994 lines
55 KiB
Python
994 lines
55 KiB
Python
import os
|
||
import io
|
||
import streamlit as st
|
||
import json
|
||
import re
|
||
import xml.etree.ElementTree as ET
|
||
from xml.dom import minidom
|
||
from datetime import datetime, timedelta
|
||
from decimal import Decimal, ROUND_HALF_UP
|
||
import fitz
|
||
import requests
|
||
from pydantic import BaseModel, Field
|
||
import pikepdf
|
||
from facturx import generate_from_binary as facturx_generate_from_binary
|
||
|
||
# ==========================================
|
||
# 0. SITZUNGS-STATUS INITIALISIEREN (ZÄHLER GEGEN WIDGET-CACHING)
|
||
# ==========================================
|
||
if "invoice_counter" not in st.session_state:
|
||
st.session_state.invoice_counter = 0
|
||
|
||
# ==========================================
|
||
# 1. API-CLIENT FÜR DAS vLLM-BACKEND
|
||
# ==========================================
|
||
class LLMClient:
|
||
|
||
def __init__(self, model="gemma431b", base_url="http://10.192.0.140:8002"):
|
||
self.model = model
|
||
|
||
base_url = base_url.strip()
|
||
if base_url.startswith("[") and ")" in base_url:
|
||
match = re.search(r"\((https?://.*?)\)", base_url)
|
||
if match:
|
||
base_url = match.group(1)
|
||
|
||
self.base_url = base_url.rstrip("/")
|
||
|
||
def chat(self, prompt, system_prompt="You are a helpful assistant.", temperature=0.0, json_schema=None):
|
||
messages = [
|
||
{"role": "system", "content": system_prompt},
|
||
{"role": "user", "content": prompt}
|
||
]
|
||
return self._openai_compatible_request(messages, temperature, json_schema)
|
||
|
||
def _openai_compatible_request(self, messages, temperature, json_schema):
|
||
url = f"{self.base_url}/v1/chat/completions"
|
||
payload = {
|
||
"model": self.model,
|
||
"messages": messages,
|
||
"temperature": temperature
|
||
}
|
||
|
||
if json_schema:
|
||
payload["guided_json"] = json.dumps(json_schema) if isinstance(json_schema, dict) else json_schema
|
||
|
||
try:
|
||
response = requests.post(url, json=payload, timeout=120)
|
||
response.raise_for_status()
|
||
data = response.json()
|
||
return data.get("choices", [{}])[0].get("message", {}).get("content", "")
|
||
except Exception as e:
|
||
return f"vLLM/OpenAI Error: {e}"
|
||
|
||
|
||
# ==========================================
|
||
# 2. METADATEN SCHREIBSCHEMA
|
||
# ==========================================
|
||
class InvoiceMetadata(BaseModel):
|
||
invoice_number: str = Field(description="Die Rechnungsnummer nach 'Nummer:' (z.B. R26-219)")
|
||
invoice_date: str = Field(description="Das Rechnungsdatum nach 'Datum:' im Format YYYY-MM-DD")
|
||
delivery_date: str = Field(description="Das Lieferdatum nach 'Lieferdatum:' im Format YYYY-MM-DD")
|
||
customer_number: str = Field(default="", description="Kundennummer, wie sie beim Empfänger hinterlegt ist (z.B. der exakte Wert nach 'Kunden-Nr:'). Wird als Käuferreferenz (BT-10) verwendet. Falls im Text keine Kundennummer erkennbar ist: leer lassen.")
|
||
buyer_reference: str = Field(default="", description="Bestellnummer des KÄUFERS für diesen Auftrag (BT-13): der exakte Wert nach 'Ihr Auftrag:' o.ä. (z.B. 05.06.2026-Herr.Knospe). NICHT die Kundennummer (die gehört in customer_number) und NICHT unsere eigene Auftragsbestätigungsnummer (die gehört in order_id). Wenn im Text keine solche Käufer-Bestellnummer erkennbar ist: leer lassen.")
|
||
recipient_name: str = Field(description="Vollständige Firma des Kunden/Empfängers (z.B. Stadtwerke Hennigsdorf GmbH)")
|
||
recipient_street: str = Field(description="Straße und Hausnummer des Kunden")
|
||
recipient_zip: str = Field(description="Postleitzahl (PLZ) des Kunden")
|
||
recipient_city: str = Field(description="Stadt des Kunden")
|
||
recipient_email: str = Field(description="E-Mail-Adresse des Kunden (LEER lassen, wenn nicht explizit im Text!)")
|
||
vat_rate_percent: float = Field(description="Der primäre Umsatzsteuersatz in Prozent (z.B. 19.0)")
|
||
|
||
delivery_note_id: str = Field(default="", description="Gefundene Lieferscheinnummer (z.B. LS26-035) VOR Position 1. Sonst leer lassen.")
|
||
order_id: str = Field(default="", description="Gefundene EIGENE Auftragsbestätigungs-/Verkaufsauftragsnummer (BT-14, z.B. AB26-080) VOR Position 1. Das ist unsere eigene Nummer, NICHT die Bestellnummer des Käufers. Sonst leer lassen.")
|
||
contract_id: str = Field(default="", description="Gefundene Nummer von Vergabeangebot, Vertrag oder Ausschreibung (z.B. AN26-130) VOR Position 1. Sonst leer lassen.")
|
||
service_start_date: str = Field(default="", description="Startdatum eines explizit genannten Einsatz- oder Leistungszeitraums VOR Position 1 im Format YYYY-MM-DD. Sonst leer.")
|
||
service_end_date: str = Field(default="", description="Enddatum eines explizit genannten Einsatz- oder Leistungszeitraums VOR Position 1 im Format YYYY-MM-DD. Sonst leer.")
|
||
invoice_note: str = Field(default="", description="Sämtlicher unstrukturierter Fließtext, Beschreibungen, Abnahmekriterien (Acceptance Criteria) oder Projektklauseln, die sich VOR Position 1 befinden. Vollständig mit Zeilenumbrüchen sichern.")
|
||
|
||
|
||
# ==========================================
|
||
# 3. MATHEMATISCHE & PARSING-FUNKTIONEN
|
||
# ==========================================
|
||
|
||
# Zentrale Geschäftskonstanten - hier zentral pflegen statt an mehreren Stellen im Code.
|
||
ZAHLUNGSZIEL_TAGE = 14
|
||
ZAHLUNGSBEDINGUNG_TEXT = f"Zahlbar innerhalb {ZAHLUNGSZIEL_TAGE} Tagen ohne Abzug."
|
||
VERKAEUFER_IBAN = "DE06120700240456207000"
|
||
|
||
|
||
def berechne_faelligkeit(invoice_date_str):
|
||
"""Fälligkeitsdatum = Rechnungsdatum + ZAHLUNGSZIEL_TAGE. Gibt 'YYYY-MM-DD' zurück.
|
||
Fällt bei unlesbarem Rechnungsdatum auf heute+Ziel zurück (nur als letzter Notnagel)."""
|
||
base = parse_to_strict_xml_date(invoice_date_str)
|
||
try:
|
||
base_date = datetime.strptime(base, "%Y-%m-%d")
|
||
except (ValueError, TypeError):
|
||
base_date = datetime.today()
|
||
return (base_date + timedelta(days=ZAHLUNGSZIEL_TAGE)).strftime("%Y-%m-%d")
|
||
|
||
|
||
def kaufmaennisches_runden(betrag, nachkommastellen=2):
|
||
if not betrag:
|
||
return 0.0
|
||
quantizer = Decimal('1.' + '0' * nachkommastellen)
|
||
return float(Decimal(f"{betrag:.6f}").quantize(quantizer, rounding=ROUND_HALF_UP))
|
||
|
||
def clean_german_float(val_str):
|
||
if not val_str:
|
||
return 0.0
|
||
s = str(val_str).strip().replace("€", "").replace(" ", "").replace("'", "").replace("\n", "").replace("\r", "")
|
||
if not s:
|
||
return 0.0
|
||
|
||
if "." in s and "," in s:
|
||
s = s.replace(".", "").replace(",", ".")
|
||
elif "," in s:
|
||
s = s.replace(",", ".")
