pip install pdfplumber openai

import pdfplumber

def extract_text_from_pdf(pdf_path):
    full_text = ""
    with pdfplumber.open(pdf_path) as pdf:
        for page in pdf.pages:
            full_text += page.extract_text() or ""
            full_text += "\n---PAGE BREAK---\n"
    return full_text

text = extract_text_from_pdf("invoice.pdf")
print(text[:2000])

import openai
import json

client = openai.OpenAI(api_key="YOUR_API_KEY")

def extract_invoice_data(text):
    prompt = f"""Extract the following fields from this invoice text.
Return ONLY valid JSON, no explanation.

Fields to extract:
- invoice_number
- date (YYYY-MM-DD format)
- vendor_name
- customer_name
- total_amount (number only)
- currency
- line_items (array of objects with: description, quantity, unit_price)

Invoice text:
{text}"""

    response = client.chat.completions.create(
        model="gpt-4-turbo",
        messages=[{"role": "user", "content": prompt}],
        response_format={"type": "json_object"}
    )
    return json.loads(response.choices[0].message.content)

data = extract_invoice_data(text)
print(json.dumps(data, ensure_ascii=False, indent=2))

import os
import csv
import pdfplumber

PDF_FOLDER = "./pdfs"
OUTPUT_CSV = "extracted_data.csv"

fieldnames = [
    "invoice_number", "date", "vendor_name",
    "customer_name", "total_amount", "currency", "source_file"
]

with open(OUTPUT_CSV, "w", newline="", encoding="utf-8") as csvfile:
    writer = csv.DictWriter(csvfile, fieldnames=fieldnames)
    writer.writeheader()

    for filename in os.listdir(PDF_FOLDER):
        if not filename.endswith(".pdf"):
            continue
        path = os.path.join(PDF_FOLDER, filename)
        text = extract_text_from_pdf(path)
        data = extract_invoice_data(text)
        data["source_file"] = filename
        flat_data = {k: data.get(k, "N/A") for k in fieldnames}
        writer.writerow(flat_data)
        print(f"Processed: {filename}")

print(f"Done. Output saved to {OUTPUT_CSV}")

name: Process PDFs Nightly

on:
  schedule:
    - cron: '0 2 * * *'
  workflow_dispatch:

jobs:
  extract:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3
      - name: Set up Python
        uses: actions/setup-python@v4
        with:
          python-version: '3.11'
      - name: Install dependencies
        run: pip install pdfplumber openai
      - name: Run extraction
        env:
          OPENAI_API_KEY: '${{ secrets.OPENAI_API_KEY }}'
        run: python extract.py
      - name: Upload CSV artifact
        uses: actions/upload-artifact@v3
        with:
          name: extracted-data
          path: extracted_data.csv