Feat: write cli feature, datamart

requirements.txt

@@ -2,3 +2,4 @@ jupyter==1.0.0
 pandas==1.5.0
 pdf-oralia==0.3.11
 pydantic==2.6.1
+click==8.1.7

scripts/__main__.py

@@ -0,0 +1,86 @@
+import logging
+from logging.config import dictConfig
+from pathlib import Path
+
+import click
+
+from .flux import consume_fluxes
+
+DATA_PATH = Path("datas/")
+assert DATA_PATH.exists()
+HISTORY_PATH = DATA_PATH / "Histoire"
+assert HISTORY_PATH.exists()
+STAGING_PATH = DATA_PATH / "staging"
+assert STAGING_PATH.exists()
+GOLD_PATH = DATA_PATH / "gold"
+assert GOLD_PATH.exists()
+MART_PATH = DATA_PATH / "datamart"
+assert MART_PATH.exists()
+
+
+@click.group()
+@click.option("--debug/--no-debug", default=False)
+def main(debug):
+    if debug:
+        logging_level = logging.DEBUG
+    else:
+        logging_level = logging.INFO
+    logging_config = dict(
+        version=1,
+        formatters={"f": {"format": "%(levelname)-8s %(name)-12s %(message)s"}},
+        handlers={
+            "h": {
+                "class": "logging.StreamHandler",
+                "formatter": "f",
+                "level": logging_level,
+            }
+        },
+        root={
+            "handlers": ["h"],
+            "level": logging_level,
+        },
+    )
+
+    dictConfig(logging_config)
+
+
+@main.command()
+def ingest():
+    from .history_stagging import FLUXES_CRG
+
+    history_crg_path = HISTORY_PATH / "CRG"
+    assert history_crg_path.exists()
+    staging_crg_path = STAGING_PATH / "CRG"
+    assert staging_crg_path.exists()
+    consume_fluxes(
+        fluxes=FLUXES_CRG,
+        origin_path=history_crg_path,
+        dest_path=staging_crg_path,
+    )
+
+
+@main.command()
+def feature():
+    from .stagging_gold import FLUXES_CRG
+
+    staging_crg_path = STAGING_PATH / "CRG"
+    assert staging_crg_path.exists()
+    gold_crg_path = GOLD_PATH / "CRG"
+    assert gold_crg_path.exists()
+
+    consume_fluxes(
+        fluxes=FLUXES_CRG(staging_crg_path),
+        origin_path=staging_crg_path,
+        dest_path=gold_crg_path,
+    )
+
+
+@main.command()
+def datamart():
+    from .gold_mart import FLUXES_LOT
+
+    consume_fluxes(fluxes=FLUXES_LOT, origin_path=GOLD_PATH, dest_path=MART_PATH)
+
+
+if __name__ == "__main__":
+    main()

scripts/flux.py

@@ -6,9 +6,6 @@ from pathlib import Path
 import pandas as pd
 from pydantic import BaseModel
 
-logger = logging.getLogger(__name__)
-logger.setLevel(logging.DEBUG)
-
 
 class Source(BaseModel):
     filename: str
@@ -23,7 +20,7 @@ class ExcelSource(Source):
 
     def get_df(self, base_path: Path) -> pd.DataFrame:
         filepath = base_path / self.filename
-        logger.debug(f"Get content of {filepath}")
+        logging.debug(f"Get content of {filepath}")
         return pd.read_excel(filepath, sheet_name=self.sheet_name)
 
 
@@ -32,7 +29,7 @@ class CSVSource(Source):
 
     def get_df(self, base_path: Path) -> pd.DataFrame:
         filepath = base_path / self.filename
-        logger.debug(f"Get content of {filepath}")
+        logging.debug(f"Get content of {filepath}")
         return pd.read_csv(filepath, **self.options)
 
 
@@ -132,15 +129,16 @@ def consume_fluxes(
     wrote_files = []
 
     for name, flux in fluxes.items():
-        print(name)
-        logger.info(f"Processing flux {name}")
+        logging.info(f"Consume {name}")
         src_df = []
         for filename, df in extract_sources(flux.sources, origin_path):
+            logging.info(f"Extracting {filename}")
             df, duplicated = split_duplicates(df, str(filename), duplicated)
             src_df.append(df)
 
+        logging.info(f"Execute {flux.transformation.function.__name__}")
         df = flux.transformation.function(src_df, **flux.transformation.extra_kwrds)
-
         files = flux.destination.write(df, dest_path, writing_func)
+        logging.info(f"{files} written")
         wrote_files += files
     return wrote_files

scripts/gold_mart.py

@@ -18,13 +18,22 @@ logger.setLevel(logging.DEBUG)
 
 
 def build_lots(dfs: list[pd.DataFrame]) -> pd.DataFrame:
-    df = dfs[0]
+    df = pd.concat(dfs)
     df = df.assign(
         Impact=df["Crédit"] - df["Débit"],
     )
     return df
 
 
+FLUXES_LOT = {
+    "Lots": Flux(
+        sources=[CSVSource(filename="CRG/crg-*.csv")],
+        transformation=Transformation(function=build_lots),
+        destination=SplitDestination(name="Lot/lot", split_column="Lot"),
+    ),
+}
+
+
 def build_pnl(dfs: list[pd.DataFrame], year: int) -> pd.DataFrame:
     df = pd.concat(dfs)
     df = df[df["Année"] == year]
@@ -40,8 +49,10 @@ def build_pnl_flux(year: int) -> Flux:
             CSVSource(filename=f"CRG/crg-{year}.csv"),
             CSVSource(filename=f"banque/banque-{year}.csv"),
         ],
-        transformation=build_pnl,
-        extra_kwrds={"year": year},
+        transformation=Transformation(
+            function=build_pnl,
+            extra_kwrds={"year": year},
+        ),
         destination=Destination(name=f"pnl/{year}"),
     )
 
