[GitHub] [arrow] jorgecarleitao commented on a change in pull request #7796: ARROW-9516: [Rust][DataFusion] refactor of column names

[GitBox <gi...@apache.org>]

jorgecarleitao commented on a change in pull request #7796:
URL: https://github.com/apache/arrow/pull/7796#discussion_r456757621



##########
File path: rust/datafusion/src/execution/context.rs
##########
@@ -995,16 +1034,16 @@ mod tests {
         let ctx = create_ctx(&tmp_dir, 1)?;
 
         let schema = Arc::new(Schema::new(vec![
-            Field::new("state", DataType::Utf8, false),
-            Field::new("salary", DataType::UInt32, false),
+            Field::new("c1", DataType::Utf8, false),
+            Field::new("c2", DataType::UInt32, false),
         ]));
 
         let plan = LogicalPlanBuilder::scan("default", "test", schema.as_ref(), None)?
             .aggregate(
-                vec![col("state")],

Review comment:
       This change was necessary because the file's schema is `c1`, `c2`, not `state, salary`. We were able to get away with this because, since the names did not mean anything, we could read a file with a given schema using another schema's field names, as long as the indexing was correct. I believe that this should not be possible by design, as it introduces situations that are non-trivial to debug.




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org