Create app.py

app.py

@@ -0,0 +1,75 @@
+from fastapi import FastAPI, File, UploadFile, Form
+from fastapi.responses import JSONResponse
+from fastapi.middleware.cors import CORSMiddleware
+import fitz  # PyMuPDF
+import io
+import re
+import base64
+
+app = FastAPI(title="Invoice Splitter API")
+
+# Allow CORS (optional but helpful for Flutter/JS frontend)
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+@app.post("/split-invoices")
+async def split_invoices(
+    file: UploadFile = File(...),
+    include_pdf: bool = Form(True),
+    initial_dpi: int = Form(300)
+):
+    try:
+        pdf_data = await file.read()
+        pdf = fitz.open(stream=pdf_data, filetype="pdf")
+
+        invoice_pattern = re.compile(r"\b[A-Z0-9]{3,10}\b")  # Example pattern
+        splits = []
+        current_invoice = None
+        current_pages = []
+
+        for page_num, page in enumerate(pdf, start=1):
+            text = page.get_text("text")
+
+            match = re.search(r"Invoice\s*No[:\s\-]*([A-Z0-9]+)", text, re.I)
+            if match:
+                invoice_no = match.group(1)
+                if current_invoice:
+                    splits.append({
+                        "invoice_no": current_invoice,
+                        "pages": current_pages.copy()
+                    })
+                    current_pages.clear()
+                current_invoice = invoice_no
+
+            current_pages.append(page_num)
+
+        if current_invoice and current_pages:
+            splits.append({"invoice_no": current_invoice, "pages": current_pages})
+
+        results = []
+        for split in splits:
+            doc = fitz.open()
+            for pno in split["pages"]:
+                doc.insert_pdf(pdf, from_page=pno-1, to_page=pno-1)
+            pdf_bytes = doc.tobytes()
+            base64_pdf = base64.b64encode(pdf_bytes).decode('utf-8') if include_pdf else None
+
+            results.append({
+                "invoice_no": split["invoice_no"],
+                "num_pages": len(split["pages"]),
+                "pages": split["pages"],
+                "pdf_base64": base64_pdf
+            })
+
+        return JSONResponse({
+            "count": len(results),
+            "parts": results
+        })
+
+    except Exception as e:
+        return JSONResponse({"error": str(e)}, status_code=500)