|
|
@ -16,6 +16,16 @@ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLI
|
|
|
|
Usage :
|
|
|
|
Usage :
|
|
|
|
transport --config <path-to-file.json> --procs <number-procs>
|
|
|
|
transport --config <path-to-file.json> --procs <number-procs>
|
|
|
|
@TODO: Create tables if they don't exist for relational databases
|
|
|
|
@TODO: Create tables if they don't exist for relational databases
|
|
|
|
|
|
|
|
example of configuration :
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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:{..}}]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
"""
|
|
|
|
"""
|
|
|
@ -46,11 +56,23 @@ if len(sys.argv) > 1:
|
|
|
|
class Post(Process):
|
|
|
|
class Post(Process):
|
|
|
|
def __init__(self,**args):
|
|
|
|
def __init__(self,**args):
|
|
|
|
super().__init__()
|
|
|
|
super().__init__()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
if 'provider' not in args['target'] :
|
|
|
|
self.PROVIDER = args['target']['type']
|
|
|
|
self.PROVIDER = args['target']['type']
|
|
|
|
self.writer = transport.factory.instance(**args['target'])
|
|
|
|
self.writer = transport.factory.instance(**args['target'])
|
|
|
|
|
|
|
|
else:
|
|
|
|
|
|
|
|
self.PROVIDER = args['target']['provider']
|
|
|
|
|
|
|
|
args['target']['context'] = 'write'
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
self.writer = transport.instance(**args['target'])
|
|
|
|
|
|
|
|
#
|
|
|
|
|
|
|
|
# If the table doesn't exists maybe create it ?
|
|
|
|
|
|
|
|
#
|
|
|
|
self.rows = args['rows']
|
|
|
|
self.rows = args['rows']
|
|
|
|
|
|
|
|
|
|
|
|
def run(self):
|
|
|
|
def run(self):
|
|
|
|
_info = {"values":self.rows} if 'couch' in self.PROVIDER else self.rows
|
|
|
|
_info = {"values":self.rows} if 'couch' in self.PROVIDER else self.rows
|
|
|
|
|
|
|
|
|
|
|
|
self.writer.write(_info)
|
|
|
|
self.writer.write(_info)
|
|
|
|
self.writer.close()
|
|
|
|
self.writer.close()
|
|
|
|
|
|
|
|
|
|
|
@ -59,7 +81,19 @@ class ETL (Process):
|
|
|
|
def __init__(self,**_args):
|
|
|
|
def __init__(self,**_args):
|
|
|
|
super().__init__()
|
|
|
|
super().__init__()
|
|
|
|
self.name = _args['id']
|
|
|
|
self.name = _args['id']
|
|
|
|
|
|
|
|
if 'provider' not in _args['source'] :
|
|
|
|
|
|
|
|
#@deprecate
|
|
|
|
self.reader = transport.factory.instance(**_args['source'])
|
|
|
|
self.reader = transport.factory.instance(**_args['source'])
|
|
|
|
|
|
|
|
else:
|
|
|
|
|
|
|
|
#
|
|
|
|
|
|
|
|
# This is the new interface
|
|
|
|
|
|
|
|
_args['source']['context'] = 'read'
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
self.reader = transport.instance(**_args['source'])
|
|
|
|
|
|
|
|
#
|
|
|
|
|
|
|
|
# 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._oargs = _args['target'] #transport.factory.instance(**_args['target'])
|
|
|
|
self.JOB_COUNT = _args['jobs']
|
|
|
|
self.JOB_COUNT = _args['jobs']
|
|
|
|
self.jobs = []
|
|
|
|
self.jobs = []
|
|
|
@ -68,6 +102,9 @@ class ETL (Process):
|
|
|
|
_args['name'] = self.name
|
|
|
|
_args['name'] = self.name
|
|
|
|
print (_args)
|
|
|
|
print (_args)
|
|
|
|
def run(self):
|
|
|
|
def run(self):
|
|
|
|
|
|
|
|
if self.cmd :
|
|
|
|
|
|
|
|
idf = self.reader.read(**self.cmd)
|
|
|
|
|
|
|
|
else:
|
|
|
|
idf = self.reader.read()
|
|
|
|
idf = self.reader.read()
|
|
|
|
idf = pd.DataFrame(idf)
|
|
|
|
idf = pd.DataFrame(idf)
|
|
|
|
idf.columns = [str(name).replace("b'",'').replace("'","").strip() for name in idf.columns.tolist()]
|
|
|
|
idf.columns = [str(name).replace("b'",'').replace("'","").strip() for name in idf.columns.tolist()]
|
|
|
@ -79,7 +116,8 @@ class ETL (Process):
|
|
|
|
try:
|
|
|
|
try:
|
|
|
|
self.log(module='write',action='partitioning')
|
|
|
|
self.log(module='write',action='partitioning')
|
|
|
|
rows = np.array_split(np.arange(idf.shape[0]),self.JOB_COUNT)
|
|
|
|
rows = np.array_split(np.arange(idf.shape[0]),self.JOB_COUNT)
|
|
|
|
|
|
|
|
#
|
|
|
|
|
|
|
|
# @TODO: locks
|
|
|
|
for i in rows :
|
|
|
|
for i in rows :
|
|
|
|
_id = 'segment #'.join([str(rows.index(i)),self.name])
|
|
|
|
_id = 'segment #'.join([str(rows.index(i)),self.name])
|
|
|
|
segment = idf.loc[i,:] #.to_dict(orient='records')
|
|
|
|
segment = idf.loc[i,:] #.to_dict(orient='records')
|
|
|
|