|
6 | 6 |
|
7 | 7 | #### Bug Fixes |
8 | 8 |
|
| 9 | +- Fixed a bug where `TRY_CAST` reader option is ignored when calling `DataFrameReader.schema().csv()`. |
| 10 | +- Fixed a bug where `event_table_telemetry` is using incorrect resource attributes. |
| 11 | +- Fixed a bug where `value_contains_null` was not propagated for `MapType` in `_as_nested` function. |
9 | 12 | - Fixed a bug where CTE optimization incorrectly deduplicated subtrees containing non-deterministic data generation functions (e.g. `uuid_string()`). |
10 | 13 | - Fixed a bug where vectorized UDFs using non-anaconda package repositories did not specify the pandas package by default. |
11 | 14 |
|
|
15 | 18 |
|
16 | 19 | - Updated the supported `pandas` versions to <=2.4 (was previously <=2.3.1). |
17 | 20 |
|
18 | | -## 1.49.0 (TBD) |
| 21 | +## 1.49.0 (2026-04-13) |
19 | 22 |
|
20 | 23 | ### Snowpark Python API Updates |
21 | 24 |
|
22 | 25 | #### New Features |
23 | 26 |
|
24 | | -- Allow user input schema when reading Parquet file on stage. |
25 | | - |
26 | | -#### Bug Fixes |
27 | | - |
28 | | -- Fixed a bug that `TRY_CAST` reader option is ignored when calling `DataFrameReader.schema().csv()`. |
| 27 | +- Allow a user-specified schema when reading Parquet files from a stage. |
29 | 28 |
|
30 | 29 | #### Improvements |
31 | 30 |
|
|
38 | 37 |
|
39 | 38 | - Fixed a bug where `Session.create_dataframe` raised `TypeError` when a `StringType` column was given a non-string Python value (e.g. `int`, `float`, `bool`, `Decimal`) for a small local relation (below the array bind threshold); `VALUES` SQL generation now coerces these types to string literals, consistent with the large-data bind-parameter path. |
40 | 39 | - Fixed a bug where `DataFrame.approxQuantile` did not accept a `Column` for the `col` parameter. |
41 | | -- Fixed a bug where `event_table_telemetry` is using incorrect resource attributes. |
42 | | -- Fixed a bug where `value_contains_null` was not propagated for `MapType` in `_as_nested` function. |
43 | 40 |
|
44 | 41 | ### Snowpark Local Testing Updates |
45 | 42 |
|
46 | 43 | #### Bug Fixes |
47 | 44 |
|
48 | 45 | - Fixed a bug where `concat` produced extra NaN rows and mismatched values after a `filter` operation in local testing. |
49 | | -- Fix a bug where `dense_rank()` would fail with `ValueError` on NULL partition values. |
50 | | -- Fixed a bug where `collect()` raised `KeyError` after `save_as_table(column_order="name")` when the source DataFrame omitted columns of `VariantType`, `MapType`, or `ArrayType` present in the target table schema in local testing. |
| 46 | +- Fixed a bug where `dense_rank()` would fail with `ValueError` on NULL partition values. |
| 47 | +- Fixed a bug where `collect()` raised `KeyError` after `save_as_table(column_order="name")` when the source DataFrame omitted columns with types `VariantType`, `MapType`, or `ArrayType` that were present in the target table schema in local testing. |
51 | 48 |
|
52 | 49 | ## 1.48.1 (2026-03-31) |
53 | 50 |
|
|
0 commit comments