Add note on programmatically creating lambdas when lazily saving a PartionedDataset (#5158)

* Add note advising caution when using loops to create lambdas

Signed-off-by: chrisschopp <[email protected]>

* Add description of change to RELEASE.md

Signed-off-by: chrisschopp <[email protected]>

---------

Signed-off-by: chrisschopp <[email protected]>
Co-authored-by: Ankita Katiyar <[email protected]>

Author: chrisschopp

RELEASE.md

@@ -5,9 +5,13 @@
 
 ## Bug fixes and other changes
 
+## Documentation changes
+* Added a note on programmatically creating lambdas when lazily saving a `PartionedDataset`.
+
 ## Community contributions
 Many thanks to the following Kedroids for contributing PRs to this release:
 * [Aseem Sangalay](https://github.com/aseemsangalay)
+* [Chris Schopp](https://github.com/chrisschopp)
 
 # Release 1.0.0
 ## Major features and improvements

docs/catalog-data/partitioned_and_incremental_datasets.md

@@ -252,6 +252,9 @@ new_partitioned_dataset:
   save_lazily: False
 ```
 
+!!! note
+    If creating lambdas in a list/dictionary comprehension or a for loop, be cautious when referencing variables defined outside the scope of the lambdas. See the [Python Programming FAQ](https://docs.python.org/3/faq/programming.html#why-do-lambdas-defined-in-a-loop-with-different-values-all-return-the-same-result) for an explanation of how this can result in unexpected values being returned from the lambdas.
+
 ## Incremental datasets
 
 [IncrementalDataset](https://docs.kedro.org/projects/kedro-datasets/en/feature-8.0/api/kedro_datasets/partitions.IncrementalDataset/) is a subclass of `PartitionedDataset`, which stores the information about the last processed partition in the so-called `checkpoint`. `IncrementalDataset` addresses the use case when partitions have to be processed incrementally, that is, each subsequent pipeline run should process just the partitions which were not processed by the previous runs.