'commit'

.idea/vcs.xml

@@ -3,5 +3,6 @@
   <component name="VcsDirectoryMappings">
     <mapping directory="$PROJECT_DIR$/../.." vcs="Git" />
     <mapping directory="$PROJECT_DIR$/.." vcs="Git" />
+    <mapping directory="$PROJECT_DIR$" vcs="Git" />
   </component>
 </project>

Apps/XinDianTu/Config/Setting.py

@@ -0,0 +1,31 @@
+
+# 采集配置
+# 滑动距离比例 (0.1 ~ 0.9)，数值越大滑动幅度越大，建议 0.3-0.5 以避免错过中间内容
+SCROLL_DISTANCE_RATIO = 0.3
+# 最大滑动/翻页次数，达到此次数后停止采集
+MAX_SCROLLS = 100
+# 默认抓取半径（公里），当检测到场站距离超过此值时停止采集
+MAX_CRAWL_DISTANCE = 50
+# 场站去重过期时间（秒），在此时间内重复出现的场站不会再次点击进入详情页
+REDIS_STATION_EXPIRE = 120
+
+# 数据库数据保留时长（天），超过此时长的历史数据（is_current=0）将被删除
+DATA_RETENTION_DAYS = 365
+
+# 等待时间配置 (秒)
+# 点击进入详情页后等待加载的时间
+WAIT_DETAIL_PAGE_LOAD = 2.5
+# 从详情页返回列表页后等待页面刷新的时间
+WAIT_BACK_TO_LIST = 1.5
+# 执行滑动操作后等待页面内容加载和稳定的时间
+WAIT_AFTER_SCROLL = 3.0
+
+# 坐标计算与安全防护
+# 屏幕顶部安全排除比例 (0.0~1.0)，此比例区域内不进行点击（避开状态栏、筛选栏等）
+SAFE_EXCLUDE_RATIO = 0.20
+# 屏幕底部安全排除比例 (0.0~1.0)，此比例区域内不进行点击（避开底部导航栏、功能按钮等）
+BOTTOM_SAFE_EXCLUDE_RATIO = 0.1
+# 默认回退屏幕宽度，当无法自动获取设备信息时使用
+FALLBACK_WIDTH = 1080
+# 默认回退屏幕高度，当无法自动获取设备信息时使用
+FALLBACK_HEIGHT = 2400

Apps/XinDianTu/Crawler.py

@@ -27,7 +27,7 @@ from Config.Config import (
     OBS_TMP_PREFIX, CDN_DOMAIN, SCROLL_DISTANCE_RATIO,
     MAX_SCROLLS, REDIS_STATION_EXPIRE,
     WAIT_DETAIL_PAGE_LOAD, WAIT_BACK_TO_LIST, WAIT_AFTER_SCROLL,
-    MAX_CRAWL_DISTANCE
+    MAX_CRAWL_DISTANCE, TEMP_IMAGE_DIR
 )
 
 # --- 用户配置区域 ---
@@ -128,7 +128,7 @@ async def get_station_list(d, service, uploader, max_scrolls=MAX_SCROLLS):
         image_uuid = str(uuid.uuid4())
         # 使用相对路径: 基于当前脚本目录下的 Images 文件夹
         base_dir = os.path.dirname(os.path.abspath(__file__))
-        save_dir = os.path.join(base_dir, "./Images")
+        save_dir = TEMP_IMAGE_DIR
         screenshot_path = take_screenshot(d, image_uuid, save_dir=save_dir)
         logger.info(f"Step [1/6] 列表页截图已完成: {screenshot_path} (耗时: {time.time() - t_shot:.2f}s)")
         
@@ -383,7 +383,7 @@ async def get_station_list(d, service, uploader, max_scrolls=MAX_SCROLLS):
                     # 使用几何特征识别 "全部时段" 按钮
                     # 临时截图
                     temp_uuid = "temp_find_expand"
-                    screenshot_path = take_screenshot(d, temp_uuid, save_dir="Temp")
+                    screenshot_path = take_screenshot(d, temp_uuid, save_dir=TEMP_IMAGE_DIR)
                     
                     # 尝试识别，将调试图片保存到 Images 目录
                     t_find = time.time()

Apps/XinDianTu/Kit.py

@@ -3,7 +3,7 @@ import os
 import cv2
 import numpy as np
 import time
-from Config.Config import BOTTOM_SAFE_EXCLUDE_RATIO
+from Config.Config import BOTTOM_SAFE_EXCLUDE_RATIO, TEMP_IMAGE_DIR
 
 logging.basicConfig(level=logging.INFO, format="%(asctime)s - %(name)s - %(levelname)s - %(message)s")
 logger = logging.getLogger(__name__)
@@ -29,7 +29,7 @@ def save_image(path, img):
         return False
 
 # 截图
-def take_screenshot(d, image_uuid, save_dir="Screenshot"):    
+def take_screenshot(d, image_uuid, save_dir=TEMP_IMAGE_DIR):    
     path = f"{save_dir}/{image_uuid}.jpg"    
     os.makedirs(save_dir, exist_ok=True)
     d.screenshot(path)    
@@ -61,7 +61,7 @@ def click_image_template(d, template_path, timeout=5.0, threshold=0.8):
     while time.time() - start_time < timeout:
         # 临时截图
         temp_uuid = "temp_click_check"
-        screenshot_path = take_screenshot(d, temp_uuid, save_dir="Temp")
+        screenshot_path = take_screenshot(d, temp_uuid, save_dir=TEMP_IMAGE_DIR)
         
         target = read_image(screenshot_path)
         if target is None:

Apps/XinDianTu/Opener.py

@@ -8,7 +8,7 @@ import uiautomator2 as u2
 from Util.Kit import take_screenshot, detect_black_agree_button, click_image_template, detect_ad_close_x, detect_any_ad_close, detect_bottom_close_circle
 from Util.ObsUtil import ObsUploader
 from Util.XinDianTuReadImageKit import XinDianTuReadImageKit
-from Config.Config import OBS_TMP_PREFIX, CDN_DOMAIN
+from Config.Config import OBS_TMP_PREFIX, CDN_DOMAIN, TEMP_IMAGE_DIR
 
 # pip install adbutils
 # 配置日志输出，方便调试和监控
@@ -29,7 +29,7 @@ async def check_and_close_ad(d):
     image_uuid = str(uuid.uuid4())
     # 使用相对路径: 基于当前脚本目录下的 Images 文件夹
     base_dir = os.path.dirname(os.path.abspath(__file__))
-    save_dir = os.path.join(base_dir, "./Images")
+    save_dir = TEMP_IMAGE_DIR
     screenshot_path = take_screenshot(d, image_uuid, save_dir=save_dir)
     logger.info(f"Step [广告检测截图] 耗时: {time.time() - t1:.4f}s")
     

Config/Config.py

@@ -1,52 +1,21 @@
-# 开发环境不同，配置信息不同
-# Doris V4
-# DORIS_HOST = "10.10.14.204"
-# DORIS_PORT = 9030
-# DORIS_FENODES = "10.10.14.204:8030"
-# REDIS_HOST = '10.10.14.14'
-# REDIS_PASSWORD = None  # 如果没有密码则设为 None
-DORIS_HOST = "www.hzkjai.com"
-DORIS_PORT = 27025
-DORIS_FENODES = "www.hzkjai.com:27024"
-REDIS_HOST = '127.0.0.1'
-REDIS_PASSWORD = "DsideaL147258369"
+# 黄海在公司内网开发时的配置信息
+DORIS_HOST = "10.10.14.204"
+DORIS_PORT = 9030
+DORIS_FENODES = "10.10.14.204:8030"
+REDIS_HOST = '10.10.14.14'
+REDIS_PASSWORD = None  # 如果没有密码则设为 None
 
-# 采集配置
-# 滑动距离比例 (0.1 ~ 0.9)，数值越大滑动幅度越大，建议 0.3-0.5 以避免错过中间内容
-SCROLL_DISTANCE_RATIO = 0.3
-# 最大滑动/翻页次数，达到此次数后停止采集
-MAX_SCROLLS = 100
-# 默认抓取半径（公里），当检测到场站距离超过此值时停止采集
-MAX_CRAWL_DISTANCE = 50
-# 场站去重过期时间（秒），在此时间内重复出现的场站不会再次点击进入详情页
-REDIS_STATION_EXPIRE = 120
-
-# 数据库数据保留时长（天），超过此时长的历史数据（is_current=0）将被删除
-DATA_RETENTION_DAYS = 365
-
-# 等待时间配置 (秒)
-# 点击进入详情页后等待加载的时间
-WAIT_DETAIL_PAGE_LOAD = 2.5
-# 从详情页返回列表页后等待页面刷新的时间
-WAIT_BACK_TO_LIST = 1.5
-# 执行滑动操作后等待页面内容加载和稳定的时间
-WAIT_AFTER_SCROLL = 3.0
-
-# 坐标计算与安全防护
-# 屏幕顶部安全排除比例 (0.0~1.0)，此比例区域内不进行点击（避开状态栏、筛选栏等）
-SAFE_EXCLUDE_RATIO = 0.20
-# 屏幕底部安全排除比例 (0.0~1.0)，此比例区域内不进行点击（避开底部导航栏、功能按钮等）
-BOTTOM_SAFE_EXCLUDE_RATIO = 0.1
-# 默认回退屏幕宽度，当无法自动获取设备信息时使用
-FALLBACK_WIDTH = 1080
-# 默认回退屏幕高度，当无法自动获取设备信息时使用
-FALLBACK_HEIGHT = 2400
+# 黄海在家开发时的配置信息
+#DORIS_HOST = "www.hzkjai.com"
+#DORIS_PORT = 27025
+#DORIS_FENODES = "www.hzkjai.com:27024"
+#REDIS_HOST = '127.0.0.1'
+#REDIS_PASSWORD = "DsideaL147258369"
 
 # 视觉模型配置
 VL_MODEL_NAME = "qwen3-vl-flash"
 VL_MODEL_NAME_AD = "qwen-vl-max"
 
-
 # 华为云配置
 OBS_AK = "WAFBGJACKDOQZDH1MKZ1"
 OBS_SK = "dlWTUbqgCICaYJG3n0Rot4jXaen2HnfFtMVxiPEo"
@@ -75,10 +44,12 @@ doris = {
     "database": DORIS_DATABASE
 }
 
-
 # REDIS 配置
 REDIS_DB = 2
 REDIS_DECODE_RESPONSES = True
 REDIS_PORT = 18890
 REDIS_MAX_CONNECTIONS = 200
 
+# 临时图片存储路径
+TEMP_IMAGE_DIR = r"d:\dsWork\aiData\Output"
+

Test/TestConflict.py

@@ -1,29 +0,0 @@
-# coding=utf-8
-import os
-import sys
-import time
-import numpy as np
-
-# Add project root to sys.path
-current_dir = os.path.dirname(os.path.abspath(__file__))
-# project_root = os.path.dirname(current_dir)
-# if project_root not in sys.path:
-#    sys.path.append(project_root)
-
-print("Importing PaddleOCR...")
-from paddleocr import PaddleOCR
-print("Importing LlmUtil...")
-# from Util.LlmUtil import get_llm_response
-print("Imports done.")
-
-def main():
-    image_path = os.path.join(current_dir, "2.jpg")
-    print(f"Initializing OCR...")
-    ocr = PaddleOCR(use_textline_orientation=True, lang="ch") 
-    print("Running OCR...")
-    result = ocr.ocr(image_path)
-    print("OCR Done.")
-    print(result)
-
-if __name__ == "__main__":
-    main()