|
||
try:
|
||
return float(s)
|
||
except ValueError:
|
||
return 0.0
|
||
|
||
def extract_invoice_items_pure_python(text):
|
||
items = []
|
||
lines = [line.strip() for line in text.split("\n") if line.strip()]
|
||
|
||
unit_pattern = re.compile(r"^(\d+(?:[\.,]\d+)?)\s*(Tag\(e\)|Std\.|Stk\.|Stunden|Stück|stk|stck|stück)$", re.IGNORECASE)
|
||
price_stop_pattern = re.compile(r"^[\d\.]+,[\d]{2}$")
|
||
|
||
for i, line in enumerate(lines):
|
||
match = unit_pattern.match(line)
|
||
if match:
|
||
raw_qty = match.group(1)
|
||
raw_unit = match.group(2)
|
||
|
||
raw_price = "0,00"
|
||
for k in range(1, 3):
|
||
if i + k < len(lines):
|
||
candidate = lines[i+k].replace("€", "").replace(" ", "").strip()
|
||
if re.match(r"^[\d.,]+$", candidate):
|
||
raw_price = candidate
|
||
break
|
||
|
||
desc_lines = []
|
||
for j in range(i - 1, -1, -1):
|
||
prev_line = lines[j]
|
||
if unit_pattern.match(prev_line) or price_stop_pattern.match(prev_line):
|
||
break
|
||
|
||
prev_lower = prev_line.lower()
|
||
if prev_lower in ["pos.", "pos", "bezeichnung", "beschreibung", "menge", "einheit", "einzelpreis", "gesamtpreis", "betrag €", "betrag", "artikel"]:
|
||
break
|
||
if any(prev_lower.startswith(w) for w in ["zwischensumme", "endbetrag", "gesamtbetrag", "umsatzsteuer"]):
|
||
break
|
||
|
||
if prev_line.isdigit() and len(prev_line) <= 2:
|
||
break
|
||
|
||
desc_lines.insert(0, prev_line)
|
||
|
||
if desc_lines:
|
||
item_name = desc_lines[0].strip()
|
||
item_description = " ".join(desc_lines[1:]).strip() if len(desc_lines) > 1 else ""
|
||
|
||
items.append({
|
||
"name": item_name,
|
||
"description": item_description,
|
||
"quantity": clean_german_float(raw_qty),
|
||
"unit": raw_unit,
|
||
"unit_price": clean_german_float(raw_price)
|
||
})
|
||
|
||
return items
|
||
|
||
def map_unece_unit_code(unit_str):
|
||
u = str(unit_str).lower()
|
||
if "tag" in u or "day" in u: return "DAY"
|
||
elif "std" in u or "stund" in u or "hour" in u or "hur" in u: return "HUR"
|
||
return "C62"
|
||
|
||
def parse_to_strict_xml_date(date_str):
|
||
"""Normalisiert ein Datum auf 'YYYY-MM-DD'. Gibt leeren String zurück, wenn das Datum
|
||
fehlt ODER nicht geparst werden kann - NICHT still auf 'heute' fallen, damit fehlerhafte
|
||
Datumsangaben sichtbar werden statt unbemerkt durchzurutschen."""
|
||
if not date_str or str(date_str).strip() == "":
|
||
return ""
|
||
clean_str = str(date_str).strip()
|
||
for fmt in ("%Y-%m-%d", "%d.%m.%Y", "%d-%m-%Y"):
|
||
try: return datetime.strptime(clean_str, fmt).strftime("%Y-%m-%d")
|
||
except ValueError: continue
|
||
return ""
|
||
|
||
def format_cii_date(date_str):
|
||
clean = parse_to_strict_xml_date(date_str)
|
||
return clean.replace("-", "")
|
||
|
||
|
||
# ==========================================
|
||
# GENERATOR A: XRECHNUNG (UBL FORMAT)
|
||
# ==========================================
|
||
def generate_compliant_xrechnung(data):
|
||
INVOICE_NS = "urn:oasis:names:specification:ubl:schema:xsd:Invoice-2"
|
||
CBC = "urn:oasis:names:specification:ubl:schema:xsd:CommonBasicComponents-2"
|
||
CAC = "urn:oasis:names:specification:ubl:schema:xsd:CommonAggregateComponents-2"
|
||
|
||
ET.register_namespace('ubl', INVOICE_NS)
|
||
ET.register_namespace('cbc', CBC)
|
||
ET.register_namespace('cac', CAC)
|
||
|
||
root = ET.Element(f"{{{INVOICE_NS}}}Invoice")
|
||
|
||
ET.SubElement(root, f"{{{CBC}}}CustomizationID").text = "urn:cen.eu:en16931:2017#compliant#urn:xeinkauf.de:kosit:xrechnung_3.0"
|
||
ET.SubElement(root, f"{{{CBC}}}ProfileID").text = "urn:fdc:peppol.eu:2017:poacc:billing:01:1.0"
|
||
|
||
ET.SubElement(root, f"{{{CBC}}}ID").text = str(data["invoice_number"])
|
||
ET.SubElement(root, f"{{{CBC}}}IssueDate").text = parse_to_strict_xml_date(data["invoice_date"])
|
||
|
||
ET.SubElement(root, f"{{{CBC}}}DueDate").text = berechne_faelligkeit(data["invoice_date"])
|
||
|
||
ET.SubElement(root, f"{{{CBC}}}InvoiceTypeCode").text = str(data.get("invoice_type_code", "380"))
|
||
|
||
if data.get("invoice_note") and str(data["invoice_note"]).strip() != "":
|
||
ET.SubElement(root, f"{{{CBC}}}Note").text = str(data["invoice_note"]).strip()
|
||
|
||
ET.SubElement(root, f"{{{CBC}}}DocumentCurrencyCode").text = "EUR"
|
||
# BT-10 Käuferreferenz: NUR eintragen, wenn tatsächlich ein Wert vorliegt. Für private/
|
||
# geschäftliche Empfänger (B2B) reicht die Kundennummer, die der Kunde bei euch hat; nur bei
|
||
# Behörden (B2G) ist eine echte Leitweg-ID vorgesehen - die kommt dann vom Empfänger selbst.
|
||
if data.get("customer_number") and str(data["customer_number"]).strip() != "":
|
||
ET.SubElement(root, f"{{{CBC}}}BuyerReference").text = str(data["customer_number"]).strip()
|
||
|
||
if data.get("service_start_date") and data.get("service_end_date"):
|
||
period_elem = ET.SubElement(root, f"{{{CAC}}}InvoicePeriod")
|
||
ET.SubElement(period_elem, f"{{{CBC}}}StartDate").text = parse_to_strict_xml_date(data["service_start_date"])
|
||
ET.SubElement(period_elem, f"{{{CBC}}}EndDate").text = parse_to_strict_xml_date(data["service_end_date"])
|
||
|
||
# BT-13 "Purchase order reference" (Bestellnummer des KÄUFERS, z.B. nach "Ihr Auftrag:") und
|
||
# BT-14 "Sales order reference" (unsere EIGENE Auftrags-/Verkaufsauftragsnummer) leben beide
|
||
# unter demselben cac:OrderReference-Element. cac:OrderReference/cbc:ID ist laut UBL-Schema
|
||
# zwingend, sobald OrderReference verwendet wird - liegt keine Käufer-Bestellnummer (BT-13)
|
||
# vor, aber eine eigene Nummer (BT-14), ist laut Peppol-Spezifikation der Platzhalter "NA"
|
||
# vorgeschrieben.
|
||
has_bt13 = data.get("buyer_reference") and str(data["buyer_reference"]).strip() != ""
|
||
has_bt14 = data.get("order_id") and str(data["order_id"]).strip() != ""
|
||
if has_bt13 or has_bt14:
|
||
order_ref = ET.SubElement(root, f"{{{CAC}}}OrderReference")
|
||
ET.SubElement(order_ref, f"{{{CBC}}}ID").text = str(data["buyer_reference"]).strip() if has_bt13 else "NA"
|
||
if has_bt14:
|
||
ET.SubElement(order_ref, f"{{{CBC}}}SalesOrderID").text = str(data["order_id"]).strip()
|
||
|
||
if data.get("delivery_note_id") and str(data["delivery_note_id"]).strip() != "":
|
||
despatch_ref = ET.SubElement(root, f"{{{CAC}}}DespatchDocumentReference")
|
||
ET.SubElement(despatch_ref, f"{{{CBC}}}ID").text = str(data["delivery_note_id"]).strip()
|
||
|
||
if data.get("contract_id") and str(data["contract_id"]).strip() != "":
|
||
contract_ref = ET.SubElement(root, f"{{{CAC}}}ContractDocumentReference")
|
||
ET.SubElement(contract_ref, f"{{{CBC}}}ID").text = str(data["contract_id"]).strip()
|
||
|
||
vat_rate = float(data['vat_rate_percent'])
|
||
vat_category = "S" if vat_rate > 0.0 else "E"
|
||
|
||
# --- SENDER (SCITECH) ---
|
||
supplier = ET.SubElement(root, f"{{{CAC}}}AccountingSupplierParty")
|
||
party_supp = ET.SubElement(supplier, f"{{{CAC}}}Party")
|
||
ET.SubElement(party_supp, f"{{{CBC}}}EndpointID", {"schemeID": "EM"}).text = "office@scitech-gmbh.de"
|
||
|
||
addr_supp = ET.SubElement(party_supp, f"{{{CAC}}}PostalAddress")
|
||
ET.SubElement(addr_supp, f"{{{CBC}}}StreetName").text = "Hermann-Schumann-Str. 4"
|
||
ET.SubElement(addr_supp, f"{{{CBC}}}CityName").text = "Hennigsdorf"
|
||
ET.SubElement(addr_supp, f"{{{CBC}}}PostalZone").text = "16761"
|
||
country_supp = ET.SubElement(addr_supp, f"{{{CAC}}}Country")
|
||
ET.SubElement(country_supp, f"{{{CBC}}}IdentificationCode").text = "DE"
|
||
|
||
tax_scheme_supp = ET.SubElement(party_supp, f"{{{CAC}}}PartyTaxScheme")
|
||
ET.SubElement(tax_scheme_supp, f"{{{CBC}}}CompanyID").text = "DE218436692"
|
||
scheme_supp = ET.SubElement(tax_scheme_supp, f"{{{CAC}}}TaxScheme")
|
||
ET.SubElement(scheme_supp, f"{{{CBC}}}ID").text = "VAT"
|
||
|
||
legal_supp = ET.SubElement(party_supp, f"{{{CAC}}}PartyLegalEntity")
|
||
ET.SubElement(legal_supp, f"{{{CBC}}}RegistrationName").text = "scitech it solutions GmbH"
