Centralize monitoring_location_id check in _get_args; trim narration

/simplify findings:

1. Move the per-function `monitoring_location_id = _check_monitoring_location_id(...)`
   into `_get_args` itself. Eliminates 13 copy-paste call sites in api.py and
   closes the bug class Copilot found twice (four functions had been missed
   when added piecemeal). New `get_*` functions inherit validation automatically.

2. Drop `_LIST_ONLY_STR_PARAMS` (a one-element `frozenset`). Inline the
   `properties` special case with a literal `elif k == "properties":` plus
   a one-line WHY comment.

3. Compress the three-paragraph narration in `_construct_api_requests` to a
   single line explaining the only non-obvious bit (`len()` instead of
   truthiness, because of numpy ndarray).

4. Add `begin_utc`/`end_utc` to `_DATE_RANGE_PARAMS`. `get_time_series_metadata`
   exposes both as range filters but the constant was missing them, so a
   two-element list would have been treated as a multi-value POST/CQL2 filter
   instead of being formatted as an ISO-8601 interval.

5. Drop the now-unused `_check_monitoring_location_id` import from api.py.

Author: thodson-usgs

dataretrieval/waterdata/api.py

@@ -27,7 +27,6 @@
 )
 from dataretrieval.waterdata.utils import (
     SAMPLES_URL,
-    _check_monitoring_location_id,
     _check_profiles,
     _default_headers,
     _get_args,
@@ -232,7 +231,6 @@ def get_daily(
         ...     last_modified="P7D",
         ... )
     """
-    monitoring_location_id = _check_monitoring_location_id(monitoring_location_id)
     service = "daily"
     output_id = "daily_id"
 
@@ -421,7 +419,6 @@ def get_continuous(
         ...     filter_lang="cql-text",
         ... )
     """
-    monitoring_location_id = _check_monitoring_location_id(monitoring_location_id)
     service = "continuous"
     output_id = "continuous_id"
 
@@ -720,7 +717,6 @@ def get_monitoring_locations(
         ...     properties=["monitoring_location_id", "state_name", "country_name"],
         ... )
     """
-    monitoring_location_id = _check_monitoring_location_id(monitoring_location_id)
     service = "monitoring-locations"
     output_id = "monitoring_location_id"
 
@@ -944,7 +940,6 @@ def get_time_series_metadata(
         ...     begin="1990-01-01/..",
         ... )
     """
-    monitoring_location_id = _check_monitoring_location_id(monitoring_location_id)
     service = "time-series-metadata"
     output_id = "time_series_id"
 
@@ -1181,7 +1176,6 @@ def get_combined_metadata(
     service = "combined-metadata"
     output_id = "combined_meta_id"
 
-    monitoring_location_id = _check_monitoring_location_id(monitoring_location_id)
     args = _get_args(locals())
 
     return get_ogc_data(args, output_id, service)
@@ -1373,7 +1367,6 @@ def get_latest_continuous(
         ...     monitoring_location_id=["USGS-05114000", "USGS-09423350"]
         ... )
     """
-    monitoring_location_id = _check_monitoring_location_id(monitoring_location_id)
     service = "latest-continuous"
     output_id = "latest_continuous_id"
 
@@ -1570,7 +1563,6 @@ def get_latest_daily(
         ...     monitoring_location_id=["USGS-05114000", "USGS-09423350"]
         ... )
     """
-    monitoring_location_id = _check_monitoring_location_id(monitoring_location_id)
     service = "latest-daily"
     output_id = "latest_daily_id"
 
@@ -1761,7 +1753,6 @@ def get_field_measurements(
         ...     time="P20Y",
         ... )
     """
-    monitoring_location_id = _check_monitoring_location_id(monitoring_location_id)
     service = "field-measurements"
     output_id = "field_measurement_id"
 
@@ -1883,7 +1874,6 @@ def get_field_measurements_metadata(
     service = "field-measurements-metadata"
     output_id = "field_series_id"
 
