disdrodb 0.1.4__py3-none-any.whl → 0.2.0__py3-none-any.whl

disdrodb/l0/readers/PARSIVEL2/ITALY/HYDROX.py

@@ -0,0 +1,239 @@
+#!/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/>.
+# -----------------------------------------------------------------------------.
+"""Reader for HYDROX PARSIVEL2 disdrometer located at Trafoi (Italy)."""
+import os
+
+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 read_old_format(filepath, logger):
+    """Read old format."""
+    ##------------------------------------------------------------------------.
+    #### Define column names
+    column_names = ["TO_SPLIT"]
+
+    ##------------------------------------------------------------------------.
+    #### Define reader options
+    reader_kwargs = {}
+
+    # - Define delimiter
+    reader_kwargs["delimiter"] = "\\n"
+
+    # - Skip first row as columns names
+    reader_kwargs["header"] = None
+
+    # - Skip header
+    reader_kwargs["skiprows"] = 0
+
+    # - Define encoding
+    reader_kwargs["encoding"] = "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'] = 'xz'
+
+    # - 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"]
+
+    ##------------------------------------------------------------------------.
+    #### 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
+    # Remove corrupted rows
+    df = df[df["TO_SPLIT"].str.count(";").isin([7, 1031])]
+
+    # Split into columns
+    df = df["TO_SPLIT"].str.split(";", expand=True, n=7)
+
+    # Assign columns names
+    names = [
+        "date",
+        "time",
+        "rainfall_rate_32bit",
+        "rainfall_accumulated_32bit",
+        "laser_amplitude",
+        "number_particles",
+        "sensor_temperature",
+        "raw_drop_number",
+    ]
+    df.columns = names
+
+    # Add datetime time column
+    df["time"] = df["date"] + "-" + df["time"]
+    df["time"] = pd.to_datetime(df["time"], format="%d.%m.%Y-%H:%M:%S", errors="coerce")
+    df = df.drop(columns=["date"])
+
+    # Correct for UTC time (from UTC+1)
+    df["time"] = df["time"] - pd.Timedelta(hours=1)
+
+    # Preprocess the raw spectrum
+    # - The '<SPECTRUM>ZERO</SPECTRUM>' indicates no drops detected
+    # --> "" generates an array of zeros in L0B processing
+    df["raw_drop_number"] = df["raw_drop_number"].str.replace("<SPECTRUM>ZERO</SPECTRUM>", "")
+
+    # Remove <SPECTRUM> and </SPECTRUM> prefix and suffix from the raw_drop_number field
+    df["raw_drop_number"] = df["raw_drop_number"].str.replace("<SPECTRUM>", "")
+    df["raw_drop_number"] = df["raw_drop_number"].str.replace("</SPECTRUM>", "")
+
+    # Add 0 before every , if , not preceded by a digit
+    # Example: ',,1,,' --> '0,0,1,0,'
+    df["raw_drop_number"] = df["raw_drop_number"].str.replace(r"(?<!\d);", "0;", regex=True)
+
+    # Replace ending 999; with 0;
+    df["raw_drop_number"] = df["raw_drop_number"].str.replace(r"999;$", "0", regex=True)
+
+    # Return the dataframe adhering to DISDRODB L0 standards
+    return df
+
+
+def read_new_format(filepath, logger):
+    """Read new NOA data format."""
+    ##------------------------------------------------------------------------.
+    #### Define column names
+    column_names = ["TO_SPLIT"]
+
+    ##------------------------------------------------------------------------.
+    #### Define reader options
+    reader_kwargs = {}
+
+    # - Define delimiter
+    reader_kwargs["delimiter"] = "\\n"
+
+    # - Skip first row as columns names
+    reader_kwargs["header"] = None
+
+    # - Skip header
+    reader_kwargs["skiprows"] = 0
+
+    # - Define encoding
+    reader_kwargs["encoding"] = "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'] = 'xz'
+
+    # - 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"]
+
+    ##------------------------------------------------------------------------.
+    #### 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
+    # Remove corrupted rows
+    df = df[df["TO_SPLIT"].str.count(";").isin([11, 1035])]
+
+    # Split into columns
+    df = df["TO_SPLIT"].str.split(";", expand=True, n=11)
+
+    # Assign columns names
+    names = [
+        "date",
+        "time",
+        "rainfall_rate_32bit",
+        "rainfall_accumulated_32bit",
+        "reflectivity_32bit",
+        "mor_visibility",
+        "laser_amplitude",
+        "number_particles",
+        "sensor_temperature",
+        "sensor_heating_current",
+        "sensor_battery_voltage",
+        "raw_drop_number",
+    ]
+    df.columns = names
+
+    # Add datetime time column
+    df["time"] = df["date"] + "-" + df["time"]
+    df["time"] = pd.to_datetime(df["time"], format="%d.%m.%Y-%H:%M:%S", errors="coerce")
+    df = df.drop(columns=["date"])
+
+    # Preprocess the raw spectrum
+    # - The '<SPECTRUM>ZERO</SPECTRUM>' indicates no drops detected
+    # --> "" generates an array of zeros in L0B processing
+    df["raw_drop_number"] = df["raw_drop_number"].str.replace("<SPECTRUM>ZERO</SPECTRUM>", "")
+
+    # Remove <SPECTRUM> and </SPECTRUM> prefix and suffix from the raw_drop_number field
+    df["raw_drop_number"] = df["raw_drop_number"].str.replace("<SPECTRUM>", "")
+    df["raw_drop_number"] = df["raw_drop_number"].str.replace("</SPECTRUM>", "")
+
+    # Add 0 before every , if , not preceded by a digit
+    # Example: ',,1,,' --> '0,0,1,0,'
+    df["raw_drop_number"] = df["raw_drop_number"].str.replace(r"(?<!\d);", "0;", regex=True)
+
+    # Replace ending 999; with 0;
+    df["raw_drop_number"] = df["raw_drop_number"].str.replace(r"999;$", "0", regex=True)
+
+    # Return the dataframe adhering to DISDRODB L0 standards
+    return df
+
+
+@is_documented_by(reader_generic_docstring)
+def reader(
+    filepath,
+    logger=None,
+):
+    """Reader."""
+    date = int(os.path.basename(filepath)[-12:-4])
+    if date > 20140000:
+        return read_new_format(filepath, logger)
+    return read_old_format(filepath, logger)

