disdrodb 0.0.20__py3-none-any.whl → 0.1.0__py3-none-any.whl

disdrodb/l0/readers/PARSIVEL/NCAR/VORTEX2_2010.py

@@ -0,0 +1,176 @@
+#!/usr/bin/env python3
+
+# -----------------------------------------------------------------------------.
+# Copyright (c) 2021-2023 DISDRODB developers
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+# -----------------------------------------------------------------------------.
+import numpy as np
+import pandas as pd
+
+from disdrodb.l0.l0_reader import is_documented_by, reader_generic_docstring
+from disdrodb.l0.l0a_processing import read_raw_text_file
+
+
+@is_documented_by(reader_generic_docstring)
+def reader(
+    filepath,
+    logger=None,
+):
+    """Reader."""
+    ##------------------------------------------------------------------------.
+    #### Define column names
+    column_names = ["TO_PARSE"]
+
+    ##------------------------------------------------------------------------.
+    #### Define reader options
+    reader_kwargs = {}
+    # - Define delimiter
+    reader_kwargs["delimiter"] = "\\n"
+
+    # - Avoid first column to become df index !!!
+    reader_kwargs["index_col"] = False
+
+    # - Define behaviour when encountering bad lines
+    reader_kwargs["on_bad_lines"] = "skip"
+
+    # Skip the first row (header)
+    reader_kwargs["skiprows"] = 0
+
+    # - Define encoding
+    reader_kwargs["encoding"] = "latin"
+
+    # - Define reader engine
+    #   - C engine is faster
+    #   - Python engine is more feature-complete
+    reader_kwargs["engine"] = "python"
+
+    # - Define on-the-fly decompression of on-disk data
+    #   - Available: gzip, bz2, zip
+    reader_kwargs["compression"] = "infer"
+
+    # - Strings to recognize as NA/NaN and replace with standard NA flags
+    #   - Already included: '#N/A', '#N/A N/A', '#NA', '-1.#IND', '-1.#QNAN',
+    #                       '-NaN', '-nan', '1.#IND', '1.#QNAN', '<NA>', 'N/A',
+    #                       'NA', 'NULL', 'NaN', 'n/a', 'nan', 'null'
+    reader_kwargs["na_values"] = ["na", "", "error"]
+
+    # Skip first row as columns names
+    reader_kwargs["header"] = None
+
+    ##------------------------------------------------------------------------.
+    #### Read the data
+    df = read_raw_text_file(
+        filepath=filepath,
+        column_names=column_names,
+        reader_kwargs=reader_kwargs,
+        logger=logger,
+    )
+
+    ##------------------------------------------------------------------------.
+    #### Adapt the dataframe to adhere to DISDRODB L0 standards
+    # Create ID and Value columns
+    df = df["TO_PARSE"].str.split(":", expand=True, n=1)
+    df.columns = ["ID", "Value"]
+
+    # Select only rows with values
+    df = df[df["Value"].astype(bool)]
+    df = df[df["Value"].apply(lambda x: x is not None)]
+
+    # Drop rows with invalid IDs
+    # - Corrupted rows
+    valid_id_str = np.char.rjust(np.arange(0, 94).astype(str), width=2, fillchar="0")
+    df = df[df["ID"].astype(str).isin(valid_id_str)]
+
+    # Create the dataframe with each row corresponding to a timestep
+    # - Group rows based on when ID values restart
+    groups = df.groupby((df["ID"].astype(int).diff() <= 0).cumsum())
+
+    # Reshape the dataframe
+    group_dfs = []
+    for _, group in groups:
+        group_df = group.set_index("ID").T
+        group_dfs.append(group_df)
+
+    # Merge each timestep dataframe
+    # --> Missing columns are infilled by NaN
+    df = pd.concat(group_dfs, axis=0)
+    df.columns = df.columns.astype(str).str.pad(width=2, side="left", fillchar="0")
+
+    # Define column names
+    column_dict = {
+        "01": "rainfall_rate_32bit",
+        "02": "rainfall_accumulated_32bit",
+        "03": "weather_code_synop_4680",
+        "04": "weather_code_synop_4677",
+        "05": "weather_code_metar_4678",
+        "06": "weather_code_nws",
+        "07": "reflectivity_32bit",
+        "08": "mor_visibility",
+        "09": "sample_interval",
+        "10": "laser_amplitude",
+        "11": "number_particles",
+        "12": "sensor_temperature",
+        # "13": "sensor_serial_number",
+        # "14": "firmware_iop",
+        # "15": "firmware_dsp",
+        "16": "sensor_heating_current",
+        "17": "sensor_battery_voltage",
+        "18": "sensor_status",
+        # "19": "start_time",
+        "20": "sensor_time",
+        "21": "sensor_date",
+        # "22": "station_name",
+        # "23": "station_number",
+        "24": "rainfall_amount_absolute_32bit",
+        "25": "error_code",
+        "30": "rainfall_rate_16bit",
+        "31": "rainfall_rate_12bit",
+        "32": "rainfall_accumulated_16bit",
+        "90": "raw_drop_concentration",
+        "91": "raw_drop_average_velocity",
+        "93": "raw_drop_number",
+    }
+
+    # Identify missing columns and add NaN
+    expected_columns = np.array(list(column_dict.keys()))
+    missing_columns = expected_columns[np.isin(expected_columns, df.columns, invert=True)].tolist()
+    if len(missing_columns) > 0:
+        for column in missing_columns:
+            df[column] = "NaN"
+
+    # Rename columns
+    df = df.rename(column_dict, axis=1)
+
+    # Keep only columns defined in the dictionary
+    df = df[list(column_dict.values())]
+
+    # Define datetime "time" column
+    df["time"] = df["sensor_date"] + "-" + df["sensor_time"]
+    df["time"] = pd.to_datetime(df["time"], format="%d.%m.%Y-%H:%M:%S", errors="coerce")
+
+    # Drop columns not agreeing with DISDRODB L0 standards
+    columns_to_drop = [
+        "sensor_date",
+        "sensor_time",
+        # "firmware_iop",
+        # "firmware_dsp",
+        # "sensor_serial_number",
+        # "station_name",
+        # "station_number",
+    ]
+    df = df.drop(columns=columns_to_drop)
+
+    # Return the dataframe adhering to DISDRODB L0 standards
+    return df