Test/TestOCR.py

@@ -1,119 +0,0 @@
-# coding=utf-8
-import os
-import sys
-from paddleocr import PaddleOCR
-import numpy as np
-
-def test_ocr():
-    # 1. 初始化 PaddleOCR
-    print("正在初始化 PaddleOCR 模型...")
-    try:
-        ocr = PaddleOCR(use_textline_orientation=True, lang="ch")
-    except Exception as e:
-        print(f"初始化失败: {e}")
-        return
-
-    # 2. 准备测试图片
-    current_dir = os.path.dirname(os.path.abspath(__file__))
-    # 优先查找 2.jpg，如果不存在则查找 1.jpg
-    image_path = os.path.join(current_dir, "2.jpg")
-    if not os.path.exists(image_path):
-        image_path = os.path.join(current_dir, "1.jpg")
-    
-    if not os.path.exists(image_path):
-        print(f"错误: 未找到测试图片: {image_path}")
-        print("请将测试图片命名为 2.jpg 或 1.jpg 并放置在 Test 目录下。")
-        return
-
-    print(f"正在识别图片: {image_path}")
-
-    # 3. 执行识别
-    try:
-        # result 是一个列表，通常包含一个 OCRResult 对象 (新版) 或列表 (旧版)
-        result = ocr.ocr(image_path)
-             
-    except Exception as e:
-        print(f"识别过程发生异常: {e}")
-        return
-
-    # 4. 输出结果
-    print("\n" + "="*20 + " 识别结果 " + "="*20)
-    sys.stdout.flush()
-    
-    # 同时输出到文件，方便查看
-    output_file = os.path.join(current_dir, "ocr_output.txt")
-    with open(output_file, "w", encoding="utf-8") as f_out:
-        if not result:
-            msg = "未识别到任何文字 (Result is empty)。"
-            print(msg)
-            f_out.write(msg + "\n")
-        else:
-            # 取出第一个结果（通常是单张图片的结果）
-            res = result[0]
-            
-            # 检查是否为 None
-            if res is None:
-                 msg = "未识别到任何文字 (Result[0] is None)。"
-                 print(msg)
-                 f_out.write(msg + "\n")
-            
-            # 情况 A: 新版 PaddleX OCRResult 对象 (表现为字典或对象)
-            elif hasattr(res, 'get') and 'rec_texts' in res:
-                 texts = res.get('rec_texts', [])
-                 scores = res.get('rec_scores', [])
-                 
-                 if not texts:
-                     msg = "未识别到任何文字 (rec_texts is empty)。"
-                     print(msg)
-                     f_out.write(msg + "\n")
-                 else:
-                     for i, text in enumerate(texts):
-                         score = scores[i] if i < len(scores) else 0.0
-                         msg = f"行 {i+1}: {text} (置信度: {score:.4f})"
-                         print(msg)
-                         f_out.write(msg + "\n")
-                         
-            # 情况 B: 对象属性访问
-            elif hasattr(res, 'rec_texts'):
-                 texts = res.rec_texts
-                 scores = res.rec_scores
-                 
-                 if not texts:
-                     msg = "未识别到任何文字 (rec_texts is empty)。"
-                     print(msg)
-                     f_out.write(msg + "\n")
-                 else:
-                     for i, text in enumerate(texts):
-                         score = scores[i] if i < len(scores) else 0.0
-                         msg = f"行 {i+1}: {text} (置信度: {score:.4f})"
-                         print(msg)
-                         f_out.write(msg + "\n")
-
-            # 情况 C: 旧版 list of lists 结构
-            elif isinstance(res, list):
-                 for idx, line in enumerate(res):
-                     try:
-                         if len(line) >= 2 and isinstance(line[1], (tuple, list)):
-                             text, score = line[1]
-                             msg = f"行 {idx+1}: {text} (置信度: {score:.4f})"
-                             print(msg)
-                             f_out.write(msg + "\n")
-                         else:
-                             msg = f"行 {idx+1}: {line} (格式未知)"
-                             print(msg)
-                             f_out.write(msg + "\n")
-                     except Exception as e:
-                         print(f"行 {idx+1} 解析失败: {e}")
-            
-            else:
-                 msg = f"无法解析结果结构: {type(res)}"
-                 print(msg)
-                 print(f"Result content: {res}")
-                 f_out.write(msg + "\n")
-                 f_out.write(f"Result content: {res}\n")
-
-    print("="*50)
-    sys.stdout.flush()
-
-if __name__ == "__main__":
-    test_ocr()