disdrodb/l0/readers/PARSIVEL2/NCAR/FARM_PARSIVEL2.py

@@ -109,17 +109,11 @@ def reader(
     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:])
-        return pd.Series({"raw_drop_concentration": c1, "raw_drop_average_velocity": c2, "raw_drop_number": c3})
-
-    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_split = df["TO_SPLIT"].str.split(",", expand=True)
+    df["raw_drop_concentration"] = df_split.iloc[:, :32].agg(",".join, axis=1)
+    df["raw_drop_average_velocity"] = df_split.iloc[:, 32:64].agg(",".join, axis=1)
+    df["raw_drop_number"] = df_split.iloc[:, 64:].agg(",".join, axis=1)
+    del df_split
 
     # Define datetime "time" column
     df["time"] = pd.to_datetime(df["time"], format="%Y-%m-%d %H:%M:%S", errors="coerce")

disdrodb/l0/readers/PARSIVEL2/NCAR/PERILS_MIPS.py

@@ -89,23 +89,10 @@ def reader(
     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:])
-        series = pd.Series(
-            {
-                "raw_drop_concentration": c1,
-                "raw_drop_average_velocity": c2,
-                # "raw_drop_number": c3,
-            },
-        )
-        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"]
+    # - raw_drop_number is missing !
+    df_split = df["TO_SPLIT"].str.split(",", expand=True)
+    df["raw_drop_concentration"] = df_split.iloc[:, :32].agg(",".join, axis=1)
+    df["raw_drop_average_velocity"] = df_split.iloc[:, 32:64].agg(",".join, axis=1)
 
