Improve databases with fatima request

queries/process_lte.py

@@ -198,12 +198,27 @@ def process_lte_data(file_path: str):
         df_lncel_lnbts, df_physical_db, on="Code_Sector", how="left"
     )
 
+    # Add Number of cells per band on MRBTS
+    df_id_mrbts_band = df_lncel_lnbts[["MRBTS", "band"]]
+    df_id_mrbts_band_grouped = (
+        df_id_mrbts_band.groupby(["MRBTS", "band"])
+        .size()
+        .unstack(fill_value=0)
+        .rename(columns=lambda x: f"Number of {x} cells on MRBTS")
+        .reset_index()
+    )
+
+    df_lncel_lnbts = pd.merge(
+        df_lncel_lnbts, df_id_mrbts_band_grouped, on="MRBTS", how="left"
+    )
+
     # Process LNCEL_FDD and LNCEL_TDD data
     df_lncel_fdd = dfs["LNCEL_FDD"]
     df_lncel_fdd.columns = df_lncel_fdd.columns.str.replace(r"[ ]", "", regex=True)
     df_lncel_fdd["ID_LNCEL"] = (
         df_lncel_fdd[["MRBTS", "LNBTS", "LNCEL"]].astype(str).apply("_".join, axis=1)
     )
+
     df_lncel_fdd = df_lncel_fdd[LNCEL_FDD_COLUMNS]
 
     df_lncel_tdd = dfs["LNCEL_TDD"]

queries/process_mrbts.py

@@ -66,16 +66,49 @@ def process_mrbts_data(
     lte_fdd_df: pd.DataFrame = UtilsVars.all_db_dfs[4]
     lte_tdd_df: pd.DataFrame = UtilsVars.all_db_dfs[5]
 
-    gsm_df = gsm_df[["ID_BCF", "site_name", "number_trx_per_bcf", "bcf_config_band"]]
+    gsm_df = gsm_df[
+        [
+            "ID_BCF",
+            "site_name",
+            "number_trx_per_bcf",
+            "bcf_config_band",
+            "G1800 TRX Per BCF",
+            "G900 TRX Per BCF",
+        ]
+    ]
     gsm_df = gsm_df.drop_duplicates(subset=["ID_BCF"], keep="first")
     gsm_df = gsm_df.rename(columns={"site_name": "gsm_name"})
 
-    wcdma_df = wcdma_df[["WBTS", "site_name", "wbts_config_band"]]
+    wcdma_df = wcdma_df[
+        [
+            "WBTS",
+            "site_name",
+            "wbts_config_band",
+            "Number of U2100 cells on WBTS",
+            "Number of U900 cells on WBTS",
+        ]
+    ]
     wcdma_df = wcdma_df.drop_duplicates(subset=["WBTS"], keep="first")
     wcdma_df = wcdma_df.rename(columns={"site_name": "wcdma_name"})
 
-    lte_fdd_df = lte_fdd_df[["MRBTS", "lnbts_name", "lte_config_band"]]
-    lte_tdd_df = lte_tdd_df[["MRBTS", "lnbts_name", "lte_config_band"]]
+    lte_fdd_df = lte_fdd_df[
+        [
+            "MRBTS",
+            "lnbts_name",
+            "lte_config_band",
+            "Number of L1800 cells on MRBTS",
+            "Number of L2600 cells on MRBTS",
+            "Number of L800 cells on MRBTS",
+        ]
+    ]
+    lte_tdd_df = lte_tdd_df[
+        [
+            "MRBTS",
+            "lnbts_name",
+            "lte_config_band",
+            "Number of L2300 cells on MRBTS",
+        ]
+    ]
     lte_df = pd.concat([lte_fdd_df, lte_tdd_df], ignore_index=True)
     lte_df = lte_df.drop_duplicates(subset=["MRBTS"], keep="first")
 

queries/process_small_bts.py

@@ -1,5 +1,7 @@
 import pandas as pd
 
+from utils.utils_vars import UtilsVars
+
 
 def process_small_bts_data(file_path: str):
     dfs = pd.read_excel(
@@ -15,7 +17,15 @@ def process_small_bts_data(file_path: str):
         pd.to_numeric(df_bts["code"], errors="coerce").fillna(0).astype(int)
     )
     df_bts["ID_BTS"] = df_bts[["BSC", "BCF", "BTS"]].astype(str).apply("_".join, axis=1)
+    df_bts["ID_BCF"] = df_bts[["BSC", "BCF"]].astype(str).apply("_".join, axis=1)
     df_bts["ID_MAL"] = df_bts[["BSC", "BTS"]].astype(str).apply("_".join, axis=1)
-    df_bts = df_bts[["ID_BTS", "ID_MAL", "code", "name"]]
+    df_bts["SectorId2"] = (
+        df_bts["sectorId"].map(UtilsVars.sector_mapping).fillna(df_bts["sectorId"])
+    )
+    df_bts["band"] = (
+        df_bts["frequencyBandInUse"].map(UtilsVars.gsm_band).fillna("not found")
+    )
+    df_bts = df_bts[["ID_BTS", "ID_MAL", "code", "name", "SectorId2", "band"]]
+    # print(df_bts.head())
 