Test/TestOcrLlm.py

@@ -1,173 +0,0 @@
-# coding=utf-8
-import os
-import sys
-import time
-import asyncio
-import json
-
-# Add project root to sys.path
-current_dir = os.path.dirname(os.path.abspath(__file__))
-project_root = os.path.dirname(current_dir)
-if project_root not in sys.path:
-    sys.path.append(project_root)
-
-from paddleocr import PaddleOCR
-from Util.LlmUtil import get_llm_response
-from Util.OcrParser import OcrParser
-import re
-
-LOG_FILE = os.path.join(current_dir, "ocr_llm_debug.txt")
-
-def log(msg):
-    print(msg)
-    sys.stdout.flush()
-    with open(LOG_FILE, "a", encoding="utf-8") as f:
-        f.write(msg + "\n")
-
-# ... imports ...
-def run_ocr_sync():
-    image_path = os.path.join(current_dir, "2.jpg")
-    if not os.path.exists(image_path):
-        image_path = os.path.join(current_dir, "1.jpg")
-    
-    log(f"Testing OCR + LLM Pipeline on: {image_path}")
-    log("-" * 50)
-
-    # --- Step 1: PaddleOCR ---
-    t_start = time.time()
-    
-    log("Initializing PaddleOCR...")
-    t_init_start = time.time()
-    try:
-        # 尝试使用轻量级模型 (Mobile) 以提升速度
-        # ocr_version='PP-OCRv4' 通常默认是 mobile
-        ocr = PaddleOCR(use_textline_orientation=True, lang="ch", ocr_version='PP-OCRv4') 
-    except Exception as e:
-        log(f"PaddleOCR Init Failed: {e}")
-        return None, None
-        
-    t_init_end = time.time()
-    log(f"PaddleOCR Init Time: {t_init_end - t_init_start:.4f}s")
-
-    log("Running OCR Inference...")
-    t_ocr_start = time.time()
-    try:
-        result = ocr.ocr(image_path)
-    except Exception as e:
-        log(f"OCR Inference Failed: {e}")
-        return None, None
-    t_ocr_end = time.time()
-    
-    ocr_text_lines = []
-    
-    # Handle different result structures
-    if not result:
-        log("OCR returned empty result.")
-    else:
-        res = result[0]
-        if res is None:
-             log("OCR result[0] is None.")
-        elif hasattr(res, 'get') and 'rec_texts' in res:
-             ocr_text_lines = res.get('rec_texts', [])
-        elif hasattr(res, 'rec_texts'):
-             ocr_text_lines = res.rec_texts
-        elif isinstance(res, list):
-             for line in res:
-                 if len(line) >= 2 and isinstance(line[1], (tuple, list)):
-                     ocr_text_lines.append(line[1][0])
-    
-    ocr_text_block = "\n".join(ocr_text_lines)
-    log(f"OCR Result ({t_ocr_end - t_ocr_start:.4f}s):")
-    log(ocr_text_block)
-    log("-" * 50)
-    
-    return ocr_text_lines, ocr_text_block, (t_ocr_start, t_ocr_end)
-
-async def run_parsing_comparison(ocr_text_lines, ocr_text_block, timing_ocr):
-    t_ocr_start, t_ocr_end = timing_ocr
-    ocr_duration = t_ocr_end - t_ocr_start
-
-    # --- Mode 1: Regex Parsing ---
-    log("Running Regex Parsing...")
-    t_regex_start = time.time()
-    try:
-        regex_data = OcrParser.parse(ocr_text_lines)
-        log("\nParsed Data (Regex):")
-        log(json.dumps(regex_data, indent=2, ensure_ascii=False))
-    except Exception as e:
-        log(f"Regex Parsing Failed: {e}")
-    t_regex_end = time.time()
-    regex_duration = t_regex_end - t_regex_start
-    log(f"Regex Parsing Time: {regex_duration:.4f}s")
-    log("-" * 50)
-
-    # --- Mode 2: LLM Parsing ---
-    log("Running LLM Parsing...")
-    
-    prompt = f"""
-    You are a data extraction assistant. Below is the OCR text recognized from a charging station list card.
-    Please extract the structured data and return it ONLY as a JSON object (no markdown, no extra text).
-    
-    Fields to extract:
-    - station_name: (String) Name of the charging station.
-    - distance: (String) Distance info (e.g., "7.4km").
-    - price: (String) Price info (e.g., "0.7111/度").
-    - tags: (List[String]) Any tags like "快", "闲3/4", "组团", "2倍积分", "P", etc.
-    - parking_info: (String) Parking related info.
-
-    OCR Text:
-    {ocr_text_block}
-    """
-
-    t_llm_start = time.time()
-    response_text = ""
-    try:
-        log("Starting LLM request...")
-        async for chunk in get_llm_response(prompt, stream=True):
-             print(chunk, end='', flush=True)
-             response_text += chunk
-        
-        print("\n")
-        log("LLM request finished.")
-        
-        t_llm_end = time.time()
-        
-        log(f"\nLLM Response ({t_llm_end - t_llm_start:.4f}s):")
-        log(response_text)
-        
-        try:
-            clean_text = response_text.replace("```json", "").replace("```", "").strip()
-            data = json.loads(clean_text)
-            log("\nParsed JSON Data:")
-            log(json.dumps(data, indent=2, ensure_ascii=False))
-        except json.JSONDecodeError:
-            log("\nFailed to parse JSON directly.")
-
-    except Exception as e:
-        log(f"LLM Error: {e}")
-        t_llm_end = time.time()
-
-    log("-" * 50)
-    log(f"Summary:")
-    log(f"OCR Time: {ocr_duration:.4f}s")
-    log(f"Regex Parsing Time: {regex_duration:.4f}s")
-    log(f"LLM Parsing Time: {t_llm_end - t_llm_start:.4f}s")
-    
-    total_regex = ocr_duration + regex_duration
-    total_llm = ocr_duration + (t_llm_end - t_llm_start)
-    
-    log(f"Total Pipeline (OCR+Regex): {total_regex:.4f}s")
-    log(f"Total Pipeline (OCR+LLM):   {total_llm:.4f}s")
-
-def main():
-    # Clear log file
-    with open(LOG_FILE, "w", encoding="utf-8") as f:
-        f.write("Starting TestOcrLlm...\n")
-        
-    ocr_lines, ocr_text, timing = run_ocr_sync()
-    if ocr_lines:
-        asyncio.run(run_parsing_comparison(ocr_lines, ocr_text, timing))
-
-if __name__ == "__main__":
-    main()
-

