pysdmx 1.10.0rc2__py3-none-any.whl → 1.10.1__py3-none-any.whl

pysdmx/__init__.py

@@ -1,3 +1,3 @@
 """Your opinionated Python SDMX library."""
 
-__version__ = "1.10.0rc2"
+__version__ = "1.10.1"

pysdmx/api/fmr/__init__.py

@@ -39,6 +39,7 @@ from pysdmx.model import (
     Dataflow,
     DataflowInfo,
     DataProvider,
+    DataStructureDefinition,
     Hierarchy,
     HierarchyAssociation,
     Metadataflow,
@@ -679,7 +680,7 @@ class RegistryClient(__BaseRegistryClient):
         agency: str = "*",
         id: str = "*",
         version: str = "+",
-    ) -> Sequence[Dataflow]:
+    ) -> Sequence[DataStructureDefinition]:
         """Get the data structures(s) matching the supplied parameters.
 
         Args:
@@ -1253,7 +1254,7 @@ class AsyncRegistryClient(__BaseRegistryClient):
         agency: str = "*",
         id: str = "*",
         version: str = "+",
-    ) -> Sequence[Dataflow]:
+    ) -> Sequence[DataStructureDefinition]:
         """Get the data structures(s) matching the supplied parameters.
 
         Args:

pysdmx/api/qb/data.py

@@ -329,6 +329,8 @@ class DataQuery(_CoreDataQuery, frozen=True, omit_defaults=True):
 
     def __get_short_v2_qs(self, api_version: ApiVersion) -> str:
         qs = ""
+        if self.components:
+            qs += self._create_component_filters(self.components)
         if self.updated_after:
             qs = super()._append_qs_param(
                 qs,

pysdmx/io/csv/__csv_aux_writer.py

@@ -3,8 +3,6 @@ from typing import List, Literal, Optional, Sequence
 
 import pandas as pd
 
-from pysdmx.errors import Invalid
-from pysdmx.io._pd_utils import _fill_na_values
 from pysdmx.io.pd import PandasDataset
 from pysdmx.model import Schema
 from pysdmx.model.dataset import ActionType
@@ -18,25 +16,6 @@ SDMX_CSV_ACTION_MAPPER = {
 }
 
 
-def _validate_schema_exists(dataset: PandasDataset) -> Schema:
-    """Validates that the dataset has a Schema defined.
-
-    Args:
-        dataset: The dataset to validate.
-
-    Returns:
-        The `Schema` from the dataset.
-
-    Raises:
-        Invalid: If the structure is not a `Schema`.
-    """
-    if not isinstance(dataset.structure, Schema):
-        raise Invalid(
-            "Dataset Structure is not a Schema. Cannot perform operation."
-        )
-    return dataset.structure
-
-
 def __write_time_period(df: pd.DataFrame, time_format: str) -> None:
     # TODO: Correct handle of normalized time format
     raise NotImplementedError("Normalized time format is not implemented yet.")
@@ -91,10 +70,8 @@ def _write_csv_2_aux(
 ) -> List[pd.DataFrame]:
     dataframes = []
     for dataset in datasets:
-        schema = _validate_schema_exists(dataset)
         # Create a copy of the dataset
         df: pd.DataFrame = copy(dataset.data)
-        df = _fill_na_values(df, schema)
         structure_ref, unique_id = dataset.short_urn.split("=", maxsplit=1)
 
         # Add additional attributes to the dataset

pysdmx/io/csv/sdmx10/reader/__init__.py

@@ -24,7 +24,7 @@ def read(input_str: str) -> Sequence[PandasDataset]:
     """
     # Get Dataframe from CSV file
     df_csv = pd.read_csv(
-        StringIO(input_str), keep_default_na=False, na_values=[]
+        StringIO(input_str), keep_default_na=False, na_values=[""]
     )
     # Drop empty columns
     df_csv = df_csv.dropna(axis=1, how="all")

pysdmx/io/csv/sdmx10/writer/__init__.py

@@ -6,12 +6,9 @@ from typing import Literal, Optional, Sequence, Union
 
 import pandas as pd
 
-from pysdmx.io._pd_utils import _fill_na_values
-from pysdmx.io.csv.__csv_aux_writer import (
-    __write_time_period,
-    _validate_schema_exists,
-)
+from pysdmx.io.csv.__csv_aux_writer import __write_time_period
 from pysdmx.io.pd import PandasDataset
+from pysdmx.model import Schema
 from pysdmx.toolkit.pd._data_utils import format_labels
 
 
@@ -47,26 +44,22 @@ def write(
     # Create a copy of the dataset
     dataframes = []
     for dataset in datasets:
-        # Validate that dataset has a proper Schema
-        schema = _validate_schema_exists(dataset)
-
         df: pd.DataFrame = copy(dataset.data)
 
-        # Fill missing values
-        df = _fill_na_values(df, schema)
-
         # Add additional attributes to the dataset
         for k, v in dataset.attributes.items():
             df[k] = v
         structure_id = dataset.short_urn.split("=")[1]
         if time_format is not None and time_format != "original":
             __write_time_period(df, time_format)
-        if labels is not None:
-            format_labels(df, labels, schema.components)
+        if labels is not None and isinstance(dataset.structure, Schema):
+            format_labels(df, labels, dataset.structure.components)
             if labels == "id":
                 df.insert(0, "DATAFLOW", structure_id)
             else:
-                df.insert(0, "DATAFLOW", f"{structure_id}:{schema.name}")
+                df.insert(
+                    0, "DATAFLOW", f"{structure_id}:{dataset.structure.name}"
+                )
         else:
             df.insert(0, "DATAFLOW", structure_id)
 
@@ -75,7 +68,8 @@ def write(
     # Concatenate the dataframes
     all_data = pd.concat(dataframes, ignore_index=True, axis=0)
 
-    all_data = all_data.astype(str)
+    # Ensure null values are represented as empty strings
+    all_data = all_data.astype(str).replace({"nan": "", "<NA>": ""})
     # If the output path is an empty string we use None
     output_path = (
         None

pysdmx/io/csv/sdmx20/reader/__init__.py

@@ -24,7 +24,7 @@ def read(input_str: str) -> Sequence[PandasDataset]:
     """
     # Get Dataframe from CSV file
     df_csv = pd.read_csv(
-        StringIO(input_str), keep_default_na=False, na_values=[]
+        StringIO(input_str), keep_default_na=False, na_values=[""]
     )
     # Drop empty columns
     df_csv = df_csv.dropna(axis=1, how="all")

pysdmx/io/csv/sdmx20/writer/__init__.py

@@ -60,7 +60,7 @@ def write(
 
     all_data = pd.concat(dataframes, ignore_index=True, axis=0)
 
