ecopipeline 1.0.0__tar.gz → 1.0.2__tar.gz

{ecopipeline-1.0.0/src/ecopipeline.egg-info → ecopipeline-1.0.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ecopipeline
-Version: 1.0.0
+Version: 1.0.2
 Summary: Contains functions for use in Ecotope Datapipelines
 Classifier: Programming Language :: Python :: 3
 Classifier: License :: OSI Approved :: GNU General Public License (GPL)

{ecopipeline-1.0.0 → ecopipeline-1.0.2}/setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name = ecopipeline
-version = 1.0.0
+version = 1.0.2
 authors = ["Carlos Bello, <bellocarlos@seattleu.edu>, Emil Fahrig <fahrigemil@seattleu.edu>, Casey Mang <cmang@seattleu.edu>, Julian Harris <harrisjulian@seattleu.edu>, Roger Tram <rtram@seattleu.edu>, Nolan Price <nolan@ecotope.com>"]
 description = Contains functions for use in Ecotope Datapipelines
 long_description = file: README.md

{ecopipeline-1.0.0 → ecopipeline-1.0.2}/src/ecopipeline/extract/__init__.py

@@ -1,3 +1,3 @@
-from .extract import get_noaa_data, json_to_df, extract_files, get_last_full_day_from_db, get_db_row_from_time, extract_new, csv_to_df, get_sub_dirs, msa_to_df, fm_api_to_df, small_planet_control_to_df, dent_csv_to_df, flow_csv_to_df, pull_egauge_data, egauge_csv_to_df, remove_char_sequence_from_csv_header, tb_api_to_df
+from .extract import get_noaa_data, json_to_df, extract_files, get_last_full_day_from_db, get_db_row_from_time, extract_new, csv_to_df, get_sub_dirs, msa_to_df, fm_api_to_df, small_planet_control_to_df, dent_csv_to_df, flow_csv_to_df, pull_egauge_data, egauge_csv_to_df, remove_char_sequence_from_csv_header, tb_api_to_df, skycentrics_api_to_df
 __all__ = ["get_noaa_data", "json_to_df", "extract_files", "get_last_full_day_from_db", "get_db_row_from_time", 'extract_new', "csv_to_df", "get_sub_dirs", "msa_to_df", "fm_api_to_df",
-           "small_planet_control_to_df","dent_csv_to_df","flow_csv_to_df","pull_egauge_data", "egauge_csv_to_df","remove_char_sequence_from_csv_header", "tb_api_to_df"]
+           "small_planet_control_to_df","dent_csv_to_df","flow_csv_to_df","pull_egauge_data", "egauge_csv_to_df","remove_char_sequence_from_csv_header", "tb_api_to_df", "skycentrics_api_to_df"]

{ecopipeline-1.0.0 → ecopipeline-1.0.2}/src/ecopipeline/extract/extract.py

@@ -15,6 +15,7 @@ import mysql.connector.errors as mysqlerrors
 import requests
 import subprocess
 import traceback
+import time
 
 
 def get_last_full_day_from_db(config : ConfigManager, table_identifier : str = "minute") -> datetime:
@@ -661,6 +662,79 @@ def egauge_csv_to_df(csv_filenames: List[str]) -> pd.DataFrame:
 
