redeployment

.dockerignore

@@ -0,0 +1,6 @@
+__pycache__/
+*.pyc
+*.pyo
+*.pyd
+.env
+temp/

Dockerfile

@@ -1,25 +1,18 @@
-FROM python:3.10
+FROM python:3.10-slim
 
 WORKDIR /app
 
 RUN apt-get update && apt-get install -y \
-  libglib2.0-0 \
-  libsm6 \
-  libxext6 \
-  libxrender1 \
-  libgomp1 \
   libgl1 \
+  libglib2.0-0 \
   && rm -rf /var/lib/apt/lists/*
 
-ENV PADDLE_PDX_DISABLE_MODEL_SOURCE_CHECK=True
-ENV OMP_NUM_THREADS=1
-
 COPY requirements.txt .
-RUN pip install --upgrade pip
+
 RUN pip install --no-cache-dir -r requirements.txt
 
-COPY app ./app
+COPY . .
 
 EXPOSE 7860
 
-CMD ["sh", "-c", "uvicorn app.main:app --host 0.0.0.0 --port ${PORT:-7860}"]
+CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860"]

app/main.py

@@ -1,28 +1,37 @@
-from fastapi import FastAPI, UploadFile, File, HTTPException
-from app.ocr_engine import OCREngine
-from app.schemas import OCRResponse, OCRLine
+from fastapi import FastAPI, UploadFile, File
+from fastapi.responses import JSONResponse
+import shutil
+import os
+import uuid
 
-app = FastAPI(
-    title="Invoice OCR API",
-    version="1.0.0"
-)
+from app.ocr import run_ocr
 
-ocr_engine = OCREngine()
+app = FastAPI(title="Invoice OCR API")
 
+UPLOAD_DIR = "temp"
+os.makedirs(UPLOAD_DIR, exist_ok=True)
 
-@app.get("/")
-def health():
-    return {"status": "OCR API running"}
 
-
-@app.post("/ocr", response_model=OCRResponse)
+@app.post("/ocr")
 async def ocr_endpoint(file: UploadFile = File(...)):
-    if not file.content_type.startswith("image/"):
-        raise HTTPException(status_code=400, detail="File must be an image")
+    try:
+        file_id = str(uuid.uuid4())
+        file_path = os.path.join(UPLOAD_DIR, f"{file_id}.jpg")
+
+        with open(file_path, "wb") as buffer:
+            shutil.copyfileobj(file.file, buffer)
+
+        result = run_ocr(file_path)
+
+        os.remove(file_path)
 
-    image_bytes = await file.read()
-    lines = ocr_engine.extract_text(image_bytes)
+        return JSONResponse({
+            "status": "success",
+            "text_blocks": result
+        })
 
-    return OCRResponse(
-        lines=[OCRLine(**line) for line in lines]
-    )
+    except Exception as e:
+        return JSONResponse({
+            "status": "error",
+            "message": str(e)
+        }, status_code=500)

app/ocr.py

@@ -0,0 +1,25 @@
+from paddleocr import PaddleOCR
+import os
+
+# Load once at startup
+ocr = PaddleOCR(
+    use_angle_cls=True,
+    lang='en',
+    use_gpu=False,
+    show_log=False
+)
+
+def run_ocr(image_path: str):
+    result = ocr.ocr(image_path, cls=True)
+
+    extracted_text = []
+    for line in result:
+        for word_info in line:
+            text = word_info[1][0]
+            confidence = float(word_info[1][1])
+            extracted_text.append({
+                "text": text,
+                "confidence": confidence
+            })
+
+    return extracted_text

app/ocr_engine.py

@@ -1,39 +0,0 @@
-from paddleocr import PaddleOCR
-import numpy as np
-import cv2
-
-
-class OCREngine:
-    def __init__(self):
-        self.ocr = PaddleOCR(
-            ocr_version="PP-OCRv4",
-            use_doc_orientation_classify=False,
-            use_doc_unwarping=False,
-            use_textline_orientation=False,
-            lang="en"
-        )
-
-    def extract_text(self, image_bytes):
-        nparr = np.frombuffer(image_bytes, np.uint8)
-        img = cv2.imdecode(nparr, cv2.IMREAD_COLOR)
-
-        if img is None:
-            return []
-
-        # Do NOT resize aggressively
-        result = self.ocr.ocr(img)
-
-        lines = []
-
-        if result and isinstance(result, list) and len(result) > 0:
-            for line in result[0]:
-                text = line[1][0]
-                score = float(line[1][1])
-
-                if text.strip():
-                    lines.append({
-                        "text": text,
-                        "confidence": score
-                    })
-
-        return lines

requirements.txt

@@ -1,7 +1,7 @@
-fastapi
-uvicorn[standard]
-paddleocr==3.4.0
-paddlepaddle==3.2.2
+fastapi==0.110.0
+uvicorn==0.29.0
+paddleocr==2.7.0.3
+paddlepaddle==2.6.1
 python-multipart
-opencv-python-headless
 pillow
+opencv-python-headless