CREATE TABLE COLUMN ENCODING clause with non-existent column passes.

[hlinnaka]

Actual behavior

postgres=# CREATE TABLE tab (
  i text,
  COLUMN non_existent ENCODING (compresstype=zlib)
)
with (appendonly=true, orientation=column);
NOTICE:  Table doesn't have 'DISTRIBUTED BY' clause -- Using column named 'i' as the Greenplum Database data distribution key for this table.
HINT:  The 'DISTRIBUTED BY' clause determines the distribution of data. Make sure column(s) chosen are the optimal data distribution key to minimize skew.
CREATE TABLE

Expected behavior

I expected an error, because there is no "non_existent" column. There is code specifically to check for that, in the validateColumnStorageEncodingClauses() function:

/*
 * Validate the sanity of column reference storage clauses.
 *
 * 1. Ensure that we only refer to columns that exist.
 * 2. Ensure that each column is referenced either zero times or once.
 */
static void
validateColumnStorageEncodingClauses(List *stenc, List *columns)
{
	...
	/*
	 * All column reference storage directives without the DEFAULT
	 * clause should refer to real columns.
	 */
	foreach(lc, stenc)
	{
		...

			if (!found)
				ereport(ERROR,
						(errcode(ERRCODE_WRONG_OBJECT_TYPE),
						 errmsg("column \"%s\" does not exist", colname)));

But that code is not being called correctly.

[hlinnaka]

The other check in validateColumnStorageEncodingClauses(), for duplicated COLUMN ENCODING directives, is also broken:

postgres=# CREATE TABLE dupetab (
  dupe text,
  column dupe encoding (compresstype=zlib),
  column dupe encoding (compresstype=none)
)
with (appendonly=true, orientation=column);
NOTICE:  Table doesn't have 'DISTRIBUTED BY' clause -- Using column named 'dupe' as the Greenplum Database data distribution key for this table.
HINT:  The 'DISTRIBUTED BY' clause determines the distribution of data. Make sure column(s) chosen are the optimal data distribution key to minimize skew.
CREATE TABLE