
+#             _info = [_item for _item in _info if '_id' in _item and _item['id'] == _args['id']]
+#             _info = _info[0] if _info else _info
+#     else:
+#         _info = _args['info']
+    
+#     if logger and type(logger) != str:
+#         ETL.logger = logger 
+#     elif logger == 'console':
+#         ETL.logger = transport.factory.instance(provider='console',context='write',lock=True)
+#     if type(_info) in [list,dict] :
+#         _info = _info if type(_info) != dict else [_info]
+#         #
+#         # The assumption here is that the objects within the list are {source,target} 
+#         jobs = []
+#         for _item in _info :
             
-            _item['jobs'] = 5 if 'procs' not in _args else int(_args['procs'])
-            _job = ETL(**_item)
+#             _item['jobs'] = 5 if 'procs' not in _args else int(_args['procs'])
+#             _job = ETL(**_item)
             
-            _job.start()
-            jobs.append(_job)
-        return jobs
+#             _job.start()
+#             jobs.append(_job)
+#         return jobs
         
-    else:
-        return None
+#     else:
+#         return None
 
-if __name__ == '__main__' :
-    _info = json.loads(open (SYS_ARGS['config']).read())
-    index = int(SYS_ARGS['index']) if 'index' in SYS_ARGS else None
-    procs = []
-    for _config in _info :
-        if 'source' in SYS_ARGS :
-            _config['source'] = {"type":"disk.DiskReader","args":{"path":SYS_ARGS['source'],"delimiter":","}}
+# if __name__ == '__main__' :
+#     _info = json.loads(open (SYS_ARGS['config']).read())
+#     index = int(SYS_ARGS['index']) if 'index' in SYS_ARGS else None
+#     procs = []
+#     for _config in _info :
+#         if 'source' in SYS_ARGS :
+#             _config['source'] = {"type":"disk.DiskReader","args":{"path":SYS_ARGS['source'],"delimiter":","}}
 
-        _config['jobs']  = 3 if 'jobs' not in SYS_ARGS else int(SYS_ARGS['jobs'])
-        etl = ETL (**_config)
-        if index is None:	
+#         _config['jobs']  = 3 if 'jobs' not in SYS_ARGS else int(SYS_ARGS['jobs'])
+#         etl = ETL (**_config)
+#         if index is None:	
             
-            etl.start()
-            procs.append(etl)
+#             etl.start()
+#             procs.append(etl)
 
-        elif _info.index(_config) == index :
+#         elif _info.index(_config) == index :
             
-            # print (_config)
-            procs = [etl]
-            etl.start()
-            break
-    #
-    #
-    N = len(procs)
-    while procs :
-        procs = [thread for thread in procs if not thread.is_done()]
-        if len(procs) < N :
-            print (["Finished ",(N-len(procs)), " remaining ", len(procs)])
-            N = len(procs)
-        time.sleep(1)
-	# print ("We're done !!")
+#             # print (_config)
+#             procs = [etl]
+#             etl.start()
+#             break
+#     #
+#     #
+#     N = len(procs)
+#     while procs :
+#         procs = [thread for thread in procs if not thread.is_done()]
+#         if len(procs) < N :
+#             print (["Finished ",(N-len(procs)), " remaining ", len(procs)])
+#             N = len(procs)
+#         time.sleep(1)
+# 	# print ("We're done !!")

transport/mongo.py

@@ -95,9 +95,15 @@ class MongoReader(Mongo,Reader):
         Mongo.__init__(self,**args)
     def read(self,**args):
         