|
||
# BT-30 Registernummer (Handelsregister) - NICHT identisch mit der Umsatzsteuer-ID (BT-31)!
|
||
ET.SubElement(legal_supp, f"{{{CBC}}}CompanyID").text = "HRB 6058 NP"
|
||
|
||
contact_supp = ET.SubElement(party_supp, f"{{{CAC}}}Contact")
|
||
ET.SubElement(contact_supp, f"{{{CBC}}}Name").text = "Geschäftsführung"
|
||
ET.SubElement(contact_supp, f"{{{CBC}}}Telephone").text = "+49 3302 4940537"
|
||
ET.SubElement(contact_supp, f"{{{CBC}}}ElectronicMail").text = "office@scitech-gmbh.de"
|
||
|
||
# --- Empfänger ---
|
||
customer = ET.SubElement(root, f"{{{CAC}}}AccountingCustomerParty")
|
||
party_cust = ET.SubElement(customer, f"{{{CAC}}}Party")
|
||
ET.SubElement(party_cust, f"{{{CBC}}}EndpointID", {"schemeID": "EM"}).text = str(data.get("recipient_email", "")).strip()
|
||
addr_cust = ET.SubElement(party_cust, f"{{{CAC}}}PostalAddress")
|
||
ET.SubElement(addr_cust, f"{{{CBC}}}StreetName").text = str(data["recipient_street"])
|
||
ET.SubElement(addr_cust, f"{{{CBC}}}CityName").text = str(data["recipient_city"])
|
||
ET.SubElement(addr_cust, f"{{{CBC}}}PostalZone").text = str(data["recipient_zip"])
|
||
country_cust = ET.SubElement(addr_cust, f"{{{CAC}}}Country")
|
||
ET.SubElement(country_cust, f"{{{CBC}}}IdentificationCode").text = "DE"
|
||
legal_cust = ET.SubElement(party_cust, f"{{{CAC}}}PartyLegalEntity")
|
||
ET.SubElement(legal_cust, f"{{{CBC}}}RegistrationName").text = str(data["recipient_name"])
|
||
|
||
if data.get("delivery_date"):
|
||
delivery = ET.SubElement(root, f"{{{CAC}}}Delivery")
|
||
ET.SubElement(delivery, f"{{{CBC}}}ActualDeliveryDate").text = parse_to_strict_xml_date(data["delivery_date"])
|
||
|
||
payment_means = ET.SubElement(root, f"{{{CAC}}}PaymentMeans")
|
||
# BT-81 Zahlungsart: Code 58 = SEPA-Überweisung (statt 30, das für Überweisungen
|
||
# außerhalb des SEPA-Raums gedacht ist - VD-Valitool-169-Hinweis).
|
||
ET.SubElement(payment_means, f"{{{CBC}}}PaymentMeansCode").text = "58"
|
||
payee_acc = ET.SubElement(payment_means, f"{{{CAC}}}PayeeFinancialAccount")
|
||
ET.SubElement(payee_acc, f"{{{CBC}}}ID").text = str(data["iban"]).replace(" ", "")
|
||
|
||
# BT-20 Zahlungsbedingungen (Freitext) - für menschliche Lesbarkeit, analog zur
|
||
# CII/ZUGFeRD-Fassung. BR-CO-25 ist bereits durch das gesetzte cbc:DueDate (BT-9) erfüllt;
|
||
# dieser Text ist zusätzliche Klarstellung, keine Pflicht.
|
||
payment_terms_ubl = ET.SubElement(root, f"{{{CAC}}}PaymentTerms")
|
||
ET.SubElement(payment_terms_ubl, f"{{{CBC}}}Note").text = ZAHLUNGSBEDINGUNG_TEXT
|
||
|
||
tax_total = ET.SubElement(root, f"{{{CAC}}}TaxTotal")
|
||
ET.SubElement(tax_total, f"{{{CBC}}}TaxAmount", {"currencyID": "EUR"}).text = f"{data['total_vat_amount']:.2f}"
|
||
|
||
tax_subtotal = ET.SubElement(tax_total, f"{{{CAC}}}TaxSubtotal")
|
||
ET.SubElement(tax_subtotal, f"{{{CBC}}}TaxableAmount", {"currencyID": "EUR"}).text = f"{data['total_net_amount']:.2f}"
|
||
ET.SubElement(tax_subtotal, f"{{{CBC}}}TaxAmount", {"currencyID": "EUR"}).text = f"{data['total_vat_amount']:.2f}"
|
||
|
||
tax_cat = ET.SubElement(tax_subtotal, f"{{{CAC}}}TaxCategory")
|
||
ET.SubElement(tax_cat, f"{{{CBC}}}ID").text = vat_category
|
||
ET.SubElement(tax_cat, f"{{{CBC}}}Percent").text = f"{vat_rate:.2f}"
|
||
|
||
tax_scheme = ET.SubElement(tax_cat, f"{{{CAC}}}TaxScheme")
|
||
ET.SubElement(tax_scheme, f"{{{CBC}}}ID").text = "VAT"
|
||
|
||
monetary_total = ET.SubElement(root, f"{{{CAC}}}LegalMonetaryTotal")
|
||
ET.SubElement(monetary_total, f"{{{CBC}}}LineExtensionAmount", {"currencyID": "EUR"}).text = f"{data['total_net_amount']:.2f}"
|
||
ET.SubElement(monetary_total, f"{{{CBC}}}TaxExclusiveAmount", {"currencyID": "EUR"}).text = f"{data['total_net_amount']:.2f}"
|
||
ET.SubElement(monetary_total, f"{{{CBC}}}TaxInclusiveAmount", {"currencyID": "EUR"}).text = f"{data['total_gross_amount']:.2f}"
|
||
ET.SubElement(monetary_total, f"{{{CBC}}}PayableAmount", {"currencyID": "EUR"}).text = f"{data['total_gross_amount']:.2f}"
|
||
|
||
for index, item in enumerate(data["items"], start=1):
|
||
line = ET.SubElement(root, f"{{{CAC}}}InvoiceLine")
|
||
ET.SubElement(line, f"{{{CBC}}}ID").text = str(index)
|
||
|
||
u_code = map_unece_unit_code(item.get("unit", "C62"))
|
||
q_val = float(item.get("quantity") or 0.0)
|
||
qty_formatted = f"{q_val:.4f}".rstrip('0').rstrip('.') if not q_val.is_integer() else f"{q_val:.2f}"
|
||
if qty_formatted == "" or qty_formatted == "0": qty_formatted = "0.00"
|
||
|
||
ET.SubElement(line, f"{{{CBC}}}InvoicedQuantity", {"unitCode": u_code}).text = qty_formatted
|
||
ET.SubElement(line, f"{{{CBC}}}LineExtensionAmount", {"currencyID": "EUR"}).text = f"{item['total_amount']:.2f}"
|
||
|
||
item_elem = ET.SubElement(line, f"{{{CAC}}}Item")
|
||
|
||
if item.get("description") and str(item["description"]).strip() != "":
|
||
ET.SubElement(item_elem, f"{{{CBC}}}Description").text = str(item["description"]).strip()
|
||
|
||
ET.SubElement(item_elem, f"{{{CBC}}}Name").text = str(item.get("name", "Dienstleistung"))
|
||
|
||
item_tax = ET.SubElement(item_elem, f"{{{CAC}}}ClassifiedTaxCategory")
|
||
ET.SubElement(item_tax, f"{{{CBC}}}ID").text = vat_category
|
||
ET.SubElement(item_tax, f"{{{CBC}}}Percent").text = f"{vat_rate:.2f}"
|
||
item_scheme = ET.SubElement(item_tax, f"{{{CAC}}}TaxScheme")
|
||
ET.SubElement(item_scheme, f"{{{CBC}}}ID").text = "VAT"
|
||
|
||
price_elem = ET.SubElement(line, f"{{{CAC}}}Price")
|
||
ET.SubElement(price_elem, f"{{{CBC}}}PriceAmount", {"currencyID": "EUR"}).text = f"{item['unit_price']:.2f}"
|
||
|
||
xml_str = ET.tostring(root, encoding="utf-8")
|
||
return minidom.parseString(xml_str).toprettyxml(indent=" ")
|
||
|
||
|
||
# ==========================================
|
||
# GENERATOR B: ZUGFeRD 2.4 (CII FORMAT UN/CEFACT)
|
||
# ==========================================
|
||
def generate_compliant_zugferd_cii(data):
|
||
RSM = "urn:un:unece:uncefact:data:standard:CrossIndustryInvoice:100"
|
||
RAM = "urn:un:unece:uncefact:data:standard:ReusableAggregateBusinessInformationEntity:100"
|
||
UDT = "urn:un:unece:uncefact:data:standard:UnqualifiedDataType:100"
|
||
QDT = "urn:un:unece:uncefact:data:standard:QualifiedDataType:100"
|
||
|
||
ET.register_namespace('rsm', RSM)
|
||
ET.register_namespace('ram', RAM)
|
||
ET.register_namespace('udt', UDT)
|
||
ET.register_namespace('qdt', QDT)
|
||
|
||
root = ET.Element(f"{{{RSM}}}CrossIndustryInvoice")
