Feat: publish tag on Matrix

.drone.yml

@@ -12,7 +12,7 @@ steps:
     image: python:3.11
     commands:
       - echo ${DRONE_TAG}
-      - sed -i "s/VERSION_PLACEHOLDER/${DRONE_TAG}/g" pyproject.toml
+      - sed -i 's/version = "[^"]*"/version = "${DRONE_TAG}"/g' pyproject.toml
       - curl -sSL https://install.python-poetry.org | python3 -
       - export PATH="/root/.local/bin:$PATH"
       - poetry --version
@@ -22,6 +22,19 @@ steps:
       PYPI_TOKEN:
         from_secret: pypi_token
 
+  - name: Notify on matrix
+    image: plugins/matrix
+    environment:
+      MATRIX_ROOMID:
+        from_secret: MATRIX_ROOMID
+      MATRIX_ACCESSTOKEN: 
+        from_secret: MATRIX_ACCESSTOKEN
+      MATRIX_USERID:
+        from_secret: MATRIX_USERID
+    settings:
+      homeserver: https://matrix.poneyworld.net
+      template: "Une nouvelle version (${DRONE_TAG}) de pdf-oralia est publiée!"
+
     when:
       event:
         include:

pdf_oralia/pages/charge.py

@@ -3,7 +3,21 @@ import re
 import numpy as np
 import pandas as pd
 
-RECAPITULATIF_DES_OPERATION = 1
+RECAPITULATIF_DES_OPERATIONS = 1
+DF_TYPES = {
+    "Fournisseur": str,
+    "RECAPITULATIF DES OPERATIONS": str,
+    "Débits": float,
+    "Crédits": float,
+    "Dont T.V.A.": float,
+    "Locatif": float,
+    "Déductible": float,
+    "immeuble": str,
+    "mois": str,
+    "annee": str,
+    "lot": str,
+}
+DEFAULT_FOURNISSEUR = "ROSIER MODICA MOTTEROZ SA"
 
 
 def is_it(page_text):
@@ -17,7 +31,7 @@ def is_it(page_text):
 
 def get_lot(txt):
     """Return lot number from "RECAPITULATIF DES OPERATIONS" """
-    regex = r"[BSM](\d+)\s-"
+    regex = r"[BSM](\d+)(?=\s*-)"
     result = re.findall(regex, txt)
     if result:
         return "{:02d}".format(int(result[0]))
@@ -27,14 +41,14 @@ def get_lot(txt):
 def keep_row(row):
     return not any(
         [
-            word.lower() in row[RECAPITULATIF_DES_OPERATION].lower()
+            word.lower() in row[RECAPITULATIF_DES_OPERATIONS].lower()
             for word in ["TOTAL", "TOTAUX", "Solde créditeur", "Solde débiteur"]
         ]
     )
 
 
 def extract(table, additionnal_fields: dict = {}):
-    """Turn table to dictionary with additionnal fields"""
+    """Turn table to dictionary with additional fields"""
     extracted = []
     header = table[0]
     for row in table[1:]:
@@ -49,10 +63,8 @@ def extract(table, additionnal_fields: dict = {}):
             for k, v in additionnal_fields.items():
                 r[k] = v
 
-            r["lot"] = get_lot(row[RECAPITULATIF_DES_OPERATION])
-
-            if "honoraire" in row[RECAPITULATIF_DES_OPERATION]:
-                r["Fournisseur"] = "IMI GERANCE"
+            if "honoraire" in row[RECAPITULATIF_DES_OPERATIONS].lower():
+                r["Fournisseur"] = DEFAULT_FOURNISSEUR
 
             extracted.append(r)
 
@@ -69,4 +81,8 @@ def table2df(tables):
         )
         df["Fournisseur"] = df["Fournisseur"].fillna(method="ffill")
         dfs.append(df)
-    return pd.concat(dfs)
+    df = pd.concat(dfs)
+
+    df["immeuble"] = df["immeuble"].apply(lambda x: x[0].capitalize())
+    df["lot"] = df["RECAPITULATIF DES OPERATIONS"].apply(get_lot)
+    return df.astype(DF_TYPES)

pdf_oralia/pages/locataire.py

@@ -1,5 +1,23 @@
+import numpy as np
 import pandas as pd
 
+DF_TYPES = {
+    "Locataires": str,
+    "Période": str,
+    "Loyers": float,
+    "Taxes": float,
+    "Provisions": float,
+    "Divers": str,
+    "Total": float,
+    "Réglés": float,
+    "Impayés": float,
+    "immeuble": str,
+    "mois": str,
+    "annee": str,
+    "Lot": str,
+    "Type": str,
+}
+
 
 def is_it(page_text):
     if "SITUATION DES LOCATAIRES" in page_text:
@@ -16,7 +34,7 @@ def is_drop(row):
 
 
 def extract(table, additionnal_fields: dict = {}):
-    """Turn table to dictionary with additionnal fields"""
+    """Turn table to dictionary with additional fields"""
     extracted = []
     header = table[0]
     for row in table[1:]:
@@ -67,6 +85,12 @@ def parse_lot(string):
     return {"Lot": "{:02d}".format(int(words[1])), "Type": " ".join(words[2:])}
 
 
+def clean_type(string):
+    if "appartement" in string.lower():
+        return string[-2:]
+    return string
+
+
 def join_row(table):
     joined = []
     for row in table:
@@ -115,8 +139,6 @@ def join_row(table):
                     }
                 )
                 joined.append(row)
-            else:
-                print(row)
 
     return joined
 
@@ -131,4 +153,14 @@ def flat_tables(tables):
 def table2df(tables):
     tables = flat_tables(tables)
     joined = join_row(tables)
-    return pd.DataFrame.from_records(joined)
+    df = pd.DataFrame.from_records(joined)
+
+    df["immeuble"] = df["immeuble"].apply(lambda x: x[0].capitalize())
+    df["Type"] = df["Type"].apply(clean_type)
+
+    numeric_cols = [k for k, v in DF_TYPES.items() if v == float]
+    df[numeric_cols] = df[numeric_cols].replace("", np.nan)
+
+    df = df.drop(df[(df["Locataires"] == "") & (df["Période"] == "")].index)
+
+    return df.astype(DF_TYPES)

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "pdf-oralia"
-version = "VERSION_PLACEHOLDER"
+version = "dev"
 description = ""
 authors = ["Bertrand Benjamin <benjamin.bertrand@opytex.org>"]
 readme = "README.md"