disdrodb/l0/readers/PARSIVEL/NCAR/VORTEX2_2010_UF.py

@@ -0,0 +1,183 @@
+#!/usr/bin/env python3
+
+# -----------------------------------------------------------------------------.
+# Copyright (c) 2021-2023 DISDRODB developers
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+# -----------------------------------------------------------------------------.
+import numpy as np
+import pandas as pd
+
+from disdrodb.l0.l0_reader import is_documented_by, reader_generic_docstring
+from disdrodb.l0.l0a_processing import read_raw_text_file
+
+
+@is_documented_by(reader_generic_docstring)
+def reader(
+    filepath,
+    logger=None,
+):
+    """Reader."""
+    ##------------------------------------------------------------------------.
+    #### Define column names
+    column_names = ["TO_PARSE"]
+
+    ##------------------------------------------------------------------------.
+    #### Define reader options
+    reader_kwargs = {}
+    # - Define delimiter
+    reader_kwargs["delimiter"] = "\\n"
+
+    # - Avoid first column to become df index !!!
+    reader_kwargs["index_col"] = False
+
+    # - Define behaviour when encountering bad lines
+    reader_kwargs["on_bad_lines"] = "skip"
+
+    # Skip the first row (header)
+    reader_kwargs["skiprows"] = 0
+
+    # - Define encoding
+    reader_kwargs["encoding"] = "latin"
+
+    # - Define reader engine
+    #   - C engine is faster
+    #   - Python engine is more feature-complete
+    reader_kwargs["engine"] = "python"
+
+    # - Define on-the-fly decompression of on-disk data
+    #   - Available: gzip, bz2, zip
+    reader_kwargs["compression"] = "infer"
+
+    # - Strings to recognize as NA/NaN and replace with standard NA flags
+    #   - Already included: '#N/A', '#N/A N/A', '#NA', '-1.#IND', '-1.#QNAN',
+    #                       '-NaN', '-nan', '1.#IND', '1.#QNAN', '<NA>', 'N/A',
+    #                       'NA', 'NULL', 'NaN', 'n/a', 'nan', 'null'
+    reader_kwargs["na_values"] = ["na", "", "error"]
+
+    # Skip first row as columns names
+    reader_kwargs["header"] = None
+
+    ##------------------------------------------------------------------------.
+    #### Read the data
+    df = read_raw_text_file(
+        filepath=filepath,
+        column_names=column_names,
+        reader_kwargs=reader_kwargs,
+        logger=logger,
+    )
+
+    ##------------------------------------------------------------------------.
+    #### Adapt the dataframe to adhere to DISDRODB L0 standards
+    # Create ID and Value columns
+    df = df["TO_PARSE"].str.split(":", expand=True, n=1)
+    df.columns = ["ID", "Value"]
+
+    # Select only rows with values
+    df = df[df["Value"].astype(bool)]
+    df = df[df["Value"].apply(lambda x: x is not None)]
+
+    # Drop rows with invalid IDs
+    # - Corrupted rows
+    valid_id_str = np.char.rjust(np.arange(0, 94).astype(str), width=2, fillchar="0")
+    df = df[df["ID"].astype(str).isin(valid_id_str)]
+
+    # Raise error if no more rows after removed corrupted ones
+    if len(df) == 0:
+        raise ValueError("No rows left after removing corrupted ones.")
+
+    # Create the dataframe with each row corresponding to a timestep
+    # - Group rows based on when ID values restart
+    groups = df.groupby((df["ID"].astype(int).diff() <= 0).cumsum())
+
+    # Reshape the dataframe
