ORC files expect UTF-8, which is a superset of ascii, in strings, char, and varchar. The only place that I know that will cause trouble if you put non-utf-8 data in strings is the statistics. The API for getting the min/max will convert to Java strings.
But back to your original point, the schema evolution should easily handle the case of string to varchar. Which version of Hive are you using? .. Owen On Thu, Jul 18, 2019 at 8:27 AM Devopam Mittra <devo...@gmail.com> wrote: > The table has data in it perhaps that is beyond ASCII. > Easier way is to go for additional column , update with data and the drop > the older one after validation of records in String type col. > > Regards > Dev > > On Thu, Jul 18, 2019, 4:44 AM William Shen <wills...@marinsoftware.com> > wrote: > >> Hi all, >> >> I assumed that it should be compatible to convert column type varchar to >> string, however, after running ALTER TABLE table CHANGE col col STRING, I >> encounter the following error when querying the column from hive: >> >> Failed with exception >> java.io.IOException:org.apache.hadoop.hive.ql.metadata.HiveException: >> java.lang.ClassCastException: >> org.apache.hadoop.hive.serde2.io.HiveVarcharWritable cannot be cast to >> org.apache.hadoop.io.Text >> >> Anyone encountered this before, or know how to work around this? >> >> Thank you! >> >> - Will >> >