-    all_data = all_data.astype(str)
+    all_data = all_data.astype(str).replace({"nan": "", "<NA>": ""})
 
     # If the output path is an empty string we use None
     output_path = (

pysdmx/io/csv/sdmx21/reader/__init__.py

@@ -24,7 +24,7 @@ def read(input_str: str) -> Sequence[PandasDataset]:
     """
     # Get Dataframe from CSV file
     df_csv = pd.read_csv(
-        StringIO(input_str), keep_default_na=False, na_values=[]
+        StringIO(input_str), keep_default_na=False, na_values=[""]
     )
     # Drop empty columns
     df_csv = df_csv.dropna(axis=1, how="all")

pysdmx/io/csv/sdmx21/writer/__init__.py

@@ -57,7 +57,7 @@ def write(
 
     all_data = pd.concat(dataframes, ignore_index=True, axis=0)
 
-    all_data = all_data.astype(str)
+    all_data = all_data.astype(str).replace({"nan": "", "<NA>": ""})
 
     # If the output path is an empty string we use None
     output_path = (

pysdmx/io/json/fusion/messages/dsd.py

@@ -80,8 +80,14 @@ class FusionAttribute(Struct, frozen=True):
     measureReferences: Optional[Sequence[str]] = None
 
     def __derive_level(self, groups: Sequence[FusionGroup]) -> str:
-        if self.attachmentLevel == "OBSERVATION":
-            return "O"
+        if self.measureReferences:
+            if (
+                len(self.measureReferences) == 1
+                and self.measureReferences[0] == "OBS_VALUE"
+            ):
+                return "O"
+            else:
+                return ",".join(self.measureReferences)
         elif self.attachmentLevel == "DATA_SET":
             return "D"
         elif self.attachmentLevel == "GROUP":

pysdmx/io/json/sdmxjson2/messages/dsd.py

@@ -126,7 +126,10 @@ class JsonAttributeRelationship(Struct, frozen=True, omit_defaults=True):
     ) -> str:
         """Returns the attachment level."""
         if measures:
-            return "O"
+            if len(measures) == 1 and measures[0] == "OBS_VALUE":
+                return "O"
+            else:
+                return ",".join(measures)
         elif self.dimensions:
             return ",".join(self.dimensions)
         elif self.group:
@@ -136,15 +139,17 @@ class JsonAttributeRelationship(Struct, frozen=True, omit_defaults=True):
             return "D"
 
     @classmethod
-    def from_model(self, rel: str) -> "JsonAttributeRelationship":
+    def from_model(
+        self, rel: str, has_measure_rel: bool = False
+    ) -> "JsonAttributeRelationship":
         """Converts a pysdmx attribute relationship to an SDMX-JSON one."""
         if rel == "D":
             return JsonAttributeRelationship(dataflow={})
-        elif rel == "O":
+        elif rel == "O" or has_measure_rel:
             return JsonAttributeRelationship(observation={})
         else:
-            dims = rel.split(",")
-            return JsonAttributeRelationship(dimensions=dims)
+            comps = rel.split(",")
+            return JsonAttributeRelationship(dimensions=comps)
 
 
 class JsonDimension(Struct, frozen=True, omit_defaults=True):
@@ -257,17 +262,29 @@ class JsonAttribute(Struct, frozen=True, omit_defaults=True):
         )
 
     @classmethod
-    def from_model(self, attribute: Component) -> "JsonAttribute":
+    def from_model(
+        self, attribute: Component, measures: Sequence[Component]
+    ) -> "JsonAttribute":
         """Converts a pysdmx attribute to an SDMX-JSON one."""
         concept = _get_concept_reference(attribute)
         usage = "mandatory" if attribute.required else "optional"
+        repr = _get_json_representation(attribute)
+
+        ids = attribute.attachment_level.split(",")  # type: ignore[union-attr]
+        comps = set(ids)
+        mids = {m.id for m in measures}
+        has_measure_rel = len(comps.intersection(mids)) > 0
         level = JsonAttributeRelationship.from_model(
-            attribute.attachment_level  # type: ignore[arg-type]
+            attribute.attachment_level,  # type: ignore[arg-type]
+            has_measure_rel,
         )
-        repr = _get_json_representation(attribute)
-        # The line below will need to be changed when we work on
-        # Measure Relationship (cf. issue #467)
-        mr = ["OBS_VALUE"] if attribute.attachment_level == "O" else None
+
+        if attribute.attachment_level == "O":
+            mr = ["OBS_VALUE"]
+        elif has_measure_rel:
+            mr = ids
+        else:
+            mr = None
 
         return JsonAttribute(
             id=attribute.id,
@@ -356,12 +373,14 @@ class JsonAttributes(Struct, frozen=True, omit_defaults=True):
 
     @classmethod
     def from_model(
-        self, attributes: Sequence[Component]
+        self, attributes: Sequence[Component], measures: Sequence[Component]
     ) -> Optional["JsonAttributes"]:
         """Converts a pysdmx list of attributes to an SDMX-JSON one."""
         if len(attributes) > 0:
             return JsonAttributes(
-                attributes=[JsonAttribute.from_model(a) for a in attributes]
+                attributes=[
+                    JsonAttribute.from_model(a, measures) for a in attributes
+                ]
             )
         else:
             return None
@@ -502,7 +521,9 @@ class JsonComponents(Struct, frozen=True, omit_defaults=True):
     ) -> "JsonComponents":
         """Converts a pysdmx components list to an SDMX-JSON one."""
         dimensions = JsonDimensions.from_model(components.dimensions)
-        attributes = JsonAttributes.from_model(components.attributes)
+        attributes = JsonAttributes.from_model(
+            components.attributes, components.measures
+        )
         measures = JsonMeasures.from_model(components.measures)
         if grps is None:
             groups = []

pysdmx/io/json/sdmxjson2/reader/doc_validation.py

@@ -88,6 +88,10 @@ def validate_sdmx_json(input_str: str) -> None:
                     lambda m: f"does not match required"
                     f" pattern {m.group(1)!r}",
                 ),
+                (
+                    r"\[\]\s+is\s+too\s+short",
+                    lambda _m: "[] should be non-empty",
+                ),
             ]
 
             msg: Optional[str] = next(

pysdmx/io/xml/__write_data_aux.py

@@ -26,6 +26,8 @@ def check_dimension_at_observation(
     for key, value in dimension_at_observation.items():
         if key not in datasets:
             raise Invalid(f"Dataset {key} not found in Message content.")
+        if value == ALL_DIM:
+            continue
         writing_validation(datasets[key])
         dataset = datasets[key]
         components = dataset.structure.components  # type: ignore[union-attr]
@@ -42,31 +44,20 @@ def check_dimension_at_observation(
     return dimension_at_observation
 