+    group_dfs = []
+    for _, group in groups:
+        group_df = group.set_index("ID").T
+        group_dfs.append(group_df)
+
+    # Merge each timestep dataframe
+    # --> Missing columns are infilled by NaN
+    df = pd.concat(group_dfs, axis=0)
+    df.columns = df.columns.astype(str).str.pad(width=2, side="left", fillchar="0")
+
+    # Define column names
+    column_dict = {
+        "01": "rainfall_rate_32bit",
+        # "02": "rainfall_accumulated_32bit",
+        # "03": "weather_code_synop_4680",
+        # "04": "weather_code_synop_4677",
+        # "05": "weather_code_metar_4678",
+        "06": "weather_code_nws",
+        "07": "reflectivity_32bit",
+        "08": "mor_visibility",
+        "09": "sample_interval",
+        # "10": "laser_amplitude",
+        "11": "number_particles",
+        "12": "sensor_temperature",
+        # "13": "sensor_serial_number",
+        # "14": "firmware_iop",
+        # "15": "firmware_dsp",
+        # "16": "sensor_heating_current",
+        # "17": "sensor_battery_voltage",
+        # "18": "sensor_status",
+        # "19": "start_time",
+        "20": "sensor_time",
+        "21": "sensor_date",
+        # "22": "station_name",
+        # "23": "station_number",
+        "24": "rainfall_amount_absolute_32bit",
+        # "25": "error_code",
+        # "30": "rainfall_rate_16_bit_30",
+        # "31": "rainfall_rate_16_bit_1200",
+        "32": "rainfall_accumulated_16bit",
+        # "90": "raw_drop_concentration",
+        # "91": "raw_drop_average_velocity",
+        "93": "raw_drop_number",
+    }
+
+    # Identify missing columns and add NaN
+    expected_columns = np.array(list(column_dict.keys()))
+    missing_columns = expected_columns[np.isin(expected_columns, df.columns, invert=True)].tolist()
+    if len(missing_columns) > 0:
+        for column in missing_columns:
+            df[column] = "NaN"
+
+    # Rename columns
+    df = df.rename(column_dict, axis=1)
+
+    # Keep only columns defined in the dictionary
+    df = df[list(column_dict.values())]
+
+    # Define datetime "time" column
+    df["time"] = df["sensor_date"] + "-" + df["sensor_time"]
+    df["time"] = pd.to_datetime(df["time"], format="%d.%m.%Y-%H:%M:%S", errors="coerce")
+
+    # Drop columns not agreeing with DISDRODB L0 standards
+    columns_to_drop = [
+        "sensor_date",
+        "sensor_time",
+        # "firmware_iop",
+        # "firmware_dsp",
+        # "sensor_serial_number",
+        # "station_name",
+        # "station_number",
+    ]
+    df = df.drop(columns=columns_to_drop)
+
+    # Stations UF4-7 have NAN at the end of the raw drop number
+    df["raw_drop_number"] = df["raw_drop_number"].str.replace("NaN;", "").iloc[0]
+
+    # Return the dataframe adhering to DISDRODB L0 standards
+    return df

disdrodb/l0/readers/{ARM/ARM_LD.py → PARSIVEL2/ARM/ARM_PARSIVEL2.py}

@@ -15,22 +15,23 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 # -----------------------------------------------------------------------------.
-"""Reader for ARM OTT Parsivel sensor."""
-from disdrodb.l0 import run_l0b_from_nc
+"""Reader for ARM OTT PARSIVEL2 sensor."""
 from disdrodb.l0.l0_reader import is_documented_by, reader_generic_docstring
+from disdrodb.l0.l0b_nc_processing import open_raw_netcdf_file, replace_custom_nan_flags, standardize_raw_dataset
 
