contractapp_forapi.py

import pdfplumber
import json
import re
from fastapi import UploadFile

# Load your mapping JSON file
with open("field_mapping_contract.json", "r", encoding="utf-8") as f:
    nested_mapping = json.load(f)

# Helper: flatten mapping (English path => Hungarian label)
def flatten_mapping(d, parent_key=''):
    items = {}
    for k, v in d.items():
        new_key = f"{parent_key}.{k}" if parent_key else k
        if isinstance(v, dict):
            items.update(flatten_mapping(v, new_key))
        elif isinstance(v, list):
            for idx, label in enumerate(v):
                items[f"{new_key}[{idx}]"] = label
        else:
            items[new_key] = v
    return items

# Helper: set nested value from flat key
def set_nested(data, key_path, value):
    parts = re.split(r'\.|\[|\]', key_path)
    parts = [p for p in parts if p != '']
    d = data
    for part in parts[:-1]:
        if part.isdigit():
            part = int(part)
            while len(d) <= part:
                d.append({})
            d = d[part]
        else:
            if part not in d:
                d[part] = [] if parts[parts.index(part)+1].isdigit() else {}
            d = d[part]
    last = parts[-1]
    if last.isdigit():
        last = int(last)
        while len(d) <= last:
            d.append(None)
        d[last] = value
    else:
        d[last] = value

# Helper: extract value using regex
def extract_value(label, text, label_list):
    if label == "Dátum":
        lines = text.splitlines()
        for i, line in enumerate(lines):
            if "Dátum" in line:
                #print(f"✅ Found line with 'Dátum': {line}")
                if i + 1 < len(lines):
                    next_line = lines[i + 1]
                    #print(f"🔍 Checking next line: {next_line}")
                    dates = re.findall(r'\d{4}-\d{2}-\d{2}', next_line)
                    if len(dates) >= 2:
                        combined = f"{dates[0]} | {dates[1]}"
                        #print(f"✅ Combined dates: {combined}")
                        return combined
                    elif len(dates) == 1:
                        return dates[0]  # fallback: only one date found
                #print("❌ No date found in next line")
                return None


    pattern = re.escape(label) + r'[:\s]*([^\n]+)'
    match = re.search(pattern, text)
    if match:
        value = match.group(1).strip()

        # Remove any prefixing numbers before the actual label
        if label == "Adószám":
            # Extract just the number after the label
            num_match = re.search(rf"{re.escape(label)}[:\s]*([0-9]+)", text)
            if num_match:
                return num_match.group(1).strip()

        # Clean up by removing the next label if it bleeds in
        for other in label_list:
            if other != label and other in value:
                value = value.split(other)[0].strip()

        return value
    return None


# Read the PDF
def read_pdf_text(file: UploadFile):
    with pdfplumber.open(file.file) as pdf:
        text = "\n".join(page.extract_text() for page in pdf.pages if page.extract_text())
    return text

# Main logic
def extract_data(file: UploadFile):
    flat_map = flatten_mapping(nested_mapping)
    text = read_pdf_text(file)
    result = {}

    for eng_path, hun_label in flat_map.items():
        val = extract_value(hun_label, text, flat_map.values())
        if val:
            set_nested(result, eng_path, val)

    result["STATUS"] = "OK"
    result["FILENAME"] = file.filename
    result["AGENCY"] = "Wavemaker Hungary Kft."

    return json.dumps(result, indent=2, ensure_ascii=False)

# Run the pipeline
#if __name__ == "__main__":
    #pdf_path = "163193_WAVEMAKER_HUNGARY_Order_WMH0245_487-2025_Outlet_Center_Parndo_Parndorf_Radio_Station_Game_atmedia_Kft.pdf"  # Change to your actual PDF file path
    #output = extract_data(pdf_path, nested_mapping)

    # Pretty print or save
    #import pprint
    #pprint.pprint(output, sort_dicts=False)

    # Optional: Save to JSON file
    # with open("output.json", "w", encoding="utf-8") as f:
    #     json.dump(output, f, ensure_ascii=False, indent=2)