0

This is the table I am trying to grab. From my HANA Database however I keep getting the below error. I know the table exists as I have pulled it in from Qlikview.

From Qlikview I pull in the table using the below syntax

"_MY_SCHEMA_"."My.Table.Name/Table_ONE"

When I try to do the same thing in R using I use the following to grab the same table 

mydate <- sqlFetch(myconn, "_MY_SCHEMA.My.Table.Name/Table_ONE")

This is the error that I get

Error in odbcTableExists(channel, sqtable)
: table not found on channel
mannym
  • 325
  • 1
  • 16
  • There are quotes around the period in the Qlikview syntax, but not when using `sqlFetch`. You may need to add escaped quotes or try a different table that a simple name to confirm the connection. – manotheshark Jan 26 '17 at 14:29

1 Answers1

1

This problem is caused by the way STRINGS are encapsulated in quotes. In this scenario this happens in two levels:

  1. R command parameter (sqlFetch ( connection-object-parameter, sql-string-parameter)
  2. HANA SQL where the literal string for the table name is required

You used the double-quotes encapsulation for both steps. In each of those two steps the strings get extracted from the parameter you used, which includes the removal of the outermost encapsulation markers which are the double-quotes here. So, after step 1 the double quotes are gone and HANA receives a string without them.
This is OK, as long as you don't actually require the double-quotes for your identifiers, but in your example you actually do. (case sensitive identifiers with special characters like / or . require the double quotes).

The solution to this is rather simple: R allows both single (') and double (") quotes to be used for string encapsulation.
Just use single quotation marks in R to hand over the string:

mydate <- sqlFetch(myconn, '_MY_SCHEMA."My.Table.Name/Table_ONE"')

Notice: you still have to encapsulate "My.Table.Name/Table_ONE" in double quotation marks so that HANA will handle this identifier correctly.

For the connection R to HANA there are also a couple of blogs out:

Lars Br.
  • 9,949
  • 2
  • 15
  • 29