disdrodb 0.1.2__py3-none-any.whl → 0.1.3__py3-none-any.whl

disdrodb/l0/readers/PARSIVEL2/SPAIN/CENER.py

@@ -0,0 +1,144 @@
+#!/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 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
+    # Define time
+    df = df["TO_PARSE"].str.split(",", n=2, expand=True)
+    df.columns = ["date", "time", "TO_PARSE"]
+    datetime_str = df["date"] + " " + df["time"]
+    df["time"] = pd.to_datetime(datetime_str, format="%d.%m.%Y %H:%M:%S", errors="coerce")
+
+    # Identify rows with integral variables
+    df_vars = df[df["TO_PARSE"].str.len() == 94]
+
+    # Split and assign column names
+    df_data = df_vars["TO_PARSE"].str.split(",", expand=True)
+    var_names = [
+        "rainfall_rate_32bit",
+        "rainfall_accumulated_32bit",
+        "weather_code_synop_4680",
+        "weather_code_synop_4677",
+        "reflectivity_32bit",
+        "mor_visibility",
+        "laser_amplitude",
+        "number_particles",
+        "sensor_temperature",
+        "sensor_heating_current",
+        "sensor_battery_voltage",
+        "sensor_status",
+        "sensor_serial_number",
+        "sensor_temperature_receiver",
+        "sensor_temperature_trasmitter",
+        "snowfall_rate",
+        "rain_kinetic_energy",
+    ]
+    df_data.columns = var_names
+    df_data["time"] = df_vars["time"]
+    df_data = df_data.drop(columns="sensor_serial_number")
+
+    # Initialize empty arrays
+    # --> 0 values array produced in L0B
+    df_data["raw_drop_concentration"] = ""
+    df_data["raw_drop_average_velocity"] = ""
+    df_data["raw_drop_number"] = ""
+
+    # Identify raw spectrum
+    df_raw_spectrum = df[df["TO_PARSE"].str.len() == 4545]
+
+    # Derive raw drop arrays
+    def split_string(s):
+        vals = [v.strip() for v in s.split(",")]
+        c1 = ",".join(vals[:32])
+        c2 = ",".join(vals[32:64])
+        c3 = ",".join(vals[64].replace("r", "").split("/"))
+        series = pd.Series(
+            {
+                "raw_drop_concentration": c1,
+                "raw_drop_average_velocity": c2,
+                "raw_drop_number": c3,
+            },
+        )
+        return series
+
+    splitted_string = df_raw_spectrum["TO_PARSE"].apply(split_string)
+    df_raw_spectrum["raw_drop_concentration"] = splitted_string["raw_drop_concentration"]
+    df_raw_spectrum["raw_drop_average_velocity"] = splitted_string["raw_drop_average_velocity"]
+    df_raw_spectrum["raw_drop_number"] = splitted_string["raw_drop_number"]
+    df_raw_spectrum = df_raw_spectrum.drop(columns=["date", "TO_PARSE"])
+
+    # Add raw array
+    df = df_data.set_index("time")
+    df_raw_spectrum = df_raw_spectrum.set_index("time")
+
+    df.update(df_raw_spectrum)
+
+    # Set back time as column
+    df = df.reset_index()
+
+    # Return the dataframe adhering to DISDRODB L0 standards
+    return df