     return df_diff
 
+def skycentrics_api_to_df(config: ConfigManager, startTime: datetime = None, endTime: datetime = None, create_csv : bool = True, time_zone: str = 'US/Pacific'):
+    """
+    Function connects to the field manager api to pull data and returns a dataframe.
+
+    Parameters
+    ----------  
+    config : ecopipeline.ConfigManager
+        The ConfigManager object that holds configuration data for the pipeline. The config manager
+        must contain information to connect to the api, i.e. the api user name and password as well as
+        the device id for the device the data is being pulled from.
+    startTime: datetime
+        The point in time for which we want to start the data extraction from. This 
+        is local time from the data's index. 
+    endTime: datetime
+        The point in time for which we want to end the data extraction. This 
+        is local time from the data's index. 
+    create_csv : bool
+        create csv files as you process such that API need not be relied upon for reprocessing
+    time_zone: str
+        The timezone for the indexes in the output dataframe as a string. Must be a string recognized as a 
+        time stamp by the pandas tz_localize() function https://pandas.pydata.org/docs/reference/api/pandas.Series.tz_localize.html
+        defaults to 'US/Pacific'
+    
+    Returns
+    ------- 
+    pd.DataFrame: 
+        Pandas Dataframe containing data from the API pull with column headers the same as the variable names in the data from the pull
+    """
+    #temporary solution while no date range available
+    
+    try:
+        df = pd.DataFrame()
+        temp_dfs = []
+        time_parser = startTime
+        while time_parser < endTime:
+            start_time_str = time_parser.strftime('%a, %d %b %H:%M:%S GMT')
+            skycentrics_token, date_str = config.get_skycentrics_token(request_str=f'GET /api/devices/{config.api_device_id}/data HTTP/1.1',date_str=start_time_str)
+            response = requests.get(f'https://api.skycentrics.com/api/devices/{config.api_device_id}/data', 
+                                headers={'Date': date_str, 'x-sc-api-token': skycentrics_token, 'Accept': 'application/json'})
+            if response.status_code == 200:
+                norm_data = pd.json_normalize(response.json(), record_path=['sensors'], meta=['time'], meta_prefix='response_')
+                if len(norm_data) != 0:
+
+                    norm_data["time_pt"] = pd.to_datetime(norm_data["response_time"])
+
+                    norm_data["time_pt"] = norm_data["time_pt"].dt.tz_convert(time_zone)
+                    norm_data = pd.pivot_table(norm_data, index="time_pt", columns="id", values="data")
+                    # Iterate over the index and round up if necessary (work around for json format from sensors)
+                    for i in range(len(norm_data.index)):
+                        if norm_data.index[i].minute == 59 and norm_data.index[i].second == 59:
+                            norm_data.index.values[i] = norm_data.index[i] + pd.Timedelta(seconds=1)
+                    temp_dfs.append(norm_data)
+            else:
+                print(f"Failed to make GET request. Status code: {response.status_code} {response.json()}")
+            time.sleep(60)
+            time_parser = time_parser + timedelta(minutes=1)
+        if len(temp_dfs) > 0:
+            df = pd.concat(temp_dfs, ignore_index=False)
+            if create_csv:
+                filename = f"{startTime.strftime('%Y%m%d%H%M%S')}.csv"
+                original_directory = os.getcwd()
+                os.chdir(config.data_directory)
+                df.to_csv(filename, index_label='time_pt')
+                os.chdir(original_directory)
+        else:
+            print("No skycentrics data retieved for time frame.")
+        return df
+        
+    except Exception as e:
+        print(f"An error occurred: {e}")
+        raise e
+    return pd.DataFrame()
+
 def fm_api_to_df(config: ConfigManager, startTime: datetime = None, endTime: datetime = None, create_csv : bool = True) -> pd.DataFrame:
     """
     Function connects to the field manager api to pull data and returns a dataframe.
@@ -776,7 +850,7 @@ def pull_egauge_data(config: ConfigManager, eGauge_ids: list, eGauge_usr : str,
     os.chdir(original_directory)
 
 def tb_api_to_df(config: ConfigManager, startTime: datetime = None, endTime: datetime = None, create_csv : bool = True, query_hours : float = 1,
-                 sensor_keys : list = [], seperate_keys : bool = False):
+                 sensor_keys : list = [], seperate_keys : bool = False, device_id_overwrite : str = None, csv_prefix : str = ""):
     """
     Function connects to the things board manager api to pull data and returns a dataframe.
 
@@ -796,6 +870,11 @@ def tb_api_to_df(config: ConfigManager, startTime: datetime = None, endTime: dat
         create csv files as you process such that API need not be relied upon for reprocessing
     query_hours : float
         number of hours to query at a time from ThingsBoard API
+
+    device_id_overwrite : str
+        Overwrites device ID for API pull
+    csv_prefix : str
+        prefix to add to the csv title
     
