批量收录新题修改中

工具v4/database_tools_2.py

@@ -433,6 +433,66 @@ def GenerateProblemListFromString(data): #从来自.tex文件的字符串生成
     return problem_list #返回一个列表, 每一项是一个由 题目内容 和 题目来源前缀 组成的元组
 
 
+def GenerateProblemListFromString202406(data): #从data字符串生成题目列表, 每个item是一道题目, 题目后有色块的准关联题目的flushright块中的信息表示交互信息(如(003214rep)表示不添加, 用003214代替; (002521s)表示和002521相同, (000354r)表示和000354相关, 返回二元组(题目内容, 交互信息列表)的字典
+    try:
+        data = re.findall(r"\\begin\{document\}([\s\S]*?)\\end\{document\}",data)[0]
+    except:
+        pass
+    data = data + "\n\\end{enumerate}"
+    data = re.sub(r"\n{2,}","\n",data)
+    # data = re.sub(r"\\definecolor[^\n]*\n","\n",data)
+    # data = re.sub(r"\\begin\{tcolorbox\}[\s\S]*?\\end\{tcolorbox\}","\n",data)
+    data = re.sub(r"\\item",r"\\enditem\\item",data)
+    data = re.sub(r"\\end\{enumerate\}",r"\\enditem",data) #切除无关信息, 保留关键信息
+    problempositions = []
+    for item in re.finditer(r"\\item([\s\S]*?)\\enditem",data):
+        problempositions.append(item.regs[1]) #确定题目内容所在位置
+    problem_list = []
+    for pos in problempositions:
+        content_raw = data[pos[0]:pos[1]].strip()
+        sim_problems = re.findall(r"\\definecolor[\s\S]*?\\end\{tcolorbox\}",content_raw)
+        for sim_p in sim_problems:
+            #处理meta待修改
+            content_raw = content_raw.replace(sim_p,"")
+        content = content_raw.strip()
+        #以下待修改----20240607
+
+        content_raw = re.sub(r"\n\%[\s\S]*$","",content_raw) #题目内容
+        content_raw = re.sub(r"\\\\$","",content_raw) # 删去题目最后一行处可能存在的\\
+        content_raw = content_raw.strip() # 删去前后多余的空格
+        meta = {}
+        if not content_raw[0] == "[": # 根据方括号内的内容生成交互信息, 这是无方括号的内容, 无meta
+            content = content_raw   
+        else:
+            same_id_list = []
+            related_id_list = []
+            unrelated_id_list = []
+            content = re.sub(r"^\[[A-Za-z\d,\s]*?\]","",content_raw).strip()
+            metaraw = re.findall(r"\[.*?\]",content_raw)[0]
+            metaraw = re.sub(r"[\[\]]","",metaraw)
+            metalist = metaraw.split(",")
+            for metaitem in metalist:
+                metaitem = metaitem.upper()
+                if metaitem.startswith("REP"):
+                    meta = {"rep":metaitem[3:].zfill(6)}
+                elif metaitem.startswith("S"):
+                    same_id_list.append(metaitem[1:].zfill(6))
+                elif metaitem.startswith("R"):
+                    related_id_list.append(metaitem[1:].zfill(6))
+                elif metaitem.startswith("U"):
+                    unrelated_id_list.append(metaitem[1:].zfill(6))
+            if not "rep" in meta:
+                meta = {"same":same_id_list,"related":related_id_list,"unrelated":unrelated_id_list}
+        subdata = data[:pos[0]] #开始寻找出处中缀
+        suflist = re.findall(r"\n(\%\s{0,}[\S]+)\n",subdata)
+        if len(suflist) == 0:
+            suffix = ""
+        else:
+            suffix = suflist[-1].replace("%","").strip()
+        problem_list.append((content,suffix,meta))
+    return problem_list #返回一个列表, 每一项是一个由 题目内容 和 题目来源前缀 和 交互信息字典 组成的元组
+
+
 def GenerateProblemListFromString2024(data): #从来自.tex文件的字符串生成题目列表, 每个item是一道题目, 新一行的%用作前缀, item后面的方括号放与题库的交互信息(如[rep3214]表示不添加, 用003214代替; [s2521;r354;u10021,20024]表示和2521相同, 和354相关, 和10021,20024无关), 返回三元组(题目内容, 题目来源前缀, 交互信息列表)的字典
     try:
         data = re.findall(r"\\begin\{document\}([\s\S]*?)\\end\{document\}",data)[0]

工具v4/批量收录新题.py

@@ -65,7 +65,7 @@ class MyWindow_bdsl(QWidget,Ui_Form):
                     output += "\n\\definecolor{mycolor}{rgb}"+colors
                     output += "\n\\begin{tcolorbox}"+f"[colback = mycolor, opacityback = 0.25, colframe = orange!10!white, breakable]\n\n"
                     output += "\\begin{flushright}\\begin{Large}\n"
-                    output += f"{simrate:.3f} \\ {id}()\n"
+                    output += f"相似度: {simrate:.3f} \\ {id}()\n"
                     output += "\\end{Large}\\end{flushright}"
                     content = pro_dict[id]
                     output += f"\n{content}\n"
@@ -89,9 +89,9 @@ class MyWindow_bdsl(QWidget,Ui_Form):
         else:
             Indexed = False
         idlistpath = "文本文件/新题收录列表.txt"
-        problems = GenerateProblemListFromString2024(data)
+        problems = GenerateProblemListFromString2024(data) #待修改（检测题目方式有变）
         # pro_dict = load_dict("../题库0.3/Problems.json")
-        nextspareid = NextSpareID(starting_id,self.database_name)
+        nextspareid = NextSpareID(starting_id,self.database_name) #待修改（检测空白块数大小）
         self.db = connect(hostname = db_host, port= db_port, username= db_user, pwd = db_pwd, db= self.database_name)
         self.cursor = self.db.cursor()
         rolled_back = False