pydartdiags 0.0.41__tar.gz → 0.0.43__tar.gz

pydartdiags-0.0.43/PKG-INFO

@@ -0,0 +1,45 @@
+Metadata-Version: 2.1
+Name: pydartdiags
+Version: 0.0.43
+Summary: Observation Sequence Diagnostics for DART
+Home-page: https://github.com/NCAR/pyDARTdiags.git
+Author: Helen Kershaw
+Author-email: Helen Kershaw <hkershaw@ucar.edu>
+Project-URL: Homepage, https://github.com/NCAR/pyDARTdiags.git
+Project-URL: Issues, https://github.com/NCAR/pyDARTdiags/issues
+Project-URL: Documentation, https://ncar.github.io/pyDARTdiags
+Classifier: Programming Language :: Python :: 3
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: pandas>=2.2.0
+Requires-Dist: numpy>=1.26
+Requires-Dist: plotly>=5.22.0
+Requires-Dist: pyyaml>=6.0.2
+
+[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
+[![codecov](https://codecov.io/gh/NCAR/pyDARTdiags/graph/badge.svg?token=VK55SQZSVD)](https://codecov.io/gh/NCAR/pyDARTdiags)
+[![PyPI version](https://badge.fury.io/py/pydartdiags.svg)](https://pypi.org/project/pydartdiags/)
+
+
+# pyDARTdiags
+
+pyDARTdiags is a Python library for obsevation space diagnostics for the Data Assimilation Research Testbed ([DART](https://github.com/NCAR/DART)).
+
+pyDARTdiags is under initial development, so please use caution.
+The MATLAB [observation space diagnostics](https://docs.dart.ucar.edu/en/latest/guide/matlab-observation-space.html) are available through [DART](https://github.com/NCAR/DART).
+
+
+pyDARTdiags can be installed through pip: https://pypi.org/project/pydartdiags/  
+Documenation : https://ncar.github.io/pyDARTdiags/
+
+## Contributing
+Contributions are welcome! If you have a feature request, bug report, or a suggestion, please open an issue on our GitHub repository.
+Please read our [Contributors Guide](https://github.com/NCAR/pyDARTdiags/blob/main/CONTRIBUTING.md) if you would like to contribute to
+pyDARTdiags.
+
+## License
+
+pyDARTdiags is released under the Apache License 2.0. For more details, see the LICENSE file in the root directory of this source tree or visit [Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0).

pydartdiags-0.0.43/README.md

@@ -0,0 +1,24 @@
+[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
+[![codecov](https://codecov.io/gh/NCAR/pyDARTdiags/graph/badge.svg?token=VK55SQZSVD)](https://codecov.io/gh/NCAR/pyDARTdiags)
+[![PyPI version](https://badge.fury.io/py/pydartdiags.svg)](https://pypi.org/project/pydartdiags/)
+
+
+# pyDARTdiags
+
+pyDARTdiags is a Python library for obsevation space diagnostics for the Data Assimilation Research Testbed ([DART](https://github.com/NCAR/DART)).
+
+pyDARTdiags is under initial development, so please use caution.
+The MATLAB [observation space diagnostics](https://docs.dart.ucar.edu/en/latest/guide/matlab-observation-space.html) are available through [DART](https://github.com/NCAR/DART).
+
+
+pyDARTdiags can be installed through pip: https://pypi.org/project/pydartdiags/  
+Documenation : https://ncar.github.io/pyDARTdiags/
+
+## Contributing
+Contributions are welcome! If you have a feature request, bug report, or a suggestion, please open an issue on our GitHub repository.
+Please read our [Contributors Guide](https://github.com/NCAR/pyDARTdiags/blob/main/CONTRIBUTING.md) if you would like to contribute to
+pyDARTdiags.
+
+## License
+
+pyDARTdiags is released under the Apache License 2.0. For more details, see the LICENSE file in the root directory of this source tree or visit [Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0).

