hiplab/data-transport
hiplab
/
data-transport
mirror of https://github.com/lnyemba/data-transport
1
1
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #6 from lnyemba/dev 

bug fix: bigquery chunk/batch sizes
This commit is contained in:
Steve L. Nyemba 2023-12-22 14:18:27 -06:00 committed by GitHub
parent ac5d64242d 81bc5a3ba1
commit 1134de9ad5
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 5 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
transport/sql.py
View File

@ -32,7 +32,7 @@ import nzpy as nz #--- netezza drivers
import sqlite3 import sqlite3
import copy import copy
import os import os
import time
class SQLRW : class SQLRW :
lock = RLock() lock = RLock()
@ -357,7 +357,7 @@ class SQLWriter(SQLRW,Writer):
# #
# Writing with schema information ... # Writing with schema information ...
rows = _info.iloc[i].to_sql(self.table,self._engine,schema=self.schema,if_exists='append',index=False) rows = _info.iloc[i].to_sql(self.table,self._engine,schema=self.schema,if_exists='append',index=False)
time.sleep(1)
else: else:
_fields = ",".join(self.fields) _fields = ",".join(self.fields)
_sql = _sql.replace(":fields",_fields) _sql = _sql.replace(":fields",_fields)
@ -495,10 +495,11 @@ class BQWriter(BigQuery,Writer):
# _df.to_gbq(**self.mode) #if_exists='append',destination_table=partial,credentials=credentials,chunksize=90000) # _df.to_gbq(**self.mode) #if_exists='append',destination_table=partial,credentials=credentials,chunksize=90000)
# #
# Let us adjust the chunking here # Let us adjust the chunking here
self._chunkks = 10 if _df.shape[0] > SQLRW.MAX_CHUNK and self._chunks == 1 else self._chunks self._chunks = 10 if _df.shape[0] > SQLRW.MAX_CHUNK and self._chunks == 1 else self._chunks
_indexes = np.array_split(np.arange(_df.shape[0]),self._chunks) _indexes = np.array_split(np.arange(_df.shape[0]),self._chunks)
for i in _indexes : for i in _indexes :
_df.iloc[i].to_gbq(**self.mode) _df.iloc[i].to_gbq(**self.mode)
time.sleep(1)
pass pass
# #
# Aliasing the big query classes allowing it to be backward compatible # Aliasing the big query classes allowing it to be backward compatible

2
transport/version.py
View File

@ -1,2 +1,2 @@
__author__ = 'The Phi Technology' __author__ = 'The Phi Technology'
__version__= '1.9.2' __version__= '1.9.3'