Update app.py

app.py

@@ -8,8 +8,8 @@ import textwrap
 import traceback
 
 # 爬虫-------------
-from save_cookie import save_cookie, get_cookie, cookie_f
-from scrap_util import getDriver, titleLocInfo, find_key_paragrap, extract_from_driver, table_record_doc
+# from save_cookie import save_cookie, get_cookie, cookie_f
+# from scrap_util import getDriver, titleLocInfo, find_key_paragrap, extract_from_driver, table_record_doc
 import helium as hm
 from postDouyin import senDouyin
 # 模型-------------
@@ -54,12 +54,12 @@ HEIGHT = 1400
 WIDTH,HEIGHT = a_.size
 
 # should load from files and build new from file
-cookie_fns = ["抖音北京人事考试","抖音广东人事考试","抖音四川人事考试","抖音浙江人事考试","抖音江苏人事考试","抖音山东人事考试","抖音河南人事考试"]
-cookie_fns = os.listdir("./cookie_list/")
-cookie_fn = cookie_fns[0]
+# cookie_fns = ["抖音北京人事考试","抖音广东人事考试","抖音四川人事考试","抖音浙江人事考试","抖音江苏人事考试","抖音山东人事考试","抖音河南人事考试"]
+# cookie_fns = os.listdir("./cookie_list/")
+# cookie_fn = cookie_fns[0]
 description = "URL--> 爬取-->解析--> 音频--> 图片--> 视频"
 
-driver = getDriver()
+# driver = getDriver()
 # sub_url ="https://www.js.msa.gov.cn/art/2023/2/24/art_11436_1391666.html"
 # hm.set_driver(driver)  # 给它一个selnuim driver
 # hm.go_to(sub_url)
@@ -84,30 +84,6 @@ def custom_predict(context, question):
     score = answer_result["score"]
     return answer, score
 
-def get_content_from_driver(task_link, content=None):
-    driver = getDriver()
-    hm.set_driver(driver)  # 给它一个selnuim driver
-    if content:
-        pass
-    hm.go_to(task_link)
-    print("hm.go_to(task_link)")
-    time.sleep(1)
-    items_ = driver.find_elements_by_xpath("//p")
-    items_ = [i.text for i in items_ if i.text != ""]
-    context_to_label =  "\n".join(items_)
-    doc = extract_from_driver(driver)
-    doc["url"] = task_link
-    doc["content"] = context_to_label
-    n_bm_sj,_ = custom_predict(context = context_to_label, question="报名时间")
-    n_fee_sj,_ = custom_predict(context = context_to_label, question="缴费时间")
-    n_ks_sj,_ = custom_predict(context = context_to_label, question="考试时间")
-    n_zkz_sj,_ = custom_predict(context = context_to_label, question="准考证时间")
-    need = [doc["title"],doc["zwlx"], doc["zwk_sheng"], doc["zwk_diqu"],
-            doc["tidy_bm_sj"], doc["tidy_fee_sj"], doc["tidy_ks_sj"], doc["tidy_zkz_sj"],context_to_label,
-            n_bm_sj, n_fee_sj, n_ks_sj, n_zkz_sj
-           ]
-    return need
-
 def image_preview(orimage=None, text="Hello Ai", x=10, y=20, w=500, h=100, bac_color = "#FFbbFF",
                   txt_color = "#000000",front="simsun.ttc", size = 50):
     if orimage is None:
@@ -473,8 +449,8 @@ with gr.Blocks() as demo:
     #              outputs=[img_files])
                  # outputs=[movie_file, img_files])
     # exit_.click(fn=exit_func)
-    login_.click(fn=loginDouyin, outputs=[login_qr])
-    login_save.click(fn=run_save_cookie, inputs = [account_name_new])
+    # login_.click(fn=loginDouyin, outputs=[login_qr])
+    # login_save.click(fn=run_save_cookie, inputs = [account_name_new])
     # post_.click(fn=mySendDouyin, inputs = [account_fn, movie_file])
     # 绑定clear点击函数
     # clear.click(fn=clear_input, inputs=[], outputs=[context, question, answer, score])