{pydartdiags-0.0.41 → pydartdiags-0.0.43}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "pydartdiags"
-version = "0.0.41"
+version = "0.0.43"
 authors = [
   { name="Helen Kershaw", email="hkershaw@ucar.edu" },
 ]

{pydartdiags-0.0.41 → pydartdiags-0.0.43}/src/pydartdiags/obs_sequence/obs_sequence.py

@@ -3,6 +3,24 @@ import datetime as dt
 import numpy as np
 import os
 import yaml
+import struct
+
+def requires_assimilation_info(func):
+    def wrapper(self, *args, **kwargs):
+        if self.has_assimilation_info:
+            return func(self, *args, **kwargs)
+        else:
+            raise ValueError("Assimilation information is required to call this function.")
+    return wrapper
+
+def requires_posterior_info(func):
+    def wrapper(self, *args, **kwargs):
+        if self.has_posterior_info:
+            return func(self, *args, **kwargs)
+        else:
+            raise ValueError("Posterior information is required to call this function.")
+    return wrapper
+
 
 class obs_sequence:
     """Create an obs_sequence object from an ascii observation sequence file.
@@ -55,22 +73,53 @@ class obs_sequence:
 
     reversed_vert = {value: key for key, value in vert.items()}
 
-    # synonyms for observation
-    synonyms_for_obs = ['NCEP BUFR observation',
-                        'AIRS observation', 
-                        'GTSPP observation', 
-                        'SST observation',
-                        'observations',
-                        'WOD observation']
     
-    def __init__(self, file):
+    def __init__(self, file, synonyms=None):
         self.loc_mod = 'None'
+        self.has_assimilation_info = False
+        self.has_posterior = False
         self.file = file
-        self.header = self.read_header(file)
+        self.synonyms_for_obs = ['NCEP BUFR observation',
+                                 'AIRS observation', 
+                                 'GTSPP observation', 
+                                 'SST observation',
+                                 'observations',
+                                 'WOD observation']
+        if synonyms:
+            if isinstance(synonyms, list):
+                self.synonyms_for_obs.extend(synonyms)
+            else:
+                self.synonyms_for_obs.append(synonyms)
+
+        if file is None:
+            # Early exit for testing purposes
+            self.df = pd.DataFrame()
+            self.types = {}
+            self.reverse_types = {}
+            self.copie_names = []
+            self.n_copies = 0
+            self.seq = []
+            self.all_obs = []
+            return
+
+        module_dir = os.path.dirname(__file__)
+        self.default_composite_types = os.path.join(module_dir,"composite_types.yaml")
+
+        if self.is_binary(file):
+            self.header = self.read_binary_header(file)
+        else:
+            self.header = self.read_header(file)
+
         self.types = self.collect_obs_types(self.header)
         self.reverse_types = {v: k for k, v in self.types.items()}
         self.copie_names, self.n_copies = self.collect_copie_names(self.header)
-        self.seq = self.obs_reader(file, self.n_copies)
+
+        if self.is_binary(file):
+            self.seq = self.obs_binary_reader(file, self.n_copies)
+            self.loc_mod = 'loc3d'  # only loc3d supported for binary, & no way to check
+        else:
+            self.seq = self.obs_reader(file, self.n_copies)
+
         self.all_obs = self.create_all_obs() # uses up the generator
         # at this point you know if the seq is loc3d or loc1d
         if self.loc_mod == 'None':
@@ -84,12 +133,16 @@ class obs_sequence:
         self.synonyms_for_obs = [synonym.replace(' ', '_') for synonym in self.synonyms_for_obs]
         rename_dict = {old: 'observation' for old in self.synonyms_for_obs  if old in self.df.columns}
         self.df = self.df.rename(columns=rename_dict)
+
         # calculate bias and sq_err is the obs_seq is an obs_seq.final
         if 'prior_ensemble_mean'.casefold() in map(str.casefold, self.columns):
-            self.df['bias'] = (self.df['prior_ensemble_mean'] - self.df['observation'])
-            self.df['sq_err'] = self.df['bias']**2  # squared error
-        module_dir = os.path.dirname(__file__)
-        self.default_composite_types = os.path.join(module_dir,"composite_types.yaml")
+            self.has_assimilation_info = True
+            self.df['prior_bias'] = (self.df['prior_ensemble_mean'] - self.df['observation'])
+            self.df['prior_sq_err'] = self.df['prior_bias']**2  # squared error
+        if 'posterior_ensemble_mean'.casefold() in map(str.casefold, self.columns):
+            self.has_posterior_info = True
+            self.df['posterior_bias'] = (self.df['posterior_ensemble_mean'] - self.df['observation'])
+            self.df['posterior_sq_err'] = self.df['posterior_bias']**2
 