 
 @is_documented_by(reader_generic_docstring)
 def reader(
-    raw_dir,
-    processed_dir,
-    station_name,
-    # Processing options
-    force=False,
-    verbose=False,
-    parallel=False,
-    debugging_mode=False,
+    filepath,
+    logger=None,
 ):
+    """Reader."""
+    ##------------------------------------------------------------------------.
+    #### Open the netCDF
+    ds = open_raw_netcdf_file(filepath=filepath, logger=logger)
+
+    ##------------------------------------------------------------------------.
+    #### Adapt the dataframe to adhere to DISDRODB L0 standards
     # Define dictionary mapping dataset variables to select and rename
     dict_names = {
         ### Dimensions
@@ -84,45 +85,21 @@ def reader(
         # 'qc_sensor_voltage':
     }
 
-    # Define dataset sanitizer
-    def ds_sanitizer_fun(ds):
-        from disdrodb.l0.l0b_nc_processing import replace_custom_nan_flags
-
-        # Replace nan flags with np.nan
-        # - ARM use the -9999 flags
-        nan_flags_variables = [
-            "sensor_temperature",
-            "laser_amplitude",
-            "mor_visibility",
-            "number_particles",
-            "weather_code_synop_4680",
-            "raw_drop_number",
-        ]
-        dict_nan_flags = {var: [-9999] for var in nan_flags_variables}
-        ds = replace_custom_nan_flags(ds, dict_nan_flags=dict_nan_flags)
-
-        # Return dataset
-        return ds
-
-    # Define glob pattern to search data files in <raw_dir>/data/<station_name>
-    glob_patterns = "*.cdf"
-
-    ####----------------------------------------------------------------------.
-    #### - Create L0A products
-    run_l0b_from_nc(
-        raw_dir=raw_dir,
-        processed_dir=processed_dir,
-        station_name=station_name,
-        # Custom arguments of the reader
-        glob_patterns=glob_patterns,
-        dict_names=dict_names,
-        ds_sanitizer_fun=ds_sanitizer_fun,
-        # Processing options
-        force=force,
-        verbose=verbose,
-        parallel=parallel,
-        debugging_mode=debugging_mode,
-    )
+    # Rename dataset variables and columns and infill missing variables
+    ds = standardize_raw_dataset(ds=ds, dict_names=dict_names, sensor_name="PARSIVEL2")
 
+    # Replace nan flags with np.nan
+    # - ARM use the -9999 flags
+    nan_flags_variables = [
+        "sensor_temperature",
+        "laser_amplitude",
+        "mor_visibility",
+        "number_particles",
+        "weather_code_synop_4680",
+        "raw_drop_number",
+    ]
+    dict_nan_flags = {var: [-9999] for var in nan_flags_variables}
+    ds = replace_custom_nan_flags(ds, dict_nan_flags=dict_nan_flags, logger=logger)
 
-# -----------------------------------------------------------------.
+    # Return the dataset adhering to DISDRODB L0B standards
+    return ds