|
||
# VD-Valitool-14: Der qdt-Namespace wird in dieser Instanz nicht benötigt, soll aber laut
|
||
# Spezifikation trotzdem am Root-Element deklariert werden.
|
||
root.set("xmlns:qdt", QDT)
|
||
|
||
ex_doc_ctx = ET.SubElement(root, f"{{{RSM}}}ExchangedDocumentContext")
|
||
|
||
# BT-23 Geschäftsprozess (PEPPOL-EN16931-R001 Pflicht). Muss laut XSD VOR
|
||
# GuidelineSpecifiedDocumentContextParameter stehen.
|
||
biz_process = ET.SubElement(ex_doc_ctx, f"{{{RAM}}}BusinessProcessSpecifiedDocumentContextParameter")
|
||
ET.SubElement(biz_process, f"{{{RAM}}}ID").text = "urn:fdc:peppol.eu:2017:poacc:billing:01:1.0"
|
||
|
||
guideline = ET.SubElement(ex_doc_ctx, f"{{{RAM}}}GuidelineSpecifiedDocumentContextParameter")
|
||
ET.SubElement(guideline, f"{{{RAM}}}ID").text = "urn:cen.eu:en16931:2017#compliant#urn:xeinkauf.de:kosit:xrechnung_3.0"
|
||
|
||
ex_doc = ET.SubElement(root, f"{{{RSM}}}ExchangedDocument")
|
||
ET.SubElement(ex_doc, f"{{{RAM}}}ID").text = str(data["invoice_number"])
|
||
ET.SubElement(ex_doc, f"{{{RAM}}}TypeCode").text = str(data.get("invoice_type_code", "380"))
|
||
|
||
issue_date = ET.SubElement(ex_doc, f"{{{RAM}}}IssueDateTime")
|
||
ET.SubElement(issue_date, f"{{{UDT}}}DateTimeString", {"format": "102"}).text = format_cii_date(data["invoice_date"])
|
||
|
||
if data.get("invoice_note") and str(data["invoice_note"]).strip() != "":
|
||
note = ET.SubElement(ex_doc, f"{{{RAM}}}IncludedNote")
|
||
ET.SubElement(note, f"{{{RAM}}}Content").text = str(data["invoice_note"]).strip()
|
||
|
||
trade_trans = ET.SubElement(root, f"{{{RSM}}}SupplyChainTradeTransaction")
|
||
|
||
vat_rate = float(data['vat_rate_percent'])
|
||
vat_category = "S" if vat_rate > 0.0 else "E"
|
||
|
||
for index, item in enumerate(data["items"], start=1):
|
||
line_item = ET.SubElement(trade_trans, f"{{{RAM}}}IncludedSupplyChainTradeLineItem")
|
||
|
||
doc_line = ET.SubElement(line_item, f"{{{RAM}}}AssociatedDocumentLineDocument")
|
||
ET.SubElement(doc_line, f"{{{RAM}}}LineID").text = str(index)
|
||
|
||
trade_prod = ET.SubElement(line_item, f"{{{RAM}}}SpecifiedTradeProduct")
|
||
ET.SubElement(trade_prod, f"{{{RAM}}}Name").text = str(item.get("name", "Dienstleistung"))
|
||
if item.get("description") and str(item["description"]).strip() != "":
|
||
ET.SubElement(trade_prod, f"{{{RAM}}}Description").text = str(item["description"]).strip()
|
||
|
||
line_agree = ET.SubElement(line_item, f"{{{RAM}}}SpecifiedLineTradeAgreement")
|
||
net_price = ET.SubElement(line_agree, f"{{{RAM}}}NetPriceProductTradePrice")
|
||
ET.SubElement(net_price, f"{{{RAM}}}ChargeAmount").text = f"{item['unit_price']:.2f}"
|
||
|
||
line_deliv = ET.SubElement(line_item, f"{{{RAM}}}SpecifiedLineTradeDelivery")
|
||
u_code = map_unece_unit_code(item.get("unit", "Stk."))
|
||
ET.SubElement(line_deliv, f"{{{RAM}}}BilledQuantity", {"unitCode": u_code}).text = f"{item['quantity']:.4f}"
|
||
|
||
line_settle = ET.SubElement(line_item, f"{{{RAM}}}SpecifiedLineTradeSettlement")
|
||
trade_tax = ET.SubElement(line_settle, f"{{{RAM}}}ApplicableTradeTax")
|
||
ET.SubElement(trade_tax, f"{{{RAM}}}TypeCode").text = "VAT"
|
||
ET.SubElement(trade_tax, f"{{{RAM}}}CategoryCode").text = vat_category
|
||
ET.SubElement(trade_tax, f"{{{RAM}}}RateApplicablePercent").text = f"{vat_rate:.2f}"
|
||
|
||
mon_sum = ET.SubElement(line_settle, f"{{{RAM}}}SpecifiedTradeSettlementLineMonetarySummation")
|
||
ET.SubElement(mon_sum, f"{{{RAM}}}LineTotalAmount").text = f"{item['total_amount']:.2f}"
|
||
|
||
header_agree = ET.SubElement(trade_trans, f"{{{RAM}}}ApplicableHeaderTradeAgreement")
|
||
# BT-10 Käuferreferenz: NUR eintragen, wenn tatsächlich ein Wert vorliegt (siehe UBL-Generator
|
||
# für die ausführliche Begründung).
|
||
if data.get("customer_number") and str(data["customer_number"]).strip() != "":
|
||
ET.SubElement(header_agree, f"{{{RAM}}}BuyerReference").text = str(data["customer_number"]).strip()
|
||
|
||
supplier_party = ET.SubElement(header_agree, f"{{{RAM}}}SellerTradeParty")
|
||
ET.SubElement(supplier_party, f"{{{RAM}}}Name").text = "scitech it solutions GmbH"
|
||
|
||
supp_legal = ET.SubElement(supplier_party, f"{{{RAM}}}SpecifiedLegalOrganization")
|
||
# BT-30 Registernummer (Handelsregister) - NICHT identisch mit der Umsatzsteuer-ID (BT-31)!
|
||
# Fest eingetragen, analog zu den übrigen fixen Verkäufer-Stammdaten.
|
||
ET.SubElement(supp_legal, f"{{{RAM}}}ID").text = "HRB 6058 NP"
|
||
|
||
# BG-6 SELLER CONTACT (BR-DE-2 Pflicht für XRechnung). Muss laut XSD nach
|
||
# SpecifiedLegalOrganization und vor PostalTradeAddress stehen.
|
||
supp_contact = ET.SubElement(supplier_party, f"{{{RAM}}}DefinedTradeContact")
|
||
ET.SubElement(supp_contact, f"{{{RAM}}}PersonName").text = "Geschäftsführung"
|
||
supp_contact_tel = ET.SubElement(supp_contact, f"{{{RAM}}}TelephoneUniversalCommunication")
|
||
ET.SubElement(supp_contact_tel, f"{{{RAM}}}CompleteNumber").text = "+49 3302 4940537"
|
||
supp_contact_mail = ET.SubElement(supp_contact, f"{{{RAM}}}EmailURIUniversalCommunication")
|
||
ET.SubElement(supp_contact_mail, f"{{{RAM}}}URIID").text = "office@scitech-gmbh.de"
|
||
|
||
supp_addr = ET.SubElement(supplier_party, f"{{{RAM}}}PostalTradeAddress")
|
||
ET.SubElement(supp_addr, f"{{{RAM}}}PostcodeCode").text = "16761"
|
||
ET.SubElement(supp_addr, f"{{{RAM}}}LineOne").text = "Hermann-Schumann-Str. 4"
|
||
ET.SubElement(supp_addr, f"{{{RAM}}}CityName").text = "Hennigsdorf"
|
||
ET.SubElement(supp_addr, f"{{{RAM}}}CountryID").text = "DE"