 
-def writing_validation(dataset: PandasDataset) -> Schema:
-    """Structural validation of the dataset.
-
-    Args:
-        dataset: The dataset to validate.
-
-    Returns:
-        The `Schema` from the dataset.
-
-    Raises:
-        Invalid: If the structure is not a `Schema` or validation fails.
-    """
+def writing_validation(dataset: PandasDataset) -> None:
+    """Structural validation of the dataset."""
     if not isinstance(dataset.structure, Schema):
         raise Invalid(
             "Dataset Structure is not a Schema. Cannot perform operation."
         )
-    schema = dataset.structure
     required_components = [
         comp.id
-        for comp in schema.components
+        for comp in dataset.structure.components
         if comp.role in (Role.DIMENSION, Role.MEASURE)
     ]
     required_components.extend(
         att.id
-        for att in schema.components.attributes
+        for att in dataset.structure.components.attributes
         if (
             att.required
             and att.attachment_level is not None
@@ -75,7 +66,7 @@ def writing_validation(dataset: PandasDataset) -> Schema:
     )
     non_required = [
         comp.id
-        for comp in schema.components
+        for comp in dataset.structure.components
         if comp.id not in required_components
     ]
     # Columns match components
@@ -91,11 +82,9 @@ def writing_validation(dataset: PandasDataset) -> Schema:
             f"Difference: {', '.join(difference)}"
         )
     # Check if the dataset has at least one dimension and one measure
-    if not schema.components.dimensions:
+    if not dataset.structure.components.dimensions:
         raise Invalid(
             "The dataset structure must have at least one dimension."
         )
-    if not schema.components.measures:
+    if not dataset.structure.components.measures:
         raise Invalid("The dataset structure must have at least one measure.")
-
-    return schema

pysdmx/io/xml/__write_structure_specific_aux.py

@@ -1,11 +1,10 @@
 # mypy: disable-error-code="union-attr"
 """Module for writing SDMX-ML 3.0 Structure Specific auxiliary functions."""
 
-from typing import Any, Dict, Hashable, List
+from typing import Any, Dict, List
 
 import pandas as pd
 
-from pysdmx.io._pd_utils import _fill_na_values
 from pysdmx.io.pd import PandasDataset
 from pysdmx.io.xml.__write_aux import (
     ABBR_MSG,
@@ -70,6 +69,9 @@ def __write_data_structure_specific(
     outfile = ""
 
     for i, (short_urn, dataset) in enumerate(datasets.items()):
+        dataset.data = dataset.data.astype(str).replace(
+            {"nan": "", "<NA>": ""}
+        )
         outfile += __write_data_single_dataset(
             dataset=dataset,
             prettyprint=prettyprint,
@@ -113,12 +115,8 @@ def __write_data_single_dataset(
     structure_urn = get_structure(dataset)
     id_structure = parse_short_urn(structure_urn).id
     sdmx_type = parse_short_urn(structure_urn).id
-
-    # Validate structure before writing
-    schema = writing_validation(dataset)
-
     # Remove nan values from DataFrame
-    dataset.data = _fill_na_values(dataset.data, schema)
+    dataset.data = dataset.data.fillna("").astype(str).replace("nan", "")
 
     nl = "\n" if prettyprint else ""
     child1 = "\t" if prettyprint else ""
@@ -141,6 +139,7 @@ def __write_data_single_dataset(
     if dim == ALL_DIM:
         data += __memory_optimization_writing(dataset, prettyprint)
     else:
+        writing_validation(dataset)
         series_codes, obs_codes, group_codes = get_codes(
             dimension_code=dim,
             structure=dataset.structure,  # type: ignore[arg-type]
@@ -231,85 +230,69 @@ def __obs_processing(data: pd.DataFrame, prettyprint: bool = True) -> str:
     return "".join(iterator)
 
 
-def __format_ser_str(
-    data_info: Dict[Any, Any], prettyprint: bool = True
+def __series_processing(
+    data: pd.DataFrame,
+    series_codes: List[str],
+    obs_codes: List[str],
+    prettyprint: bool = True,
 ) -> str:
-    """Formats the series as key=value pairs."""
-    child2 = "\t\t" if prettyprint else ""
-    child3 = "\t\t\t" if prettyprint else ""
-    nl = "\n" if prettyprint else ""
+    def __generate_series_str() -> str:
+        """Generates the series item with its observations."""
+        out_list: List[str] = []
+        data.groupby(by=series_codes)[obs_codes].apply(
+            lambda x: __format_dict_ser(out_list, x)
+        )
 
-    out_element = f"{child2}<Series "
+        return "".join(out_list)
 
-    for k, v in data_info.items():
-        if k != "Obs":
-            out_element += f"{k}={__escape_xml(str(v))!r} "
+    def __format_dict_ser(
+        output_list: List[str],
+        obs: Any,
+    ) -> Any:
+        """Formats the series as key=value pairs."""
+        # Creating the observation dict,
+        # we always get the first element on Series
+        # as we are grouping by it
+        data_dict["Series"][0]["Obs"] = obs.to_dict(orient="records")
+        output_list.append(__format_ser_str(data_dict["Series"][0]))
+        # We remove the data for series as it is no longer necessary
+        del data_dict["Series"][0]
 
-    out_element += f">{nl}"
+    def __format_ser_str(data_info: Dict[Any, Any]) -> str:
+        """Formats the series as key=value pairs."""
+        child2 = "\t\t" if prettyprint else ""
+        child3 = "\t\t\t" if prettyprint else ""
+        nl = "\n" if prettyprint else ""
 
-    for obs in data_info["Obs"]:
-        out_element += f"{child3}<Obs "
+        out_element = f"{child2}<Series "
 
-        for k, v in obs.items():
-            out_element += f"{k}={__escape_xml(str(v))!r} "
+        for k, v in data_info.items():
+            if k != "Obs":
+                out_element += f"{k}={__escape_xml(str(v))!r} "
 
-        out_element += f"/>{nl}"
+        out_element += f">{nl}"
+
+        for obs in data_info["Obs"]:
+            out_element += f"{child3}<Obs "
 
-    out_element += f"{child2}</Series>{nl}"
+            for k, v in obs.items():
+                out_element += f"{k}={__escape_xml(str(v))!r} "
 
