I am trying to fetch larger dataset from teradata using dask and sqlalchmey. I am able to apply single whereclause and able to fetch data.below is the working code
td_engine = create_engine(connString)
metadata = MetaData()
t = Table(
"table",
metadata,
Column("c1"),
schema="schema",
)
sql = select([t]).where(
t.c.c1 == 'abc',
)
)
start = perf_counter()
df = dd.read_sql_table(sql, connString, index_col="c1",schema="schema")
end = perf_counter()
print("Time taken to execute the code {}".format(end - start))
print(df.head())
but when I am trying to apply and in whereclause I am getting error
sql = select([t]).where(
and_(
t.c.c1 == 'abc',
t.c.c2 == 'xyz'
)
)
More context would be helpful. If you simply need to execute the query, have you considered using the pandas read_sql function and composing the SQL request yourself?
import teradatasql
import pandas as pd
with teradatasql.connect(host="whomooz",user="guest",password="please") as con:
df = pd.read_sql("select c1 from mytable where c1='abc' and c2='xyz'", con)
print(df.head())
Or is there a specific need to use the pandas functions to construct the SQL request?
Related
I've got my data put into an SQLite3 database, and now I'm trying to work on a little script to access data I want for given dates. I got the SELECT statement to work with the date ranges, but I can't seem to add another condition to fine tune the search.
db columns id, date, driverid, drivername, pickupStop, pickupPkg, delStop, delPkg
What I've got so far:
import pandas as pd
import sqlite3
sql_data = 'driverperformance.sqlite'
conn = sqlite3.connect(sql_data)
cur = conn.cursor()
date_start = "2021-12-04"
date_end = "2021-12-10"
df = pd.read_sql_query("SELECT DISTINCT drivername FROM DriverPerf WHERE date BETWEEN :dstart and :dend", params={"dstart": date_start, "dend": date_end}, con=conn)
drivers = df.values.tolist()
for d in drivers:
driverDF = pd.read_sql_query("SELECT * FROM DriverPerf WHERE drivername = :driver AND date BETWEEN :dstart and :dend", params={"driver": d, "dstart": date_start, "dend": date_end}, con=conn)
I've tried a few different versions of the "WHERE drivername" part but it always seems to fail.
Thanks!
If I'm not mistaken, drivers will be a list of lists. Have you tried
.... params={"driver": d[0] ....
I have several large pandas dataframes (about 30k+ rows) and need to upload a different version of them daily to a MS SQL Server db. I am trying to do so with the to_sql pandas function. On occasion, it will work. Other times, it will fail - silently - as if the code uploaded all of the data despite not having uploaded a single row.
Here is my code:
class SQLServerHandler(DataBaseHandler):
...
def _getSQLAlchemyEngine(self):
'''
Get an sqlalchemy engine
from the connection string
The fast_executemany fails silently:
https://stackoverflow.com/questions/48307008/pandas-to-sql-doesnt-insert-any-data-in-my-table/55406717
'''
# escape special characters as required by sqlalchemy
dbParams = urllib.parse.quote_plus(self.connectionString)
# create engine
engine = sqlalchemy.create_engine(
'mssql+pyodbc:///?odbc_connect={}'.format(dbParams))
return engine
#logExecutionTime('Time taken to upload dataframe:')
def uploadData(self, tableName, dataBaseSchema, dataFrame):
'''
Upload a pandas dataFrame
to a database table <tableName>
'''
engine = self._getSQLAlchemyEngine()
dataFrame.to_sql(
tableName,
con=engine,
index=False,
if_exists='append',
method='multi',
chunksize=50,
schema=dataBaseSchema)
Switching the method to None seems to work properly but the data takes an insane amount of time to upload (30+ mins). Having multiple tables (20 or so) a day of this size discards this solution.
The proposed solution here to add the schema as a parameter doesn't work. Neither does creating a sqlalchemy session and passsing it to the con parameter with session.get_bind().