     Returns
     ------- 
@@ -804,16 +883,17 @@ def tb_api_to_df(config: ConfigManager, startTime: datetime = None, endTime: dat
         Will return with index in UTC so needs to be converted after to appropriate timezone
     """
     df = pd.DataFrame()
+    api_device_id = device_id_overwrite if not device_id_overwrite is None else config.api_device_id
     if len(sensor_keys) <= 0:
         token = config.get_thingsboard_token()
-        key_list = _get_tb_keys(config, token)
+        key_list = _get_tb_keys(token, api_device_id)
         if len(key_list) <= 0:
-            raise Exception(f"No sensors available at ThingsBoard site with id {config.api_device_id}")
-        return tb_api_to_df(config, startTime, endTime, create_csv, query_hours, key_list, seperate_keys)
+            raise Exception(f"No sensors available at ThingsBoard site with id {api_device_id}")
+        return tb_api_to_df(config, startTime, endTime, create_csv, query_hours, key_list, seperate_keys, device_id_overwrite, csv_prefix)
     if seperate_keys:
         df_list = []
         for sensor_key in sensor_keys:
-            df_list.append(tb_api_to_df(config, startTime, endTime, False, query_hours, [sensor_key], False))
+            df_list.append(tb_api_to_df(config, startTime, endTime, False, query_hours, [sensor_key], False, device_id_overwrite, csv_prefix))
         df = pd.concat(df_list)
     else:    
     # not seperate_keys:
@@ -826,13 +906,13 @@ def tb_api_to_df(config: ConfigManager, startTime: datetime = None, endTime: dat
         if endTime - timedelta(hours=query_hours) > startTime:
             time_diff = endTime - startTime
             midpointTime = startTime + time_diff / 2
-            df_1 = tb_api_to_df(config, startTime, midpointTime, query_hours=query_hours, sensor_keys=sensor_keys, create_csv=False)#True if startTime >= datetime(2025,7,13,9) and startTime <= datetime(2025,7,13,10) else csv_pass_down)
-            df_2 = tb_api_to_df(config, midpointTime, endTime, query_hours=query_hours, sensor_keys=sensor_keys,create_csv=False)#True if endTime >= datetime(2025,7,13,9) and endTime <= datetime(2025,7,13,10) else csv_pass_down)
+            df_1 = tb_api_to_df(config, startTime, midpointTime, query_hours=query_hours, sensor_keys=sensor_keys, create_csv=False, device_id_overwrite = device_id_overwrite)#True if startTime >= datetime(2025,7,13,9) and startTime <= datetime(2025,7,13,10) else csv_pass_down)
+            df_2 = tb_api_to_df(config, midpointTime, endTime, query_hours=query_hours, sensor_keys=sensor_keys,create_csv=False, device_id_overwrite = device_id_overwrite)#True if endTime >= datetime(2025,7,13,9) and endTime <= datetime(2025,7,13,10) else csv_pass_down)
             df = pd.concat([df_1, df_2])
             df = df.sort_index()
             df = df.groupby(df.index).mean()
         else:
-            url = f'https://thingsboard.cloud/api/plugins/telemetry/DEVICE/{config.api_device_id}/values/timeseries'
+            url = f'https://thingsboard.cloud/api/plugins/telemetry/DEVICE/{api_device_id}/values/timeseries'
             token = config.get_thingsboard_token()
             key_string = ','.join(sensor_keys)
             params = {
@@ -844,7 +924,6 @@ def tb_api_to_df(config: ConfigManager, startTime: datetime = None, endTime: dat
                 'interval' : '0',
                 'agg' : 'NONE'
             }
-
             # Headers
             headers = {
                 'accept': 'application/json',
@@ -855,14 +934,6 @@ def tb_api_to_df(config: ConfigManager, startTime: datetime = None, endTime: dat
                 response = requests.get(url, headers=headers, params=params)
                 if response.status_code == 200:
                     response_json = response.json()
-                    # if create_csv:
-                    #     json_filename = f"{startTime.strftime('%Y%m%d%H%M%S')}.json"
-                    #     print(f"filename: {json_filename}, url: {url}, params: {params}")
-                    #     original_directory = os.getcwd()
-                    #     os.chdir(config.data_directory)
-                    #     with open(json_filename, 'w') as f:
-                    #         json.dump(response_json, f, indent=4)  # indent=4 makes it human-readable
-                    #     os.chdir(original_directory)
                         
                     data = {}
                     for key, records in response_json.items():
@@ -886,7 +957,7 @@ def tb_api_to_df(config: ConfigManager, startTime: datetime = None, endTime: dat
                 df = pd.DataFrame()
     # save to file
     if create_csv:
-        filename = f"{startTime.strftime('%Y%m%d%H%M%S')}.csv"
+        filename = f"{csv_prefix}{startTime.strftime('%Y%m%d%H%M%S')}.csv"
         original_directory = os.getcwd()
         os.chdir(config.data_directory)
         df.to_csv(filename, index_label='time_pt')
@@ -900,8 +971,8 @@ def _get_float_value(value):
     except (ValueError, TypeError):
         return None
     
-def _get_tb_keys(config: ConfigManager, token : str) -> List[str]:
-    url = f'https://thingsboard.cloud/api/plugins/telemetry/DEVICE/{config.api_device_id}/keys/timeseries'
+def _get_tb_keys(token : str, api_device_id : str) -> List[str]:
+    url = f'https://thingsboard.cloud/api/plugins/telemetry/DEVICE/{api_device_id}/keys/timeseries'
 
     # Headers
     headers = {

{ecopipeline-1.0.0 → ecopipeline-1.0.2}/src/ecopipeline/transform/transform.py

@@ -157,20 +157,29 @@ def _rm_cols(col, bounds_df):  # Helper function for remove_outliers
     """
     Function will take in a pandas series and bounds information
     stored in a dataframe, then check each element of that column and set it to nan
-    if it is outside the given bounds. 
+    if it is outside the given bounds.
 