-    return out_element
+            out_element += f"/>{nl}"
 
+        out_element += f"{child2}</Series>{nl}"
 
-def __build_series_dict(
-    data: pd.DataFrame, series_codes: List[str]
-) -> Dict[str, List[Dict[Hashable, Any]]]:
-    """Build series dictionary from data."""
-    if not series_codes:
-        return {"Series": [{}] if not data.empty else []}
-    return {
+        return out_element
+
+    # Getting each datapoint from data and creating dict
+    data = data.sort_values(series_codes, axis=0)
+    data_dict = {
         "Series": data[series_codes]
         .drop_duplicates()
         .reset_index(drop=True)
         .to_dict(orient="records")
     }
 
+    out = __generate_series_str()
 
-def __process_series_observations(
-    data: pd.DataFrame,
-    series_codes: List[str],
-    obs_codes: List[str],
-    data_dict: Dict[str, List[Dict[Hashable, Any]]],
-    prettyprint: bool = True,
-) -> str:
-    """Process series and their observations into XML string."""
-    out_list: List[str] = []
-
-    def append_series_with_obs(obs: Any) -> str:
-        """Append series with observations to output list."""
-        data_dict["Series"][0]["Obs"] = obs.to_dict(orient="records")
-        result = __format_ser_str(data_dict["Series"][0], prettyprint)
-        out_list.append(result)
-        del data_dict["Series"][0]
-        return result
-
-    if not series_codes:
-        if not data.empty:
-            append_series_with_obs(data[obs_codes])
-    else:
-        data.groupby(by=series_codes)[obs_codes].apply(append_series_with_obs)
-
-    return "".join(out_list)
-
-
-def __series_processing(
-    data: pd.DataFrame,
-    series_codes: List[str],
-    obs_codes: List[str],
-    prettyprint: bool = True,
-) -> str:
-    """Write series to SDMX-ML Structure-Specific format."""
-    data = data.sort_values(series_codes, axis=0)
-    data_dict = __build_series_dict(data, series_codes)
-    return __process_series_observations(
-        data, series_codes, obs_codes, data_dict, prettyprint
-    )
+    return out

pysdmx/io/xml/sdmx21/writer/generic.py

@@ -2,11 +2,10 @@
 """Module for writing SDMX-ML 2.1 Generic data messages."""
 
 from pathlib import Path
-from typing import Any, Dict, Hashable, List, Optional, Sequence, Tuple, Union
+from typing import Any, Dict, List, Optional, Sequence, Tuple, Union
 
 import pandas as pd
 
-from pysdmx.io._pd_utils import _fill_na_values
 from pysdmx.io.format import Format
 from pysdmx.io.pd import PandasDataset
 from pysdmx.io.xml.__write_aux import (
@@ -123,6 +122,7 @@ def __write_data_generic(
 
     for short_urn, dataset in datasets.items():
         writing_validation(dataset)
+        dataset.data = dataset.data.fillna("").astype(str)
         outfile += __write_data_single_dataset(
             dataset=dataset,
             prettyprint=prettyprint,
@@ -160,8 +160,7 @@ def __write_data_single_dataset(
     outfile = ""
     structure_urn = get_structure(dataset)
     id_structure = parse_short_urn(structure_urn).id
-    schema = writing_validation(dataset)
-    dataset.data = _fill_na_values(dataset.data, schema)
+    dataset.data = dataset.data.fillna("").astype(str).replace("nan", "")
 
     nl = "\n" if prettyprint else ""
     child1 = "\t" if prettyprint else ""
@@ -348,14 +347,9 @@ def __series_processing(
 ) -> str:
     def __generate_series_str() -> str:
         out_list: List[str] = []
-        group_cols = series_codes + series_att_codes
-        if not group_cols:
-            if not data.empty:
-                __format_dict_ser(out_list, data)
-        else:
-            data.groupby(by=group_cols)[data.columns].apply(
-                lambda x: __format_dict_ser(out_list, x)
-            )
+        data.groupby(by=series_codes + series_att_codes)[data.columns].apply(
+            lambda x: __format_dict_ser(out_list, x)
+        )
 
         return "".join(out_list)
 
@@ -365,14 +359,13 @@ def __series_processing(
     ) -> Any:
         obs_data = group_data[obs_codes + obs_att_codes].copy()
         data_dict["Series"][0]["Obs"] = obs_data.to_dict(orient="records")
-        if series_att_codes:
-            data_dict["Series"][0].update(
-                {
-                    k: v
-                    for k, v in group_data[series_att_codes].iloc[0].items()
-                    if k in series_att_codes
-                }
-            )
+        data_dict["Series"][0].update(
+            {
+                k: v
+                for k, v in group_data[series_att_codes].iloc[0].items()
+                if k in series_att_codes
+            }
+        )
         output_list.append(
             __format_ser_str(
                 data_info=data_dict["Series"][0],
@@ -387,17 +380,12 @@ def __series_processing(
 
     # Getting each datapoint from data and creating dict
     data = data.sort_values(series_codes, axis=0)
-    if not series_codes:
-        data_dict: Dict[str, List[Dict[Hashable, Any]]] = {
-            "Series": [{}] if not data.empty else []
-        }
-    else:
-        data_dict = {
-            "Series": data[series_codes]
-            .drop_duplicates()
-            .reset_index(drop=True)
-            .to_dict(orient="records")
-        }
+    data_dict = {
+        "Series": data[series_codes]
+        .drop_duplicates()
+        .reset_index(drop=True)
+        .to_dict(orient="records")
+    }
 
     out = __generate_series_str()
 

pysdmx/model/concept.py

@@ -35,8 +35,6 @@ class DataType(str, Enum):
     """True or False."""
     COUNT = "Count"
     """A simple incrementing integer type."""
-    DATA_SET_REFERENCE = "DataSetReference"
-    """Reference to a data set."""
     DATE = "GregorianDay"
     """A ISO 8601 date (e.g. ``2011-06-17``)."""
     DATE_TIME = "DateTime"
@@ -49,24 +47,12 @@ class DataType(str, Enum):
     """A decimal number (8 bytes)."""
     DURATION = "Duration"
     """An ISO 8601 duration."""
-    EXCLUSIVE_VALUE_RANGE = "ExclusiveValueRange"
-    """A range of values excluding boundaries."""
     FLOAT = "Float"
     """A decimal number (4 bytes)."""
-    GEOSPATIAL_INFORMATION = "GeospatialInformation"
-    """Geospatial data format."""
     GREGORIAN_TIME_PERIOD = "GregorianTimePeriod"
     """This is the union of YEAR, YEAR_MONTH, and DATE."""
-    IDENTIFIABLE_REFERENCE = "IdentifiableReference"
-    """Reference to an identifiable object."""
-    INCLUSIVE_VALUE_RANGE = "InclusiveValueRange"
-    """A range of values including boundaries."""
-    INCREMENTAL = "Incremental"
-    """An integer type that increases sequentially."""
     INTEGER = "Integer"
     """A whole number (4 bytes)."""
-    KEY_VALUES = "KeyValues"
-    """Key values reference."""
     LONG = "Long"
     """A whole number (8 bytes)."""
     MONTH = "Month"
@@ -99,8 +85,6 @@ class DataType(str, Enum):
     """A string (as immutable sequence of Unicode code points)."""
     TIME = "Time"
     """An ISO 8601 time (e.g. ``12:50:42``)."""
-    TIMES_RANGE = "TimesRange"
-    """A range of time periods."""
     URI = "URI"
     """A uniform resource identifier, such as a URL."""
     XHTML = "XHTML"