|
||
|
||
# BT-34 Elektronische Adresse des Verkäufers (PEPPOL-EN16931-R020 Pflicht). Muss laut
|
||
# XSD nach PostalTradeAddress und vor SpecifiedTaxRegistration stehen.
|
||
supp_uri = ET.SubElement(supplier_party, f"{{{RAM}}}URIUniversalCommunication")
|
||
ET.SubElement(supp_uri, f"{{{RAM}}}URIID", {"schemeID": "EM"}).text = "office@scitech-gmbh.de"
|
||
|
||
supp_tax = ET.SubElement(supplier_party, f"{{{RAM}}}SpecifiedTaxRegistration")
|
||
ET.SubElement(supp_tax, f"{{{RAM}}}ID", {"schemeID": "VA"}).text = "DE218436692"
|
||
|
||
customer_party = ET.SubElement(header_agree, f"{{{RAM}}}BuyerTradeParty")
|
||
ET.SubElement(customer_party, f"{{{RAM}}}Name").text = str(data["recipient_name"])
|
||
|
||
cust_addr = ET.SubElement(customer_party, f"{{{RAM}}}PostalTradeAddress")
|
||
ET.SubElement(cust_addr, f"{{{RAM}}}PostcodeCode").text = str(data["recipient_zip"])
|
||
ET.SubElement(cust_addr, f"{{{RAM}}}LineOne").text = str(data["recipient_street"])
|
||
ET.SubElement(cust_addr, f"{{{RAM}}}CityName").text = str(data["recipient_city"])
|
||
ET.SubElement(cust_addr, f"{{{RAM}}}CountryID").text = "DE"
|
||
|
||
# BT-49 Elektronische Adresse des Käufers (PEPPOL-EN16931-R010 Pflicht). Muss laut
|
||
# XSD nach PostalTradeAddress stehen. KEIN erfundener Fallback - fehlt die Adresse, wird das
|
||
# (leere) Pflichtfeld sichtbar und die UI blockiert den Export mit Warnung.
|
||
cust_uri = ET.SubElement(customer_party, f"{{{RAM}}}URIUniversalCommunication")
|
||
ET.SubElement(cust_uri, f"{{{RAM}}}URIID", {"schemeID": "EM"}).text = str(data.get("recipient_email", "")).strip()
|
||
|
||
if data.get("order_id") and str(data["order_id"]).strip() != "":
|
||
# BT-14 "Sales order reference" - unsere EIGENE Auftrags-/Verkaufsauftragsnummer.
|
||
# In CII ein eigenständiges Element. Muss laut XSD VOR BuyerOrderReferencedDocument stehen.
|
||
order_ref = ET.SubElement(header_agree, f"{{{RAM}}}SellerOrderReferencedDocument")
|
||
ET.SubElement(order_ref, f"{{{RAM}}}IssuerAssignedID").text = str(data["order_id"]).strip()
|
||
|
||
if data.get("buyer_reference") and str(data["buyer_reference"]).strip() != "":
|
||
# BT-13 "Purchase order reference" - die Bestellnummer des KÄUFERS (z.B. nach
|
||
# "Ihr Auftrag:"), NICHT unsere eigene Nummer (die ist BT-14/order_id oben).
|
||
buyer_order_ref = ET.SubElement(header_agree, f"{{{RAM}}}BuyerOrderReferencedDocument")
|
||
ET.SubElement(buyer_order_ref, f"{{{RAM}}}IssuerAssignedID").text = str(data["buyer_reference"]).strip()
|
||
|
||
if data.get("contract_id") and str(data["contract_id"]).strip() != "":
|
||
contract_ref = ET.SubElement(header_agree, f"{{{RAM}}}ContractReferencedDocument")
|
||
ET.SubElement(contract_ref, f"{{{RAM}}}IssuerAssignedID").text = str(data["contract_id"]).strip()
|
||
|
||
header_deliv = ET.SubElement(trade_trans, f"{{{RAM}}}ApplicableHeaderTradeDelivery")
|
||
if data.get("delivery_date"):
|
||
act_deliv = ET.SubElement(header_deliv, f"{{{RAM}}}ActualDeliverySupplyChainEvent")
|
||
occ_date = ET.SubElement(act_deliv, f"{{{RAM}}}OccurrenceDateTime")
|
||
ET.SubElement(occ_date, f"{{{UDT}}}DateTimeString", {"format": "102"}).text = format_cii_date(data["delivery_date"])
|
||
|
||
# HINWEIS: "DeliveryNoteReferencedDocument" existiert im EN16931/Factur-X-Schema NICHT
|
||
# (im offiziellen XSD nicht vorhanden). Das korrekte, schema-konforme Element für eine
|
||
# Lieferschein-/Lieferavis-Referenz ist "DespatchAdviceReferencedDocument" und muss laut
|
||
# XSD-Sequenz NACH ActualDeliverySupplyChainEvent stehen.
|
||
if data.get("delivery_note_id") and str(data["delivery_note_id"]).strip() != "":
|
||
dn_ref = ET.SubElement(header_deliv, f"{{{RAM}}}DespatchAdviceReferencedDocument")
|
||
ET.SubElement(dn_ref, f"{{{RAM}}}IssuerAssignedID").text = str(data["delivery_note_id"]).strip()
|
||
|
||
header_settle = ET.SubElement(trade_trans, f"{{{RAM}}}ApplicableHeaderTradeSettlement")
|
||
ET.SubElement(header_settle, f"{{{RAM}}}InvoiceCurrencyCode").text = "EUR"
|
||
|
||
pay_means = ET.SubElement(header_settle, f"{{{RAM}}}SpecifiedTradeSettlementPaymentMeans")
|
||
# BT-81 Zahlungsart: Code 58 = SEPA-Überweisung. Code 30 ist für Überweisungen AUSSERHALB
|
||
# des SEPA-Raums gedacht (VD-Valitool-169-Hinweis) - da hier immer eine IBAN verwendet wird,
|
||
# ist 58 der korrekte Code.
|
||
ET.SubElement(pay_means, f"{{{RAM}}}TypeCode").text = "58"
|
||
payee_acc = ET.SubElement(pay_means, f"{{{RAM}}}PayeePartyCreditorFinancialAccount")
|
||
ET.SubElement(payee_acc, f"{{{RAM}}}IBANID").text = str(data["iban"]).replace(" ", "")
|
||
|
||
header_tax = ET.SubElement(header_settle, f"{{{RAM}}}ApplicableTradeTax")
|
||
ET.SubElement(header_tax, f"{{{RAM}}}CalculatedAmount").text = f"{data['total_vat_amount']:.2f}"
|
||
ET.SubElement(header_tax, f"{{{RAM}}}TypeCode").text = "VAT"
|
||
ET.SubElement(header_tax, f"{{{RAM}}}BasisAmount").text = f"{data['total_net_amount']:.2f}"
|
||
ET.SubElement(header_tax, f"{{{RAM}}}CategoryCode").text = vat_category
|
||
ET.SubElement(header_tax, f"{{{RAM}}}RateApplicablePercent").text = f"{vat_rate:.2f}"
|
||
|
||
if data.get("service_start_date") and data.get("service_end_date"):
|
||
period = ET.SubElement(header_settle, f"{{{RAM}}}BillingSpecifiedPeriod")
|
||
start_pt = ET.SubElement(period, f"{{{RAM}}}StartDateTime")
|
||
ET.SubElement(start_pt, f"{{{UDT}}}DateTimeString", {"format": "102"}).text = format_cii_date(data["service_start_date"])
|
||
end_pt = ET.SubElement(period, f"{{{RAM}}}EndDateTime")
|
||
ET.SubElement(end_pt, f"{{{UDT}}}DateTimeString", {"format": "102"}).text = format_cii_date(data["service_end_date"])
|
||
|
||
# BT-9 Fälligkeitsdatum (BR-CO-25 Pflicht: bei positivem Zahlbetrag muss entweder ein
|
||
# Fälligkeitsdatum ODER Zahlungsbedingungen angegeben werden). Zentrale Berechnung, analog
|
||
# zur XRechnung(UBL)-Generierung.
|
||
due_date_str = berechne_faelligkeit(data["invoice_date"]).replace("-", "")
|
||
payment_terms = ET.SubElement(header_settle, f"{{{RAM}}}SpecifiedTradePaymentTerms")
|
||
ET.SubElement(payment_terms, f"{{{RAM}}}Description").text = ZAHLUNGSBEDINGUNG_TEXT
|
||
due_date_elem = ET.SubElement(payment_terms, f"{{{RAM}}}DueDateDateTime")
|
||
ET.SubElement(due_date_elem, f"{{{UDT}}}DateTimeString", {"format": "102"}).text = due_date_str
|
||
|
||
header_sum = ET.SubElement(header_settle, f"{{{RAM}}}SpecifiedTradeSettlementHeaderMonetarySummation")
|
||
ET.SubElement(header_sum, f"{{{RAM}}}LineTotalAmount").text = f"{data['total_net_amount']:.2f}"
|
||
ET.SubElement(header_sum, f"{{{RAM}}}TaxBasisTotalAmount").text = f"{data['total_net_amount']:.2f}"
|
||
ET.SubElement(header_sum, f"{{{RAM}}}TaxTotalAmount", {"currencyID": "EUR"}).text = f"{data['total_vat_amount']:.2f}"
|
||
ET.SubElement(header_sum, f"{{{RAM}}}GrandTotalAmount").text = f"{data['total_gross_amount']:.2f}"
|
||
ET.SubElement(header_sum, f"{{{RAM}}}DuePayableAmount").text = f"{data['total_gross_amount']:.2f}"
|
||
|
||
xml_str = ET.tostring(root, encoding="utf-8")
|
||
return minidom.parseString(xml_str).toprettyxml(indent=" ")
