How does PIG 0.7.0 handle schema of data? Sometimes, 1) there is a missing
column in the input data to be loaded and the total number of columns to be
read is smaller than that specified in the schema and 2) particularly sometimes
there is a missing column but however the total # of columns to be read is
greater than that specified in the schema, which makes it hard for a load
function to detect this missing column actually.
As I know, there are some discussion on this in jira, s.a. PIG-1274 and
PIG-1188. So, how the schema is respected in both cases?
Thanks,
-Michael
