Skriptet dhe dokumentat

convert.py

@@ -0,0 +1,49 @@
+import pandas as pd
+from os import listdir
+from os.path import isfile, join
+from docx.api import Document
+from enum import Enum
+
+
+class Klinika(Enum):
+    KIRURGJI = 'kirurgji'
+    OBS = 'OBS-GYN'
+    URGJENCA = 'urgjenca'
+
+
+def convert(klinika: str):
+    print(klinika)
+    columns = []
+    filtered_columns = []
+    docx_path = './Word/'
+    xlsx_path = './Excel/'
+    list_files = [f for f in listdir(docx_path) if isfile(join(docx_path, f))]
+    list_paths = []
+    df = pd.DataFrame()
+    for file in list_files:
+        if klinika in file and "lock" not in file:
+            list_paths.append(join(docx_path, file))
+    for path in list_paths:
+        document = Document(path)
+        for table in document.tables:
+            for cell in table.rows[0].cells:
+                columns.append(cell.text.strip().replace('\n', ' '))
+            RowA = table.rows[0]
+            table_element = table._tbl
+            table_element.remove(RowA._tr)
+            for row in table.rows:
+                text = [cell.text for cell in row.cells]
+                df = df.append([text], ignore_index=True)
+        print(f'{path} Done')
+
+    for word in columns:
+        if word not in filtered_columns:
+            filtered_columns.append(word)
+    print(f'{len(filtered_columns)} Columns found')
+    print('writing to excel...')
+    df.columns = filtered_columns
+    excel_path = join(xlsx_path, f"raw_{klinika}_2016-2019.xlsx")
+    df.to_excel(excel_path, index=False)
+
+
+convert(Klinika.OBS.value)

llogarit_moshen.py

@@ -0,0 +1,122 @@
+import pandas as pd
+from datetime import datetime
+import calendar
+from os.path import join
+import time
+from enum import Enum
+
+
+class Klinika(Enum):
+    KIRURGJI = 'kirurgji'
+    OBS = 'OBS-GYN'
+    URGJENCA = 'urgjenca'
+
+
+klinika = Klinika.KIRURGJI.value
+
+xlsx_path = './Excel/'
+excel_path = join(xlsx_path, "raw_{klinika}_2016-2019.xlsx")
+
+df = pd.read_excel(excel_path, sheet_name='Sheet1')
+df = df.reset_index()
+df['MOSHA'] = [None] * 2384
+df['ERROR'] = [None] * 2384
+i = 0
+datat_pattern = "%d/%m/%y"
+data_shtri_korigj = []
+datelindja_korigj = []
+
+
+def llogarit_moshen(datelindja: datetime, data_shtri: datetime):
+    return data_shtri.year - datelindja.year - ((data_shtri.month, data_shtri.day) < (datelindja.month, datelindja.day))
+
+
+def konverto_data(datat: str, i: int):
+    # i +=1
+    datat_pattern = "%d/%m/%y"
+    try:
+        if "2572/81" in datat or datat == "?" or datat == "nan" or datat == "Pensionist":
+            df.iat[i, df.columns.get_loc('MOSHA')] = "ERROR"
+            df.iat[i, df.columns.get_loc('ERROR')] = datat
+            return ""
+        if "vj" in datat:
+            datat = datat.replace("vjec", "")
+            datat = datat.replace("vj", "")
+            df.iat[i, df.columns.get_loc('MOSHA')] = int(datat)
+            return "DEKLARUAR"
+        if "muaj" in datat:
+            datat = datat.replace("muaj", "")
+            datat = int(datat) / 12
+            df.iat[i, df.columns.get_loc('MOSHA')] = float(datat)
+            return "DEKLARUAR"
+        datat = datat.split('/')
+        if len(datat) == 1:
+            datat.insert(0, '1')
+            datat.insert(1, '1')
+        if datat[0] == "0":
+            datat[0] = "1"
+        if len(datat) < 3 and len(datat) > 1:
+            datat.insert(0, '1')
+        if (len(datat)) > 2:
+            if datat[2] in ['1', '2', '3', '4', '5', '6', '7', '8', '9']:
+                datat[2] = f"0{datat[2]}"
+            if int(datat[2]) > 99:
+                datat_pattern = "%d/%m/%Y"
+            # Detyrohem te bej nje supozim qe nuk ka datelindje 1919 e poshte, bazuar mbi datat e shtrimit qe jane max 2019
+            elif int(datat[2]) > 19 and int(datat[2]) < 100:
+                datat[2] = f"19{datat[2]}"
+                datat_pattern = "%d/%m/%Y"
+            else:
+                datat_pattern = "%d/%m/%y"
+        datat = "/".join(datat)
+        try:
+            datat = datetime.strptime(datat, datat_pattern)
+            return datat
+        except ValueError as e:
+            print(f"{i}: {e} -- {datat}")
+            return None
+
+    except TypeError as e:
+        print(f"{i}: {e} -- {datat}")
+        return None
+
+
+def pastro_string(datat: str, i: int):
+    datat = str(datat)
+    datat = datat.strip()
+    datat = datat.replace(" ", "")
+    datat = datat.replace(".", "/")
+    datat = datat.replace("'", "")
+    datat = konverto_data(datat, i)
+    return datat
+
+
+for index, row in df.iterrows():
+    data_shtri = row['DATA E SHTRIMIT']
+    datelindja = row['DATELINDJA']
+    emer = row['EMER']
+    data_shtri = pastro_string(data_shtri, i)
+    datelindja = pastro_string(datelindja, i)
+    if data_shtri is None or datelindja is None:
+        print(f"{i}: {data_shtri} --{emer}-- {datelindja}")
+        break
+    if type(data_shtri) is datetime and type(datelindja) is datetime:
+        mosha = llogarit_moshen(datelindja, data_shtri)
+        if mosha == 0:
+            df.iat[i, df.columns.get_loc('MOSHA')] = "ERROR"
+            df.iat[i, df.columns.get_loc('ERROR')] = mosha
+        else:
+            df.iat[i, df.columns.get_loc('MOSHA')] = mosha
+    data_shtri_korigj.append(data_shtri)
+    datelindja_korigj.append(datelindja)
+    i += 1
+
+df.insert(16, "DT_SHTRIMI_KORIGJ", data_shtri_korigj)
+df.insert(17, "DATELINDJA_KORIGJ", datelindja_korigj)
+df.drop(columns=df.columns[0], axis=1, inplace=True)
+
+current_GMT = time.gmtime()
+ts = calendar.timegm(current_GMT)
+
+excel_path = join(xlsx_path, "{klinika}_{ts}.xlsx")
+df.to_excel(excel_path, index=False)