Test/TestOcrRegex.py

@@ -1,107 +0,0 @@
-# coding=utf-8
-import re
-import json
-
-def parse_ocr_lines(lines):
-    result = {
-        "station_name": "",
-        "distance": "",
-        "price": None,
-        "piles": [],
-        "parking": "",
-        "tags": []
-    }
-    
-    # Pre-process lines: remove confidence scores for parsing
-    clean_lines = []
-    for line in lines:
-        # Remove prefix "行 X: "
-        text = re.sub(r"^行\s*\d+:\s*", "", line)
-        # Remove suffix "(置信度: ...)"
-        text = re.sub(r"\s*\(置信度:.*?\)\s*$", "", text)
-        clean_lines.append(text.strip())
-
-    # 1. Station Name
-    for line in clean_lines:
-        # Skip empty or short noise
-        if len(line) < 2: continue
-        # Skip if starts with special chars
-        if line.startswith("（") or line.startswith("("): continue
-        # Skip if contains typical attribute keywords
-        if "km" in line.lower() or "/度" in line or "￥" in line: continue
-        # Skip if strictly numeric (unlikely for name)
-        if re.match(r"^\d+$", line): continue
-        
-        result["station_name"] = line
-        break
-        
-    # 2. Distance
-    for line in clean_lines:
-        m = re.search(r"(\d+(\.\d+)?)\s*km", line, re.IGNORECASE)
-        if m:
-            result["distance"] = m.group(0)
-            break
-            
-    # 3. Price (Standard)
-    # Look for "0.xxxx/度"
-    for line in clean_lines:
-        m = re.search(r"(\d+\.\d+)(?=/度)", line)
-        if m:
-            result["price"] = float(m.group(1))
-            break
-    
-    # 4. Piles
-    current_type = "未知"
-    for line in clean_lines:
-        if "快" in line: current_type = "快"
-        elif "慢" in line: current_type = "慢"
-        elif "超" in line: current_type = "超"
-        
-        # Match "闲3/4" or "3/4"
-        # Regex: optional "闲", int, /, int
-        m = re.search(r"(?:闲)?(\d+)/(\d+)", line)
-        if m:
-            # Check if it looks like a price (contains dot)
-            if "." in line: continue
-            
-            free = int(m.group(1))
-            total = int(m.group(2))
-            result["piles"].append({
-                "type": current_type,
-                "free": free,
-                "total": total
-            })
-            
-    # 5. Parking
-    for line in clean_lines:
-        if "停车" in line:
-            # Clean up leading punctuation
-            cleaned = re.sub(r"^[·\.\sP]+", "", line)
-            result["parking"] = cleaned
-            break
-            
-    return result
-
-if __name__ == "__main__":
-    # User provided sample data
-    sample_input = [
-        "行 1: 长春市绿园区雁鸣湖公共充电站 (置信度: 0.9963)",
-        "行 2: （… (置信度: 0.6244)",
-        "行 3: 7.4km (置信度: 0.9975)",
-        "行 4: 0.7111/度 (置信度: 0.9450)",
-        "行 5: 快 (置信度: 0.9987)",
-        "行 6: 闲3/4 (置信度: 0.9941)",
-        "行 7: ￥ (置信度: 0.8734)",
-        "行 8: 组团 (置信度: 0.9995)",
-        "行 9: 2倍积分 (置信度: 0.9997)",
-        "行 10: P (置信度: 0.9929)",
-        "行 11: ·收费停车：以场地实际收费为准 (置信度: 0.9736)"
-    ]
-
-    print("--- Input Data ---")
-    for l in sample_input:
-        print(l)
-        
-    parsed = parse_ocr_lines(sample_input)
-    print("\n--- Parsed Result (Regex) ---")
-    print(json.dumps(parsed, ensure_ascii=False, indent=2))

