Merge pull request #71 from opera-adt/dev

v1.0.11

Author: cmarshak

CHANGELOG.md

@@ -6,6 +6,15 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [PEP 440](https://www.python.org/dev/peps/pep-0440/)
 and uses [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [1.0.11] - 2026-01-27
+
+### Changed
+* Defaults for max_pre_img per burst set to defaults (4, 3, 3) in enumeration of many products and single product. This aligns with standard inputs for DIST-S1.
+
+### Fixed
+* Casting time strings into timestamps using pandas 2.0+ was attaching precision that was breaking schema validation with different precision (I believe pandas default is nano-second precision and ASF distributes with microsecond precision). Using `coerce=True` for the schema (forcing it to cast into the correct precision) and forcing data to have UTC time-zones fixes this issue with validation.
+
+
 ## [1.0.10] - 2026-01-07
 
 ### Added

README.md

@@ -76,7 +76,8 @@ Yields:
 <details>
   <summary>Output</summary>
 
-```[{'mgrs_tile_id': '19HBD',
+```
+[{'mgrs_tile_id': '19HBD',
   'post_acq_date': '2023-11-05',
   'track_number': 91,
   'post_acq_timestamp': '2023-11-05 23:36:49+00:00'},

src/dist_s1_enumerator/asf.py

@@ -135,7 +135,7 @@ def get_rtc_s1_ts_metadata_by_burst_ids(
     properties_f = [
         {
             'opera_id': p['sceneName'],
-            'acq_dt': pd.to_datetime(p['startTime']),
+            'acq_dt': pd.Timestamp(p['startTime'], tz='UTC'),
             'track_number': p['pathNumber'],
             'polarizations': p['polarization'],
             'all_urls': [p['url']] + p['additionalUrls'],

src/dist_s1_enumerator/dist_enum.py

@@ -231,7 +231,7 @@ def enumerate_dist_s1_products(
     df_rtc_ts: gpd.GeoDataFrame,
     mgrs_tile_ids: list[str],
     lookback_strategy: str = 'multi_window',
-    max_pre_imgs_per_burst: int = (5, 5, 5),
+    max_pre_imgs_per_burst: int = (4, 3, 3),
     min_pre_imgs_per_burst: int = 1,
     tqdm_enabled: bool = True,
     delta_lookback_days: int = 365,

src/dist_s1_enumerator/dist_enum_inputs.py

@@ -32,7 +32,7 @@ def enumerate_dist_s1_workflow_inputs(
     start_acq_dt: datetime | pd.Timestamp | str | None = None,
     stop_acq_dt: datetime | pd.Timestamp | str | None = None,
     lookback_strategy: str = 'multi_window',
-    max_pre_imgs_per_burst: int | list[int] | tuple[int, ...] = (5, 5, 5),
+    max_pre_imgs_per_burst: int | list[int] | tuple[int, ...] = (4, 3, 3),
     min_pre_imgs_per_burst: int = 1,
     delta_lookback_days: int | list[int] | tuple[int, ...] = 365,
     delta_window_days: int = 365,
@@ -60,7 +60,7 @@ def enumerate_dist_s1_workflow_inputs(
         Lookback strategy to use, by default 'multi_window'. Options are
         'immediate_lookback' or 'multi_window'.
     max_pre_imgs_per_burst : int | list[int] | tuple[int, ...], optional
-        Maximum number of pre-images per burst to include, by default 10.
+        Maximum number of pre-images per burst to include, by default (4, 3, 3).
         If lookback strategy is 'multi_window':
             - this is interpreted as the maximum number of pre-images on each anniversary date.
             - tuple/list of integers are provided, each int represents the maximum number of pre-images on each

src/dist_s1_enumerator/tabular_models.py

@@ -27,7 +27,7 @@
     {
         'opera_id': Column(str, required=True),
         'jpl_burst_id': Column(str, required=True),
-        'acq_dt': Column(DateTime(tz='UTC'), required=True),
+        'acq_dt': Column(DateTime(tz='UTC'), coerce=True, required=True),
         'acq_date_for_mgrs_pass': Column(str, required=False),
         'polarizations': Column(str, required=True),
         'track_number': Column(int, required=True),