|
||
|
||
|
||
# Pfad zum sRGB-ICC-Profil, das für ein PDF/A-3-konformes OutputIntent benötigt wird.
|
||
# (PDF/A verlangt zwingend ein eingebettetes Farbprofil - ohne das gilt die PDF nicht als PDF/A,
|
||
# selbst wenn XMP-Metadaten und eingebettetes XML korrekt sind.)
|
||
SRGB_ICC_PATH = os.path.join(os.path.dirname(os.path.abspath(__file__)), "sRGB.icc")
|
||
|
||
|
||
def add_pdfa_output_intent(pdf_bytes: bytes, icc_path: str = SRGB_ICC_PATH) -> bytes:
|
||
"""
|
||
Fügt der PDF ein OutputIntent mit sRGB-ICC-Profil hinzu, falls noch keines vorhanden ist.
|
||
Das ist zwingend erforderlich, damit eine PDF als PDF/A-3(B) gilt (ZUGFeRD-Pflicht).
|
||
"""
|
||
pdf = pikepdf.open(io.BytesIO(pdf_bytes))
|
||
try:
|
||
if "/OutputIntents" not in pdf.Root or len(pdf.Root.get("/OutputIntents", [])) == 0:
|
||
if not os.path.isfile(icc_path):
|
||
raise FileNotFoundError(
|
||
f"sRGB-ICC-Profil nicht gefunden unter '{icc_path}'. "
|
||
"Für echte PDF/A-3-Konformität muss diese Datei mit ausgeliefert werden."
|
||
)
|
||
with open(icc_path, "rb") as f:
|
||
icc_bytes = f.read()
|
||
icc_stream = pikepdf.Stream(pdf, icc_bytes)
|
||
icc_stream["/N"] = 3
|
||
icc_stream["/Alternate"] = pikepdf.Name("/DeviceRGB")
|
||
output_intent = pdf.make_indirect(pikepdf.Dictionary({
|
||
"/Type": pikepdf.Name("/OutputIntent"),
|
||
"/S": pikepdf.Name("/GTS_PDFA1"),
|
||
"/OutputConditionIdentifier": pikepdf.String("sRGB IEC61966-2.1"),
|
||
"/Info": pikepdf.String("sRGB IEC61966-2.1"),
|
||
"/DestOutputProfile": icc_stream,
|
||
}))
|
||
pdf.Root["/OutputIntents"] = pikepdf.Array([output_intent])
|
||
buf = io.BytesIO()
|
||
pdf.save(buf)
|
||
return buf.getvalue()
|
||
finally:
|
||
pdf.close()
|
||
|
||
|
||
def fix_xrechnung_xmp_conformance(pdf_bytes: bytes) -> bytes:
|
||
"""
|
||
Die factur-x-Bibliothek schreibt standardmäßig fx:ConformanceLevel = 'EN 16931' und
|
||
fx:Version = '1.0' in die XMP-Metadaten (generischer Factur-X-Standard). Da unsere CII-XML
|
||
aber explizit das KoSIT-XRechnung-Profil deklariert (GuidelineID .../xrechnung_3.0), erwarten
|
||
Validatoren hier stattdessen 'XRECHNUNG' und '3.0' - sonst schlagen z. B.
|
||
VD-Valitool-H-022/H-024 fehl, weil PDF-Metadaten und eingebettetes XML nicht zusammenpassen.
|
||
"""
|
||
pdf = pikepdf.open(io.BytesIO(pdf_bytes))
|
||
try:
|
||
if "/Metadata" in pdf.Root:
|
||
xmp_bytes = pdf.Root.Metadata.read_bytes()
|
||
xmp_str = xmp_bytes.decode("utf-8")
|
||
xmp_str = xmp_str.replace(
|
||
"<fx:ConformanceLevel>EN 16931</fx:ConformanceLevel>",
|
||
"<fx:ConformanceLevel>XRECHNUNG</fx:ConformanceLevel>",
|
||
).replace(
|
||
"<fx:Version>1.0</fx:Version>",
|
||
"<fx:Version>3.0</fx:Version>",
|
||
)
|
||
pdf.Root.Metadata.write(xmp_str.encode("utf-8"))
|
||
buf = io.BytesIO()
|
||
pdf.save(buf)
|
||
return buf.getvalue()
|
||
finally:
|
||
pdf.close()
|
||
|
||
|
||
# ==========================================
|
||
# 4. API CLIENT INITIALISIERUNG
|
||
# ==========================================
|
||
def get_cached_llm():
|
||
if "llm_instance" not in st.session_state:
|
||
llm_url = os.environ.get('VLLM_URL', 'http://10.192.0.140:8002')
|
||
llm_model = os.environ.get('LLM_MODEL', 'gemma431b')
|
||
st.session_state.llm_instance = LLMClient(model=llm_model, base_url=llm_url)
|
||
return st.session_state.llm_instance
|
||
|
||
|
||
# ==========================================
|
||
# 5. STREAMLIT FRONTEND
|
||
# ==========================================
|
||
st.set_page_config(page_title="PDF zu E-Rechnung", layout="wide")
|
||
st.title("📄 PDF zu E-Rechnung Konverter")
|
||
|
||
llm = get_cached_llm()
|
||
col1, col2 = st.columns(2)
|
||
|
||
with col1:
|
||
st.subheader("1. Dokument hochladen")
|
||
uploaded_file = st.file_uploader("Wähle eine Rechnungs-PDF", type=["pdf"])
|
||
pdf_text = ""
|
||
if uploaded_file is not None:
|
||
doc = fitz.open(stream=uploaded_file.read(), filetype="pdf")
|
||
for page in doc:
|
||
pdf_text += page.get_text("text") + "\n"
|
||
st.success("PDF eingelesen!")
|
||
with st.expander("Gefundener Text"): st.text(pdf_text)
|
||
|
||
with col2:
|
||
st.subheader("2. Strukturierte Daten validieren")
|
||
if "extracted_data" not in st.session_state: st.session_state.extracted_data = None
|
||
|
||
if uploaded_file is not None and pdf_text.strip() != "":
|
||
if st.button("Daten extrahieren", width="stretch"):
|
||
with st.spinner("Extrahiere Daten über vLLM-API..."):
|
||
|
||
schema_dict = InvoiceMetadata.model_json_schema()
|
||
|
||
sys_prompt = (
|
||
"Du bist ein präziser OCR-Parser für deutsche B2B-Rechnungen. "
|
||
"Extrahiere die Metadaten fehlerfrei als valides JSON-Objekt. "
|
||
"Das JSON MUSS zwingend exakt folgende Schlüssel (Keys) enthalten:\n"
|
||
"- invoice_number\n"
|
||
"- invoice_date (im Format YYYY-MM-DD)\n"
|
||
"- delivery_date (im Format YYYY-MM-DD)\n"
|
||
"- buyer_reference\n"
|
||
"- customer_number\n"
|
||
"- recipient_name\n"
|
||
"- recipient_street\n"
|
||
"- recipient_zip\n"
|
||
"- recipient_city\n"
|
||
"- recipient_email\n"
|
||
"- vat_rate_percent (als Zahl, z.B. 19.0)\n"
|
||
"- delivery_note_id\n"
|
||
"- order_id\n"
|
||
"- contract_id\n"
|
||
"- service_start_date\n"
|
||
"- service_end_date\n"
|
||
"- invoice_note\n\n"
|
||
"WICHTIGE REGEL: Analysiere den Text direkt VOR der ersten Tabellenposition '1'. "
|
||
"Sortiere alle dort vorlaufenden Zeilen vollständig in die dafür vorgesehenen ID-, Datums- und Freitext-Felder ein. "
|
||
"Schreibe diese vorlaufenden Metadaten NIEMALS in die Beschreibung der ersten Position hinein!\n\n"
|
||
"Antworte ausschließlich mit dem reinen JSON-Objekt. Schreibe keinen anderen Text."