-    Args: 
-        col: pd.Series 
+    Args:
+        col: pd.Series
             Pandas dataframe column from data being processed
         bounds_df: pd.DataFrame
             Pandas dataframe indexed by the names of the columns from the dataframe that col came from. There should be at least
             two columns in this dataframe, lower_bound and upper_bound, for use in removing outliers
-    Returns: 
-        None 
+    Returns:
+        None
     """
     if (col.name in bounds_df.index):
-        c_lower = float(bounds_df.loc[col.name]["lower_bound"])
-        c_upper = float(bounds_df.loc[col.name]["upper_bound"])
+        c_lower = bounds_df.loc[col.name]["lower_bound"]
+        c_upper = bounds_df.loc[col.name]["upper_bound"]
+
+        # Skip if both bounds are NaN
+        if pd.isna(c_lower) and pd.isna(c_upper):
+            return
+
+        # Convert bounds to float, handling NaN values
+        c_lower = float(c_lower) if not pd.isna(c_lower) else -np.inf
+        c_upper = float(c_upper) if not pd.isna(c_upper) else np.inf
+
         col.mask((col > c_upper) | (col < c_lower), other=np.NaN, inplace=True)
 
 # TODO: remove_outliers STRETCH GOAL: Functionality for alarms being raised based on bounds needs to happen here.
@@ -784,7 +793,7 @@ def convert_on_off_col_to_bool(df: pd.DataFrame, column_names: list) -> pd.DataF
     pd.DataFrame: Dataframe with specified columns converted from Celsius to Farenhiet.
     """
     
-    mapping = {'ON': True, 'OFF': False}
+    mapping = {'ON': True, 'OFF': False, 'On': True, 'Off': False}
     
     for column_name in column_names: 
         df[column_name] = df[column_name].map(mapping).where(df[column_name].notna(), df[column_name])
@@ -1175,6 +1184,9 @@ def join_to_hourly(hourly_data: pd.DataFrame, noaa_data: pd.DataFrame) -> pd.Dat
     pd.DataFrame:
         A single, joined dataframe
     """
+    #fixing pipelines for new years
+    if 'OAT_NOAA' in noaa_data.columns and not noaa_data['OAT_NOAA'].notnull().any():
+        return hourly_data
     out_df = hourly_data.join(noaa_data)
     return out_df
 

{ecopipeline-1.0.0 → ecopipeline-1.0.2}/src/ecopipeline/utils/ConfigManager.py

@@ -4,6 +4,9 @@ import mysql.connector
 import mysql.connector.cursor
 import requests
 from datetime import datetime
+import base64
+import hashlib
+import hmac
 
 class ConfigManager:
     """
@@ -56,6 +59,8 @@ class ConfigManager:
         self.data_directory = data_directory
         self.api_usr = None
         self.api_pw = None
+        self.api_token = None
+        self.api_secret = None
         self.api_device_id = None
         if self.data_directory is None:
             configured_data_method = False
@@ -74,6 +79,11 @@ class ConfigManager:
                     self.api_pw = configure.get('data', 'api_pw')
                     self.api_device_id = configure.get('data','device_id')
                     configured_data_method = True
+                elif 'api_token' in configure['data'] and 'api_secret' in configure['data']:
+                    self.api_token = configure.get('data', 'api_token')
+                    self.api_secret = configure.get('data', 'api_secret')
+                    self.api_device_id = configure.get('data','device_id')
+                    configured_data_method = True
             if not configured_data_method:
                 raise Exception('data configuration section missing or incomplete in configuration file.')
 
@@ -261,4 +271,13 @@ class ConfigManager:
     def get_fm_device_id(self) -> str:
         if self.api_device_id is None:
             raise Exception("Field Manager device ID has not been configured.")
-        return self.api_device_id
+        return self.api_device_id
+    
+    def get_skycentrics_token(self, request_str = 'GET /api/devices/ HTTP/1.', date_str : str = None) -> tuple:
+        if date_str is None:
+            date_str = datetime.utcnow().strftime('%a, %d %b %H:%M:%S GMT')
+        signature = base64.b64encode(hmac.new(self.api_secret.encode(),
+            '{}\n{}\n{}\n{}'.format(request_str, date_str, '', hashlib.md5(''.encode()).hexdigest()).encode(),
+            hashlib.sha1).digest())
+        token = '{}:{}'.format(self.api_token, signature.decode())
+        return token, date_str

{ecopipeline-1.0.0 → ecopipeline-1.0.2/src/ecopipeline.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ecopipeline
-Version: 1.0.0
+Version: 1.0.2
 Summary: Contains functions for use in Ecotope Datapipelines
 Classifier: Programming Language :: Python :: 3
 Classifier: License :: OSI Approved :: GNU General Public License (GPL)