hydroserverpy 0.5.0b3__tar.gz → 0.5.0b5__tar.gz

{hydroserverpy-0.5.0b3/src/hydroserverpy.egg-info → hydroserverpy-0.5.0b5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: hydroserverpy
-Version: 0.5.0b3
+Version: 0.5.0b5
 Requires-Python: <4,>=3.9
 License-File: LICENSE
 Requires-Dist: requests>=2

{hydroserverpy-0.5.0b3 → hydroserverpy-0.5.0b5}/setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name = hydroserverpy
-version = 0.5.0b3
+version = 0.5.0b5
 
 [options]
 package_dir = 

{hydroserverpy-0.5.0b3 → hydroserverpy-0.5.0b5}/src/hydroserverpy/api/models/etl/data_source.py

@@ -1,8 +1,8 @@
+import requests
 import tempfile
 from typing import Union, List, Optional, TYPE_CHECKING
 from uuid import UUID
 from pydantic import BaseModel, Field
-from urllib.request import urlopen
 from hydroserverpy.etl_csv.hydroserver_etl_csv import HydroServerETLCSV
 from .orchestration_system import OrchestrationSystem
 from .orchestration_configuration import OrchestrationConfigurationFields
@@ -135,14 +135,17 @@ class DataSource(HydroServerModel, DataSourceFields, OrchestrationConfigurationF
                 )
                 loader.run()
         elif self.settings["extractor"]["type"] == "HTTP":
-            with tempfile.NamedTemporaryFile(mode="w") as temp_file:
-                with urlopen(self.settings["extractor"]["urlTemplate"]) as response:
-                    chunk_size = 1024 * 1024 * 10  # Use a 10mb chunk size.
-                    while True:
-                        chunk = response.read(chunk_size)
-                        if not chunk:
-                            break
-                        temp_file.write(chunk)
+            with tempfile.NamedTemporaryFile(mode="w+") as temp_file:
+                response = requests.get(
+                    self.settings["extractor"]["urlTemplate"],
+                    stream=True,
+                    timeout=60,
+                )
+                response.raise_for_status()
+                chunk_size = 1024 * 1024 * 10  # Use a 10mb chunk size.
+                for chunk in response.iter_content(chunk_size=chunk_size):
+                    if chunk:
+                        temp_file.write(chunk.decode("utf-8"))
                 temp_file.seek(0)
                 loader = HydroServerETLCSV(
                     self._connection, data_file=temp_file, data_source=self

{hydroserverpy-0.5.0b3 → hydroserverpy-0.5.0b5}/src/hydroserverpy/api/models/sta/datastream.py

@@ -141,20 +141,6 @@ class Datastream(HydroServerModel, DatastreamFields):
             _connection=_connection, _model_ref="datastreams", _uid=_uid, **data
         )
 
-        self._workspace_id = str(
-            data.get("workspace_id")
-            or data.get("workspaceId")
-            or data["properties"]["workspace"]["id"]
-        )
-        self._processing_level_id = str(
-            data.get("workspace_id")
-            or data.get("workspaceId")
-            or data["properties"]["processingLevelId"]
-        )
-        self._unit_id = str(
-            data.get("unit_id") or data.get("unitId") or data["properties"]["unitId"]
-        )
-
         self._workspace = None
         self._thing = None
         self._observed_property = None
@@ -167,7 +153,8 @@ class Datastream(HydroServerModel, DatastreamFields):
         """The workspace this datastream belongs to."""
 
         if self._workspace is None:
-            self._workspace = self._connection.workspaces.get(uid=self._workspace_id)
+            datastream = self._connection.request("get", f"/api/data/datastreams/{str(self.uid)}").json()
+            self._workspace = self._connection.workspaces.get(uid=datastream["workspaceId"])
 
         return self._workspace
 
@@ -246,7 +233,8 @@ class Datastream(HydroServerModel, DatastreamFields):
         """The unit this datastream uses."""
 
         if self._unit is None:
-            self._unit = self._connection.units.get(uid=self._unit_id)
+            datastream = self._connection.request("get", f"/api/data/datastreams/{str(self.uid)}").json()
+            self._unit = self._connection.units.get(uid=datastream["unitId"])
             self._original_data["unit"] = self._unit
 
         return self._unit
@@ -263,9 +251,8 @@ class Datastream(HydroServerModel, DatastreamFields):
         """The processing level of this datastream."""
 