     # Define datetime time column
     df["year"] = df["year"].str.replace(".0", "")

disdrodb/l0/readers/PARSIVEL2/NCAR/RELAMPAGO_PARSIVEL2.py

@@ -90,20 +90,11 @@ def reader(
     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")
+    # Create the dataframe where each row corresponds to a timestep
+    df["_group"] = (df["ID"].astype(int).diff() <= 0).cumsum()
+    df = df.pivot(index="_group", columns="ID")  # noqa
+    df.columns = df.columns.get_level_values("ID")
+    df = df.reset_index(drop=True)
 
     # Assign column names
     column_dict = {

disdrodb/l0/readers/PARSIVEL2/NCAR/SNOWIE_PJ.py

@@ -81,19 +81,11 @@ def reader(
     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)
+    # Create the dataframe where each row corresponds to a timestep
+    df["_group"] = (df["ID"].astype(int).diff() <= 0).cumsum()
+    df = df.pivot(index="_group", columns="ID")  # noqa
+    df.columns = df.columns.get_level_values("ID")
+    df = df.reset_index(drop=True)
 
     # Assign column names
     column_dict = {
@@ -122,9 +114,14 @@ def reader(
         # "23": "station_number",
         "24": "rainfall_amount_absolute_32bit",
         "25": "error_code",
+        # "26": "sensor_temperature_pcb",
+        # "27": "sensor_temperature_receiver",
+        # "28": "sensor_temperature_trasmitter",
         "30": "rainfall_rate_16_bit_30",
         "31": "rainfall_rate_16_bit_1200",
         "32": "rainfall_accumulated_16bit",
+        # "34": "rain_kinetic_energy",
+        # "35": "snowfall_rate",
         "90": "raw_drop_concentration",
         "91": "raw_drop_average_velocity",
         "93": "raw_drop_number",

disdrodb/l0/readers/PARSIVEL2/NCAR/SNOWIE_SB.py

@@ -81,19 +81,11 @@ def reader(
     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)
+    # Create the dataframe where each row corresponds to a timestep
+    df["_group"] = (df["ID"].astype(int).diff() <= 0).cumsum()
+    df = df.pivot(index="_group", columns="ID")  # noqa
+    df.columns = df.columns.get_level_values("ID")
+    df = df.reset_index(drop=True)
 
     # Assign column names
     column_dict = {
@@ -122,9 +114,14 @@ def reader(
         # "23": "station_number",
         "24": "rainfall_amount_absolute_32bit",
         "25": "error_code",
+        # "26": "sensor_temperature_pcb",
+        # "27": "sensor_temperature_receiver",
+        # "28": "sensor_temperature_trasmitter",
         "30": "rainfall_rate_16_bit_30",
         "31": "rainfall_rate_16_bit_1200",
         "32": "rainfall_accumulated_16bit",
+        # "34": "rain_kinetic_energy",
+        # "35": "snowfall_rate",
         "90": "raw_drop_concentration",
         "91": "raw_drop_average_velocity",
         "93": "raw_drop_number",