I am using:
ODBC Driver 17 for SQL Server
pandas 1.2.1
sqlalchemy 1.3.22
pyodbc 4.0.30
Does anyone know how to make it raise an exception if it fails?
Or why it is not uploading any data?
In rebuttal to this answer, if to_sql() was to fall victim to the issue described in
SQL Server does not finish execution of a large batch of SQL statements
then it would have to be constructing large anonymous code blocks of the form
-- Note no SET NOCOUNT ON;
INSERT INTO gh_pyodbc_262 (id, txt) VALUES (0, 'row0');
INSERT INTO gh_pyodbc_262 (id, txt) VALUES (1, 'row1');
INSERT INTO gh_pyodbc_262 (id, txt) VALUES (2, 'row2');
…
and that is not what to_sql() is doing. If it were, then it would start to fail well below 1_000 rows, at least on SQL Server 2017 Express Edition:
import pandas as pd
import pyodbc
import sqlalchemy as sa
print(pyodbc.version) # 4.0.30
table_name = "gh_pyodbc_262"
num_rows = 400
print(f" num_rows: {num_rows}") # 400
cnxn = pyodbc.connect("DSN=mssqlLocal64", autocommit=True)
crsr = cnxn.cursor()
crsr.execute(f"TRUNCATE TABLE {table_name}")
sql = "".join(
[
f"INSERT INTO {table_name} ([id], [txt]) VALUES ({i}, 'row{i}');"
for i in range(num_rows)
]
)
crsr.execute(sql)
row_count = crsr.execute(f"SELECT COUNT(*) FROM {table_name}").fetchval()
print(f"row_count: {row_count}") # 316
Using to_sql() for that same operation works
import pandas as pd
import pyodbc
import sqlalchemy as sa
print(pyodbc.version) # 4.0.30
table_name = "gh_pyodbc_262"
num_rows = 400
print(f" num_rows: {num_rows}") # 400
df = pd.DataFrame(
[(i, f"row{i}") for i in range(num_rows)], columns=["id", "txt"]
)
engine = sa.create_engine(
"mssql+pyodbc://#mssqlLocal64", fast_executemany=True
)
df.to_sql(
table_name,
engine,
index=False,
if_exists="replace",
)
with engine.connect() as conn:
row_count = conn.execute(
sa.text(f"SELECT COUNT(*) FROM {table_name}")
).scalar()
print(f"row_count: {row_count}") # 400
and indeed will work for thousands and even millions of rows. (I did a successful test with 5_000_000 rows.)
Ok, this seems to be an issue with SQL Server itself.
SQL Server does not finish execution of a large batch of SQL statements
I have a spark script that needs to make 60 api calls for every row. Currently I am using BigQuery as a data warehouse. I was wondering if there was a way I can use either the BigQuery API or BigQuery Storage API to query the database from my udf? Maybe a way to perform batch queries? Would pandas-gbq be a better solution? Each query that I need to make per row is a select count(*) from dataset.table where {...} query.
Currently I am using the big query client as shown in the code snippet below, but I am not sure if this is the best way to utilize my resources. Apologies if the code is not done properly for this use case, I am new to spark and BigQuery.
def clients():
os.environ['GOOGLE_APPLICATION_CREDENTIALS'] = '/home/hadoop/credentials.json'
credentials, your_project_id = google.auth.default(
scopes=["https://www.googleapis.com/auth/cloud-platform"]
)
# Make clients.
bqclient = bigquery.Client(
credentials=credentials,
project=your_project_id,
)
bqstorageclient = bigquery_storage_v1beta1.BigQueryStorageClient(
credentials=credentials
)
return bqclient, bqstorageclient
def query_cache(query):
bqclient, bqstorageclient = clients()
dataframe = (
bqclient.query(query)
.result()
.to_dataframe(bqstorage_client=bqstorageclient)
)
return dataframe['f0_'][0]
#pandas_udf(schema(), PandasUDFType.GROUPED_MAP)
def calc_counts(df):
query = "select count(*) from dataset.table where ...{some column filters}..."
