diff --git a/requirements.txt b/requirements.txt
new file mode 100644
index 0000000..65a5a79
--- /dev/null
+++ b/requirements.txt
@@ -0,0 +1,4 @@
+jupyter==1.0.0 
+pandas==1.5.0
+pdf-oralia==0.3.11
+pydantic==2.6.1
diff --git a/scripts/history_stagging.py b/scripts/history_stagging.py
new file mode 100644
index 0000000..b1fc422
--- /dev/null
+++ b/scripts/history_stagging.py
@@ -0,0 +1,206 @@
+import logging
+from collections.abc import Callable
+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
+    sheet_name: str = ""
+
+
+class Flux(BaseModel):
+    sources: list[Source]
+    transformation: Callable
+    extra_kwrds: dict = {}
+
+
+def to_csv(df, dest_basename):
+    dest = dest_basename.parent / (dest_basename.name + ".csv")
+    if dest.exists():
+        df.to_csv(dest, mode="a", header=False, index=False)
+    else:
+        df.to_csv(dest, index=False)
+    return dest
+
+
+def write_split_by(
+    df: pd.DataFrame, column: str, dest_path: Path, writing_func
+) -> list[Path]:
+    wrote_files = []
+
+    for col_value in df[column].unique():
+        filtered_df = df[df[column] == col_value]
+
+        dest_basename = dest_path / f"{col_value}"
+        dest = writing_func(filtered_df, dest_basename)
+        wrote_files.append(dest)
+
+    return wrote_files
+
+
+def extract_sources(sources: list[Source], base_path: Path = Path()):
+    for src in sources:
+        filepath = base_path / src.filename
+        assert filepath.exists
+        yield src.filename, pd.read_excel(filepath, sheet_name=src.sheet_name)
+
+
+def split_duplicates(
+    df, origin: str, duplicated: dict[str, pd.DataFrame]
+) -> [pd.DataFrame, dict[str, pd.DataFrame]]:
+    duplicates = df.duplicated()
+    no_duplicates = df[~duplicates]
+    duplicated[origin] = df[duplicates]
+    return no_duplicates, duplicated
+
+
+def crg(history_path: Path, staging_path: Path, metadatas: dict, writing_func=to_csv):
+    duplicated = {}
+    wrote_files = []
+
+    for name, metadata in metadatas.items():
+        logger.debug(f"Processing {name}")
+        src_df = []
+        for filename, df in extract_sources(metadata.sources, history_path):
+            df, duplicated = split_duplicates(df, str(filename), duplicated)
+            src_df.append(df)
+
+        df = metadata.transformation(src_df, **metadata.extra_kwrds)
+
+        files = write_split_by(df, "Année", staging_path, writing_func)
+        wrote_files += files
+    return wrote_files
+
+
+def trans_2017_2021(dfs, **kwrds):
+    df, cat = dfs
+    cat_drop = list(cat[cat["Nouvelles"] == "NE PAS IMPORTER"]["Anciennes"])
+    cat_trans = cat[cat["Nouvelles"] != "NE PAS IMPORTER"]
+
+    trans = {}
+    for _, (old, new) in cat_trans.iterrows():
+        trans[old] = new
+
+    df = df[~df["Categorie"].isin(cat_drop)]
+
+    df = df.assign(
+        Immeuble=df["immeuble"],
+        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),
+        Année=df["Date"].astype(str).str.slice(0, 4),
+        Mois=df["Date"].astype(str).str.slice(5, 7),
+        Catégorie=df["Categorie"].replace(trans),
+        Fournisseur="",
+    )
+
+    return df
+
+
+def trans_2022_charge(dfs, **kwrds):
+    df = dfs[0]
+    df = df.assign(
+        Immeuble=df["immeuble"],
+        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),
+        Année=df["annee"],
+        Mois=df["mois"],
+        Catégorie=df["Catégorie Charge"],
+        # Catégorie=df["Catégorie Charge"].replace(trans),
+        Fournisseur="",
+        Régie="Oralia - Gelas",
+        Libellé="",
+    )
+    return df
+
+
+def trans_2022_loc(dfs, **kwrds):
+    df = dfs[0]
+    df = df.assign(
+        Immeuble=df["immeuble"],
+        Porte=df["lot"],
+        Débit=0,
+        Crédit=df["Réglés"].fillna(0),
+        Lot=df["immeuble"].astype(str)[0] + df["lot"].astype("str").str.zfill(2),
+        Année=df["annee"],
+        Mois=df["mois"],
+        Catégorie="Loyer Charge",
+        Fournisseur="",
+        Régie="Oralia - Gelas",
+        Libellé="",
+    )
+    return df
+
+
+def trans_2023(dfs, year, **kwrds):
+    df = dfs[0]
+    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),
+        Année=year,
+    )
+    return df
+
+
+METADATAS = {
+    "2017 2021 - charge et locataire.xlsx": Flux(
+        sources=[
+            Source(
+                filename="2017 2021 - charge et locataire.xlsx", sheet_name="DB CRG"
+            ),
+            Source(
+                filename="2017 2021 - charge et locataire.xlsx",
+                sheet_name="Catégories",
+            ),
+        ],
+        transformation=trans_2017_2021,
+    ),
+    "2022 - charge.xlsx": Flux(
+        sources=[
+            Source(filename="2022 - charge.xlsx", sheet_name="Sheet1"),
+        ],
+        transformation=trans_2022_charge,
+    ),
+    "2022 - locataire.xlsx": Flux(
+        sources=[
+            Source(filename="2022 - locataire.xlsx", sheet_name="Sheet1"),
+        ],
+        transformation=trans_2022_loc,
+    ),
+    "2023 - charge et locataire.xlsx": Flux(
+        sources=[
+            Source(
+                filename="2023 - charge et locataire.xlsx",
+                sheet_name="DB CRG 2023 ...",
+            ),
+        ],
+        transformation=trans_2023,
+        extra_kwrds={"year": 2023},
+    ),
+}
+
+if __name__ == "__main__":
+    data_path = Path("datas/")
+    assert data_path.exists()
+    history_path = data_path / "Histoire"
+    assert history_path.exists()
+    history_crg_path = history_path / "CRG"
+    assert history_crg_path.exists()
+
+    staging_path = data_path / "staging"
+    assert staging_path.exists()
+    staging_crg_path = staging_path / "CRG"
+    assert staging_crg_path.exists()
+
+    crg_files = crg(history_crg_path, staging_crg_path, METADATAS)
+    print(crg_files)