Test/TestPaddleOCRKit.py

@@ -1,118 +0,0 @@
-# coding=utf-8
-import os
-import sys
-import cv2
-import json
-import time
-
-# Add project root to sys.path
-current_dir = os.path.dirname(os.path.abspath(__file__))
-project_root = os.path.dirname(current_dir)
-if project_root not in sys.path:
-    sys.path.append(project_root)
-
-from Util import Kit
-from Util.PaddleOCRKit import get_ocr_kit
-
-def test_integration():
-    image_path = os.path.join(current_dir, "2.jpg")
-    if not os.path.exists(image_path):
-        print(f"Image not found: {image_path}")
-        return
-
-    print(f"Testing integration on: {image_path}")
-    
-    # 1. Generate Metadata using Kit
-    print("Running Kit.crop_cards_from_image...")
-    Kit.crop_cards_from_image(image_path, output_dir=current_dir)
-    
-    json_path = image_path.replace(".jpg", ".json")
-    json_metadata = {}
-    
-    if os.path.exists(json_path):
-        with open(json_path, 'r', encoding='utf-8') as f:
-            json_metadata = json.load(f)
-            
-    if not json_metadata.get("cards"):
-        print("Kit failed to find cards (expected for single card image). Mocking metadata.")
-        img = cv2.imread(image_path)
-        h, w = img.shape[:2]
-        json_metadata = {
-            "cards": [
-                {
-                    "id": 1,
-                    "rect": [0, 0, w, h],
-                    "click_point": [w//2, h//2]
-                }
-            ]
-        }
-        
-    print(f"Loaded metadata with {len(json_metadata.get('cards', []))} cards.")
-
-    # 2. Run OCR Logic (Simulating Crawler.py)
-    print("Running OCR Logic...")
-    
-    ocr_kit = get_ocr_kit()
-    
-    original_img = cv2.imread(image_path)
-    stations = []
-    
-    t_start = time.time()
-    
-    if "cards" in json_metadata and original_img is not None:
-        h_img, w_img = original_img.shape[:2]
-        
-        for card in json_metadata["cards"]:
-            rect = card.get("rect")
-            if not rect: continue
-            x1, y1, x2, y2 = rect
-            
-            # 边界检查
-            x1 = max(0, min(x1, w_img))
-            x2 = max(0, min(x2, w_img))
-            y1 = max(0, min(y1, h_img))
-            y2 = max(0, min(y2, h_img))
-            
-            if x2 <= x1 or y2 <= y1: continue
-            
-            # 裁剪卡片
-            cropped_card = original_img[y1:y2, x1:x2]
-            
-            # 识别
-            parsed_data = ocr_kit.recognize(cropped_card)
-            print(f"Parsed Data: {json.dumps(parsed_data, indent=2, ensure_ascii=False)}")
-            
-            if parsed_data and parsed_data.get("station_name"):
-                 # 格式化数据
-                 piles_list = parsed_data.get("piles", [])
-                 piles_str_parts = []
-                 for p in piles_list:
-                     p_type = p.get("type", "")
-                     p_free = p.get("free", 0)
-                     p_total = p.get("total", 0)
-                     piles_str_parts.append(f"{p_type}:{p_free}/{p_total}")
-                 
-                 piles_str = " ".join(piles_str_parts)
-                 
-                 station_info = {
-                     "station_name": parsed_data.get("station_name"),
-                     "price": str(parsed_data.get("price")) if parsed_data.get("price") is not None else "",
-                     "piles": piles_str,
-                     "distance": parsed_data.get("distance", ""),
-                     "uia_center_x": card["click_point"][0],
-                     "uia_center_y": card["click_point"][1],
-                     "tags": parsed_data.get("tags", []),
-                     "parking_info": parsed_data.get("parking", "")
-                 }
-                 stations.append(station_info)
-
-    t_end = time.time()
-    
-    print("-" * 50)
-    print(f"Total Processing Time: {t_end - t_start:.4f}s")
-    print(f"Found {len(stations)} stations:")
-    for s in stations:
-        print(json.dumps(s, indent=2, ensure_ascii=False))
-
-if __name__ == "__main__":
-    test_integration()

Test/ocr_llm_debug.txt

@@ -1,78 +0,0 @@
-Starting TestOcrLlm...
-Testing OCR + LLM Pipeline on: D:\dsWork\YltProject\dsCrawler\Test\2.jpg
---------------------------------------------------
-Initializing PaddleOCR...
-PaddleOCR Init Time: 1.7990s
-Running OCR Inference...
-OCR Result (1.7636s):
-长春市绿园区雁鸣湖公共充电站
-..
-17.4km
-0.7111/度
-
-快
-闲3/4
-组团
-2倍积分
-P
-收费停车：以场地实际收费为准
---------------------------------------------------
-Running Regex Parsing...
-
-Parsed Data (Regex):
-{
-  "station_name": "长春市绿园区雁鸣湖公共充电站",
-  "distance": "17.4km",
-  "price": 0.7111,
-  "piles": [
-    {
-      "type": "快",
-      "free": 3,
-      "total": 4
-    }
-  ],
-  "parking": "收费停车：以场地实际收费为准",
-  "tags": [
-    "组团",
-    "2倍积分",
-    "P"
-  ]
-}
-Regex Parsing Time: 0.0015s
---------------------------------------------------
-Running LLM Parsing...
-Starting LLM request...
-LLM request finished.
-
-LLM Response (3.6250s):
-```json
-{
-    "station_name": "长春市绿园区雁鸣湖公共充电站",
-    "distance": "17.4km",
-    "price": "0.7111/度",
-    "tags": ["快", "闲3/4", "组团", "2倍积分", "P"],
-    "parking_info": "收费停车：以场地实际收费为准"
-}
-```
-
-Parsed JSON Data:
-{
-  "station_name": "长春市绿园区雁鸣湖公共充电站",
-  "distance": "17.4km",
-  "price": "0.7111/度",
-  "tags": [
-    "快",
-    "闲3/4",
-    "组团",
-    "2倍积分",
-    "P"
-  ],
-  "parking_info": "收费停车：以场地实际收费为准"
-}
---------------------------------------------------
-Summary:
-OCR Time: 1.7636s
-Regex Parsing Time: 0.0015s
-LLM Parsing Time: 3.6250s
-Total Pipeline (OCR+Regex): 1.7652s
-Total Pipeline (OCR+LLM):   5.3886s

Test/ocr_llm_result.txt

@@ -1,27 +0,0 @@
-OCR Result:
-长春市绿园区雁鸣湖公共充电站
-（…
-7.4km
-0.7111/度
-快
-闲3/4
-￥
-组团
-2倍积分
-P
-·收费停车：以场地实际收费为准
-
-LLM Response:
-{
-  "station_name": "长春市绿园区雁鸣湖公共充电站",
-  "distance": "7.4km",
-  "price": "0.7111/度",
-  "tags": ["快", "闲3/4", "组团", "2倍积分", "P"],
-  "parking_info": "收费停车：以场地实际收费为准"
-}
-
---------------------------------------------------
-Summary:
-OCR Time: 15.7442s
-LLM Time: 3.1676s
-Total Pipeline Time (excluding init): 18.9118s