     return df_bts

queries/process_trx.py

@@ -12,6 +12,8 @@ TRX_COLUMNS = [
     "number_trx_per_cell",
     "number_trx_per_bcf",
     "number_trx_per_site",
+    "G1800 TRX Per BCF",
+    "G900 TRX Per BCF",
 ]
 
 
@@ -21,11 +23,16 @@ TRX_BTS_COLUMNS = [
     "BTS",
     "TRX",
     "ID_BTS",
+    "ID_BCF",
     "number_trx_per_cell",
     "number_trx_per_bcf",
     "number_trx_per_site",
+    "G1800 TRX Per BCF",
+    "G900 TRX Per BCF",
     "code",
     "name",
+    "SectorId2",
+    "band",
     "adminState",
     "bbUnitSupportsEdge",
     "channel0Maio",
@@ -182,10 +189,35 @@ def process_trx_with_bts_name(file_path: str):
         "Signal"
     ].transform("sum")
 
-    df_trx_bts_name = df_trx_bts_name[TRX_BTS_COLUMNS]
+    # Avoir les TRX par bande et par secteur et BCF sous forme concaténée comme 5/3/3
+
+    trx_band_bcfid = (
+        df_trx_bts_name.groupby(["ID_BCF", "band", "SectorId2"])
+        .size()
+        .reset_index(name="count")
+    )
+
+    # Étape 2 : pivot avec valeurs manquantes = 0
+    pivot_df = trx_band_bcfid.pivot_table(
+        index=["ID_BCF", "band"], columns="SectorId2", values="count", fill_value=0
+    )
 
-    # UtilsVars.all_db_dfs.append(df_trx_bts_name)
+    # Étape 3 : s’assurer que toutes les colonnes de secteur (1, 2, 3) existent
+    for s in [1, 2, 3]:
+        if s not in pivot_df.columns:
+            pivot_df[s] = 0
 
+    # 🟢 Conversion en entier AVANT concaténation
+    pivot_df[[1, 2, 3]] = pivot_df[[1, 2, 3]].astype(int)
+
+    pivot_df["concat"] = pivot_df[[1, 2, 3]].astype(str).agg("/".join, axis=1)
+
+    final = pivot_df["concat"].unstack().reset_index().rename_axis(None, axis=1)
+    final.columns = ["ID_BCF", "G1800 TRX Per BCF", "G900 TRX Per BCF"]
+
+    df_trx_bts_name = pd.merge(df_trx_bts_name, final, on="ID_BCF", how="left")
+
+    df_trx_bts_name = df_trx_bts_name[TRX_BTS_COLUMNS]
     return df_trx_bts_name
 
 

queries/process_wcdma.py

@@ -161,7 +161,19 @@ def process_wcdma_data(file_path: str):
     df_wncel["code_wcel"] = (
         df_wncel[["CODE", "WNCEL"]].astype(str).apply("_".join, axis=1)
     )
+
     df_wncel = df_wncel[WNCEL_COLUMNS]
+    # Number of cell per Band per WBTS
+
+    df_id_wbts_band = df_wcel[["ID_WBTS", "band"]]
+    df_id_wbts_band_grouped = (
+        df_id_wbts_band.groupby(["ID_WBTS", "band"])
+        .size()
+        .unstack(fill_value=0)
+        .rename(columns=lambda x: f"Number of {x} cells on WBTS")
+        .reset_index()
+    )
+    print(df_id_wbts_band_grouped.head())
 
     # Merge dataframes
     df_wcel_bcf = pd.merge(df_wcel, df_wbts, on="ID_WBTS", how="left")
@@ -169,6 +181,7 @@ def process_wcdma_data(file_path: str):
     df_3g = pd.merge(df_wcel_bcf, df_wncel, on="code_wcel", how="left")
 
     df_3g = df_3g[WCEL_COLUMNS]
+    df_3g = pd.merge(df_3g, df_id_wbts_band_grouped, on="ID_WBTS", how="left")
 
     df_physical_db = get_physical_db()
     df_3g = pd.merge(df_3g, df_band, on="code", how="left")