Admin
c176c7ce39
New tool for processing damage protocols: - Tab 1: Generate pre-filled PDF from Schadenlaufzettel.docx - Tab 2: Analyze filled PDF forms and extract dropdown selections - Tab 3: Generate Vorbericht.docx from filled PDF data Backend: Flask API with pypdf and python-docx Frontend: Vanilla JS SPA with drag-drop file upload 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
137 lines
4.6 KiB
Python
137 lines
4.6 KiB
Python
"""
|
|
Schadenprotokoll Processors - PDF/DOCX Verarbeitung
|
|
"""
|
|
from docx import Document
|
|
from docx.oxml.ns import qn
|
|
from pypdf import PdfReader, PdfWriter
|
|
import zipfile
|
|
import xml.etree.ElementTree as ET
|
|
|
|
|
|
def parse_laufzettel(docx_path: str) -> dict:
|
|
"""Extrahiert Schadensdaten aus dem Laufzettel.docx"""
|
|
data = {}
|
|
|
|
with zipfile.ZipFile(docx_path, "r") as z:
|
|
xml_content = z.read("word/document.xml")
|
|
root = ET.fromstring(xml_content)
|
|
|
|
ns = {"w": "http://schemas.openxmlformats.org/wordprocessingml/2006/main"}
|
|
|
|
for sdt in root.findall(".//w:sdt", ns):
|
|
alias = sdt.find(".//w:alias", ns)
|
|
if alias is not None:
|
|
alias_val = alias.get(qn("w:val"))
|
|
texts = sdt.findall(".//w:t", ns)
|
|
content = " ".join([t.text for t in texts if t.text]).strip()
|
|
|
|
field_map = {
|
|
"Schadennummer": "Schadennummer",
|
|
"Versicherungsnehmer": "Versicherungsnehmer:",
|
|
"Versicherer": "Versicherer:",
|
|
"Datum": "Datum:",
|
|
"Schadenort": "Schadenort:",
|
|
"Schadenart": "Schadenart:",
|
|
"e-Mail": "e-Mail VN:",
|
|
}
|
|
|
|
if alias_val in field_map:
|
|
data[field_map[alias_val]] = content
|
|
elif alias_val == "Straße":
|
|
data["Adresse:"] = content
|
|
elif alias_val == "Ort":
|
|
if "Adresse:" in data:
|
|
data["Adresse:"] += ", " + content
|
|
else:
|
|
data["Adresse:"] = content
|
|
|
|
return data
|
|
|
|
|
|
def fill_pdf(template_path: str, output_path: str, data: dict) -> str:
|
|
"""Befuellt das PDF-Template mit den Schadensdaten"""
|
|
reader = PdfReader(template_path)
|
|
writer = PdfWriter()
|
|
|
|
for page in reader.pages:
|
|
writer.add_page(page)
|
|
if "/Annots" in page:
|
|
writer.update_page_form_field_values(
|
|
writer.pages[-1], data, auto_regenerate=False
|
|
)
|
|
|
|
with open(output_path, "wb") as f:
|
|
writer.write(f)
|
|
|
|
return output_path
|
|
|
|
|
|
def analyze_pdf(pdf_path: str) -> dict:
|
|
"""Liest alle Formularfelder aus dem ausgefuellten PDF"""
|
|
reader = PdfReader(pdf_path)
|
|
result = {"textfields": {}, "dropdowns": {}, "checkboxes": {}}
|
|
|
|
for page_num, page in enumerate(reader.pages):
|
|
if "/Annots" not in page:
|
|
continue
|
|
|
|
for annot in page["/Annots"]:
|
|
obj = annot.get_object()
|
|
field_type = obj.get("/FT", "")
|
|
field_name = obj.get("/T", f"Feld_{page_num}")
|
|
|
|
if field_type == "/Tx":
|
|
value = obj.get("/V", "")
|
|
result["textfields"][field_name] = str(value) if value else ""
|
|
|
|
elif field_type == "/Ch":
|
|
value = obj.get("/V", "")
|
|
options = []
|
|
if "/Opt" in obj:
|
|
for opt in obj["/Opt"]:
|
|
if isinstance(opt, list):
|
|
options.append(str(opt[1]) if len(opt) > 1 else str(opt[0]))
|
|
else:
|
|
options.append(str(opt))
|
|
result["dropdowns"][field_name] = {
|
|
"selected": str(value) if value else "",
|
|
"options": options
|
|
}
|
|
|
|
elif field_type == "/Btn":
|
|
value = obj.get("/V", "/Off")
|
|
result["checkboxes"][field_name] = value != "/Off"
|
|
|
|
return result
|
|
|
|
|
|
def generate_vorbericht(pdf_data: dict, template_path: str, output_path: str) -> str:
|
|
"""Generiert Vorbericht.docx aus PDF-Daten"""
|
|
doc = Document(template_path)
|
|
|
|
tf = pdf_data.get("textfields", {})
|
|
replacements = {
|
|
"Schadennummer": tf.get("Schadennummer", "xx"),
|
|
"Versicherer": tf.get("Versicherer:", "xx"),
|
|
"Versicherungsnehmer": tf.get("Versicherungsnehmer:", "xx"),
|
|
"Adresse": tf.get("Adresse:", "xx"),
|
|
}
|
|
|
|
# Dropdown-Antworten fuer Sachverhalt sammeln
|
|
dropdowns = pdf_data.get("dropdowns", {})
|
|
sachverhalt = []
|
|
for key in sorted(dropdowns.keys()):
|
|
val = dropdowns[key].get("selected", "")
|
|
if val and val not in ["Was ist beschädigt?", "Schadenursache?", ""]:
|
|
sachverhalt.append(val)
|
|
|
|
for para in doc.paragraphs:
|
|
text = para.text
|
|
for key, value in replacements.items():
|
|
if key.lower() in text.lower() and "xx" in text:
|
|
para.text = text.replace("xx", value, 1)
|
|
break
|
|
|
|
doc.save(output_path)
|
|
return output_path
|