use hbase

Related pages: setups

See setupsfor installation guides, and happybase configuration.

python2.7

import happybase

#Connect hbase thrift service on port 9090, hbase thrift server must be running.

connection=happybase.Connection("192.168.56.1")

connection=happybase.Connection("192.168.56.1", table_prefix='ns1') #for specific namespace

connection.open()

#Connection pool

pool = happybase.ConnectionPool(size=3, host='192.168.56.1')

with pool.connection() as connection:

    print(connection.tables())

#Connection pool for specific namespace(table_prefix)

pool = happybase.ConnectionPool(size=3, host='192.168.56.1', table_prefix='projectNameX')

with pool.connection() as connection:

    connection.tables()

#Create table #If connection is for specific namespace then tables will be created with 'namespace_' as prefix

connection.create_table('hbtable', {'cf1' : dict(max_versions=3), 'cf2' : dict() } )

#get table instance

table=connection.table('hbtable')

#put rows in table

table.put(b'1', {b'cf1:name': b'Rahul', b'cf2:employer':b'emdee Digitronics Pvt. Ltd', b'cf1:age':b'31'})

table.put(b'2', {b'cf1:name': b'Rohit', b'cf2:employer':b'ugam', b'cf1:age':b'31'})

#multiple puts:

with table.batch() as b:

    b.put(b'3', {b'cf1:name': b'name3', b'cf1:age': b'age3'})

    b.put(b'4', {b'cf1:col1': b'name4', b'cf1:age': b'age4'})

    b.put(b'4', {b'cf1:name': b'name4', b'cf1:age': b'age4'})

#get specific row

row=table.row(b"1")

print(row[b'cf1:name'])

print(row[b'cf2:employer'])

for data in row:

    print(str(data) + "===" + str(row[data]))

#get specific rows

rows=table.rows([b"1", b'2'])

for key, datas in rows:

    print(key)

    for data in datas:

        print(str(data) + "===" + str(datas[data]))

#get specific qualifiers only

row = table.row(b'1', columns=[b'cf1:name', b'cf2:employer'])

for data in row:

    print(str(data) + "===" + str(row[data]))

#get all qualifiers of specific columnFamily

row = table.row(b'1', columns=[b'cf1'])

for data in row:

    print(str(data) + "===" + str(row[data]))

#get values earlier than given timestamp

row = table.row(b'1', timestamp=1492920395629) #helpful in retrieving older version of data

for data in row:

    print(str(data) + "===" + str(row[data]))

#obtain timestamp of insertion

row = table.row(b'1', columns=[b'cf1:name'], include_timestamp=True)

value, timestamp = row[b'cf1:name']

print(value, timestamp)

#scan all rows in table

for key, data in table.scan():

    print(key, data)

#scan rowkeys starting from

for key, data in table.scan(row_start=b'2'):

    print(key, data)

for key, data in table.scan(row_stop=b'2'):

    print(key, data)

for key, data in table.scan(row_start=b'1', row_stop=b'2'):

    print(key, data)

for key, data in table.scan(row_prefix=b'1'):

    print(key, data)

#delete a row's qualifier

table.delete(b'4', {b'cf1:col1'})

table.delete(b'4')

##Counters

print(table.counter_inc(b'5', b'cf1:counter'))  # prints 1

print(table.counter_inc(b'5', b'cf1:counter'))  # prints 1

print(table.counter_inc(b'5', b'cf1:counter'))  # prints 1

print(table.counter_dec(b'5', b'cf1:counter'))  # prints 1

print(table.counter_inc(b'5', b'cf1:counter', value=3)) last counter + 3

print(table.counter_get(b'5', b'cf1:counter'))

print(table.counter_set(b'5', b'cf1:counter', 6))

print(table.counter_get(b'5', b'cf1:counter'))

print(table.counter_set(b'5', b'cf1:counter2', 6))

print(table.counter_get(b'5', b'cf1:counter2'))

Google Sites

Report abuse