pysdmx/model/dataflow.py

@@ -101,10 +101,11 @@ class Component(
     one of: *D* (for Dataset), *O* (for Observation), any string identifying a
     component ID (FREQ) or comma-separated list of component IDs
     (FREQ,REF_AREA). The latter can be used to identify the dimension, group
-    or series to which the attribute is attached. The attachment level of a
-    component may vary with the statistical domain, i.e. a component attached
-    to a series in a particular domain may be attached to, say, the dataset in
-    another domain.
+    or series to which the attribute is attached. It can also be used to
+    identify the measure(s) to which the attribute relates, in case multiple
+    measures are defined. The attachment level of a component may vary with the
+    statistical domain, i.e. a component attached to a series in a particular
+    domain may be attached to, say, the dataset in another domain.
 
     The *codes* field indicates the expected (i.e. allowed) set of values a
     component can take within a particular domain. In addition to
@@ -128,7 +129,9 @@ class Component(
             Attributes can be attached at different levels such as
             D (for dataset-level attributes), O (for observation-level
             attributes) or a combination of dimension IDs, separated by
-            commas, for series- and group-level attributes).
+            commas, for series- and group-level attributes, as well as for
+            attributes attached to one or more measures, when multiple
+            measures are defined).
             A post_init check makes this attribute mandatory for attributes.
         array_def: Any additional constraints for array types.
         urn: The URN of the component.

{pysdmx-1.10.0rc2.dist-info → pysdmx-1.10.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pysdmx
-Version: 1.10.0rc2
+Version: 1.10.1
 Summary: Your opinionated Python SDMX library
 License: Apache-2.0
 License-File: LICENSE