|
||
)
|
||
prompt_meta = f"Rechnungstext:\n{pdf_text}"
|
||
|
||
res_meta_text = llm.chat(
|
||
prompt=prompt_meta,
|
||
system_prompt=sys_prompt,
|
||
temperature=0.0,
|
||
json_schema=schema_dict
|
||
)
|
||
|
||
clean_json_text = res_meta_text.strip()
|
||
backtick_trigger = chr(96) * 3
|
||
if clean_json_text.startswith(backtick_trigger):
|
||
lines = clean_json_text.splitlines()
|
||
if lines[0].startswith(backtick_trigger):
|
||
lines = lines[1:]
|
||
if lines and lines[-1].startswith(backtick_trigger):
|
||
lines = lines[:-1]
|
||
clean_json_text = "\n".join(lines).strip()
|
||
|
||
try:
|
||
final_data = json.loads(clean_json_text)
|
||
except Exception as e:
|
||
st.error(f"Fehler beim Parsen der API-Antwort: {e}")
|
||
st.code(res_meta_text)
|
||
st.stop()
|
||
|
||
if final_data.get("recipient_email") == "office@scitech-gmbh.de":
|
||
final_data["recipient_email"] = ""
|
||
|
||
final_data["items"] = extract_invoice_items_pure_python(pdf_text)
|
||
|
||
st.session_state.invoice_counter += 1
|
||
st.session_state.extracted_data = final_data
|
||
|
||
if st.session_state.extracted_data:
|
||
data = st.session_state.extracted_data
|
||
|
||
# --- BT-3 RECHNUNGSTYP DROPDOWN-MENÜ ---
|
||
inv_type_ui = st.selectbox(
|
||
"Rechnungstyp (BT-3)",
|
||
options=["380 (Handelsrechnung)", "381 (Gutschrift)", "384 (Korrekturrechnung)", "326 (Teilrechnung)"],
|
||
index=0,
|
||
help="Bestimmt den steuerrechtlichen Belegtyp im XML-Header."
|
||
)
|
||
selected_type_code = inv_type_ui.split(" ")[0]
|
||
|
||
inv_num = st.text_input("Rechnungsnummer (BT-1)", value=data.get("invoice_number", ""))
|
||
inv_date = st.text_input("Rechnungsdatum (BT-2, YYYY-MM-DD)", value=data.get("invoice_date", ""))
|
||
inv_delivery = st.text_input("Lieferdatum (BT-72, YYYY-MM-DD)", value=data.get("delivery_date", ""))
|
||
cust_num = st.text_input(
|
||
"Kundennummer / Leitweg-ID (BT-10)",
|
||
value=data.get("customer_number", ""),
|
||
help="Eure Kundennummer beim Empfänger (Standardfall für B2B) oder eine Leitweg-ID, falls "
|
||
"der Empfänger eine Behörde ist. Sonst leer lassen - das Feld wird dann im XML komplett "
|
||
"weggelassen."
|
||
)
|
||
buyer_ref = st.text_input(
|
||
"Ihr Auftrag / Bestellnummer (BT-13)",
|
||
value=data.get("buyer_reference", ""),
|
||
help="Die vom KÄUFER vorgegebene Bestellnummer für diesen Auftrag (z.B. nach 'Ihr Auftrag:'). "
|
||
"NICHT die Kundennummer (die gehört ins Feld oben) und NICHT die eigene Auftragsbestätigungs-"
|
||
"/Verkaufsauftragsnummer (die gehört ins Feld 'Auftragsbestätigung / Verkaufsauftrag (BT-14)' "
|
||
"weiter unten). Sonst leer lassen."
|
||
)
|
||
|
||
# UI-Eingabefelder für die Daten vor Position 1 (BT-22 etc.)
|
||
st.markdown("---")
|
||
st.markdown("**Globale Referenzen & Vorlauftexte:**")
|
||
c_ref1, c_ref2, c_ref3 = st.columns(3)
|
||
with c_ref1:
|
||
dn_id = st.text_input("Lieferscheinnummer (BT-16)", value=data.get("delivery_note_id", ""))
|
||
with c_ref2:
|
||
o_id = st.text_input("Auftragsbestätigung / Verkaufsauftrag (BT-14)", value=data.get("order_id", ""))
|
||
with c_ref3:
|
||
c_id = st.text_input("Vertrag / Vergabeangebot (BT-12)", value=data.get("contract_id", ""))
|
||
|
||
c_p1, c_p2 = st.columns(2)
|
||
with c_p1:
|
||
s_start = st.text_input("Leistungszeitraum Start (BT-73, YYYY-MM-DD)", value=data.get("service_start_date", ""))
|
||
with c_p2:
|
||
s_end = st.text_input("Leistungszeitraum Ende (BT-74, YYYY-MM-DD)", value=data.get("service_end_date", ""))
|
||
|
||
i_note = st.text_area("Globale Rechnungserläuterung / Freitext (BT-22)", value=data.get("invoice_note", ""), height=150)
|
||
st.markdown("---")
|
||
|
||
# Adressdaten
|
||
c_s, c_r = st.columns(2)
|
||
with c_s:
|
||
st.markdown("**Absender (Verkäufer):**")
|
||
st.text_input("Firma (Sender) (BT-27)", value="scitech it solutions GmbH", disabled=True)
|
||
st.text_input("E-Mail (Sender) (BT-34)", value="office@scitech-gmbh.de", disabled=True)
|
||
st.text_input("Telefon (Sender) (BT-42)", value="+49 3302 4940537", disabled=True)
|
||
st.text_input("Straße (Sender) (BT-35)", value="Hermann-Schumann-Str. 4", disabled=True)
|
||
st.text_input("PLZ (Sender) (BT-38)", value="16761", disabled=True)
|
||
st.text_input("Stadt (Sender) (BT-37)", value="Hennigsdorf", disabled=True)
|
||
st.text_input("Umsatzsteuer-ID (Sender) (BT-31)", value="DE218436692", disabled=True)
|
||
st.text_input("Registernummer (Sender) (BT-30)", value="HRB 6058 NP", disabled=True)
|
||
st.text_input("IBAN (Sender) (BT-84)", value=VERKAEUFER_IBAN, disabled=True)
|
||
with c_r:
|
||
st.markdown("**Empfänger (Kunde):**")
|
||
r_name = st.text_input("Firma (Kunde) (BT-44)", value=data.get("recipient_name", ""))
|
||
r_email = st.text_input("E-Mail (Kunde) (BT-49)", value=data.get("recipient_email", ""))
|
||
r_street = st.text_input("Straße (Kunde) (BT-50)", value=data.get("recipient_street", ""))
|
||
r_zip = st.text_input("PLZ (Kunde) (BT-53)", value=data.get("recipient_zip", ""))
|
||
r_city = st.text_input("Stadt (Kunde) (BT-52)", value=data.get("recipient_city", ""))
|
||
|
||
st.markdown("### 3. Positionen")
|
||
|
||
if st.button("➕ Neue Position hinzufügen"):
|
||
data["items"].append({"name": "Neue Dienstleistung", "description": "", "quantity": 1.0, "unit": "Stk.", "unit_price": 0.0})
|
||
st.rerun()
|
||
|
||
edited_items = []
|
||
inv_id = st.session_state.invoice_counter
|
||
|
||
for idx, item in enumerate(data.get("items", [])):
|
||
with st.expander(f"📋 Position {idx+1}: {item.get('name', 'Dienstleistung')}", expanded=True):
|
||
|
||
c_del, _ = st.columns([1, 4])
|
||
with c_del:
|
||
if st.button("🗑️ Position löschen", key=f"del_btn_{inv_id}_{idx}"):
|
||
data["items"].pop(idx)
|
||
st.rerun()
|
||
|
||
c_name, c_desc = st.columns([1, 1])
|
||
with c_name:
|
||
item_name = st.text_input("Artikelname (BT-153)", value=item.get("name", ""), key=f"item_name_{inv_id}_{idx}")
|
||
with c_desc:
|
||
desc_text = item.get("description", "")
|
||
estimated_lines = desc_text.count("\n") + (len(desc_text) // 45) + 1
|
||
calc_height = max(68, min(400, estimated_lines * 24))
|
||
item_description = st.text_area("Zusatzbeschreibung (BT-154)", value=desc_text, key=f"item_desc_{inv_id}_{idx}", height=calc_height)
|
||
|
||
c_qty, c_unit, c_price, c_total = st.columns(4)
|
||
with c_qty:
|
||
item_quantity = st.number_input("Menge (BT-129)", value=float(item.get("quantity") or 0.0), format="%.3f", step=0.1, key=f"item_qty_{inv_id}_{idx}")
|
||
with c_unit:
|
||
item_unit = st.text_input("Einheit (BT-130)", value=item.get("unit", "Stk."), key=f"item_unit_{inv_id}_{idx}")
|
||
with c_price:
|
||
item_price = st.number_input("Einzelpreis (BT-146, €)", value=float(item.get("unit_price") or 0.0), format="%.2f", key=f"item_price_{inv_id}_{idx}")
|
||
with c_total:
|
||
raw_item_total = item_quantity * item_price
|
||
item_total_amount = kaufmaennisches_runden(raw_item_total)
|
||
st.markdown("**Gesamtbetrag (€):**")
|
||
st.markdown(f"**{item_total_amount:.2f} €**")
|
||
|
||
edited_items.append({
|
||
"name": item_name,
|
||
"description": item_description,
|
||
"quantity": item_quantity,
|
||
"unit": item_unit,
|
||
"unit_price": item_price,
|
||
"total_amount": item_total_amount
|
||
})
|
||
|
||
st.markdown("---")
|
||
vat_rate = st.number_input("Steuersatz (BT-152, %)", value=float(data.get("vat_rate_percent") or 19.0))