Test/ocr_output.txt

@@ -1,11 +0,0 @@
-行 1: 长春市绿园区雁鸣湖公共充电站 (置信度: 0.9963)
-行 2: （… (置信度: 0.6244)
-行 3: 7.4km (置信度: 0.9975)
-行 4: 0.7111/度 (置信度: 0.9450)
-行 5: 快 (置信度: 0.9987)
-行 6: 闲3/4 (置信度: 0.9941)
-行 7: ￥ (置信度: 0.8734)
-行 8: 组团 (置信度: 0.9995)
-行 9: 2倍积分 (置信度: 0.9997)
-行 10: P (置信度: 0.9929)
-行 11: ·收费停车：以场地实际收费为准 (置信度: 0.9736)

Test/test02.py

@@ -1,9 +0,0 @@
-import uiautomator2 as u2
-
-d = u2.connect()
-
-d.reset_uiautomator()
-d.sleep(3)
-xml = d.dump_hierarchy()  # 导出当前UI结构（XML）
-with open("ui_hierarchy5_2.xml", "w", encoding="utf-8") as f:
-    f.write(xml)

Test/test03.py

@@ -1,14 +0,0 @@
-from Util.ObsUtil import ObsUploader
-
-from Config.Config import OBS_TMP_PREFIX
-
-uploader = ObsUploader()
-
-object_key = OBS_TMP_PREFIX + "/1d3eb56c-942e-42d3-8993-f1ea8ad7d97b.jpg"
-success, result = uploader.upload_file(
-    object_key=object_key,
-    file_path="Screenshot/1d3eb56c-942e-42d3-8993-f1ea8ad7d97b.jpg"
-)
-
-print(success)
-print(result)

Test/test04.py

@@ -1,12 +0,0 @@
-import uiautomator2 as u2
-
-d = u2.connect()
-
-# 启动应用（以微信为例）
-d.app_start("com.tencent.mm")
-
-# 截微信图保存
-d.screenshot("home.jpg")
-
-# 关闭应用
-d.app_stop("com.tencent.mm")

Test/ui_hierarchy5_2.xml

@@ -1,45 +0,0 @@
-<?xml version='1.0' encoding='UTF-8' standalone='yes' ?>
-
-<hierarchy rotation="0">
-
-  <node index="0" text="" resource-id="" class="android.widget.FrameLayout" package="com.android.systemui" content-desc="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[0,0][1080,72]" drawing-order="0" hint="" display-id="0">
-
-    <node index="0" text="" resource-id="com.android.systemui:id/status_bar_launch_animation_container" class="android.widget.FrameLayout" package="com.android.systemui" content-desc="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[0,0][1080,72]" drawing-order="1" hint="" display-id="0" />
-
-    <node index="1" text="" resource-id="com.android.systemui:id/status_bar_container" class="android.widget.FrameLayout" package="com.android.systemui" content-desc="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[0,0][1080,72]" drawing-order="2" hint="" display-id="0">
-
-      <node index="0" text="" resource-id="com.android.systemui:id/status_bar" class="android.widget.FrameLayout" package="com.android.systemui" content-desc="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[0,0][1080,72]" drawing-order="1" hint="" display-id="0">
-
-        <node index="0" text="" resource-id="com.android.systemui:id/status_bar_contents" class="android.widget.LinearLayout" package="com.android.systemui" content-desc="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[0,0][1020,72]" drawing-order="2" hint="" display-id="0">
-
-          <node index="0" text="" resource-id="" class="android.widget.FrameLayout" package="com.android.systemui" content-desc="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[24,0][510,72]" drawing-order="1" hint="" display-id="0">
-
-            <node index="0" text="" resource-id="com.android.systemui:id/status_bar_left_side" class="android.widget.LinearLayout" package="com.android.systemui" content-desc="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[24,0][510,72]" drawing-order="2" hint="" display-id="0">
-
-              <node index="0" text="9:15" resource-id="com.android.systemui:id/clock" class="android.widget.TextView" package="com.android.systemui" content-desc="上午9:15" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[24,0][113,72]" drawing-order="2" hint="" display-id="0" />
-
-              <node index="1" text="" resource-id="com.android.systemui:id/notification_icon_area" class="android.widget.FrameLayout" package="com.android.systemui" content-desc="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[113,0][510,72]" drawing-order="4" hint="" display-id="0">
-
-                <node index="0" text="" resource-id="com.android.systemui:id/notification_icon_area_inner" class="android.widget.LinearLayout" package="com.android.systemui" content-desc="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[113,0][510,72]" drawing-order="1" hint="" display-id="0">
-
-                  <node index="0" text="" resource-id="com.android.systemui:id/notificationIcons" class="android.widget.FrameLayout" package="com.android.systemui" content-desc="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[113,0][510,72]" drawing-order="1" hint="" display-id="0" />
-
-                </node>
-
-              </node>
-
-            </node>
-
-          </node>
-
-          <node index="1" text="" resource-id="com.android.systemui:id/system_icon_area" class="android.widget.LinearLayout" package="com.android.systemui" content-desc="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[510,0][996,72]" drawing-order="4" hint="" display-id="0">
-
-            <node index="0" text="" resource-id="com.android.systemui:id/system_icons" class="android.widget.LinearLayout" package="com.android.systemui" content-desc="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[510,0][996,72]" drawing-order="1" hint="" display-id="0">
-
-              <node index="0" text="" resource-id="com.android.systemui:id/statusIcons" class="android.widget.LinearLayout" package="com.android.systemui" content-desc="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[510,0][973,72]" drawing-order="1" hint="" display-id="0">
-
-                <node index="0" text="" resource-id="com.android.systemui:id/wifi_combo" class="android.widget.FrameLayout" package="com.android.systemui" content-desc="WLAN 信号满格。" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[857,3][910,69]" drawing-order="14" hint="" display-id="0">
-
-                  <node index="0" text="" resource-id="com.android.systemui:id/wifi_group" class="android.widget.LinearLayout" package="com.android.systemui" content-desc="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[865,3][910,69]" drawing-order="1" hint="" display-id="0">
-
-                    <node index="0" text="" resource-id="com.android.systemui:id/wifi_combo" class="android.widget.FrameLayout" package="com.android.systemui" content-desc="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" scrollable="false" long-clickable="false" password="false" selected="false" visible-to-user="true" bounds="[865,13][910,58]" drawing-order="2" hint="" display-id="0">