disdrodb/l0/readers/PARSIVEL2/BRAZIL/CHUVA_PARSIVEL2.py

@@ -0,0 +1,163 @@
+#!/usr/bin/env python3
+# -----------------------------------------------------------------------------.
+# Copyright (c) 2021-2023 DISDRODB developers
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+# -----------------------------------------------------------------------------.
+import numpy as np
+import pandas as pd
+
+from disdrodb.l0.l0_reader import is_documented_by, reader_generic_docstring
+from disdrodb.l0.l0a_processing import read_raw_text_file
+
+
+@is_documented_by(reader_generic_docstring)
+def reader(
+    filepath,
+    logger=None,
+):
+    """Reader."""
+    ##------------------------------------------------------------------------.
+    #### Define column names
+    column_names = ["TO_PARSE"]
+
+    ##------------------------------------------------------------------------.
+    #### Define reader options
+    reader_kwargs = {}
+    # - Define delimiter
+    reader_kwargs["delimiter"] = "\\n"
+    # - Skip first row as columns names
+    # - Define encoding
+    reader_kwargs["encoding"] = "latin"  # "ISO-8859-1"
+    # - Avoid first column to become df index !!!
+    reader_kwargs["index_col"] = False
+    # - Define behaviour when encountering bad lines
+    reader_kwargs["on_bad_lines"] = "skip"
+    # - Define reader engine
+    #   - C engine is faster
+    #   - Python engine is more feature-complete
+    reader_kwargs["engine"] = "python"
+    # - Define on-the-fly decompression of on-disk data
+    #   - Available: gzip, bz2, zip
+    reader_kwargs["compression"] = "infer"
+    # - Strings to recognize as NA/NaN and replace with standard NA flags
+    #   - Already included: '#N/A', '#N/A N/A', '#NA', '-1.#IND', '-1.#QNAN',
+    #                       '-NaN', '-nan', '1.#IND', '1.#QNAN', '<NA>', 'N/A',
+    #                       'NA', 'NULL', 'NaN', 'n/a', 'nan', 'null'
+    reader_kwargs["na_values"] = ["na", "", "error"]
+
+    ##------------------------------------------------------------------------.
+    #### Read the data
+    df = read_raw_text_file(
+        filepath=filepath,
+        column_names=column_names,
+        reader_kwargs=reader_kwargs,
+        logger=logger,
+    )
+
+    ##------------------------------------------------------------------------.
+    #### Adapt the dataframe to adhere to DISDRODB L0 standards
+    # Create ID and Value columns
+    df = df["TO_PARSE"].str.split(":", expand=True, n=1)
+    df.columns = ["ID", "Value"]
+
+    # Select only rows with values
+    df = df[df["Value"].astype(bool)]
+    df = df[df["Value"].apply(lambda x: x is not None)]
+
+    # Drop rows with invalid IDs
+    # - Corrupted rows
+    valid_id_str = np.char.rjust(np.arange(0, 94).astype(str), width=2, fillchar="0")
+    df = df[df["ID"].astype(str).isin(valid_id_str)]
+
+    # Create the dataframe with each row corresponding to a timestep
+    # - Group rows based on when ID values restart
+    groups = df.groupby((df["ID"].astype(int).diff() <= 0).cumsum())
+
+    # Reshape the dataframe
+    group_dfs = []
+    for _, group in groups:
+        group_df = group.set_index("ID").T
+        group_dfs.append(group_df)
+
+    # Merge each timestep dataframe
+    # --> Missing columns are infilled by NaN
+    df = pd.concat(group_dfs, axis=0)
+    df.columns = df.columns.astype(str).str.pad(width=2, side="left", fillchar="0")
+
+    # Define available column names
+    column_dict = {
+        "01": "rainfall_rate_32bit",
+        "02": "rainfall_accumulated_32bit",
+        "03": "weather_code_synop_4680",
+        "04": "weather_code_synop_4677",
+        "05": "weather_code_metar_4678",
+        "06": "weather_code_nws",
+        "07": "reflectivity_32bit",
+        "08": "mor_visibility",
+        "09": "sample_interval",
+        "10": "laser_amplitude",
+        "11": "number_particles",
+        "12": "sensor_temperature",
+        # "13": "sensor_serial_number",
+        # "14": "firmware_iop",
+        # "15": "firmware_dsp",
+        "16": "sensor_heating_current",
+        "17": "sensor_battery_voltage",
+        "18": "sensor_status",
+        # "19": "start_time",
+        "20": "sensor_time",
+        "21": "sensor_date",
+        # "22": "station_name",
+        # "23": "station_number",
+        "24": "rainfall_amount_absolute_32bit",
+        "25": "error_code",
+        "30": "rainfall_rate_16_bit_30",
+        "31": "rainfall_rate_16_bit_1200",
+        "32": "rainfall_accumulated_16bit",
+        "90": "raw_drop_concentration",
+        "91": "raw_drop_average_velocity",
+        "93": "raw_drop_number",
+    }
+
+    # Identify missing columns and add NaN
+    expected_columns = np.array(list(column_dict.keys()))
+    missing_columns = expected_columns[np.isin(expected_columns, df.columns, invert=True)].tolist()
+    if len(missing_columns) > 0:
+        for column in missing_columns:
+            df[column] = "NaN"
+
+    # Rename columns
+    df = df.rename(column_dict, axis=1)
+
+    # Keep only columns defined in the dictionary
+    df = df[list(column_dict.values())]
+
+    # Define datetime "time" column
+    df["time"] = df["sensor_date"] + "-" + df["sensor_time"]
+    df["time"] = pd.to_datetime(df["time"], format="%d.%m.%Y-%H:%M:%S", errors="coerce")
+
+    # Drop columns not agreeing with DISDRODB L0 standards
+    columns_to_drop = [
+        "sensor_date",
+        "sensor_time",
+        # "firmware_iop",
+        # "firmware_dsp",
+        # "sensor_serial_number",
+        # "station_name",
+        # "station_number",
+    ]
+    df = df.drop(columns=columns_to_drop)
+
+    return df