     def create_all_obs(self):
         """ steps through the generator to create a
@@ -128,16 +181,44 @@ class obs_sequence:
                 raise ValueError("Neither 'loc3d' nor 'loc1d' could be found in the observation sequence.")
         typeI = obs.index('kind') # type of observation
         type_value = obs[typeI + 1]
-        data.append(self.types[type_value]) # observation type
+        if not self.types:
+            data.append('Identity')
+        else:
+            data.append(self.types[type_value]) # observation type
+            
         # any observation specific obs def info is between here and the end of the list
+        # can be obs_def & external forward operator
+        metadata = obs[typeI+2:-2]
+        obs_def_metadata, external_metadata = self.split_metadata(metadata)
+        data.append(obs_def_metadata)
+        data.append(external_metadata)
+
         time = obs[-2].split()
         data.append(int(time[0])) # seconds
         data.append(int(time[1])) # days
         data.append(convert_dart_time(int(time[0]), int(time[1]))) # datetime   # HK todo what is approprate for 1d models?
         data.append(float(obs[-1])) # obs error variance ?convert to sd?
-        
+
         return data
 
+    @staticmethod
+    def split_metadata(metadata):
+        """
+        Split the metadata list at the first occurrence of an element starting with 'externalF0'.
+
+        Args:
+            metadata (list of str): The metadata list to be split.
+
+        Returns:
+            tuple: Two sublists, the first containing elements before 'externalF0', and the second
+                containing 'externalF0' and all elements after it. If 'externalF0' is not found,
+                the first sublist contains the entire metadata list, and the second is empty.
+        """
+        for i, item in enumerate(metadata):
+            if item.startswith('external_FO'):
+                return metadata[:i], metadata[i:]
+        return metadata, []
+
     def list_to_obs(self, data):
         obs = []
         obs.append('OBS        ' + str(data[0]))  # obs_num lots of space
@@ -149,10 +230,16 @@ class obs_sequence:
             obs.append('   '.join(map(str, data[self.n_copies+2:self.n_copies+5])) + '   ' + str(self.reversed_vert[data[self.n_copies+5]]) )  # location x, y, z, vert
             obs.append('kind') # this is type of observation
             obs.append(self.reverse_types[data[self.n_copies + 6]])  # observation type
+            # Convert metadata to a string and append
+            obs.extend(data[self.n_copies + 7])  # metadata
         elif self.loc_mod == 'loc1d':
             obs.append(data[self.n_copies+2])  # 1d location
             obs.append('kind') # this is type of observation
             obs.append(self.reverse_types[data[self.n_copies + 3]])  # observation type
+            # Convert metadata to a string and append
+            metadata = ' '.join(map(str, data[self.n_copies + 4:-4]))
+            if metadata:
+                obs.append(metadata)  # metadata
         obs.append(' '.join(map(str, data[-4:-2])))  # seconds, days
         obs.append(data[-1])  # obs error variance
 
@@ -251,15 +338,83 @@ class obs_sequence:
         elif self.loc_mod == 'loc1d':
             heading.append('location')
         heading.append('type')
+        heading.append('metadata')
+        heading.append('external_FO')
         heading.append('seconds')
         heading.append('days')
         heading.append('time')
         heading.append('obs_err_var')
         return heading
 