-        if 'mongo' in args or 'cmd' in args:
+        if 'mongo' in args or 'cmd' in args or 'pipeline' in args:
             #
             # @TODO:
+            cmd = {}
+            if 'pipeline' in args :
+                cmd['pipeline']= args['pipeline']
+            if 'aggregate' not in cmd :
+                cmd['aggregate'] = self.collection
+            if 'pipeline' not in args or 'aggregate' not in cmd :
                 cmd = args['mongo'] if 'mongo' in args else args['cmd']
             if "aggregate" in cmd :
                 if "allowDiskUse" not in cmd :

transport/nextcloud.py

@@ -0,0 +1,80 @@
+"""
+We are implementing transport to and from nextcloud (just like s3)
+"""
+import os
+import sys
+from transport.common import Reader,Writer
+import pandas as pd
+from io import StringIO
+import json
+import nextcloud_client as nextcloud
+
+class Nextcloud :
+    def __init__(self,**_args):
+        pass
+        self._delimiter = None
+        self._handler = nextcloud.Client(_args['url'])
+        _uid = _args['uid']
+        _token = _args['token']
+        self._uri = _args['folder'] if 'folder' in _args else './'
+        if self._uri.endswith('/') :
+            self._uri = self._uri[:-1]
+        self._file = None if 'file' not in _args else _args['file']
+        self._handler.login(_uid,_token)
+    def close(self):
+        try:
+            self._handler.logout()
+        except Exception as e:
+            pass
+        
+    
+class NextcloudReader(Nextcloud,Reader):
+    def __init__(self,**_args):
+        # self._file = [] if 'file' not in _args else _args['file']
+        super().__init__(**_args)
+        pass
+    def read(self,**_args):
+        _filename = self._file if 'file' not in _args else _args['file']
+        #
+        # @TODO: if _filename is none, an exception should be raised
+        #
+        _uri = '/'.join([self._uri,_filename])
+        if self._handler.get_file(_uri) :
+            #
+            #
+            _info = self._handler.file_info(_uri)
+            _content = self._handler.get_file_contents(_uri).decode('utf8')
+            if _info.get_content_type() == 'text/csv' :
+                #
+                # @TODO: enable handling of csv, xls, parquet, pickles
+                _file = StringIO(_content)
+                return pd.read_csv(_file)
+            else:
+                #
+                # if it is neither a structured document like csv, we will return the content as is
+                return _content
+        return None     
+class NextcloudWriter (Nextcloud,Writer):
+    """
+    This class will write data to an instance of nextcloud
+    """
+    def __init__(self,**_args)    :
+        super().__init__(**_args)
+        self
+    def write(self,_data,**_args):
+        """
+        This function will upload a file to a given destination 
+        :file   has the uri of the location of the file
+        """
+        _filename = self._file if 'file' not in _args else _args['file']
+        _uri = '/'.join([self._uri,_filename])
+        if type(_data) == pd.DataFrame :
+            f = StringIO()
+            _data.to_csv(f,index=False)
+            _content = f.getvalue()
+        elif type(_data) == dict :
+            _content = json.dumps(_data)
+        else:
+            _content = str(_data)
+        self._handler.put_file_contents(_uri,_content)
+

transport/providers.py

@@ -1,4 +1,4 @@
-from transport.common import Reader, Writer,Console #, factory
+# from transport.common import Reader, Writer,Console #, factory
 from transport import disk
 import sqlite3
 from transport import s3 as s3
@@ -8,6 +8,9 @@ from transport import mongo as mongo
 from transport import sql as sql
 from transport import etl as etl
 from transport import qlistener
+from transport import bricks
+from transport import session
+from transport import nextcloud
 import psycopg2 as pg
 import mysql.connector as my
 from google.cloud import bigquery as bq
@@ -26,12 +29,14 @@ SQLITE = 'sqlite'
 SQLITE3= 'sqlite'
 REDSHIFT = 'redshift'
 NETEZZA = 'netezza'
-MYSQL = 'mysql'
+MYSQL = 'mysql+mysqlconnector'
 RABBITMQ = 'rabbitmq'
 MARIADB  = 'mariadb'
 COUCHDB = 'couch'
 CONSOLE = 'console'
 ETL = 'etl'
+NEXTCLOUD = 'nextcloud'
+
 #
 # synonyms of the above
 BQ 		= BIGQUERY
@@ -45,18 +50,50 @@ AWS_S3  = 's3'
 RABBIT = RABBITMQ
 