@@ -58,18 +69,15 @@ if __name__ == "__main__":
     mart_path = data_path / "datamart"
     assert mart_path.exists()
 
-    lot_fluxes = {
-        "Lots": Flux(
-            sources=[CSVSource(filename="CRG/crg-*.csv")],
-            transformation=Transformation(function=build_lots),
-            destination=SplitDestination(name="Lot/lot", split_column="Lot"),
-        ),
-    }
+    files = consume_fluxes(
+        fluxes=FLUXES_LOT, origin_path=gold_path, dest_path=mart_path
+    )
+
     years = list(range(2017, 2024))
     # pnl_fluxes = {f"pnl-{year}": build_pnl_flux(year) for year in years}
     pnl_fluxes = {}
 
     files = consume_fluxes(
-        fluxes={**lot_fluxes, **pnl_fluxes}, origin_path=gold_path, dest_path=mart_path
+        fluxes=pnl_fluxes, origin_path=gold_path, dest_path=mart_path
     )
     print(files)

scripts/history_stagging.py

@@ -22,6 +22,12 @@ def extract_cat(cat: pd.DataFrame):
     return trans, cat_drop
 
 
+def lot_naming(value):
+    if str(value).isnumeric():
+        return str(value).zfill(2)
+    return "PC"
+
+
 def trans_2017_2021(
     dfs: list[pd.DataFrame], stagging_columns: list[str], **kwrds
 ) -> pd.DataFrame:
@@ -35,7 +41,7 @@ def trans_2017_2021(
         Porte=df["porte"],
         Débit=df["Débit"].fillna(0),
         Crédit=df["Crédit"].fillna(0),
-        Lot=df["immeuble"].astype(str) + df["porte"].astype("str").str.zfill(2),
+        Lot=df["porte"].apply(lot_naming),
         Année=df["Date"].astype(str).str.slice(0, 4),
         Mois=df["Date"].astype(str).str.slice(5, 7),
         Catégorie=df["Categorie"].replace(cat_trans),
@@ -54,7 +60,7 @@ def trans_2022_charge(
         Porte=df["lot"],
         Débit=df["Débits"].fillna(0),
         Crédit=df["Crédits"].fillna(0),
-        Lot=df["immeuble"].astype(str)[0] + df["lot"].astype("str").str.zfill(2),
+        Lot=df["lot"].apply(lot_naming),
         Année=df["annee"],
         Mois=df["mois"],
         Catégorie=df["Catégorie Charge"],
@@ -75,7 +81,7 @@ def trans_2022_loc(
         Porte=df["lot"],
         Débit=0,
         Crédit=df["Réglés"].fillna(0),
-        Lot=df["immeuble"].astype(str)[0] + df["lot"].astype("str").str.zfill(2),
+        Lot=df["lot"].apply(lot_naming),
         Année=df["annee"],
         Mois=df["mois"],
         Catégorie="Loyer Charge",
@@ -93,7 +99,7 @@ def trans_2023(
     df = df.assign(
         Débit=df["Débit"].fillna(0),
         Crédit=df["Crédit"].fillna(0),
-        Lot=df["Immeuble"].astype(str) + df["Porte"].astype("str").str.zfill(2),
+        Lot=lot_naming(df["Porte"]),
         Année=year,
     )
     return df[stagging_columns]
@@ -113,7 +119,7 @@ STAGGING_COLUMNS = [
     "Crédit",
 ]
 
-FLUXES = {
+FLUXES_CRG = {
     "2017 2021 - charge et locataire.xlsx": Flux(
         sources=[
             ExcelSource(
@@ -179,7 +185,7 @@ if __name__ == "__main__":
     assert staging_crg_path.exists()
 
     crg_files = consume_fluxes(
-        fluxes=FLUXES,
+        fluxes=FLUXES_CRG,
         origin_path=history_crg_path,
         dest_path=staging_crg_path,
     )

scripts/stagging_gold.py

@@ -14,6 +14,7 @@ def feature_crg(dfs: list[pd.DataFrame]) -> pd.DataFrame:
     df = dfs[0]
     df = df.assign(
         Impact=df["Crédit"] - df["Débit"],
+        Lot=df["Immeuble"].astype(str) + df["Lot"].astype("str"),
     )
     return df
 
@@ -47,6 +48,12 @@ def build_crg_fluxes(
     return fluxes
 
 
+def FLUXES_CRG(staging_crg_path: Path):
+    return build_crg_fluxes(
+        crg_path=staging_crg_path, pattern="*.csv", transformation=feature_crg
+    )
+
+
 if __name__ == "__main__":
     data_path = Path("datas/")
     assert data_path.exists()
@@ -61,10 +68,9 @@ if __name__ == "__main__":
     gold_crg_path = gold_path / "CRG"
     assert gold_crg_path.exists()
 
-    fluxes = build_crg_fluxes(
-        crg_path=staging_crg_path, pattern="*.csv", transformation=feature_crg
-    )
     crg_files = consume_fluxes(
-        fluxes=fluxes, origin_path=staging_crg_path, dest_path=gold_crg_path
+        fluxes=FLUXES_CRG(staging_crg_path),
+        origin_path=staging_crg_path,
+        dest_path=gold_crg_path,
     )
     print(crg_files)