[SPARK-9505] DataFrames : Mysql JDBC not support column names with special characters

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Duplicate
• Affects Version/s: 1.3.0
• Component/s: Spark Core, SQL
• Assignee: Unassigned
• Reporter: Pangjiu
• Created: 31/Jul/15 09:46
• Updated: 28/Dec/15 12:51
• Resolved: 28/Dec/15 12:51

Description

HI all,

I had above issue on connect to mySQL database through SQLContext. If the mySQL table's column name contains special characters like #[ ] %, it throw exception : "You have an error in your SQL syntax".

Below is coding:

Class.forName("com.mysql.jdbc.Driver").newInstance()

val url = "jdbc:mysql://localhost:3306/sakila?user=root&password=xxx"
val driver = "com.mysql.jdbc.Driver"

val sqlContext = new SQLContext(sc)

val output = { sqlContext.load("jdbc", Map
(
"url" -> url,
"driver" -> driver,
"dbtable" -> "(SELECT `ID`, `NAME%` FROM `agent`) AS tableA "
)
)
}

Hope dataframes for sqlContext can support for special characters very soon as this become a stopper now.

Thanks

Issue Links

• duplicates: SPARK-8616 SQLContext doesn't handle tricky column names when loading from JDBC
• duplicates: SPARK-12437 Reserved words (like table) throws error when writing a data frame to JDBC
• relates to: SPARK-12437 Reserved words (like table) throws error when writing a data frame to JDBC