import requests
from requests.adapters import HTTPAdapter
from urllib3.util.retry import Retry
# 定義超時(shí)配置
TIMEOUT = (10, 30)  # (connect_timeout, read_timeout)

def create_session_with_retry(
    total=3,
    backoff_factor=1,
    status_forcelist=[429, 500, 502, 503, 504]
):
    """
    創(chuàng)建帶有重試機(jī)制的requests會(huì)話
    """
    session = requests.Session()
    retry_strategy = Retry(
        total=total,                    # 總重試次數(shù)（含首次）
        status_forcelist=status_forcelist,  # 觸發(fā)重試的狀態(tài)碼
        method_whitelist=["POST"],      # 允許重試的HTTP方法
        backoff_factor=backoff_factor, # 退避因子：等待時(shí)間為 {backoff_factor} * (2 ^ (重試次數(shù) - 1))
        raise_on_redirect=False,
        raise_on_status=False
    )
    adapter = HTTPAdapter(max_retries=retry_strategy)
    session.mount("http://", adapter)
    session.mount("https://", adapter)
    return session

import time
import logging
# 配置日志
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
def ocr_recognize(image_path, api_url="http://localhost:8080/ocr"):
    """
    調(diào)用CRNN OCR服務(wù)進(jìn)行文字識(shí)別
    """
    session = create_session_with_retry(total=3, backoff_factor=1)
    try:
        with open(image_path, 'rb') as f:
            files = {'image': f}
            start_time = time.time()
            response = session.post(
                api_url,
                files=files,
                timeout=TIMEOUT
            )
            end_time = time.time()
        logger.info(f"? 請(qǐng)求成功 | 耗時(shí): {end_time - start_time:.2f}s | 狀態(tài)碼: {response.status_code}")
        if response.status_code == 200:
            result = response.json()
            return result.get("text", ""), result.get("confidence", [])
        else:
            logger.error(f"? 服務(wù)返回錯(cuò)誤狀態(tài): {response.status_code}, 內(nèi)容: {response.text}")
            return None, []
    except requests.exceptions.Timeout:
        logger.error("? 請(qǐng)求超時(shí)：請(qǐng)檢查網(wǎng)絡(luò)或增加read_timeout")
    except requests.exceptions.ConnectionError as e:
        logger.error(f"?? 連接失?。簕e}")
    except requests.exceptions.RequestException as e:
        logger.error(f"?? 其他請(qǐng)求異常：{e}")
    finally:
        session.close()
    return None, []

text, confidences = ocr_recognize("invoice.jpg")
if text:
    print("識(shí)別結(jié)果：", text)

import os

def get_timeout_by_size(image_path):
    file_size_kb = os.path.getsize(image_path) / 1024
    if file_size_kb < 100:
        return (10, 15)
    elif file_size_kb < 500:
        return (10, 25)
    else:
        return (10, 40)  # 大圖預(yù)留更多時(shí)間

session = create_session_with_retry()

for img_path in image_list:
    timeout = get_timeout_by_size(img_path)
    # ... 使用同一session發(fā)送請(qǐng)求

import uuid

headers = {
    "X-Request-ID": str(uuid.uuid4())
}
response = session.post(api_url, files=files, timeout=timeout, headers=headers)

from concurrent.futures import ThreadPoolExecutor, as_completed

def batch_ocr(images, max_workers=3):
    results = {}
    with ThreadPoolExecutor(max_workers=max_workers) as executor:
        future_to_img = {
            executor.submit(ocr_recognize, img): img for img in images
        }
        for future in as_completed(future_to_img):
            img = future_to_img[future]
            try:
                text, _ = future.result()
                results[img] = text
            except Exception as e:
                results[img] = f"Error: {e}"
    return results

import requests
from requests.adapters import HTTPAdapter
from urllib3.util.retry import Retry
import logging
import time
# --- 配置區(qū) ---
API_URL = "http://your-ocr-service:8080/ocr"
TIMEOUT_BASE = (10, 30)
RETRY_TOTAL = 3
BACKOFF_FACTOR = 1
MAX_WORKERS = 3
# --- 日志 ---
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
# --- 工具函數(shù) ---
def create_retry_session():
    session = requests.Session()
    retry = Retry(
        total=RETRY_TOTAL,
        backoff_factor=BACKOFF_FACTOR,
        status_forcelist=[429, 500, 502, 503, 504],
        allowed_methods=["POST"]
    )
    adapter = HTTPAdapter(max_retries=retry)
    session.mount("http://", adapter)
    session.mount("https://", adapter)
    return session
def ocr_request(image_path):
    session = create_retry_session()
    try:
        with open(image_path, 'rb') as f:
            files = {'image': f}
            resp = session.post(API_URL, files=files, timeout=TIMEOUT_BASE)
        if resp.status_code == 200:
            return resp.json().get("text", "")
        else:
            logger.warning(f"Status {resp.status_code}: {resp.text}")
            return None
    except Exception as e:
        logger.error(f"Request failed: {e}")
        return None
    finally:
        session.close()