 QLISTENER = 'qlistener'
-
+QUEUE = QLISTENER
+DATABRICKS= 'databricks+connector'
 DRIVERS  = {PG:pg,REDSHIFT:pg,MYSQL:my,MARIADB:my,NETEZZA:nz,SQLITE:sqlite3}
-CATEGORIES ={'sql':[NETEZZA,PG,MYSQL,REDSHIFT,SQLITE,MARIADB],'nosql':[MONGODB,COUCHDB],'cloud':[BIGQUERY],'file':[FILE],
-             'queue':[RABBIT,QLISTENER],'memory':[CONSOLE,QLISTENER],'http':[HTTP]}
+CATEGORIES ={'sql':[NETEZZA,PG,MYSQL,REDSHIFT,SQLITE,MARIADB],'nosql':[MONGODB,COUCHDB],'cloud':[NEXTCLOUD,S3,BIGQUERY,DATABRICKS],'file':[FILE],
+             'queue':[RABBIT,QLISTENER],'memory':[CONSOLE,QUEUE],'http':[HTTP]}
 
-READ = {'sql':sql.SQLReader,'nosql':{MONGODB:mongo.MongoReader,COUCHDB:couch.CouchReader},'cloud':sql.BigQueryReader,
+READ = {'sql':sql.SQLReader,'nosql':{MONGODB:mongo.MongoReader,COUCHDB:couch.CouchReader},
+        'cloud':{BIGQUERY:sql.BigQueryReader,DATABRICKS:bricks.BricksReader,NEXTCLOUD:nextcloud.NextcloudReader},
         'file':disk.DiskReader,'queue':{RABBIT:queue.QueueReader,QLISTENER:qlistener.qListener},
-        'cli':{CONSOLE:Console},'memory':{CONSOLE:Console}
+        # 'cli':{CONSOLE:Console},'memory':{CONSOLE:Console},'http':session.HttpReader
+        }
+WRITE = {'sql':sql.SQLWriter,'nosql':{MONGODB:mongo.MongoWriter,COUCHDB:couch.CouchWriter},
+         'cloud':{BIGQUERY:sql.BigQueryWriter,DATABRICKS:bricks.BricksWriter,NEXTCLOUD:nextcloud.NextcloudWriter},
+         'file':disk.DiskWriter,'queue':{RABBIT:queue.QueueWriter,QLISTENER:qlistener.qListener},
+        #  'cli':{CONSOLE:Console},
+        #  'memory':{CONSOLE:Console}, 'http':session.HttpReader
+         
         }
-WRITE = {'sql':sql.SQLWriter,'nosql':{MONGODB:mongo.MongoWriter,COUCHDB:couch.CouchWriter},'cloud':sql.BigQueryWriter,
-         'file':disk.DiskWriter,'queue':{RABBIT:queue.QueueWriter,QLISTENER:qlistener.qListener},'cli':{CONSOLE:Console},'memory':{CONSOLE:Console}
+# SQL_PROVIDERS = [POSTGRESQL,MYSQL,NETEZZA,MARIADB,SQLITE]
+PROVIDERS = {
+    FILE:{'read':disk.DiskReader,'write':disk.DiskWriter},
+    SQLITE:{'read':disk.SQLiteReader,'write':disk.SQLiteWriter,'driver':sqlite3},
+    'sqlite3':{'read':disk.SQLiteReader,'write':disk.SQLiteWriter,'driver':sqlite3},
+    
+    POSTGRESQL:{'read':sql.SQLReader,'write':sql.SQLWriter,'driver':pg,'default':{'host':'localhost','port':5432}},
+    NETEZZA:{'read':sql.SQLReader,'write':sql.SQLWriter,'driver':nz,'default':{'port':5480}},
+    REDSHIFT:{'read':sql.SQLReader,'write':sql.SQLWriter,'driver':pg,'default':{'host':'localhost','port':5432}},
+    RABBITMQ:{'read':queue.QueueReader,'writer':queue.QueueWriter,'context':queue.QueueListener,'default':{'host':'localhost','port':5432}},
+    
+    MYSQL:{'read':sql.SQLReader,'write':sql.SQLWriter,'driver':my,'default':{'host':'localhost','port':3306}},
+    MARIADB:{'read':sql.SQLReader,'write':sql.SQLWriter,'driver':my,'default':{'host':'localhost','port':3306}},
+    
+    S3:{'read':s3.s3Reader,'write':s3.s3Writer},
+    BIGQUERY:{'read':sql.BigQueryReader,'write':sql.BigQueryWriter},
+    DATABRICKS:{'read':bricks.BricksReader,'write':bricks.BricksWriter},
+    NEXTCLOUD:{'read':nextcloud.NextcloudReader,'write':nextcloud.NextcloudWriter},
+
+    QLISTENER:{'read':qlistener.qListener,'write':qlistener.qListener,'default':{'host':'localhost','port':5672}},
+    CONSOLE:{'read':qlistener.Console,"write":qlistener.Console},
+    HTTP:{'read':session.HttpReader,'write':session.HttpWriter},
     
