.. role:: bash(code) :language: bash

===== Python connector for SQream DB

Version: 2.1.4a3

Supported SQream DB versions: 1.13 onward

Installing

Install with pip, by running :bash:pip install pysqream.

Usage example:

.. code-block:: python

## Import and establish a connection  
#  ---------------------------------   
import pysqream

# version information
print pysqream.version_info()

con = pysqream.Connector()
# Connection parameters: IP, Port, Database, Username, Password, Clustered, Timeout(sec)
sqream_connection_params = '127.0.0.1', 5000, 'master', 'sqream', 'sqream', False, 30
con.connect(*sqream_connection_params)


## Run queries using the API 
#  -------------------------     
# Create a table
statement = 'create or replace table table_name (int_column int)'
con.prepare(statement)
con.execute()
con.close()

# Insert sample data
statement = 'insert into table_name(int_column) values (5), (6)'
con.prepare(statement)
con.execute()
con.close()

# Retreive data
statement = 'select int_column from table_name'
con.prepare(statement)
con.execute()
con.next_row()

# Pull out the actual data
first_row_int = con.get_int(1)
con.next_row()
second_row_int = con.get_int(1)
con.next_row()
print (first_row_int, second_row_int)
con.close()


## After running all statements
#  ----------------------------
con.close_connection()

Example of GET data loop:

.. code-block:: python

# Here we create the according table by
# executing a "create or replace table table_name (int_column int, varchar_column varchar(10))" statement

row1 = []
row2 = []

statement = 'select int_column, varchar_column from table_name'
con.prepare(statement)
con.execute()

while con.next_row():
    row1.append(con.get_int(1))
    row2.append(con.get_string(2))

con.close()
con.close_connection()

Example of a SET data loop for data loading:

.. code-block:: python

# here we create the according table by executing a 
# "create or replace table table_name (int_column int, varchar_column varchar(10))" statement

row1 = [1,2,3]
row2 = ["s1","s2","s3"]
length_of_arrays = 3

# each interogation symbol represent a column to which the network insertion can push
statement = 'insert into table_name(int_column, varchar_column) values(?, ?)' 
con.prepare(statement)
con.execute()

for idx in range(length_of_arrays):
    con.set_int(1, row1[idx])      # we put a value at column 1 of the table
    con.set_varchar(2, row2[idx])  # we put a value at column 2 of the table
    con.next_row()

con.close()
con.close_connection()

Example inserting from a csv to SQream

.. code-block:: python

def insert_from_csv(con,table_name,csv_filename, field_delimiter = ',', null_markers = []):

    # get info on the columns for the insert statement

    # you can get this info after preparing the insert, but we need to at
    # least know the number of columns to be able to construct the insert
    # statement

    with pysqream.sqream_run(con,f"select * from {table_name} limit 0") as con:
        column_types = con.get_column_types()


    def parse_datetime(v):
        try:
            return datetime.datetime.strptime(row[i], '%Y-%m-%d %H:%M:%S.%f')
        except ValueError:
            try: 
                return datetime.datetime.strptime(row[i], '%Y-%m-%d %H:%M:%S')
            except ValueError:
                return datetime.datetime.strptime(row[i], '%Y-%m-%d')