         if self._processing_level is None:
-            self._processing_level = self._connection.processinglevels.get(
-                uid=self._processing_level_id
-            )
+            datastream = self._connection.request("get", f"/api/data/datastreams/{str(self.uid)}").json()
+            self._processing_level = self._connection.processinglevels.get(uid=datastream["processingLevelId"])
             self._original_data["processing_level"] = self._processing_level
 
         return self._processing_level
@@ -285,13 +272,10 @@ class Datastream(HydroServerModel, DatastreamFields):
         """Refresh this datastream from HydroServer."""
 
         self._workspace = None
-        self._workspace_id = None
         self._thing = None
         self._observed_property = None
         self._unit = None
-        self._unit_id = None
         self._processing_level = None
-        self._processing_level_id = None
         self._sensor = None
         super()._refresh()
 
@@ -336,3 +320,23 @@ class Datastream(HydroServerModel, DatastreamFields):
             uid=self.uid,
             observations=observations,
         )
+
+    # TODO: Find a better long-term solution for this issue.
+    def sync_phenomenon_end_time(self):
+        """Ensures the phenomenon_end_time field matches the actual end time of the observations."""
+
+        response = self._connection.request(
+            "get", f"/api/data/datastreams/{str(self.uid)}/observations",
+            params={
+                "order": "desc",
+                "page": 1,
+                "page_size": 1
+            }
+        ).json()
+
+        if len(response["phenomenon_time"]) > 0:
+            self.phenomenon_end_time = datetime.fromisoformat(response["phenomenon_time"][0])
+        else:
+            self.phenomenon_end_time = None
+
+        self.save()

{hydroserverpy-0.5.0b3 → hydroserverpy-0.5.0b5}/src/hydroserverpy/etl_csv/hydroserver_etl_csv.py

@@ -1,4 +1,5 @@
 import csv
+import math
 import logging
 import croniter
 import pandas as pd
@@ -51,7 +52,7 @@ class HydroServerETLCSV:
         self._file_header_error = False
         self._file_timestamp_error = False
 
-        self._chunk_size = 10000
+        self._chunk_size = 1000
         self._observations = {}
 
     def run(self):
@@ -124,6 +125,9 @@ class HydroServerETLCSV:
         timestamp = self._parse_row_timestamp(row)
 
         for datastream in self._datastreams.values():
+            if index == self._data_source.settings["transformer"]["dataStartRow"]:
+                datastream.sync_phenomenon_end_time()
+
             if str(datastream.uid) not in self._datastream_start_row_indexes.keys():
                 if (
                     not datastream.phenomenon_end_time
@@ -138,14 +142,27 @@ class HydroServerETLCSV:
                 if str(datastream.uid) not in self._observations.keys():
                     self._observations[str(datastream.uid)] = []
 
+                raw_result = row[
+                    self._datastream_column_indexes[
+                        self._datastream_mapping[str(datastream.uid)]
+                    ]
+                ]
+
+                if isinstance(raw_result, (int, float)):
+                    result = raw_result
+                else:
+                    try:
+                        result = float(raw_result)
+                    except (TypeError, ValueError):
+                        result = datastream.no_data_value
+
+                if math.isnan(result):
+                    result = datastream.no_data_value
+
                 self._observations[str(datastream.uid)].append(
                     {
                         "phenomenon_time": timestamp,
-                        "result": row[
-                            self._datastream_column_indexes[
-                                self._datastream_mapping[str(datastream.uid)]
-                            ]
-                        ],
+                        "result": result,
                     }
                 )
 
@@ -282,8 +299,9 @@ class HydroServerETLCSV:
                         uid=datastream_id,
                         observations=observations_df,
                     )
-                except HTTPError:
+                except HTTPError as e:
                     failed_datastreams.append(datastream_id)
+                    logger.error(f"Failed to POST observations to datastream: {str(datastream_id)} - {e}")
 
             elif datastream_id in self._failed_datastreams:
                 logger.info(
@@ -308,13 +326,13 @@ class HydroServerETLCSV:
 
         if self._data_source.crontab is not None:
             next_run = croniter.croniter(
-                self._data_source.crontab, datetime.now()
+                self._data_source.crontab, datetime.now(timezone.utc)
             ).get_next(datetime)
         elif (
             self._data_source.interval is not None
             and self._data_source.interval_units is not None
         ):
-            next_run = datetime.now() + timedelta(
+            next_run = datetime.now(timezone.utc) + timedelta(
                 **{self._data_source.interval_units: self._data_source.interval}
             )
         else:

{hydroserverpy-0.5.0b3 → hydroserverpy-0.5.0b5/src/hydroserverpy.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: hydroserverpy
-Version: 0.5.0b3
+Version: 0.5.0b5
 Requires-Python: <4,>=3.9
 License-File: LICENSE
 Requires-Dist: requests>=2