refactor(baidu): 重构搜索结果的获取和解析逻辑，及返回的结果样式 (close #35)

PicImageSearch/baidu.py

@@ -1,3 +1,4 @@
+import re
 from json import loads as json_loads
 from pathlib import Path
 from typing import Any, Dict, Optional, Union
@@ -25,8 +26,13 @@ async def search(
             )
         else:
             raise ValueError("url or file is required")
-        resp_text, resp_url, _ = await self.post(
+        resp_text, _, _ = await self.post(
             "https://graph.baidu.com/upload", params=params, data=data
         )
-        resp_text, resp_url, _ = await self.get((json_loads(resp_text))["data"]["url"])
-        return BaiDuResponse(resp_text, resp_url)
+        next_url = (json_loads(resp_text))["data"]["url"]
+        resp_text, resp_url, _ = await self.get(next_url)
+        next_url = (re.search(r'"firstUrl":"([^"]+)"', resp_text)[1]).replace(r"\/", "/")  # type: ignore
+        resp_text, _, _ = await self.get(next_url)
+        next_url = (json_loads(resp_text))["data"]["ajaxTextUrl"]
+        resp_text, _, _ = await self.get(next_url)
+        return BaiDuResponse(json_loads(resp_text), resp_url)

PicImageSearch/model/baidu.py

@@ -1,45 +1,19 @@
-import json
-import re
-from typing import Any, Dict, List, Optional
+from typing import Any, Dict, List
 
 
 class BaiDuItem:
     def __init__(self, data: Dict[str, Any]):
         self.origin: Dict[str, Any] = data  # 原始数据
-        self.page_title: str = data["fromPageTitle"]  # 页面标题
-        self.title: str = data["title"][0]  # 标题
-        self.abstract: str = data["abstract"]  # 说明文字
-        self.image_src: str = data["image_src"]  # 图片地址
-        self.url: str = data["url"]  # 图片所在网页地址
-        self.img_list: List[str] = data.get("imgList", [])  # 其他图片地址列表
+        self.similarity: float = float(f"{float(data['simi']) * 100:.2f}")
+        self.title: str = data["fromPageTitle"]  # 页面标题
+        self.thumbnail: str = data["thumbUrl"]  # 图片地址
+        self.url: str = data["fromUrl"]  # 图片所在网页地址
 
 
 class BaiDuResponse:
-    def __init__(self, resp_text: str, resp_url: str):
+    def __init__(self, resp_json: Dict[str, Any], resp_url: str):
         self.url: str = resp_url  # 搜索结果地址
         self.similar: List[Dict[str, Any]] = []  # 相似结果返回值
-        self.raw: List[BaiDuItem] = []  # 来源结果返回值
-        # 原始数据
-        self.origin: List[Dict[str, Any]] = json.loads(
-            re.search(r"cardData = (.+);window\.commonData", resp_text)[1]  # type: ignore
-        )
-        self.same: Optional[Dict[str, Any]] = {}
-        for i in self.origin:
-            setattr(self, i["cardName"], i)
-        if self.same:
-            self.raw = [BaiDuItem(x) for x in self.same["tplData"]["list"]]
-            info = self.same["extData"]["showInfo"]
-            del info["other_info"]
-            for y in info:
-                for z in info[y]:
-                    try:
-                        self.similar[info[y].index(z)][y] = z
-                    except IndexError:
-                        self.similar.append({y: z})
-        # 获取所有卡片名
-        self.item: List[str] = [
-            attr
-            for attr in dir(self)
-            if not callable(getattr(self, attr))
-            and not attr.startswith(("__", "origin", "raw", "same", "url"))
-        ]
+        self.origin: Dict[str, Any] = resp_json  # 原始数据
+        # 来源结果返回值
+        self.raw: List[BaiDuItem] = [BaiDuItem(i) for i in resp_json["data"]["list"]]

demo/demo_baidu.py

@@ -31,17 +31,12 @@ def test_sync() -> None:
 
 def show_result(resp: BaiDuResponse) -> None:
     # logger.info(resp.origin)  # 原始数据
-    logger.info(resp.item)
-    if resp.same:  # 存在来源结果
-        # logger.info(resp.raw[0].origin)
-        logger.info(resp.raw[0].page_title)
-        logger.info(resp.raw[0].title)
-        logger.info(resp.raw[0].abstract)
-        logger.info(resp.raw[0].url)
-        logger.info(resp.raw[0].image_src)
-        logger.info(resp.raw[0].img_list)
-    else:
-        logger.info(resp.similar)
+    logger.info(resp.url)
+    # logger.info(resp.raw[0].origin)
+    logger.info(resp.raw[0].similarity)
+    logger.info(resp.raw[0].title)
+    logger.info(resp.raw[0].url)
+    logger.info(resp.raw[0].thumbnail)
     logger.info("-" * 50)