+    @requires_assimilation_info    
+    def select_by_dart_qc(self, dart_qc):
+        """
+        Selects rows from a DataFrame based on the DART quality control flag.
+
+        Parameters:
+            df (DataFrame): A pandas DataFrame.
+            dart_qc (int): The DART quality control flag to select.
+
+        Returns:
+            DataFrame: A DataFrame containing only the rows with the specified DART quality control flag.
+
+        Raises:
+            ValueError: If the DART quality control flag is not present in the DataFrame.
+        """
+        if dart_qc not in self.df['DART_quality_control'].unique():
+            raise ValueError(f"DART quality control flag '{dart_qc}' not found in DataFrame.")
+        else:
+            return self.df[self.df['DART_quality_control'] == dart_qc]
+
+    @requires_assimilation_info
+    def select_failed_qcs(self):
+        """
+        Select rows from the DataFrame where the DART quality control flag is greater than 0.
+
+        Returns:
+            pandas.DataFrame: A DataFrame containing only the rows with a DART quality control flag greater than 0.
+        """
+        return self.df[self.df['DART_quality_control'] > 0]
+
+    @requires_assimilation_info
+    def possible_vs_used(self):
+        """
+        Calculates the count of possible vs. used observations by type.
+
+        This function takes a DataFrame containing observation data, including a 'type' column for the observation
+        type and an 'observation' column. The number of used observations ('used'), is the total number
+        minus the observations that failed quality control checks (as determined by the `select_failed_qcs` function).
+        The result is a DataFrame with each observation type, the count of possible observations, and the count of
+        used observations.
+
+        Returns:
+            pd.DataFrame: A DataFrame with three columns: 'type', 'possible', and 'used'. 'type' is the observation type,
+            'possible' is the count of all observations of that type, and 'used' is the count of observations of that type
+            that passed quality control checks.
+        """
+        possible = self.df.groupby('type')['observation'].count()
+        possible.rename('possible', inplace=True)
+        
+        failed_qcs = self.select_failed_qcs().groupby('type')['observation'].count()
+        used = possible - failed_qcs.reindex(possible.index, fill_value=0)
+        used.rename('used', inplace=True)
+        
+        return pd.concat([possible, used], axis=1).reset_index()
+
+
+    @staticmethod
+    def is_binary(file):
+        """Check if a file is binary file."""
+        with open(file, 'rb') as f:
+            chunk = f.read(1024)
+            if b'\0' in chunk:
+                return True
+        return False
+
+
     @staticmethod
     def read_header(file):
-        """Read the header and number of lines in the header of an obs_seq file"""
+        """Read the header and number of lines in the header of an ascii obs_seq file"""
         header = []
         with open(file, 'r') as f:
             for line in f:
@@ -270,6 +425,118 @@ class obs_sequence:
                     header.append(line.strip())
         return header
 
+    @staticmethod
+    def read_binary_header(file):
+        """Read the header and number of lines in the header of a binary obs_seq file from Fortran output"""
+        header = []
+        linecount = 0
+        obs_types_definitions = -1000
+        num_obs = 0
+        max_num_obs = 0 
+        # need to get:
+        #   number of obs_type_definitions
+        #   number of copies
+        #   number of qcs
+        with open(file, 'rb') as f:
+            while True:
+                # Read the record length
+                record_length = obs_sequence.read_record_length(f)
+                if record_length is None:
+                    break
+                record = f.read(record_length)
+                if not record: # end of file
+                    break
+
+                # Read the trailing record length (should match the leading one)
+                obs_sequence.check_trailing_record_length(f, record_length)
+
+                linecount += 1
+
+                if linecount == 3: 
+                    obs_types_definitions = struct.unpack('i', record)[0]
+                    continue               
+
+                if linecount == 4+obs_types_definitions:
+                    num_copies, num_qcs, num_obs, max_num_obs = struct.unpack('iiii', record)[:16]
+                    break
+            
+            # Go back to the beginning of the file
+            f.seek(0)
+            
+            for _ in range(2):
+                record_length = obs_sequence.read_record_length(f)
+                if record_length is None:
+                    break
+
+                record = f.read(record_length)
+                if not record:  # end of file
+                    break
+
+                obs_sequence.check_trailing_record_length(f, record_length)  
+                header.append(record.decode('utf-8').strip())                          
+
+            header.append(str(obs_types_definitions))
+
+            # obs_types_definitions
+            for _ in range(3,4+obs_types_definitions):
+                 # Read the record length
+                record_length = obs_sequence.read_record_length(f)
+                if record_length is None:
+                    break
+
+                # Read the actual record
+                record = f.read(record_length)
+                if not record:  # end of file
+                    break
+
+                obs_sequence.check_trailing_record_length(f, record_length) 
+
+                if _ == 3:
+                    continue # num obs_types_definitions
+                # Read an integer and a string from the record
+                integer_value = struct.unpack('i', record[:4])[0]
+                string_value = record[4:].decode('utf-8').strip()
+                header.append(f"{integer_value} {string_value}")
+
+            header.append(f"num_copies:   {num_copies}  num_qc:   {num_qcs}")
+            header.append(f"num_obs: {num_obs}  max_num_obs: {max_num_obs}")
+           
+            #copie names
+            for _ in range(5+obs_types_definitions, 5+obs_types_definitions+num_copies+num_qcs+1):
+                 # Read the record length
+                record_length = obs_sequence.read_record_length(f)
+                if record_length is None:
+                    break
+
+                # Read the actual record
+                record = f.read(record_length)
+                if not record:
+                    break
+
+                obs_sequence.check_trailing_record_length(f, record_length) 
+
+                if _ == 5+obs_types_definitions:
+                    continue
+
+                # Read the whole record as a string
+                string_value = record.decode('utf-8').strip()
+                header.append(string_value)
+
+            # first and last obs
+            # Read the record length 
+            record_length = obs_sequence.read_record_length(f)
+
+            # Read the actual record
+            record = f.read(record_length)
+ 
+            obs_sequence.check_trailing_record_length(f, record_length) 
+
+            # Read the whole record as a two integers
+            first, last = struct.unpack('ii', record)[:8]
+            header.append(f"first: {first} last: {last}")
+
+        return header
+
     @staticmethod
     def collect_obs_types(header):
         """Create a dictionary for the observation types in the obs_seq header"""
@@ -299,7 +566,7 @@ class obs_sequence:
 
