sir3stoolkit.mantle package
Submodules
sir3stoolkit.mantle.alternative_models module
sir3stoolkit.mantle.dataframes module
Created on Fri Aug 29 09:22:31 2025
@author: Jablonski
- class sir3stoolkit.mantle.dataframes.Dataframes_SIR3S_Model[source]
Bases:
SIR3S_ModelThis class is supposed to extend the general SIR3S_Model class with the possibility of using pandas dataframes when working with SIR 3S. Getting dataframes, inserting elements via dataframes, running algorithms on dataframes should be made possible.
- class DataFrames(df_node: 'pd.DataFrame' = <factory>, df_pipe: 'pd.DataFrame' = <factory>)[source]
Bases:
object- df_node: DataFrame
- df_pipe: DataFrame
- apply_metadata_property_updates(element_type: str | Enum, updates_df: DataFrame, properties_new: List[str] | None = None, tag: str | None = '_new') DataFrame[source]
Apply metadata updates for a single property using a DataFrame with keys and new values.
Expects: - One key column (default: ‘tk’), and - Exactly one column named ‘<metadata_property>_new’ (or pass property_name to target one explicitly).
- Parameters:
element_type (Union[str, Enum]) – The element type to update (e.g., self.ObjectTypes.Pipe).
updates_df (pd.DataFrame) – Input with at least: - key column (default ‘tk’), and - one ‘<property>_new’ column (e.g., ‘diameter_mm_new’).
property_name (Optional[str], default None) – If given, we will look for the column f”{property_name}_new”. If None, we auto-detect a single ‘*_new’ column.
on (str, default 'tk') – Name of the key column in updates_df. If it’s the index, it will be reset.
create_missing (bool, default False) – If True, allow creating metadata rows that don’t exist yet (your set-logic decides how).
dry_run (bool, default False) – If True, do NOT apply; just return a normalized summary of what would be changed.
allow_na (bool, default True) – If False, rows with NaN in the ‘<property>_new’ column will be dropped (skipped).
- Returns:
A summary dataframe with columns: [‘tk’, ‘property’, ‘new_value’] (+ ‘status’ when dry_run) representing the intended updates.
- Return type:
pd.DataFrame
- generate_dataframes()[source]
Manually generate and populate all relevant dataframes.
Inlcude optional param with ElemenTypes to only create certain dataframes
- generate_element_metadata_dataframe(element_type: str | Enum, properties: List[str] | None = None, geometry: bool | None = False, end_nodes: bool | None = False) DataFrame[source]
Generate a dataframe with metadata (static) properties for all devices of a given element type.
- Parameters:
element_type (Enum or str) – The element type (e.g., self.ObjectTypes.Node).
properties (list[str], optional) – If properties=None => All available properties will be used If properties=[] => No properties will be used
geometry (bool, optional) – If True, includes geometric information for each element in the dataframe. Note that this is still a regular DataFrame no GeoDataFrame, to convert it you need a crs values. Adds a ‘geometry’ column containing spatial data (WKT represenation eg. POINT (x y)) Default is False.
end_nodes (bool, optional) – If True and supported by the element type, includes tks of end nodes as cols (fkKI, fkKK, fkKI2, fkKK2) in the dataframe. Default is False.
- Returns:
Dataframe with one row per device (tk) and columns for requested metadata properties, geometry and end nodes. Columns: [“tk”, <metadata_props>]
- Return type:
pd.DataFrame
- generate_element_results_dataframe(element_type: str | Enum, properties: List[str] | None = None, timestamps: List[str] | None = None) DataFrame[source]
Generate a dataframe with RESULT (time-dependent) properties for all devices and timestamps.
- Parameters:
element_type (Enum or str) – The element type (e.g., self.ObjectTypes.Node).
properties (list[str], optional) – List of RESULT property names (vectors) to include. If None, includes ALL available result properties.
timestamps (list[str], optional) – List of timestamps to include. If None, uses all simulation timestamps (or STAT if no sims).
- Returns:
Dataframe with one row per (timestamp, tk) and columns for requested result properties. Columns: [“timestamp”, “tk”, <result_props>]
- Return type:
pd.DataFrame