Added an experimental filter for minimally networked persons

Jupyter/NetworkX/archive_prod/ui_elements.py

@@ -1,4 +1,5 @@
 """Dash elements."""
+import os
 
 import pandas as pd
 import plotly.graph_objs as go
@@ -60,7 +61,7 @@ def get_finance_data(session: Session) -> pd.DataFrame:
 
 
 @cached(  # type: ignore
-    cache=TTLCache(maxsize=1, ttl=300),
+    cache=TTLCache(maxsize=1, ttl=int(os.getenv("PYTHON_CACHE_TTL", "3600"))),
     key=lambda session: 0 if session is None else str(session.bind),
 )
 def get_options(session: Session | None) -> dict[int, str]:

src/aki_prj23_transparenzregister/ui/data_elements.py

@@ -106,7 +106,7 @@ def get_finance_data_of_one_company(session: Session, company_id: int) -> pd.Dat
 
 
 @cached(  # type: ignore
-    cache=TTLCache(maxsize=1, ttl=300),
+    cache=TTLCache(maxsize=1, ttl=int(os.getenv("PYTHON_CACHE_TTL", "3600"))),
     key=lambda session: 0 if session is None else str(session.bind),
 )
 def get_options(session: Session | None) -> dict[int, str]:

src/aki_prj23_transparenzregister/ui/pages/home.py

@@ -56,7 +56,7 @@ dash.register_page(
 
 def person_relation_type_filter() -> list[str]:
     """Returns a Numpy Array of String with Person relation types."""
-    return get_all_person_relations()["relation_type"].unique().tolist()
+    return get_all_person_relations(False)["relation_type"].unique().tolist()
 
 
 def company_relation_type_filter() -> list[str]:
@@ -408,9 +408,10 @@ def layout() -> html:
 def update_graph_data(
     person_relation_type: frozenset[str] | None = None,
     company_relation_type: frozenset[str] | None = None,
+    drop_min_person_links: bool = False,
 ) -> tuple[nx.Graph, pd.DataFrame, dict, list]:
     """_summary_."""
-    person_df = get_all_person_relations()
+    person_df = get_all_person_relations(drop_min_person_links)
     company_df = get_all_company_relations()
 
     person_relation = filter_relation_type(person_df, person_relation_type)

src/aki_prj23_transparenzregister/utils/networkx/networkx_data.py

@@ -1,8 +1,10 @@
 """Module to receive and filter Data for working with NetworkX."""
+import os
 from functools import lru_cache
 
 import networkx as nx
 import pandas as pd
+from cachetools import TTLCache, cached
 from sqlalchemy.orm import aliased
 
 from aki_prj23_transparenzregister.ui.session_handler import SessionHandler
@@ -24,6 +26,10 @@ COLOR_COMPANY = "#006250"
 COLOR_PERSON = "#ff5200"
 
 
+@cached(  # type: ignore
+    cache=TTLCache(maxsize=100, ttl=int(os.getenv("PYTHON_CACHE_TTL", "3600"))),
+    key=lambda session, company_id: hash((company_id, str(session.bind))),
+)
 def find_all_company_relations() -> pd.DataFrame:
     """_summary_.
 
@@ -64,7 +70,7 @@ def find_all_company_relations() -> pd.DataFrame:
 
 
 def get_all_company_relations() -> pd.DataFrame:
-    """This Methods makes a Database Request for all Companies and their relations, modifies the ID Column and returns the Result as an DataFrame.
+    """Makes a Database Request for all Companies and their relations, modifies the ID Column and returns the Result as an DataFrame.
 
     Returns:
         DataFrame: DataFrame with all Relations between Companies.
@@ -102,7 +108,11 @@ def get_all_company_relations() -> pd.DataFrame:
     return company_relations
 
 
-def get_all_person_relations() -> pd.DataFrame:
+@cached(  # type: ignore
+    cache=TTLCache(maxsize=100, ttl=int(os.getenv("PYTHON_CACHE_TTL", "3600"))),
+    key=lambda session, company_id: hash((company_id, str(session.bind))),
+)
+def get_all_person_relations(drop_min_person_links: bool) -> pd.DataFrame:
     """These method makes a Database Request for all Persons and their relations, modifies the ID Column and returns the Result as an DataFrame.
 
     Returns:
@@ -137,7 +147,8 @@ def get_all_person_relations() -> pd.DataFrame:
     person_relations["id_person"] = person_relations["id_person"].apply(
         lambda x: f"p_{x}"
     )
-
+    if drop_min_person_links:
+        return person_relations.groupby("id_person").filter(lambda x: len(x) > 1)
     return person_relations
 
 
@@ -379,7 +390,7 @@ def get_all_metrics_from_id(company_id: int) -> pd.Series:
         pd.DataFrame: _description_
     """
     # Get Data
-    person_df = get_all_person_relations()
+    person_df = get_all_person_relations(False)
     company_df = get_all_company_relations()
 
     # Create Edge and Node List from data
@@ -416,7 +427,7 @@ def get_relations_number_from_id(id: str) -> tuple[int, int, int]:
         tuple[int,int,int]: _description_
     """
     # Get Data
-    person_df = get_all_person_relations()
+    person_df = get_all_person_relations(False)
     company_df = get_all_company_relations()
 
     # Create Edge and Node List from data
@@ -457,7 +468,7 @@ def get_relations_until_level_3(id: str) -> tuple[dict, list]:
        tuple[dict, list]: nodes, edges
     """
     # Get Data
-    person_df = get_all_person_relations()
+    person_df = get_all_person_relations(False)
     company_df = get_all_company_relations()
 
     # Create Edge and Node List from data

src/aki_prj23_transparenzregister/utils/transfer_news.py

@@ -58,7 +58,7 @@ def extract_domain(url: str) -> str:
     return urlparse(url).netloc
 
 
-@cached(cache=TTLCache(5000, ttl=300), key=lambda ner_name, _: hash(ner_name))  # type: ignore
+@cached(cache=TTLCache(5000, ttl=int(os.getenv("PYTHON_CACHE_TTL", "3600"))), key=lambda ner_name, _: hash(ner_name))  # type: ignore
 def get_company_id(ner_name: str, db: Session) -> int | None:
     """Finds the company id to an entity found via NER.