Hive load CSV with commas in quoted fields

问题:

I am trying to load a CSV file into a Hive table like so:

CREATE TABLE mytable
(
num1 INT,
text1 STRING,
num2 INT,
text2 STRING
)
ROW FORMAT DELIMITED FIELDS TERMINATED BY ",";

LOAD DATA LOCAL INPATH '/data.csv'
OVERWRITE INTO TABLE mytable;    

The csv is delimited by an comma (,) and looks like this:

1, "some text, with comma in it", 123, "more text"

This will return corrupt data since there is a ',' in the first string.
Is there a way to set an text delimiter or make Hive ignore the ',' in strings?

I can't change the delimiter of the csv since it gets pulled from an external source.

回答1:

The problem is that Hive doesn't handle quoted texts. You either need to pre-process the data by changing the delimiter between the fields (e.g: with a Hadoop-streaming job) or you can also give a try to use a custom CSV SerDe which uses OpenCSV to parse the files.

回答2:

If you can re-create or parse your input data, you can specify an escape character for the CREATE TABLE:

ROW FORMAT DELIMITED FIELDS TERMINATED BY "," ESCAPED BY '\\';

Will accept this line as 4 fields

1,some text\, with comma in it,123,more text

回答3:

As of Hive 0.14, the CSV SerDe is a standard part of the Hive install

ROW FORMAT SERDE 'org.apache.hadoop.hive.serde2.OpenCSVSerde'

(See: https://cwiki.apache.org/confluence/display/Hive/CSV+Serde)

回答4:

keep the delimiter in single quotes it will work.

ROW FORMAT DELIMITED 
FIELDS TERMINATED BY ','
LINES TERMINATED BY '\n';

This will work

回答5:

Add a backward slash in FIELDS TERMINATED BY '\;'

For Example:

CREATE  TABLE demo_table_1_csv
COMMENT 'my_csv_table 1'
ROW FORMAT DELIMITED
FIELDS TERMINATED BY '\;'
LINES TERMINATED BY '\n'
STORED AS TEXTFILE
LOCATION 'your_hdfs_path'
AS 
select a.tran_uuid,a.cust_id,a.risk_flag,a.lookback_start_date,a.lookback_end_date,b.scn_name,b.alerted_risk_category,
CASE WHEN (b.activity_id is not null ) THEN 1 ELSE 0 END as Alert_Flag 
FROM scn1_rcc1_agg as a LEFT OUTER JOIN scenario_activity_alert as b ON a.tran_uuid = b.activity_id;

I have tested it, and it worked.