df['count'] = df.apply(query_cache, args=(query), axis=1)
The simpler option is to use the spark-bigquery-connector, which let you query BigQuery directly and get the result as a Spark dataframe. Converting this dataframe into pandas is then simple:
spark_df = spark.read.format('bigquery').option('table', table).load()
pandas_df = spark_df.toPandas()
I am trying to read a large table (10-15M rows) from a database into pandas dataframe and I'm using the following code:
def read_sql_tmpfile(query, db_engine):
with tempfile.TemporaryFile() as tmpfile:
copy_sql = "COPY ({query}) TO STDOUT WITH CSV {head}".format(
query=query, head="HEADER"
)
conn = db_engine.raw_connection()
cur = conn.cursor()
cur.copy_expert(copy_sql, tmpfile)
tmpfile.seek(0)
df = pandas.read_csv(tmpfile)
return df
I can use this if I have a simple query like this and I pass this into above func:
'''SELECT * from hourly_data'''
But what if I want to pass some variable into this query i.e.
'''SELECT * from hourly_data where starttime >= %s '''
Now where do I pass the parameter?
You cannot use parameters with COPY. Unfortunately that extends to the query you use inside COPY, even if you could use parameters with the query itself.
You will have to construct a query string including the parameter (beware of SQL injection) and use that with COPY.
I am trying to write a program in Python3 that will run a query on a table in Microsoft SQL and put the results into a Pandas DataFrame.
My first try of this was the below code, but for some reason I don't understand the columns do not appear in the order I ran them in the query and the order they appear in and the labels they are given as a result change, stuffing up the rest of my program:
import pandas as pd, pyodbc
result_port_mapl = []
# Use pyodbc to connect to SQL Database
con_string = 'DRIVER={SQL Server};SERVER='+ <server> +';DATABASE=' +
<database>
cnxn = pyodbc.connect(con_string)
cursor = cnxn.cursor()
# Run SQL Query
cursor.execute("""
SELECT <field1>, <field2>, <field3>
FROM result
""")
# Put data into a list
for row in cursor.fetchall():
temp_list = [row[2], row[1], row[0]]
result_port_mapl.append(temp_list)
# Make list of results into dataframe with column names
## FOR SOME REASON HERE row[1] AND row[0] DO NOT CONSISTENTLY APPEAR IN THE
## SAME ORDER AND SO THEY ARE MISLABELLED
result_port_map = pd.DataFrame(result_port_mapl, columns={'<field1>', '<field2>', '<field3>'})
I have also tried the following code
import pandas as pd, pyodbc
# Use pyodbc to connect to SQL Database
con_string = 'DRIVER={SQL Server};SERVER='+ <server> +';DATABASE=' + <database>
cnxn = pyodbc.connect(con_string)
cursor = cnxn.cursor()
# Run SQL Query
cursor.execute("""
SELECT <field1>, <field2>, <field3>
FROM result
""")
# Put data into DataFrame
# This becomes one column with a list in it with the three columns
# divided by a comma
result_port_map = pd.DataFrame(cursor.fetchall())
# Get column headers
# This gives the error "AttributeError: 'pyodbc.Cursor' object has no
# attribute 'keys'"
result_port_map.columns = cursor.keys()
If anyone could suggest why either of those errors are happening or provide a more efficient way to do it, it would be greatly appreciated.
Thanks
If you just use read_sql? Like:
import pandas as pd, pyodbc
con_string = 'DRIVER={SQL Server};SERVER='+ <server> +';DATABASE=' + <database>
cnxn = pyodbc.connect(con_string)
query = """
SELECT <field1>, <field2>, <field3>
FROM result
"""
result_port_map = pd.read_sql(query, cnxn)
result_port_map.columns.tolist()