{pysdmx-1.10.0rc2.dist-info → pysdmx-1.10.1.dist-info}/RECORD

@@ -1,5 +1,5 @@
 pysdmx/__extras_check.py,sha256=Tmluui2OuJVyJB6a1Jl0PlrRjpsswhtCjAqtRLOSero,2059
-pysdmx/__init__.py,sha256=Q03oQ7Jwt3q1RWCdX2lHnJ8s7cd-qe2_l47m90Dqdgo,71
+pysdmx/__init__.py,sha256=76qTEXwql_RGDSAgomUh23-P8FW79kx6Pcvc8iKZR8s,68
 pysdmx/api/__init__.py,sha256=8lRaF6kEO51ehl0fmW_pHLvkN_34TtEhqhr3oKo6E6g,26
 pysdmx/api/dc/__init__.py,sha256=oPU32X8CRZy4T1to9mO5KMqMwxQsVI424dPqai-I8zI,121
 pysdmx/api/dc/_api.py,sha256=poy1FYFXnF6maBGy5lpOodf32-7QQjH8PCBNDkuOXxQ,7747
@@ -10,12 +10,12 @@ pysdmx/api/dc/query/_parsing_util.py,sha256=pUc5z6sijGmJZsLilAxiPsCSRIO7l2iznzL3
 pysdmx/api/dc/query/_py_parser.py,sha256=_kVUk6Xu5jZdclng1F6eDSZS2-ok_yncI1y1q5lYpBU,1502
 pysdmx/api/dc/query/_sql_parser.py,sha256=vQjhSyt6qA4jAHchkq4XXVhNPtKjKSVLzhoPkUyhJKk,1561
 pysdmx/api/dc/query/util.py,sha256=9HALmvlgVCckaMTVG7sEFbAw_mBwfbL4K-Pac2KPSYw,915
-pysdmx/api/fmr/__init__.py,sha256=-0v6VBGt49k7-TDUMPfu7v6RHn6br10BUiMPneQjQjM,50473
+pysdmx/api/fmr/__init__.py,sha256=uXD2dOY2lJSY6CoracIh8c1_7wWiJcsQ5ebIEIbHumk,50532
 pysdmx/api/fmr/maintenance.py,sha256=AnR1beyL6nsoDM5LmDLXnRMW5JvhGHXTADf_INSNgUg,5920
 pysdmx/api/gds/__init__.py,sha256=BBl75wEdcz9zPMfk6kAGHitRW39S774VL9xifMZ_uHs,11560
 pysdmx/api/qb/__init__.py,sha256=vxdMJFFg4J_KWADrnzj_8KcU0UlwpJPdx0yiW3QJo9Y,1498
 pysdmx/api/qb/availability.py,sha256=2yPHnTXu_jACNKNhhtXMkxVkfLK1Ewa5ucGAbRxvC5o,10181
-pysdmx/api/qb/data.py,sha256=RUcSQBZYkqlrD2ehrGxUWUEAyjbfl5CSgjQ6BwN3hus,22049
+pysdmx/api/qb/data.py,sha256=hQayPQnN_ebBgIoC1xDei2uTbyv0UsoM4qyZyGaNz78,22143
 pysdmx/api/qb/gds.py,sha256=Z2KhP6m09_oWI2CbYRhlTsx8VLC-_UZaQEOEqX94SOw,4975
 pysdmx/api/qb/refmeta.py,sha256=--h0QOvaLGILT_6GBAZ2Ld5aqAELEW-PTsHNcj7YoG8,10677
 pysdmx/api/qb/registration.py,sha256=IURlmXcXQi8e-w5YXCgRNs07EQJZJ2bOdZb7M_k5iZ8,7132
@@ -25,19 +25,18 @@ pysdmx/api/qb/structure.py,sha256=0m_Fmp410Rfjdv0ehLS8ivwccwn-xfBkcIdYayu8pgg,17
 pysdmx/api/qb/util.py,sha256=at2Sb2kVltSTDc1gKiqG6HtIFhjSx-Msbe--wCvRbQI,3667
 pysdmx/errors.py,sha256=9bfujYykzfGMa1TuUOmH9QqghnZGOo556fvbKH2jFa8,3295
 pysdmx/io/__init__.py,sha256=96ZCY1PfcWp_q2Nlo2tHJAK31sH_b05v9UkbR0vGdg0,180
-pysdmx/io/_pd_utils.py,sha256=NgGhjn9clB0xD_3hsOzwXci8ix9dLe2Lt2DAZ9Tkyzw,2555
 pysdmx/io/csv/__csv_aux_reader.py,sha256=2RGzhga_VDnh2OVX_Bo-rR2hgAQxHXa7zt7-D5MVBu0,3994
-pysdmx/io/csv/__csv_aux_writer.py,sha256=4PlnMuzv_XUoJnZCS8GjqzTpjsSf79bmD9KTmTH24KE,4779
+pysdmx/io/csv/__csv_aux_writer.py,sha256=lCPl6hpyp12xhXlyimRlxxOT33RcgXDGTQZb8JguksI,4118
 pysdmx/io/csv/__init__.py,sha256=53f2rPkwILigOqArgRQOOwLk-ha6zVTe4EypIsR7K6I,107
 pysdmx/io/csv/sdmx10/__init__.py,sha256=NAAm_yodK-gzkuzewGQeYpF3f5nZmDA4vWGfT2KGTWc,38
-pysdmx/io/csv/sdmx10/reader/__init__.py,sha256=l7nK6Q-VpOhd_VFYNxvd05b8GxyNQXpI7aBPQYvy4LM,2733
-pysdmx/io/csv/sdmx10/writer/__init__.py,sha256=_6u3ANN84JW1wb-9YpgzKSXcpMdUBRmDtFK2fVG_r7I,2959
+pysdmx/io/csv/sdmx10/reader/__init__.py,sha256=uGc-sv4YXHteVQZPTdkVUkVZ6iKY7h7Fg56dw7VZ2UU,2735
+pysdmx/io/csv/sdmx10/writer/__init__.py,sha256=d-kLcP711k1nmG_D4whDxqWCzODRT7HTqk95N-jXBK8,2923
 pysdmx/io/csv/sdmx20/__init__.py,sha256=6_YCb4iuUWJRS9y0KSdf4ebNKblSlnTTzNC5c19kNk8,38
-pysdmx/io/csv/sdmx20/reader/__init__.py,sha256=DJMLkE4YKlBCxK4R38R3JXkd3pfiYtNa9HB8PIim0cQ,2844
-pysdmx/io/csv/sdmx20/writer/__init__.py,sha256=vaeaRT1qtAl3dkkzL2EeCrkpLmeL_r3ivqqakVDxAn0,2456
+pysdmx/io/csv/sdmx20/reader/__init__.py,sha256=PmVXd8QXvXej6XSDAPsIc8VptLk69NK37-wunHH7Pvc,2846
+pysdmx/io/csv/sdmx20/writer/__init__.py,sha256=puksYRzcog3wv9JGWA--6rvv9aRAn86Vsv1CyA7Em-c,2489
 pysdmx/io/csv/sdmx21/__init__.py,sha256=I3_dwi4A4if62_mwEjqbOa-F7mhoIMf0D6szpDf3W7c,38
-pysdmx/io/csv/sdmx21/reader/__init__.py,sha256=hoXkOJM8prZZ6QESG3ZjReN2P-8pGryN6CPeEYtrqjw,2864
-pysdmx/io/csv/sdmx21/writer/__init__.py,sha256=B-0ZYFxUm1cmv0-rwDaE4TmIE0fnAeojNPIXXP2ATXQ,2451
+pysdmx/io/csv/sdmx21/reader/__init__.py,sha256=J1cCkZh3klgZZWjdQ_U1zkfzT_DVzQmdreGZhN33SLs,2866
+pysdmx/io/csv/sdmx21/writer/__init__.py,sha256=CH8Nm7hqvXyN6XM_D2nJRmbKj6CJV-X1QzSF0WJrs0E,2484
 pysdmx/io/format.py,sha256=EO-PyYpiU0WswvEGA5UHokladxPezcwBUo1AJTqxp1Q,5250
 pysdmx/io/input_processor.py,sha256=P1_jKegrOyV7EaZLjLrq8fX2u1EI7gPBJoKvlBCNkP0,6967
 pysdmx/io/json/fusion/messages/__init__.py,sha256=ac2jWfjGGBcfoSutiKy68LzqwNp_clt2RzmJOaYCxL0,2142
@@ -47,7 +46,7 @@ pysdmx/io/json/fusion/messages/concept.py,sha256=m4lTyncSIriFXWWupE-zLxUVPx3Xrg5
 pysdmx/io/json/fusion/messages/constraint.py,sha256=dPkzhCWN49Y9ReSZPRFTdM6GWc0rU2BZTyFfWsqlX34,615
 pysdmx/io/json/fusion/messages/core.py,sha256=GdzF3TNUGrB0gxuaaSpk9LaYqcdy_M6L2azExZQfM0Q,4843
 pysdmx/io/json/fusion/messages/dataflow.py,sha256=lsaMPjmA-KiM51I78wrONfNHyvfBSeAll5Sp0jmTezc,2972
-pysdmx/io/json/fusion/messages/dsd.py,sha256=9QO0IzDSa3sHRTt6tvRAE1tAieugxB55tJ0KmeHf230,10136
+pysdmx/io/json/fusion/messages/dsd.py,sha256=hnVOY0GZSuYQFCZd-hcjGd_YEacUue_9S4YjpSJJigM,10341
 pysdmx/io/json/fusion/messages/map.py,sha256=TPsCFuUfk5Jhhe7CNvEoHuFNZFpHhvNiYFWeIEUx-sc,7695
 pysdmx/io/json/fusion/messages/metadataflow.py,sha256=Js4j8lUF9ZwqL7lJUrfrjk9tmBmRQPt8qxdrfnZ6R5E,1374
 pysdmx/io/json/fusion/messages/mpa.py,sha256=WvcHn7Pa_UBHxkZbfSzIxc8qeeMfTWThxgCRHuioXFY,1494
@@ -73,7 +72,7 @@ pysdmx/io/json/sdmxjson2/messages/concept.py,sha256=x7BoG6AaziZGNjxeypwy_lsFTmdH
 pysdmx/io/json/sdmxjson2/messages/constraint.py,sha256=TCg-Z6ZkZSzlhjvaZebKk1wL_CPhmJzyKjEkE4FPkAc,9055
 pysdmx/io/json/sdmxjson2/messages/core.py,sha256=qF0fscWY1yRxmN-4s2UweEWqbDMSioaUPaxlYEo8ouY,10137
