[jira] [Resolved] (SPARK-21355) JSON datasource failed to parse a value having newline in its value

["Hyukjin Kwon (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/SPARK-21355?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Hyukjin Kwon resolved SPARK-21355.
----------------------------------
    Resolution: Invalid

I am resolving this per https://stackoverflow.com/a/42073.

{quote}
This is of course correct, but I'd like to add the reason for having to do this: the JSON spec at ietf.org/rfc/rfc4627.txt contains this sentence in section 2.5: "All Unicode characters may be placed within the quotation marks except for the characters that must be escaped: quotation mark, reverse solidus, and the control characters (U+0000 through U+001F)." Since a newline is a control character, it must be escaped.
{quote}

> JSON datasource failed to parse a value having newline in its value
> -------------------------------------------------------------------
>
>                 Key: SPARK-21355
>                 URL: https://issues.apache.org/jira/browse/SPARK-21355
>             Project: Spark
>          Issue Type: Bug
>          Components: SQL
>    Affects Versions: 2.2.0
>            Reporter: Hyukjin Kwon
>            Priority: Minor
>
> I guess this is a rather corner case. I found this during testing SPARK-21289.
> It looks a bug in Jackson.
> The codes below failed to parse newline in the value.
> {code}
> scala> spark.read.json(Seq("{ \"f\": \"a\nb\"}", "{ \"f\": \"abc\"}").toDS).show()
> +---------------+----+
> |_corrupt_record|   f|
> +---------------+----+
> |  { "f": "a
> b"}|null|
> |           null| abc|
> +---------------+----+
> {code}
> Meaning this also does not work
> with the JSON files as below:
> {code}
> {"f": "
> d",  "f0": 3}
> {code}
> {code}
> scala> spark.read.option("multiLine", true).json("tmp.json").show()
> +--------------------+
> |     _corrupt_record|
> +--------------------+
> |{"f": "
> d",  "f0"...|
> +--------------------+
> {code}
> Of course, the codes below work:
> {code}
> scala> spark.read.json(Seq("{ \"f\": \"ab\"}", "{ \"f\": \"abc\"}").toDS).show()
> +---+
> |  f|
> +---+
> | ab|
> |abc|
> +---+
> {code}



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

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