Tools/T1_CreateTable.py

@@ -8,7 +8,6 @@ logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(name)s - %(level
 logger = logging.getLogger(__name__)
 
 # 将项目根目录添加到系统路径，以便能够导入 DbKit 和 Config 等模块
-# DbKit/doris_ddl.py -> DbKit -> Root (2 levels up)
 project_root = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 if project_root not in sys.path:
     sys.path.append(project_root)
@@ -31,7 +30,7 @@ async def init_tables():
 
     logger.info("正在读取 SQL 文件...")
     # SQL 文件路径：相对于当前脚本所在目录 (DbKit/)
-    sql_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), '', 'doris_ddl.sql')
+    sql_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), '', 'Sql/doris_ddl.sql')
     
     logger.info(f"使用 SQL 文件: {sql_path}")
 

Tools/T2_ScreenShot.py

@@ -5,6 +5,8 @@ import os
 from datetime import datetime
 import uiautomator2 as u2
 
+from Config.Config import TEMP_IMAGE_DIR
+
 # 配置日志输出
 logging.basicConfig(
     level=logging.INFO,
@@ -22,14 +24,8 @@ async def take_screenshot():
     
     # 执行截图
     timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
-    filename = f"current_{timestamp}.jpg"
-    
-    # 获取目标目录: 项目根目录下的 Temp/DevTools
-    project_root = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
-    target_dir = os.path.join(project_root, "Temp", "DevTools")
-    os.makedirs(target_dir, exist_ok=True)
-    filepath = os.path.join(target_dir, filename)
-    
+    filename = f"Screenshot_{timestamp}.jpg"
+    filepath = os.path.join(TEMP_IMAGE_DIR, filename)
     logger.info(f"正在拍照（截图）并保存至: {filepath}")
     d.screenshot(filepath)
     

Util/DevTools/crop_test.py

@@ -1,73 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-基于行扫描和统计特征的卡片截取工具
-"""
-import sys
-import os
-
-# 添加项目根目录到系统路径
-project_root = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
-sys.path.append(project_root)
-from Util import Kit
-
-def clean_directory(dir_path):
-    """
-    清理现场：删除目录下除纯数字.jpg以外的所有文件
-    """
-    print(f"Cleaning directory: {dir_path}")
-    if not os.path.exists(dir_path):
-        return
-
-    for filename in os.listdir(dir_path):
-        path = os.path.join(dir_path, filename)
-        if not os.path.isfile(path):
-            continue
-            
-        name, ext = os.path.splitext(filename)
-        # 保留条件：后缀是 .jpg 且文件名是纯数字
-        # [修改] 同时保留 _flag.jpg, _vl.jpg 供查看
-        # [修改] 保留 .json 文件
-        if ext.lower() == ".json":
-            continue
-
-        if ext.lower() == ".jpg" and (name.isdigit() or name.endswith("_flag") or name.endswith("_vl")):
-            continue
-            
-        # 清理旧的 _for_vl.jpg
-        if name.endswith("_for_vl"):
-            pass # Let it fall through to delete
-        else:
-            # 其他非生成的文件，可能需要保留吗？
-            # 这里的逻辑是清理 output 目录，假设该目录下只有生成的图片
-            # 如果是原始图片（比如 1.jpg），不能删！
-            pass
-            
-        # 原始图片保护 (文件名比较短，通常是 1.jpg, 2.jpg 等)
-        if len(name) <= 2 and name.isdigit():
-             continue
-
-        try:
-            os.remove(path)
-            print(f"  Deleted: {filename}")
-        except Exception as e:
-            print(f"  Error deleting {filename}: {e}")
-
-def crop_cards(img_path):
-    Kit.crop_cards_from_image(img_path)
-
-if __name__ == "__main__":
-    test_files = [
-        r"d:\dsWork\dsProject\dsCrawler\Tools\Images\1.jpg",
-        r"d:\dsWork\dsProject\dsCrawler\Tools\Images\2.jpg",
-        r"d:\dsWork\dsProject\dsCrawler\Tools\Images\3.jpg",
-        r"d:\dsWork\dsProject\dsCrawler\Tools\Images\4.jpg"
-    ]
-    
-    # 在测试前清理现场
-    if test_files:
-        # 假设所有图片都在同一个文件夹
-        target_dir = os.path.dirname(test_files[0])
-        clean_directory(target_dir)
-        
-    for f in test_files:
-        crop_cards(f)