You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Xiao Li (JIRA)" <ji...@apache.org> on 2018/07/17 20:48:00 UTC
[jira] [Updated] (SPARK-23723) New encoding option for json
datasource
[ https://issues.apache.org/jira/browse/SPARK-23723?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Xiao Li updated SPARK-23723:
----------------------------
Summary: New encoding option for json datasource (was: New charset option for json datasource)
> New encoding option for json datasource
> ---------------------------------------
>
> Key: SPARK-23723
> URL: https://issues.apache.org/jira/browse/SPARK-23723
> Project: Spark
> Issue Type: Sub-task
> Components: SQL
> Affects Versions: 2.4.0
> Reporter: Maxim Gekk
> Assignee: Maxim Gekk
> Priority: Major
> Fix For: 2.4.0
>
>
> Currently JSON Reader can read json files in different charset/encodings. The JSON Reader uses the jackson-json library to automatically detect the charset of input text/stream. Here you can see the method which detects encoding: [https://github.com/FasterXML/jackson-core/blob/master/src/main/java/com/fasterxml/jackson/core/json/ByteSourceJsonBootstrapper.java#L111-L174]
>
> The detectEncoding method checks the BOM ([https://en.wikipedia.org/wiki/Byte_order_mark]) at the beginning of a text. The BOM can be in the file but it is not mandatory. If it is not present, the auto detection mechanism can select wrong charset. And as a consequence of that, the user cannot read the json file. *The proposed option will allow to bypass the auto detection mechanism and set the charset explicitly.*
>
> The charset option is already exposed as a CSV option: [https://github.com/apache/spark/blob/master/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/csv/CSVOptions.scala#L87-L88] . I propose to add the same option for JSON.
>
> Regarding to JSON Writer, *the charset option will give to the user opportunity* to read json files in charset different from UTF-8, modify the dataset and *write results back to json files in the original encoding.* At the moment it is not possible to do because the result can be saved in UTF-8 only.
--
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