|
||
|
||
# Summen mit Decimal bilden, damit die Netto-Summe exakt der Summe der (bereits gerundeten)
|
||
# Positionsbeträge entspricht. Reine float-Summation kann minimal driften und dann zu
|
||
# BR-CO-10-Validierungsfehlern führen (Summe der Zeilen != TaxExclusiveAmount).
|
||
net_dec = sum((Decimal(str(item["total_amount"])) for item in edited_items), Decimal("0"))
|
||
vat_dec = (net_dec * Decimal(str(vat_rate)) / Decimal("100")).quantize(Decimal("0.01"), rounding=ROUND_HALF_UP)
|
||
gross_dec = (net_dec + vat_dec).quantize(Decimal("0.01"), rounding=ROUND_HALF_UP)
|
||
computed_net_amount = float(net_dec.quantize(Decimal("0.01"), rounding=ROUND_HALF_UP))
|
||
computed_vat_amount = float(vat_dec)
|
||
computed_gross_amount = float(gross_dec)
|
||
|
||
st.markdown("**Berechnete Summen:**")
|
||
c_net_ui, c_vat_ui, c_gross_ui = st.columns(3)
|
||
with c_net_ui: st.metric("Netto-Gesamt (€)", f"{computed_net_amount:.2f}")
|
||
with c_vat_ui: st.metric("MwSt-Gesamt (€)", f"{computed_vat_amount:.2f}")
|
||
with c_gross_ui: st.metric("Brutto-Zahlbetrag (€)", f"{computed_gross_amount:.2f}")
|
||
|
||
# --- EXPORT-SEKTION DROPDOWN-AUSWAHL ---
|
||
st.markdown("---")
|
||
export_format = st.selectbox(
|
||
"💾 Export-Format wählen",
|
||
options=["XRechnung v3.0.2 (Reines UBL XML)", "ZUGFeRD v2.4 (Hybrid-PDF mit CII XML)"],
|
||
help="XRechnung generiert ein reines, valides UBL-XML. ZUGFeRD bettet ein schema-konformes CII-XML (factur-x.xml) direkt in deine PDF ein."
|
||
)
|
||
|
||
final_invoice_dict = {
|
||
"invoice_type_code": selected_type_code,
|
||
"invoice_number": inv_num, "invoice_date": inv_date, "delivery_date": inv_delivery,
|
||
"buyer_reference": buyer_ref, "customer_number": cust_num, "recipient_name": r_name, "recipient_email": r_email,
|
||
"recipient_street": r_street, "recipient_zip": r_zip, "recipient_city": r_city,
|
||
"items": edited_items, "total_net_amount": computed_net_amount, "total_vat_amount": computed_vat_amount,
|
||
"vat_rate_percent": vat_rate, "total_gross_amount": computed_gross_amount, "iban": VERKAEUFER_IBAN,
|
||
"delivery_note_id": dn_id,
|
||
"order_id": o_id,
|
||
"contract_id": c_id,
|
||
"service_start_date": s_start,
|
||
"service_end_date": s_end,
|
||
"invoice_note": i_note
|
||
}
|
||
|
||
# --- PFLICHTFELD-VALIDIERUNG vor dem Export ---
|
||
# Sammelt blockierende Probleme (Export nicht möglich), damit fehlerhafte/fehlende
|
||
# Pflichtangaben sichtbar werden, statt still mit Platzhaltern oder falschen Werten
|
||
# in die Rechnung zu wandern.
|
||
blocker = []
|
||
if not str(inv_num).strip():
|
||
blocker.append("Rechnungsnummer (BT-1) fehlt.")
|
||
if parse_to_strict_xml_date(inv_date) == "":
|
||
blocker.append("Rechnungsdatum (BT-2) fehlt oder ist unlesbar (erwartet: YYYY-MM-DD). "
|
||
"Bitte korrigieren - es wird KEIN Datum automatisch eingesetzt.")
|
||
if not str(r_email).strip():
|
||
blocker.append("E-Mail des Kunden (BT-49) fehlt. Diese elektronische Adresse ist für "
|
||
"XRechnung/ZUGFeRD ein Pflichtfeld (PEPPOL-EN16931-R010) und wird NICHT "
|
||
"mit einem Platzhalter befüllt. Bitte eintragen.")
|
||
if inv_delivery and parse_to_strict_xml_date(inv_delivery) == "":
|
||
blocker.append("Lieferdatum (BT-72) ist unlesbar (erwartet: YYYY-MM-DD). Bitte korrigieren oder leeren.")
|
||
if not edited_items:
|
||
blocker.append("Die Rechnung enthält keine Position (mindestens eine ist erforderlich).")
|
||
|
||
if blocker:
|
||
st.warning("⚠️ Export blockiert - folgende Pflichtangaben fehlen oder sind fehlerhaft:\n\n"
|
||
+ "\n".join(f"- {b}" for b in blocker))
|
||
|
||
export_disabled = len(blocker) > 0
|
||
|
||
if "UBL XML" in export_format:
|
||
if st.button("🔒 XRechnung exportieren", type="primary", width="stretch", disabled=export_disabled):
|
||
xml_data = generate_compliant_xrechnung(final_invoice_dict)
|
||
st.download_button(label="📥 Download XRechnung XML", data=xml_data, file_name=f"XRechnung_{inv_num}.xml", mime="text/xml")
|
||
else:
|
||
if st.button("🔒 ZUGFeRD PDF exportieren", type="primary", width="stretch", disabled=export_disabled):
|
||
try:
|
||
cii_xml_string = generate_compliant_zugferd_cii(final_invoice_dict)
|
||
|
||
uploaded_file.seek(0)
|
||
original_pdf_bytes = uploaded_file.read()
|
||
|
||
# 1) factur-x übernimmt: XSD-Validierung der CII-XML, korrektes Einbetten
|
||
# (Filespec mit AFRelationship=/Data, /F + /UF, Subtype text/xml) und die
|
||
# komplette PDF/A-3-konforme XMP-Metadatenstruktur (pdfaid:part/conformance,
|
||
# fx:-Schema-Erweiterung). Bricht die XML-Validierung fehl, wird hier ein
|
||
# klarer Fehler geworfen statt einer stillen, ungültigen PDF.
|
||
zugferd_pdf_bytes = facturx_generate_from_binary(
|
||
original_pdf_bytes,
|
||
cii_xml_string.encode("utf-8"),
|
||
flavor="factur-x",
|
||
level="en16931",
|
||
check_xsd=True,
|
||
check_schematron=False, # benötigt optional 'saxonche' für Schematron-Regeln
|
||
afrelationship="data",
|
||
)
|
||
|
||
# 2) Zusätzliches OutputIntent (sRGB-ICC-Profil) ergänzen. Das ist für echte
|
||
# PDF/A-3-Konformität zwingend, wird von factur-x aber nicht selbst gesetzt.
|
||
zugferd_pdf_bytes = add_pdfa_output_intent(zugferd_pdf_bytes)
|
||
zugferd_pdf_bytes = fix_xrechnung_xmp_conformance(zugferd_pdf_bytes)
|
||
|
||
st.download_button(
|
||
label="📥 Download ZUGFeRD Hybrid-PDF",
|
||
data=zugferd_pdf_bytes,
|
||
file_name=f"ZUGFeRD_{inv_num}.pdf",
|
||
mime="application/pdf"
|
||
)
|
||
st.info(
|
||
"ℹ️ Hinweis: Die eingebettete CII-XML wurde gegen das offizielle EN16931-XSD-Schema "
|
||
"validiert. Für eine vollständige PDF/A-3-Zertifizierung (z. B. Schriftarten-Einbettung "
|
||
"der Ursprungs-PDF) empfiehlt sich zusätzlich eine Prüfung mit veraPDF."
|
||
)
|
||
except Exception as e:
|
||
st.error(f"Fehler bei der ZUGFeRD-Generierung: {e}")
|
||
|
||
# ==========================================
|
||
# FOOTER
|
||
# ==========================================
|
||
st.markdown("---")
|
||
st.caption(
|
||
"📝 **Hinweise:**\n\n"
|
||
"**Human-in-the-Loop:** "
|
||
"Das Tool dient als Automations-Assistent. Vor dem finalen XML-Export ist eine visuelle Endkontrolle der extrahierten Daten "
|
||
"(insbesondere Rechnungsnummer und Empfängeradresse) durch den Sachbearbeiter zwingend erforderlich.\n\n"
|
||
"**GoBD-Konformität:** "
|
||
"Das Tool generiert ausschließlich das strukturierte XML-Format. Für die Einhaltung der GoBD-Richtlinien "
|
||
"ist der Betreiber der Software selbst verantwortlich. \n\n"
|
||
) |