     @staticmethod
     def obs_reader(file, n):
-        """Reads the obs sequence file and returns a generator of the obs"""
+        """Reads the ascii obs sequence file and returns a generator of the obs"""
         previous_line = ''
         with open(file, 'r') as f:     
             for line in f:
@@ -339,6 +606,115 @@ class obs_sequence:
                                 previous_line = next_line
                         yield obs
 
+    @staticmethod
+    def check_trailing_record_length(file, expected_length):
+            """Reads and checks the trailing record length from the binary file written by Fortran.
+
+            Parameters: 
+                file (file): The file object.
+                expected_length (int): The expected length of the trailing record.
+
+               Assuming 4 bytes:
+               | Record Length (4 bytes) | Data (N bytes) | Trailing Record Length (4 bytes) |
+            """
+            trailing_record_length_bytes = file.read(4)
+            trailing_record_length = struct.unpack('i', trailing_record_length_bytes)[0]
+            if expected_length != trailing_record_length:
+                raise ValueError("Record length mismatch in Fortran binary file")
+
+    @staticmethod
+    def read_record_length(file):
+        """Reads and unpacks the record length from the file."""
+        record_length_bytes = file.read(4)
+        if not record_length_bytes:
+            return None  # End of file
+        return struct.unpack('i', record_length_bytes)[0]
+
+
+    def obs_binary_reader(self, file, n):
+        """Reads the obs sequence binary file and returns a generator of the obs"""
+        header_length = len(self.header)
+        with open(file, 'rb') as f:
+            # Skip the first len(obs_seq.header) lines
+            for _ in range(header_length-1):
+                # Read the record length
+                record_length = obs_sequence.read_record_length(f)
+                if record_length is None: # End of file
+                    break
+
+                # Skip the actual record
+                f.seek(record_length, 1)
+
+                # Skip the trailing record length
+                f.seek(4, 1)
+
+            obs_num = 0
+            while True:
+                obs = []
+                obs_num += 1
+                obs.append(f"OBS        {obs_num}")              
+                for _ in range(n): # number of copies
+                    # Read the record length
+                    record_length = obs_sequence.read_record_length(f)
+                    if record_length is None:
+                        break
+                    # Read the actual record (copie)
+                    record = f.read(record_length)
+                    obs.append(struct.unpack('d', record)[0])
+
+                    # Read the trailing record length (should match the leading one)
+                    obs_sequence.check_trailing_record_length(f, record_length)
+            
+                # linked list info
+                record_length = obs_sequence.read_record_length(f)
+                if record_length is None:
+                    break
+
+                record = f.read(record_length)
+                int1, int2, int3 = struct.unpack('iii', record[:12])
+                linked_list_string = f"{int1:<12} {int2:<10} {int3:<12}"
+                obs.append(linked_list_string)
+
+                obs_sequence.check_trailing_record_length(f, record_length)
+
+                # location (note no location header "loc3d" or "loc1d" for binary files)
+                obs.append('loc3d')
+                record_length = obs_sequence.read_record_length(f)
+                record = f.read(record_length)
+                x,y,z,vert = struct.unpack('dddi', record[:28])
+                location_string = f"{x} {y} {z} {vert}" 
+                obs.append(location_string)            
+
+                obs_sequence.check_trailing_record_length(f, record_length)
+                
+                #   kind (type of observation) value
+                obs.append('kind')
+                record_length_bytes = f.read(4)
