bq. is it available in version 1.2.6? If you were talking about hbase-spark/src/main/scala/org/apache/hadoop/hbase/spark/HBaseContext.scala , it is not in 1.2.6 since hbase-spark module was not part of 1.2.6 release.
FYI On Mon, Jun 11, 2018 at 2:08 AM, Kang Minwoo <[email protected]> wrote: > Thank you for giving me a good way. > But HBaseContext does not seem to exist in version 1.2.6 > is it available in version 1.2.6? > > And the problem that I can not use HBaseContext is that I am using > CustomTableInputFormat which extends TableInputFormat. > > Best regards, > Minwoo Kang > > ________________________________________ > 보낸 사람: Juan Jose Escobar <[email protected]> > 보낸 날짜: 2018년 6월 9일 토요일 18:36 > 받는 사람: [email protected] > 제목: Re: Odd cell result > > Hello, > > Are you trying to read exported files or similar? Otherwise I think you > need to indicate the format of the data you are reading. I think what you > want to do is easier like this: > > val sparkConf = new SparkConf() > .... > val sc = new SparkContext(sparkConf) > > val conf = HBaseConfiguration.create() > val hbaseContext = new HBaseContext(sc, conf) > val scan = new Scan() > // ... scan config > val rdd = hbaseContext.hbaseRDD(TableName.valueOf(tableName), scan) > rdd.count() > > or use a Spark-HBase connector which encapsulates the details > > Regards > > > On Sat, Jun 9, 2018 at 8:48 AM, Kang Minwoo <[email protected]> > wrote: > > > 1) I am using just InputFormat. (I do not know it is the right answer to > > the question.) > > > > 2) code snippet > > > > ``` > > val rdd = sc.newAPIHadoopFile(...) > > rdd.count() > > ``` > > > > 3) hbase version 1.2.6 > > > > Best regards, > > Minwoo Kang > > > > ________________________________________ > > 보낸 사람: Ted Yu <[email protected]> > > 보낸 날짜: 2018년 6월 8일 금요일 20:01 > > 받는 사람: hbase-user > > 제목: Re: Odd cell result > > > > Which connector do you use for Spark 2.1.2 ? > > > > Is there any code snippet which may reproduce what you experienced ? > > > > Which hbase release are you using ? > > > > Thanks > > > > On Fri, Jun 8, 2018 at 1:50 AM, Kang Minwoo <[email protected]> > > wrote: > > > > > Hello, Users > > > > > > I recently met an unusual situation. > > > That is the cell result does not contain column family. > > > > > > I thought the cell is the smallest unit where data could be transferred > > in > > > HBase. > > > But cell does not contain column family means the cell is not the > > smallest > > > unit. > > > I'm wrong? > > > > > > It occurred in Spark 2.1.2 and did not occur in MR. > > > And now it is not reappearance. > > > > > > Best regards, > > > Minwoo Kang > > > > > >
