It seems that Vanilla Apache Hive 0.13 sorts in a yet another different way than what we expect and also differently from Cloudera Imapla (another Hive flavor). SELECT BQT1.SmallA.StringNum FROM BQT1.SmallA order by BQT1.SmallA.StringNum 1: -1 2: -10 3: -11 4: -12 5: -13 6: -14 7: -15 8: -16 9: -17 10: -18 11: -19 12: -2 13: -20 14: -21 15: -22 16: -24 17: -3 18: -4 19: -5 20: -6 21: -8 22: -9 23: 0 24: 1 25: 10 26: 11 27: 12 28: 13 29: 14 30: 15 31: 16 32: 17 33: 18 34: 19 35: 2 36: 20 37: 21 38: 22 39: 23 40: 24 41: 3 42: 4 43: 5 44: 6 45: 7 46: 8 47: null 48: null 49: null 50: null
Upstream feature was added to help with sorting options: https://issues.jboss.org/browse/TEIID-3156 Since we are moving more into European / Asian territories where different locales will occur, this is something that probably should back port, right?
Yes, it should be back ported. It is better to use TEIID-3156 than disable the sorting completely. Otherwise it would be in some rollup patch probably so it is better to back port it now.
Steven Hawkins <shawkins> updated the status of jira TEIID-3142 to Resolved
Steven Hawkins <shawkins> updated the status of jira TEIID-3142 to Closed
Steven Hawkins <shawkins> updated the status of jira TEIID-3156 to Closed