disdrodb/l0/readers/PARSIVEL2/SPAIN/CR1000DL.py

@@ -0,0 +1,201 @@
+#!/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 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
+
+
+def reader_parsivel(filepath, logger):
+    """Reader for Parsivel CR1000 Data Logger file."""
+    ##------------------------------------------------------------------------.
+    #### Define column names
+    column_names = [
+        "time",
+        "RECORD",
+        "rainfall_rate_32bit",
+        "rainfall_accumulated_32bit",
+        "weather_code_synop_4680",
+        "weather_code_synop_4677",
+        "reflectivity_32bit",
+        "mor_visibility",
+        "laser_amplitude",
+        "number_particles",
+        "sensor_temperature",
+        "sensor_heating_current",
+        "sensor_battery_voltage",
+        "sensor_status",
+        "rain_kinetic_energy",
+        "V_Batt_Min",
+    ]
+
+    ##------------------------------------------------------------------------.
+    #### Define reader options
+    reader_kwargs = {}
+    # - Define delimiter
+    reader_kwargs["delimiter"] = ","
+    # - Skip first row as columns names
+    reader_kwargs["header"] = None
+    # - Skip first 3 rows
+    reader_kwargs["skiprows"] = 4
+    # - 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
+    # Define time
+    df["time"] = pd.to_datetime(df["time"], format="%Y-%m-%d %H:%M:%S", errors="coerce")
+
+    # Drop columns not agreeing with DISDRODB L0 standards
+    df = df.drop(columns=["RECORD", "V_Batt_Min"])
+    return df
+
+
+def reader_spectrum(filepath, logger):
+    """Reader for Spectrum CR1000 Data Logger file."""
+    ##------------------------------------------------------------------------.
+    #### Define column names
+    column_names = ["TO_PARSE"]
+
+    ##------------------------------------------------------------------------.
+    #### Define reader options
+    reader_kwargs = {}
+    # - Define delimiter
+    reader_kwargs["delimiter"] = "\\n"
+    # - Skip first row as columns names
+    reader_kwargs["header"] = None
+    # - Skip first 3 rows
+    reader_kwargs["skiprows"] = 4
+    # - 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
+    # Define time
+    df = df["TO_PARSE"].str.split(",", n=2, expand=True)
+    df.columns = ["time", "RECORD", "TO_PARSE"]
+
+    df["time"] = pd.to_datetime(df["time"].str.strip('"'), format="%Y-%m-%d %H:%M:%S", errors="coerce")
+
+    # Derive raw drop arrays
+    def split_string(s):
+        vals = [v.strip() for v in s.split(",")]
+        c1 = ",".join(vals[:32])  # -10
+        c1 = c1.replace("-10", "0")
+        c2 = "0,0," + ",".join(vals[32:62])
+        c3 = ",".join(vals[62:])
+        series = pd.Series(
+            {
+                "raw_drop_concentration": c1,
+                "raw_drop_average_velocity": c2,
+                "raw_drop_number": c3,
+            },
+        )
+        return series
+
+    splitted_string = df["TO_PARSE"].apply(split_string)
+    df["raw_drop_concentration"] = splitted_string["raw_drop_concentration"]
+    df["raw_drop_average_velocity"] = splitted_string["raw_drop_average_velocity"]
+    df["raw_drop_number"] = splitted_string["raw_drop_number"]
+
+    # Drop columns not agreeing with DISDRODB L0 standards
+    df = df.drop(columns=["TO_PARSE", "RECORD"])
+    return df
+
+
+@is_documented_by(reader_generic_docstring)
+def reader(
+    filepath,
+    logger=None,
+):
+    """Reader."""
+    # Retrieve Spectrum filepath
+    spectrum_filepath = filepath.replace("Parsivel", "Spectre")
+
+    # Read integral variables
+    df = reader_parsivel(filepath, logger=logger)
+
+    # Initialize empty arrays
+    # --> 0 values array produced in L0B
+    df["raw_drop_concentration"] = ""
+    df["raw_drop_average_velocity"] = ""
+    df["raw_drop_number"] = ""
+
+    # Read raw spectrum for corresponding timesteps
+    df_raw_spectrum = reader_spectrum(spectrum_filepath, logger=logger)
+
+    # Add raw array to df
+    df = df.set_index("time")
+    df_raw_spectrum = df_raw_spectrum.set_index("time")
+    df.update(df_raw_spectrum)
+
+    # Set back time as column
+    df = df.reset_index()
+
+    # Return the dataframe adhering to DISDRODB L0 standards
+    return df