+                record_length = struct.unpack('i', record_length_bytes)[0]
+                record = f.read(record_length)
+                kind = f"{struct.unpack('i', record)[0]}"
+                obs.append(kind)
+ 
+                obs_sequence.check_trailing_record_length(f, record_length)
+
+                # time (seconds, days)
+                record_length = obs_sequence.read_record_length(f)
+                record = f.read(record_length)
+                seconds, days = struct.unpack('ii', record)[:8]
+                time_string = f"{seconds} {days}"
+                obs.append(time_string)
+
+                obs_sequence.check_trailing_record_length(f, record_length)
+                
+                # obs error variance
+                record_length = obs_sequence.read_record_length(f)
+                record = f.read(record_length)
+                obs.append(struct.unpack('d', record)[0])
+ 
+                obs_sequence.check_trailing_record_length(f, record_length)
+
+                yield obs
+
     def composite_types(self, composite_types='use_default'):
         """
         Set up and construct composite types for the DataFrame.
@@ -406,65 +782,6 @@ def convert_dart_time(seconds, days):
     """
     time = dt.datetime(1601,1,1) + dt.timedelta(days=days, seconds=seconds)
     return time
-    
-def select_by_dart_qc(df, dart_qc):
-    """
-    Selects rows from a DataFrame based on the DART quality control flag.
-
-    Parameters:
-        df (DataFrame): A pandas DataFrame.
-        dart_qc (int): The DART quality control flag to select.
-
-    Returns:
-        DataFrame: A DataFrame containing only the rows with the specified DART quality control flag.
-
-    Raises:
-        ValueError: If the DART quality control flag is not present in the DataFrame.
-    """
-    if dart_qc not in df['DART_quality_control'].unique():
-        raise ValueError(f"DART quality control flag '{dart_qc}' not found in DataFrame.")
-    else:
-        return df[df['DART_quality_control'] == dart_qc]
-
-def select_failed_qcs(df):
-    """
-    Selects rows from a DataFrame where the DART quality control flag is greater than 0.
-
-    Parameters:
-        df (DataFrame): A pandas DataFrame.
-
-    Returns:
-        DataFrame: A DataFrame containing only the rows with a DART quality control flag greater than 0.
-    """
-    return df[df['DART_quality_control'] > 0]
-
-def possible_vs_used(df):
-    """
-    Calculates the count of possible vs. used observations by type.
-
-    This function takes a DataFrame containing observation data, including a 'type' column for the observation
-    type and an 'observation' column. The number of used observations ('used'), is the total number
-    minus the observations that failed quality control checks (as determined by the `select_failed_qcs` function).
-    The result is a DataFrame with each observation type, the count of possible observations, and the count of
-    used observations.
-
-    Parameters:
-        df (pd.DataFrame): A DataFrame with at least two columns: 'type' for the observation type and 'observation'
-        for the observation data. It may also contain other columns required by the `select_failed_qcs` function
-        to determine failed quality control checks.
-
-    Returns:
-        pd.DataFrame: A DataFrame with three columns: 'type', 'possible', and 'used'. 'type' is the observation type,
-        'possible' is the count of all observations of that type, and 'used' is the count of observations of that type
-        that passed quality control checks.
-
-    """
-    possible = df.groupby('type')['observation'].count()
-    possible.rename('possible', inplace=True)
-    used = df.groupby('type')['observation'].count() - select_failed_qcs(df).groupby('type')['observation'].count()
-    used.rename('used', inplace=True)
-    return pd.concat([possible, used], axis=1).reset_index()
-
 
 def construct_composit(df_comp, composite, components):
     """
@@ -504,3 +821,5 @@ def construct_composit(df_comp, composite, components):
     merged_df = merged_df.drop(columns=[col for col in merged_df.columns if col.endswith('_v')])
 
     return merged_df
+
+