Python calls HBase simple instances, and python calls hbase
A new engineer is not familiar with HBase, java is not familiar, and python is okay. I suggest that he consider using thrift call of HBase to complete the current work.
First, install thrift
Download thrift. here, we use thrift-0.7.0-dev.tar.gz.
tar xzf thrift-0.7.0-dev.tar.gzcd thrift-0.7.0-devsudo ./configure --with-cpp=no --with-ruby=nosudo makesudo make install
Then, go to the HBase source code package and find
Src/main/resources/org/apache/hadoop/hbase/thrift/
Run
Thrift -- gen py Hbase. thrift
Mv gen-py/hbase // usr/lib/python2.4/site-packages/(may vary with python versions)
I have written some calling scripts for your reference.
from unittest import TestCase, mainfrom thrift import Thriftfrom thrift.transport import TSocketfrom thrift.transport import TTransportfrom thrift.protocol import TBinaryProtocolfrom hbase import Hbasefrom hbase.ttypes import ColumnDescriptor, Mutation, BatchMutationclass HBaseTester: def __init__(self, netloc, port, table="staftesttable"): self.tableName = table self.transport = TTransport.TBufferedTransport( TSocket.TSocket(netloc, port)) self.protocol = TBinaryProtocol.TBinaryProtocol(self.transport) self.client = Hbase.Client(self.protocol) self.transport.open() tables = self.client.getTableNames() if self.tableName not in tables: self.__createTable() def __del__(self): self.transport.close() def __createTable(self): name = ColumnDescriptor(name='name') foo = ColumnDescriptor(name='foo') self.client.createTable(self.tableName, [name,foo]) def put(self,key,name,foo): name = Mutation(column="name:v", value=name) foo = Mutation(column="foo:v",value=foo) self.client.mutateRow(self.tablename,key,[name,foo]) def scanner(self,column): scanner = client.scannerOpen(self.tablename,"",[column])r = client.scannerGet(scanner)result= [] while r: print r[0] result.append(r[0]) r = client.scannerGet(scanner) print "Scanner finished" return result class TestHBaseTester(TestCase): def setUp(self): self.writer = HBaseTester("localhost", 9090) def tearDown(self): name = self.writer.tableName client = self.writer.client client.disableTable(name) client.deleteTable(name) def testCreate(self): tableName = self.writer.tableName client = self.writer.client self.assertTrue(self.writer.tableName in client.getTableNames()) columns =['name:','foo:'] for i in client.getColumnDescriptors(tableName): self.assertTrue(i in columns) def testPut(self): self.writer.put("r1","n1","f1") self.writer.put("r2","n2","f2") self.writer.put("r3","n3","") self.writer.scanner("name:") if __name__ == "__main__": main()
The simple example of calling HBase in python above is all the content shared by Alibaba Cloud xiaobian. I hope you can give us a reference and support the help house.