disdrodb/l0/readers/PARSIVEL2/SPAIN/LIAISE.py

@@ -0,0 +1,137 @@
+#!/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 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
+    # Define time
+    df = df["TO_PARSE"].str.split(",", n=2, expand=True)
+    df.columns = ["date", "time", "TO_PARSE"]
+    datetime_str = df["date"] + " " + df["time"]
+    df["time"] = pd.to_datetime(datetime_str, format="%d.%m.%Y %H:%M:%S", errors="coerce")
+
+    # Identify rows with integral variables
+    df_vars = df[df["TO_PARSE"].str.len() == 61]
+
+    # Split and assign column names
+    df_data = df_vars["TO_PARSE"].str.split(",", expand=True)
+    var_names = [
+        "rainfall_rate_32bit",
+        "rainfall_accumulated_32bit",
+        "weather_code_synop_4680",
+        "weather_code_synop_4677",
+        "reflectivity_32bit",
+        "mor_visibility",
+        "laser_amplitude",
+        "number_particles",
+        "sensor_temperature",
+        "sensor_heating_current",
+        "sensor_battery_voltage",
+    ]
+    df_data.columns = var_names
+    df_data["time"] = df_vars["time"]
+
+    # Initialize empty arrays
+    # --> 0 values array produced in L0B
+    df_data["raw_drop_concentration"] = ""
+    df_data["raw_drop_average_velocity"] = ""
+    df_data["raw_drop_number"] = ""
+
+    # Identify raw spectrum
+    df_raw_spectrum = df[df["TO_PARSE"].str.len() == 4545]
+
+    # Derive raw drop arrays
+    def split_string(s):
+        vals = [v.strip() for v in s.split(",")]
+        c1 = ",".join(vals[:32])
+        c2 = ",".join(vals[32:64])
+        c3 = ",".join(vals[64].replace("r", "").split("/"))
+        series = pd.Series(
+            {
+                "raw_drop_concentration": c1,
+                "raw_drop_average_velocity": c2,
+                "raw_drop_number": c3,
+            },
+        )
+        return series
+
+    splitted_string = df_raw_spectrum["TO_PARSE"].apply(split_string)
+    df_raw_spectrum["raw_drop_concentration"] = splitted_string["raw_drop_concentration"]
+    df_raw_spectrum["raw_drop_average_velocity"] = splitted_string["raw_drop_average_velocity"]
+    df_raw_spectrum["raw_drop_number"] = splitted_string["raw_drop_number"]
+    df_raw_spectrum = df_raw_spectrum.drop(columns=["date", "TO_PARSE"])
+
+    # Add raw array
+    df = df_data.set_index("time")
+    df_raw_spectrum = df_raw_spectrum.set_index("time")
+
+    df.update(df_raw_spectrum)
+
+    # Set back time as column
+    df = df.reset_index()
+
+    # Return the dataframe adhering to DISDRODB L0 standards
+    return df

disdrodb/l0/readers/PARSIVEL2/{NETHERLANDS/DELFT.py → USA/C3WE.py}

@@ -29,33 +29,34 @@ def reader(
     """Reader."""
     ##------------------------------------------------------------------------.
     #### Define column names
-    column_names = ["time", "epoch_time", "TO_BE_PARSED"]
+    column_names = ["TO_PARSE"]
 
     ##------------------------------------------------------------------------.
     #### Define reader options
     reader_kwargs = {}
-    # - Define delimiter
-    reader_kwargs["delimiter"] = ";"
-    # - Skip first row as columns names
+    # Skip first row as columns names
     reader_kwargs["header"] = None
-    # - Avoid first column to become df index
+    # Skip file with encoding errors
+    reader_kwargs["encoding_errors"] = "ignore"
+    # - 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"
-    # - Define parser engine
+    # - 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"
-    # reader_kwargs['zipped'] = False
-    # reader_kwargs['zipped'] = True
     # - 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", "-.-", " NA"]
+    reader_kwargs["na_values"] = ["na", "", "error", "NA", "-.-"]
+
     ##------------------------------------------------------------------------.
     #### Read the data
     df = read_raw_text_file(
@@ -67,98 +68,77 @@ def reader(
 
     ##------------------------------------------------------------------------.
     #### Adapt the dataframe to adhere to DISDRODB L0 standards
-    # NOTE: Station 8 has all raw_drop_number corrupted, so it can't be used
-
-    # Remove rows that have a corrupted "TO_BE_PARSED" column
-    # - PAR001, PAR002 have length 3726 (no station_name)
-    # - PAR007 have length 3736 ()
-    df = df.loc[df["TO_BE_PARSED"].astype(str).str.len() >= 3726]
-    df = df.loc[df["TO_BE_PARSED"].astype(str).str.len() <= 3736]
+    # Define 'time' datetime
 
-    # Convert 'time' column to datetime
-    df_time = pd.to_datetime(df["time"], format="%Y%m%d-%H%M%S", errors="coerce")
+    # Split the columns
+    df["TO_PARSE"].iloc[0:5].str.split(";", n=16, expand=True).iloc[0]
 
-    # Strip values from start and end of the string
-    df["TO_BE_PARSED"] = df["TO_BE_PARSED"].str.lstrip("b'").str.rstrip("'").str.rstrip("\\r\\n'")  # noqa: B005
-
-    # Split the column 'TO_BE_PARSED'
-    df_to_parse = df["TO_BE_PARSED"].str.split(";", expand=True, n=99)
-
-    # Retrieve DISDRODB compliant columns
-    df = df_to_parse.iloc[:, 0:35]
+    df = df["TO_PARSE"].str.split(";", n=16, expand=True)
 
