Github user hvanhovell commented on a diff in the pull request:

    https://github.com/apache/spark/pull/12537#discussion_r60756114
  
    --- Diff: 
sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/parser/TableIdentifierParserSuite.scala
 ---
    @@ -30,13 +64,22 @@ class TableIdentifierParserSuite extends SparkFunSuite {
         // Illegal names.
         intercept[ParseException](parseTableIdentifier(""))
         intercept[ParseException](parseTableIdentifier("d.q.g"))
    +  }
     
    +  test("table identifier - keywords") {
         // SQL Keywords.
    -    val keywords = Seq("select", "from", "where", "left", "right")
    +    val keywords = Seq("select", "from", "where") ++ 
hiveNonReservedRegression
         keywords.foreach { keyword =>
           intercept[ParseException](parseTableIdentifier(keyword))
           assert(TableIdentifier(keyword) === 
parseTableIdentifier(s"`$keyword`"))
           assert(TableIdentifier(keyword, Option("db")) === 
parseTableIdentifier(s"db.`$keyword`"))
         }
       }
    +
    +  test("table identifier - non reserved keywords") {
    +    // Hive keywords are allowed.
    +    hiveNonReservedKeyword.foreach { nonReserved =>
    +      assert(TableIdentifier(nonReserved) === 
parseTableIdentifier(nonReserved))
    +    }
    +  }
    --- End diff --
    
    @gatorsmile thanks for filing the PR.
    
    The current identifier parsing is probably as lenient as it gets. We really 
cannot add much more keywords without breaking the parsing process, for 
instance when `LEFT` is made nonReserved the parser will turn ```select * from 
a left join b on a.id = b.id``` into an `INNER` join because it interprets 
`LEFT` as the alias for table `a`. Union/Intersect/Except are also examples of 
this. I do like the idea of better error messages and I don't know if ANTLR has 
facilities for this. Lets do this in a different PR/JIRA if we want to 
investigate this.
    
    We want to support names that start with an underscore. This is mainly 
because of  a Datasets of tuples; column names will all start with an 
underscore. Hive disallows underscores because `CharSetName`s start with an 
underscore.
    



---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to