    # insert the csv
    qstring = ",".join(['?'] * len(column_types))
    with pysqream.sqream_run(con, f"insert into {table_name} values ({qstring})") as con:
        with open(csv_filename, mode='r') as csv_file:
            csv_reader = csv.reader(csv_file, delimiter=field_delimiter)
            for row in csv_reader:
                for i,(t,v) in enumerate(zip(column_types, row)):
                    ii = i + 1
                    if row[i] in null_markers:
                        con.set_null(ii)
                    elif t.tid == 'Tinyint':
                        con.set_ubyte(ii, int(row[i]))
                    elif t.tid == "Smallint":
                        con.set_short(ii, int(row[i]))
                    elif t.tid == "Int":
                        con.set_int(ii, int(row[i]))
                    elif t.tid == "Bigint":
                        con.set_long(ii, int(row[i]))
                    elif t.tid == "Real":
                        con.set_float(ii, float(row[i]))
                    elif t.tid == "Float":
                        con.set_double(ii, float(row[i]))
                    elif t.tid == "Date":
                        dt = datetime.datetime.strptime(row[i], "%Y-%m-%d")
                        dt = datetime.date(dt.year, dt.month, dt.day)
                        con.set_date(ii, dt)
                    elif t.tid == "DateTime":
                        dt = parse_datetime(row[i])
                        con.set_datetime(ii, dt)
                    elif t.tid == "Varchar":
                        con.set_varchar(ii, row[i])
                    elif t.tid == "NVarchar":
                        con.set_nvarchar(ii, row[i])
                con.next_row()

Example saving the results of a query to a csv file

.. code-block:: python

def save_query(con, query, csv_filename, field_delimiter, null_marker):

    with pysqream.sqream_run(con, query) as con:
        column_types = con.get_column_types()
        with open(csv_filename, 'x', newline='') as csvfile:
            wr = csv.writer(csvfile, delimiter=field_delimiter,quoting=csv.QUOTE_MINIMAL)
            while con.next_row():
                csv_row = []
                for i,t in enumerate(column_types):
                    ii = i + 1
                    if con.is_null(ii):
                        csv_row.append(null_marker)
                    elif t.tid == 'Tinyint':
                        csv_row.append(con.get_ubyte(ii))
                    elif t.tid == "Smallint":
                        csv_row.append(con.get_short(ii))
                    elif t.tid == "Int":
                        csv_row.append(con.get_int(ii))
                    elif t.tid == "Bigint":
                        csv_row.append(con.get_long(ii))
                    elif t.tid == "Real":
                        csv_row.append(con.get_float(ii))
                    elif t.tid == "Float":
                        csv_row.append(con.get_double(ii))
                    elif t.tid == "Date":
                        csv_row.append(con.get_date(ii))
                    elif t.tid == "DateTime":
                        csv_row.append(con.get_datetime(ii))
                    elif t.tid == "Varchar":
                        csv_row.append(con.get_varchar(ii))
                    elif t.tid == "NVarchar":
                        csv_row.append(con.get_nvarchar(ii))
                wr.writerow(csv_row)

API Reference

All functions are accessed through the Connector class imported from pysqream.py:

Initialization - Termination

.. code-block:: python

import pysqream
con = pysqream.Connector()

# arg types are: string, integer, string, string, string, boolean, integer
con.connect(ip, port, database, username, password, clustered, timeout) 

# closes the statement (to do after execute + necessary fetch/put to close the statement and be 
# able to open another one through prepare())
con.close() 

# closes the connection completely, destructing the socket, a call to "connect(..)" needs to be done do continue
con.close_connection() 

High level protocol functions

.. code-block:: python

con.prepare(statement) #string of the query to run
con.execute()

# if the statement is an insert it produces a put and for select it produces a fetch, rows are 
# incremented through that function (see Usage example)
con.next_row() 

Get column based data

By column id or column name (integer or string)

.. code-block:: python

is_null(col_id_or_col_name)
get_bool(col_id_or_col_name)
get_ubyte(col_id_or_col_name)
get_short(col_id_or_col_name)
get_int(col_id_or_col_name)
get_long(col_id_or_col_name)
get_float(col_id_or_col_name)
get_double(col_id_or_col_name)
get_date(col_id_or_col_name)
get_datetime(col_id_or_col_name)
get_varchar(col_id_or_col_name)
get_nvarchar(col_id_or_col_name)

Set column based data

By column id

.. code-block:: python

set_null(col)
set_bool(col, val)
set_ubyte(col, val)
set_short(col, val)
set_int(col, val)
set_long(col, val)
set_float(col, val)
set_double(col, val)
set_date(col, val)
set_datetime(col, val)
set_varchar(col, val)
set_nvarchar(col, val)