 pysdmx/io/json/sdmxjson2/messages/dataflow.py,sha256=wjeq9yexTa012AtGdZsZflp3WQ6fP-3kas-gxADTFeQ,6256
-pysdmx/io/json/sdmxjson2/messages/dsd.py,sha256=CMnFZ97oXEJEIO-OnX-uZkSpPN-f5E6jbJo_q8WrYHE,19854
+pysdmx/io/json/sdmxjson2/messages/dsd.py,sha256=ZLFISJYtpP4cPQwcIjMkEGVPBE4Yy6cCKvwGw0xjFh4,20431
 pysdmx/io/json/sdmxjson2/messages/map.py,sha256=ZB7XPX6nUcu2MMHAsgwAR0nxlbEQF7YasplhlS5K9EI,16774
 pysdmx/io/json/sdmxjson2/messages/metadataflow.py,sha256=1hKCyzTEAvB_MOgmtjBObC9RVNSge7Sick6nQMwl17o,2994
 pysdmx/io/json/sdmxjson2/messages/mpa.py,sha256=ryoQCvOvd2j6gPdGOEML4bc-NXUSetuKNOfmd9Ogn2s,3050
@@ -85,7 +84,7 @@ pysdmx/io/json/sdmxjson2/messages/schema.py,sha256=JwFYjgvhK_1NN5KQIUYNb0ul4ywQh
 pysdmx/io/json/sdmxjson2/messages/structure.py,sha256=c0dyTJK49UpGvL1iLFaFLYFkT89kzvXwk65qd_j-Y1U,12738
 pysdmx/io/json/sdmxjson2/messages/vtl.py,sha256=C-JQY1_W8SrJd2lLdUGCmQO9Br1pdqdT8WmB1K4e_yY,35284
 pysdmx/io/json/sdmxjson2/reader/__init__.py,sha256=RbNnZSrGQa4OE0HBWJau9tPFSQbDklcKZaBWOzxEw4I,1629
-pysdmx/io/json/sdmxjson2/reader/doc_validation.py,sha256=PsY_VEJOuEtXj7pRgabiEbWBSWjTlK2oN-ayU0XIXC4,3680
+pysdmx/io/json/sdmxjson2/reader/doc_validation.py,sha256=Li0N5UmTKJy_FWT5vfwt7DEtp8xiYaXccDRBna_Q0Jw,3822
 pysdmx/io/json/sdmxjson2/reader/metadata.py,sha256=FT9CEWjrVfUK4vTEqs-f2ZO6jWeRRkEHCjsHMNgKQp0,1230
 pysdmx/io/json/sdmxjson2/reader/structure.py,sha256=PKHvH_lY2XJtKkg5rGbGSHyYu_raGLrBuaEy8BKr6U0,1209
 pysdmx/io/json/sdmxjson2/writer/__init__.py,sha256=DZGkas1ghei4p6SZsIQI1LPToS-d8F1Nx75MC8reT7g,270
@@ -104,8 +103,8 @@ pysdmx/io/xml/__structure_aux_reader.py,sha256=50UPOCk2XnCU4J1hQNAXiGL2n8QPXdf4z
 pysdmx/io/xml/__structure_aux_writer.py,sha256=0i08hvFw2TfRElaGAeTwfsOaOpw8XWBlZ_zWdxmLPkM,45612
 pysdmx/io/xml/__tokens.py,sha256=M0x-tgoh6_pzL_BP-MArCu3w0NO-AUS6bR-W6BIEJG8,6891
 pysdmx/io/xml/__write_aux.py,sha256=c3HgDMey8nBXyeT_yU8PWdk25bgYyX49R21fLv8CgZc,15534
-pysdmx/io/xml/__write_data_aux.py,sha256=ebuqtz97wa7scNM7cO0A_Cr40TXmRm3qqYbOjALj6wY,3547
-pysdmx/io/xml/__write_structure_specific_aux.py,sha256=zNep8HYFYnBYjOKZZa7PWRhc60rnRMmnfJh39QMaPtg,9292
+pysdmx/io/xml/__write_data_aux.py,sha256=mnJpooA4koqGbKhF-6eonT3drQ-qCuwwP-lfZLHKqVQ,3403
+pysdmx/io/xml/__write_structure_specific_aux.py,sha256=reRDVw4Xwag0ODyMzm9uOk9WJ_e1ELxAPYHSMUUDJBQ,8919
 pysdmx/io/xml/config.py,sha256=R24cczVkzkhjVLXpv-qfEm88W3_QTqVt2Qofi8IvJ5Y,93
 pysdmx/io/xml/doc_validation.py,sha256=WXDhte96VEAeZMMHJ0Y68WW8HEoOhEiOYEnbGP5Zwjw,1795
 pysdmx/io/xml/header.py,sha256=My03uhWD3AkfTwfUqiblmLIZuqd7uvIEYsOial6TClg,5971
@@ -118,7 +117,7 @@ pysdmx/io/xml/sdmx21/reader/structure_specific.py,sha256=S3-gLmaBFjBRIr25qQtlrao
 pysdmx/io/xml/sdmx21/reader/submission.py,sha256=8daiBW-sIVGaB6lYwHqJNkLI7IixMSydCK-0ZO8ri4I,1711
 pysdmx/io/xml/sdmx21/writer/__init__.py,sha256=QQGFAss26njCC4eKYxhBcI9LYm5NHuJaAJGKCrIrL80,31
 pysdmx/io/xml/sdmx21/writer/error.py,sha256=0wkX7K_n2oZNkOKg_zpl9Id82qP72Lqof-T-ZLGoZ1M,353
-pysdmx/io/xml/sdmx21/writer/generic.py,sha256=_ouKoVndG7Jj8_EAvUXPK1RbDKdKemh0kSRaDPFjTHo,16214
+pysdmx/io/xml/sdmx21/writer/generic.py,sha256=8_kUMMUiIFVdKMyhpR2LKDyfiinBohO_aL_6GZpOInY,15786
 pysdmx/io/xml/sdmx21/writer/structure.py,sha256=S3qoNgXxrakn2V4NLdL5U5mAA16XisI0PpJDuxqalFE,2084
 pysdmx/io/xml/sdmx21/writer/structure_specific.py,sha256=iXc1J-RzoKyRznvgGgdTSeUfyqZLouI8CtSq2YhGBWI,2877
 pysdmx/io/xml/sdmx30/__init__.py,sha256=8BScJFEgLy8DoUreu2RBUtxjGjKyClkKBI_Qtarbk-Y,38
@@ -140,9 +139,9 @@ pysdmx/model/__base.py,sha256=M1O-uT8RqeKZpGT09HD6ifjPl0F9ORxlRYra3fn8qCM,15182
 pysdmx/model/__init__.py,sha256=UPZtum_PF-nPPQa21Bq1doUXLCoU-yRGPh45ZXdUI9k,5493
 pysdmx/model/category.py,sha256=ksYIOGPHgZI619RhmRXZUXHP_juY9im40tWzR2yuMEc,6796
 pysdmx/model/code.py,sha256=Wu6rEXeZf_XA0aBrDXgN-3yvySAHH7SAjrWliFlmC24,12799
-pysdmx/model/concept.py,sha256=aEVUZVeS2NUzvQ0MZLjFT7iKRFHzhH6oC2vmH7kHLXI,10325
+pysdmx/model/concept.py,sha256=mQfqJdtWc10WdTKX_Mw7Znw65cN3QO-kCar9MWYeWO4,9645
 pysdmx/model/constraint.py,sha256=MwI_GLKzwkuo0BzAsgcnDeB-b7bq8qqwHNte5JjCEFA,1960
-pysdmx/model/dataflow.py,sha256=9EzGn-EDm1OQa52N5ep8VApZoj7lHWfIs-W5tnBP9FY,23954
+pysdmx/model/dataflow.py,sha256=IyVBWIKSkl6Qm5R2a6n_uPveUPsIpaYUMc_4hcwJ4Uw,24196
 pysdmx/model/dataset.py,sha256=Lbr7tYonGHD3NZUD-M9hK2puaEAluOVPG2DbkOohzMM,4861
 pysdmx/model/gds.py,sha256=QrnmI8Hn--C95gGXCeUeWwhn-Ur7DuT08Cg7oPJIEVI,4976
 pysdmx/model/map.py,sha256=9a3hl6efq_5kAYuJWkepoQOkao8Eqk17N69JGyRfxsk,17506
@@ -164,7 +163,7 @@ pysdmx/util/__init__.py,sha256=m_XWRAmVJ7F6ai4Ckrj_YuPbhg3cJZAXeZrEThrL88k,3997
 pysdmx/util/_date_pattern_map.py,sha256=IS1qONwVHbTBNIFCT0Rqbijj2a9mYvs7onXSK6GeQAQ,1620
 pysdmx/util/_model_utils.py,sha256=nQ1yWBt-tZYDios9xvRvJ7tHq6A8_RoGdY1wi7WGz2w,3793
 pysdmx/util/_net_utils.py,sha256=nOTz_x3FgFrwKh42_J70IqYXz9duQkMFJWtejZXcLHs,1326
-pysdmx-1.10.0rc2.dist-info/METADATA,sha256=ljfxmPVZvRL2pcDovZhz986lhAuMwF6mlwgm1qEoA14,4852
-pysdmx-1.10.0rc2.dist-info/WHEEL,sha256=zp0Cn7JsFoX2ATtOhtaFYIiE2rmFAD4OcMhtUki8W3U,88
-pysdmx-1.10.0rc2.dist-info/licenses/LICENSE,sha256=3XTNDPtv2RxDUNkQzn9MIWit2u7_Ob5daMLEq-4pBJs,649
-pysdmx-1.10.0rc2.dist-info/RECORD,,
+pysdmx-1.10.1.dist-info/METADATA,sha256=tXCxkMOnWoEJWTBRH4ZvzO1k20lm9cxDIneSSfQ8pzE,4849
+pysdmx-1.10.1.dist-info/WHEEL,sha256=zp0Cn7JsFoX2ATtOhtaFYIiE2rmFAD4OcMhtUki8W3U,88
+pysdmx-1.10.1.dist-info/licenses/LICENSE,sha256=3XTNDPtv2RxDUNkQzn9MIWit2u7_Ob5daMLEq-4pBJs,649
+pysdmx-1.10.1.dist-info/RECORD,,