disdrodb/l0/readers/PARSIVEL2/PHILIPPINES/PAGASA.py

@@ -0,0 +1,232 @@
+#!/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/>.
+# -----------------------------------------------------------------------------.
+"""DISDRODB reader for PANGASA PARSIVEL2 raw text data."""
+# import os
+# import tempfile
+# from disdrodb.utils.compression import unzip_file_on_terminal
+
+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
+from disdrodb.utils.logger import log_error
+
+
+def read_txt_file(file, filename, logger):
+    """Parse a single txt file within the daily zip 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
+    # - 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=file,
+        column_names=column_names,
+        reader_kwargs=reader_kwargs,
+        logger=logger,
+    )
+    ##--------------------------------\----------------------------------------.
+    #### Adapt the dataframe to adhere to DISDRODB L0 standards
+    # Empty file, return None
+    if len(df) == 0:
+        raise ValueError(f"{filename} is empty.")
+
+    # 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 where each row corresponds to a timestep
+    df["_group"] = (df["ID"].astype(int).diff() <= 0).cumsum()
+    df = df.pivot(index="_group", columns="ID")  # noqa
+    df.columns = df.columns.get_level_values("ID")
+    df = df.reset_index(drop=True)
+
+    # Assign 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",
+        "26": "sensor_temperature_pcb",
+        "27": "sensor_temperature_receiver",
+        "28": "sensor_temperature_trasmitter",
+        "30": "rainfall_rate_16_bit_30",
+        "31": "rainfall_rate_16_bit_1200",
+        "32": "rainfall_accumulated_16bit",
+        # "33": "reflectivity_16bit",
+        "34": "rain_kinetic_energy",
+        "35": "snowfall_rate",
+        # 60: "number_particles_all",
+        # 61: "list_particles",
+        "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
+    time_str = filename.split("_")[-1].replace(".txt", "")
+    df["time"] = pd.to_datetime(time_str, format="%Y%m%d%H%M%S", errors="coerce")
+
+    # Keep only rows with valid raw_drop_number
+    df = df[df["raw_drop_number"].str.count(";") == 1024]
+    if len(df) == 0:
+        raise ValueError("Invalid raw drop number field.")
+
+    # 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
+
+
+@is_documented_by(reader_generic_docstring)
+def reader(
+    filepath,
+    logger=None,
+):
+    """Reader."""
+    import zipfile
+
+    # return read_txt_file(file=filepath,
+    #                      filename=os.path.basename(filepath),
+    #                      logger=logger,
+    #                      )
+
+    # ---------------------------------------------------------------------.
+    #### Iterate over all files (aka timesteps) in the daily zip archive
+    # - Each file contain a single timestep !
+    # list_df = []
+    # with tempfile.TemporaryDirectory() as temp_dir:
+    #     # Extract all files
+    #     unzip_file_on_terminal(filepath, temp_dir)
+
+    #     # Walk through extracted files
+    #     for root, _, files in os.walk(temp_dir):
+    #         for filename in sorted(files):
+    #             if filename.endswith(".txt"):
+    #                 full_path = os.path.join(root, filename)
+    #                 try:
+    #                     df = read_txt_file(file=full_path, filename=filename, logger=logger)
+    #                     if df is not None:
+    #                         list_df.append(df)
+    #                 except Exception as e:
+    #                     msg = f"An error occurred while reading {filename}: {e}"
+    #                     log_error(logger=logger, msg=msg, verbose=True)
+
+    list_df = []
+    with zipfile.ZipFile(filepath, "r") as zip_ref:
+        filenames = sorted(zip_ref.namelist())
+        for filename in filenames:
+            if filename.endswith(".txt"):
+                # Open file
+                with zip_ref.open(filename) as file:
+                    try:
+                        df = read_txt_file(file=file, filename=filename, logger=logger)
+                        if df is not None:
+                            list_df.append(df)
+                    except Exception as e:
+                        msg = f"An error occurred while reading {filename}. The error is: {e}."
+                        log_error(logger=logger, msg=msg, verbose=True)
+
+    # Check the zip file contains at least some non.empty files
+    if len(list_df) == 0:
+        raise ValueError(f"{filepath} contains only empty files!")
+
+    # Concatenate all dataframes into a single one
+    df = pd.concat(list_df)
+
+    # ---------------------------------------------------------------------.
+    return df

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

@@ -111,31 +111,19 @@ def reader(
     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_split = df["TO_PARSE"].str.split(",", expand=True)
+    df_raw_spectrum["raw_drop_concentration"] = df_split.iloc[:, :32].agg(",".join, axis=1)
+    df_raw_spectrum["raw_drop_average_velocity"] = df_split.iloc[:, 32:64].agg(",".join, axis=1)
+    df_raw_spectrum["raw_drop_number"] = df_split.iloc[:, 64:].agg(",".join, axis=1)
     df_raw_spectrum = df_raw_spectrum.drop(columns=["date", "TO_PARSE"])
+    del df_split
 
     # Add raw array
     df = df_data.set_index("time")
     df_raw_spectrum = df_raw_spectrum.set_index("time")
 
     df.update(df_raw_spectrum)
+    del df_raw_spectrum
 
     # Set back time as column
     df = df.reset_index()