     # Assign column names
-    column_names = [
-        "rainfall_rate_32bit",
-        "rainfall_accumulated_32bit",
-        "weather_code_synop_4680",
-        "weather_code_synop_4677",
-        "weather_code_metar_4678",
-        "weather_code_nws",
-        "reflectivity_32bit",
-        "mor_visibility",
-        "sample_interval",
-        "laser_amplitude",
-        "number_particles",
-        "sensor_temperature",
+    names = [
         "sensor_serial_number",
-        "firmware_iop",
-        "firmware_dsp",
+        "sensor_status",
+        "laser_amplitude",
         "sensor_heating_current",
         "sensor_battery_voltage",
-        "sensor_status",
-        "sensor_time_measurement_start",
+        "dummy_date",
         "sensor_time",
         "sensor_date",
-        "station_name",
-        "station_number",
-        "rainfall_amount_absolute_32bit",
-        "error_code",
-        "sensor_temperature_pcb",
-        "sensor_temperature_receiver",
-        "sensor_temperature_trasmitter",
-        "rainfall_rate_16_bit_30",
-        "rainfall_rate_16_bit_1200",
+        "sensor_temperature",
+        "number_particles",
+        "rainfall_rate_32bit",
+        "reflectivity_32bit",
         "rainfall_accumulated_16bit",
-        "reflectivity_16bit",
-        "rain_kinetic_energy",
-        "snowfall_rate",
-        "number_particles_all",
-        # "number_particles_all_detected",
+        "mor_visibility",
+        "weather_code_synop_4680",
+        "weather_code_synop_4677",
+        "TO_SPLIT",
     ]
-    df.columns = column_names
-
-    # Add time column
-    df["time"] = df_time
-
-    # Retrieve raw_drop_concentration
-    df["raw_drop_concentration"] = df_to_parse.iloc[:, 35:67].apply(
-        lambda x: ",".join(x.dropna().astype(str)),
-        axis=1,
-    )
-    # Retrieve raw_drop_average_velocity
-    df["raw_drop_average_velocity"] = df_to_parse.iloc[:, 67:99].apply(
-        lambda x: ",".join(x.dropna().astype(str)),
-        axis=1,
-    )
-
-    # Retrieve raw_drop_number
-    df_raw_drop_number = df_to_parse.iloc[:, 99].squeeze()
-    df_raw_drop_number = df_raw_drop_number.str.replace(r"(\w{3})", r"\1,", regex=True)
-    df["raw_drop_number"] = df_raw_drop_number
+    df.columns = names
+
+    # Derive raw drop arrays
+    def split_string(s):
+        vals = [v.strip() for v in s.split(";")]
+        c1 = ";".join(vals[:32])
+        c2 = ";".join(vals[32:64])
+        c3 = ";".join(vals[64:1088])
+        c4 = vals[1088]
+        c5 = vals[1089]
+        series = pd.Series(
+            {
+                "raw_drop_concentration": c1,
+                "raw_drop_average_velocity": c2,
+                "raw_drop_number": c3,
+                "rain_kinetic_energy": c4,
+                "CHECK_EMPTY": c5,
+            },
+        )
+        return series
+
+    splitted_string = df["TO_SPLIT"].apply(split_string)
+    df["raw_drop_concentration"] = splitted_string["raw_drop_concentration"]
+    df["raw_drop_average_velocity"] = splitted_string["raw_drop_average_velocity"]
+    df["raw_drop_number"] = splitted_string["raw_drop_number"]
+    df["rain_kinetic_energy"] = splitted_string["rain_kinetic_energy"]
+    df["CHECK_EMPTY"] = splitted_string["CHECK_EMPTY"]
+
+    # Ensure valid observation
+    df = df[df["CHECK_EMPTY"] == ""]
+
+    # Add the time column
+    time_str = df["sensor_date"] + "-" + df["sensor_time"]
+    df["time"] = pd.to_datetime(time_str, format="%d.%m.%Y-%H:%M:%S", errors="coerce")
 
     # Drop columns not agreeing with DISDRODB L0 standards
     columns_to_drop = [
-        "firmware_iop",
-        "firmware_dsp",
-        "sensor_time_measurement_start",
-        "sensor_time",
+        "dummy_date",
         "sensor_date",
-        "station_name",
-        "station_number",
-        "sensor_serial_number",
+        "sensor_time",
         "sensor_serial_number",
-        # "number_particles_all_detected",
+        "rainfall_accumulated_16bit",  # unexpected format
+        "CHECK_EMPTY",
+        "TO_SPLIT",
     ]
     df = df.drop(columns=columns_to_drop)