pysdmx/io/_pd_utils.py

@@ -1,83 +0,0 @@
-import pandas as pd
-
-from pysdmx.errors import Invalid
-from pysdmx.model.concept import DataType
-from pysdmx.model.dataflow import Schema
-
-NUMERIC_TYPES = {
-    DataType.BIG_INTEGER,
-    DataType.COUNT,
-    DataType.DECIMAL,
-    DataType.DOUBLE,
-    DataType.FLOAT,
-    DataType.INCREMENTAL,
-    DataType.INTEGER,
-    DataType.LONG,
-    DataType.SHORT,
-}
-
-
-def _fill_na_values(data: pd.DataFrame, structure: Schema) -> pd.DataFrame:
-    """Fills missing values in the DataFrame based on the component type.
-
-    Numeric components are filled with "NaN".
-    Other components are filled with "#N/A".
-    If the structure does not have components,
-    all missing values are filled with "".
-
-    Args:
-        data: The DataFrame to fill.
-        structure: The structure definition (´Schema´).
-
-    Returns:
-        The DataFrame with filled missing values.
-
-    Raises:
-        Invalid: If the structure does not have components.
-    """
-    for component in structure.components:
-        if component.id in data.columns:
-            if component.dtype in NUMERIC_TYPES:
-                data[component.id] = (
-                    data[component.id].astype(object).fillna("NaN")
-                )
-            else:
-                data[component.id] = (
-                    data[component.id].astype(object).fillna("#N/A")
-                )
-
-    return data
-
-
-def _validate_explicit_null_values(
-    data: pd.DataFrame, structure: Schema
-) -> None:
-    """Validates that explicit null values are correct for the component type.
-
-    Numeric components must not contain "#N/A".
-    Non-numeric components must not contain "NaN".
-
-    Args:
-        data: The DataFrame to validate.
-        structure: The structure definition (´Schema´).
-
-    Raises:
-        Invalid: If invalid null values are found.
-    """
-    for component in structure.components:
-        if component.id in data.columns:
-            series = data[component.id].astype(str)
-            if component.dtype in NUMERIC_TYPES:
-                # Numeric: #N/A is invalid
-                if series.isin(["#N/A"]).any():
-                    raise Invalid(
-                        f"Invalid null value '#N/A' in numeric component "
-                        f"'{component.id}'."
-                    )
-            else:
-                # Non-numeric: NaN is invalid
-                if series.isin(["NaN"]).any():
-                    raise Invalid(
-                        f"Invalid null value 'NaN' in non-numeric component "
-                        f"'{component.id}'."
-                    )