+    MONGODB:{'read':mongo.MongoReader,'write':mongo.MongoWriter,'default':{'port':27017,'host':'localhost'}},
+    COUCHDB:{'read':couch.CouchReader,'writer':couch.CouchWriter,'default':{'host':'localhost','port':5984}},
+    ETL :{'read':etl.Transporter,'write':etl.Transporter}
 }
 DEFAULT = {PG:{'host':'localhost','port':5432},MYSQL:{'host':'localhost','port':3306}}
 DEFAULT[MONGODB] = {'port':27017,'host':'localhost'}

transport/qlistener.py

@@ -40,3 +40,8 @@ class qListener :
         _q = qListener._queue[_id]
         _q.put(_data)
         _q.join()
+class Console (qListener):
+    def __init__(self,**_args):
+        super().__init__(callback=print)
+        
+        # self.callback = print

transport/session.py

@@ -1,54 +1,60 @@
 from flask import request, session
 from datetime import datetime
 import re
-from common import Reader, Writer
+from transport.common import Reader, Writer
 import json
+import requests
+from io import StringIO
+import pandas as pd
 
-class HttpRequestReader(Reader):
+
+class HttpReader(Reader):
 	"""
 	This class is designed to read data from an Http request file handler provided to us by flask
 	The file will be heald in memory and processed accordingly
 	NOTE: This is inefficient and can crash a micro-instance (becareful)
 	"""
 
-	def __init__(self,**params):
-		self.file_length = 0
-		try:
-			
-			#self.file = params['file']	
-			#self.file.seek(0, os.SEEK_END)
-			#self.file_length = self.file.tell()
+	def __init__(self,**_args):
+		self._url = _args['url']
+		self._headers = None if 'headers' not in _args else _args['headers']
 					
-			#print 'size of file ',self.file_length
-			self.content = params['file'].readlines()
-			self.file_length = len(self.content)
-		except Exception as e:
-			print ("Error ... ",e)
-			pass
+	# def isready(self):
+	# 	return self.file_length > 0
+	def format(self,_response):
+		_mimetype= _response.headers['Content-Type']
+		if _mimetype == 'text/csv' or 'text/csv':
+			_content = _response.text
+			return pd.read_csv(StringIO(_content))
+		#
+		# @TODO: Add support for excel, JSON and other file formats that fit into a data-frame
+		#
 		
-	def isready(self):
-		return self.file_length > 0
-	def read(self,size =-1):
-		i = 1
-		for row in self.content:
-			i += 1
-			if size == i:
-				break
-			yield row
+		return _response.text
+	def read(self,**_args):
+		if self._headers :
+			r = requests.get(self._url,headers = self._headers)
+		else:
+			r = requests.get(self._url,headers = self._headers)
+		return self.format(r)
 		
