[jira] [Created] (SPARK-25506) Spark CSV multiline with CRLF

["eugen yushin (JIRA)" <ji...@apache.org>]

eugen yushin created SPARK-25506:
------------------------------------

             Summary: Spark CSV multiline with CRLF
                 Key: SPARK-25506
                 URL: https://issues.apache.org/jira/browse/SPARK-25506
             Project: Spark
          Issue Type: Improvement
          Components: Input/Output
    Affects Versions: 2.3.1, 2.2.0
         Environment: spark 2.2.0 and 2.3.1
scala 2.11.8
            Reporter: eugen yushin


Spark produces empty rows (or ']' when printing via call to `collect`) dealing with '\r' character at the end of each line in CSV file. Note, no fields are escaped in original input file.
{code:java}
val multilineDf = sparkSession.read
  .format("csv")
  .options(Map("header" -> "true", "inferSchema" -> "false", "escape" -> "\"", "multiLine" -> "true"))
  .load("src/test/resources/multiLineHeader.csv")

val df = sparkSession.read
  .format("csv")
  .options(Map("header" -> "true", "inferSchema" -> "false", "escape" -> "\""))
  .load("src/test/resources/multiLineHeader.csv")

multilineDf.show()
multilineDf.collect().foreach(println)

df.show()
df.collect().foreach(println)
{code}
Result:
{code:java}
+----+-----+
|
+----+-----+
|
|
+----+-----+

]
]

+----+----+
|col1|col2|
+----+----+
|   1|   1|
|   2|   2|
+----+----+

[1,1]
[2,2]
{code}
Input file:
{code:java}
cat -vt src/test/resources/multiLineHeader.csv
col1,col2^M
1,1^M
2,2^M
{code}



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org