Python-Projekte/gcstruct.py

import pandas as pd
import xml.etree.ElementTree as ET
import json
import re
from bs4 import BeautifulSoup


config = ["SKR51", "Herkunft_KST", "Absatzkanal", "Kostentraeger", "Marke", "Standort", "Manuelle_Konten"]
columns = ["Konto_Nr", "Konto_Bezeichnung", "Konto_Art", "Kostenstelle", "STK", "Konto_1", "Konto_2", "Konto_3", "Konto_4", "Konto_5"]

def get_tree(node, parents):
    result = []
    for child in node:
        parents.append(child.attrib['Name'])
        result.append({ "id": ";".join(parents) + ";"*(10 - len(parents)), "text": child.attrib['Name'], "children": get_tree(child, parents) })
        parents.pop()
    return result

def structure_from_tree(node):
    result = []
    result.append(node['id'])
    for child in node['children']:
        result.extend(structure_from_tree(child))
    return result

def xml_from_tree (xml_node, tree_node):
    for child in tree_node['children']:
        element = ET.SubElement(xml_node, "Ebene")
        element.set("Name", child['text'])
        xml_from_tree(element, child)

def split_it(text, index):
    try:
        return re.findall(r"([^;]+) - ([^;]*);;", text)[0][index]
    except:
        return ""
  

def get_structure_and_tree ():
    df = pd.read_csv("gcstruct/Kontenrahmen/Kontenrahmen.csv", sep=";", encoding="ansi", decimal=",", converters={i: str for i in range(0, 200)}) #, index_col="Konto_Nr")
    for i, structure in enumerate(config):
        first = i*10 + 1
        df[structure] = df['Ebene'+str(first)] + ";" + df['Ebene'+str(first+1)] + ";" + df['Ebene'+str(first+2)] + ";" + df['Ebene'+str(first+3)] + ";" + df['Ebene'+str(first+4)] + ";" + df['Ebene'+str(first+5)] + ";" + df['Ebene'+str(first+6)] + ";" + df['Ebene'+str(first+7)] + ";" + df['Ebene'+str(first+8)] + ";" + df['Ebene'+str(first+9)]
        df['LetzteEbene' + str(i+1) + '_Nr'] = df[structure].apply(lambda x: split_it(x, 0))
        df['LetzteEbene' + str(i+1) + '_Bez'] = df[structure].apply(lambda x: split_it(x, 1))

    df = df[columns + config + ['LetzteEbene' + str(i+1) + '_Nr' for i in range(len(config))] + ['LetzteEbene' + str(i+1) + '_Bez' for i in range(len(config))]]
    json_result = { "Kontenrahmen": df.to_dict("records") }

    # df2 = pd.read_csv("gcstruct/Strukturen/Kontenrahmen.csv/SKR51.csv", sep=";", encoding="ansi", decimal=",", converters={i: str for i in range(0, 200)})
    # print(df2.head())

    for i, structure in enumerate(config):
        tree = ET.parse("gcstruct/Xml/" + structure + ".xml")
        json_result[structure] = get_tree(tree.getroot(), [])
        
    json.dump(json_result, open("gcstruct/SKR51.json", "w"), indent=2)

def post_structure_and_tree ():
    json_post = json.load(open("gcstruct/SKR51.json", "r"))

    # Kontenrahmen.csv
    ebenen = ["Ebene" + str(i) for i in range(1, len(config)*10+1)]
    header = ";".join(columns + ebenen)
    cols = columns + config
    with open("gcstruct/Kontenrahmen/Kontenrahmen_out.csv", "w", encoding="ansi") as f:
        f.write(header + "\n")
        for row in json_post['Kontenrahmen']:
            f.write(";".join([row[e] for e in cols]) + "\n")
    # print(header)
    # xml und evtl. Struktur.csv
    for i, structure in enumerate(config):
        with open("gcstruct/Strukturen/Kontenrahmen.csv/" + structure + "_out.csv", "w", encoding="ansi") as f:
            f.write(";".join(["Ebene" + str(i*10 + j) for j in range(1, 11)]) + "\n")
            rows = structure_from_tree({"id": ";"*9, "children": json_post[structure] })
            f.write("\n".join(rows))

        #with open("gcstruct/Strukturen/Kontenrahmen.csv/" + structure + "_2.csv", "w", encoding="ansi") as f:
        root = ET.Element("Ebene")
        root.set("Name", structure)
        xml_from_tree(root, {"id": ";"*9, "children": json_post[structure] })

        with open("gcstruct/Xml/" + structure + "_out.xml", "w", encoding="utf-8") as f:
            f.write(BeautifulSoup(ET.tostring(root), "xml").prettify())

get_structure_and_tree()
# post_structure_and_tree()