-class HttpSessionWriter(Writer):
+class HttpWriter(Writer):
 	"""
-		This class is designed to write data to a session/cookie
+	This class is designed to submit data to an endpoint (url)
 	"""
-	def __init__(self,**params):
+	def __init__(self,**_args):
 		"""
 			@param key	required session key
 		"""
-		self.session = params['queue']
-		self.session['sql'] = []
-		self.session['csv'] = []
-		self.tablename = re.sub('..+$','',params['filename'])
-		self.session['uid'] = params['uid']
+		self._url = _args['url']
+		self._name = _args['name']
+		self._method = 'post' if 'method' not in _args else _args['method']
+		
+		# self.session = params['queue']
+		# self.session['sql'] = []
+		# self.session['csv'] = []
+		# self.tablename = re.sub('..+$','',params['filename'])
+		# self.session['uid'] = params['uid']
 		#self.xchar = params['xchar']
 			
 		
@@ -57,10 +63,26 @@ class HttpSessionWriter(Writer):
 		return "".join(["INSERT INTO :table VALUES('",values,"');\n"]).replace(':table',self.tablename)		
 	def isready(self):
 		return True
-	def write(self,**params):
-		label = params['label']
-		row = params ['row']
+	def write(self,_data,**_args):
+		#
+		#
+		_method = self._method if 'method' not in _args else _args['method']
+		_method = _method.lower()
+		_mimetype = 'text/csv'
+		if type(_data) == dict :
+			_mimetype = 'application/json'
+			_content = _data
+		else:
+			_content = _data.to_dict(orient='records')
+		_headers = {'Content-Type':_mimetype}
+		_pointer = getattr(requests,_method)
+		
+		_pointer ({self._name:_content},headers=_headers)
+
+
+		# label = params['label']
+		# row = params ['row']
 		
-		if label == 'usable':
-			self.session['csv'].append(self.format(row,','))
-			self.session['sql'].append(self.format_sql(row))
+		# if label == 'usable':
+		# 	self.session['csv'].append(self.format(row,','))
+		# 	self.session['sql'].append(self.format_sql(row))

transport/sql.py

@@ -291,17 +291,17 @@ class SQLWriter(SQLRW,Writer):
         """
         # inspect = False if 'inspect' not in _args else _args['inspect']
         # cast = False if 'cast' not in _args else _args['cast']
-        if not self.fields :
-            if type(info) == list :
-                _fields = info[0].keys()
-            elif type(info) == dict :
-                _fields = info.keys()
-            elif type(info) == pd.DataFrame :
-                _fields = info.columns.tolist()
-
-            # _fields = info.keys() if type(info) == dict else info[0].keys()
-            _fields = list (_fields)
-            self.init(_fields)
+        # if not self.fields :
+        #     if type(info) == list :
+        #         _fields = info[0].keys()
+        #     elif type(info) == dict :
+        #         _fields = info.keys()
+        #     elif type(info) == pd.DataFrame :
+        #         _fields = info.columns.tolist()
+            
+        #     # _fields = info.keys() if type(info) == dict else info[0].keys()
+        #     # _fields = list (_fields)
+        #     self.init(_fields)
             
         try:
             table = _args['table'] if 'table' in _args else self.table
@@ -431,8 +431,8 @@ class BQReader(BigQuery,Reader) :
         
         super().__init__(**_args)    
     def apply(self,sql):
-        self.read(sql=sql)
-        pass
+        return self.read(sql=sql)
+        
     def read(self,**_args):
         SQL = None
         table = self.table if 'table' not in _args else _args['table']

transport/version.py

@@ -1,2 +1,2 @@
 __author__ = 'The Phi Technology'
-__version__= '1.8.2'
+__version__= '1.9.2'