-    monitoring_location_id = _check_monitoring_location_id(monitoring_location_id)
     args = _get_args(locals())
 
     return get_ogc_data(args, output_id, service)
@@ -2004,7 +1994,6 @@ def get_peaks(
     service = "peaks"
     output_id = "peak_id"
 
-    monitoring_location_id = _check_monitoring_location_id(monitoring_location_id)
     args = _get_args(locals())
 
     return get_ogc_data(args, output_id, service)
@@ -2536,7 +2525,6 @@ def get_stats_por(
         ... )
     """
     # Build argument dictionary, omitting None values
-    monitoring_location_id = _check_monitoring_location_id(monitoring_location_id)
     params = _get_args(locals(), exclude={"expand_percentiles"})
 
     return get_stats_data(
@@ -2666,7 +2654,6 @@ def get_stats_date_range(
         ... )
     """
     # Build argument dictionary, omitting None values
-    monitoring_location_id = _check_monitoring_location_id(monitoring_location_id)
     params = _get_args(locals(), exclude={"expand_percentiles"})
 
     return get_stats_data(
@@ -2839,7 +2826,6 @@ def get_channel(
         ...     monitoring_location_id="USGS-02238500",
         ... )
     """
-    monitoring_location_id = _check_monitoring_location_id(monitoring_location_id)
     service = "channel-measurements"
     output_id = "channel_measurements_id"
 

dataretrieval/waterdata/utils.py

@@ -149,7 +149,9 @@ def _switch_properties_id(properties: list[str] | None, id_name: str, service: s
 # string list. Used by ``_construct_api_requests`` to keep them out of the
 # POST/CQL2 multi-value path and to route them through ``_format_api_dates``,
 # and by ``_NO_NORMALIZE_PARAMS`` to bypass string-iterable normalization.
-_DATE_RANGE_PARAMS = frozenset({"datetime", "last_modified", "begin", "end", "time"})
+_DATE_RANGE_PARAMS = frozenset(
+    {"datetime", "last_modified", "begin", "begin_utc", "end", "end_utc", "time"}
+)
 
 
 def _parse_datetime(value: str) -> datetime | None:
@@ -468,11 +470,7 @@ def _construct_api_requests(
             dates = service == "daily" and i != "last_modified"
             params[i] = _format_api_dates(params[i], date=dates)
 
-    # Join bbox/properties into the comma-separated form the OGC API expects.
-    # For ``bbox`` use ``len() > 0`` so ``numpy.ndarray`` inputs don't trip
-    # the ambiguous truth-value error; for ``properties`` the truthy check is
-    # right because ``_get_args`` always materializes it to a list (and
-    # ``_switch_properties_id`` further upstream returns ``[]`` for None).
+    # `len()` instead of truthiness: a numpy ndarray would raise on `if bbox:`.
     if bbox is not None and len(bbox) > 0:
         params["bbox"] = ",".join(map(str, bbox))
     if properties:
@@ -1203,12 +1201,6 @@ def _check_profiles(
     "thresholds",
 }
 
-# Param names that must be a list of strings (never a single string).
-# A single string passed in would iterate as characters in
-# ``_construct_api_requests``'s ``",".join(...)`` step, producing a
-# malformed URL. ``_get_args`` wraps single-string input into a list.
-_LIST_ONLY_STR_PARAMS = frozenset({"properties"})
-
 
 def _normalize_str_iterable(
     value: str | Iterable[str] | None,
@@ -1344,9 +1336,10 @@ def _get_args(
     for k, v in local_vars.items():
         if k in to_exclude or v is None:
             continue
-        if k in _LIST_ONLY_STR_PARAMS:
-            # Wrap a single string so the downstream `",".join(...)` doesn't
-            # iterate it as characters.
+        if k == "monitoring_location_id":
+            args[k] = _check_monitoring_location_id(v)
+        elif k == "properties":
+            # `",".join(properties)` would iterate a bare string as characters.
             args[k] = [v] if isinstance(v, str) else _normalize_str_iterable(v, k)
         elif (
             k in _NO_NORMALIZE_PARAMS