Merge commit '6e7ad7e32522a88f1d71a15307ca856fe3c3f495'

工具v4/database_tools_2.py

@@ -393,6 +393,18 @@ def NextSpareID(num,database): #返回adict中下一个空闲的题号
     mydb.close()
     return num
 
+def NextSpareIDseries(num,cursor,length): #返回adict中以num起始, 连续长度至少为length的下一个空闲的题号
+    cursor.execute("SELECT ID FROM problems;")
+    idlist = [ret[0] for ret in cursor.fetchall()]
+    idset = set(idlist)
+    num = int(num)
+    numset = set([str(num+i).zfill(6) for i in range(length)])
+    while len(idset & numset) >= 1:
+        numset.remove(str(num).zfill(6))
+        numset.add(str(num+length).zfill(6))
+        num += 1
+    return num
+
 def NextSpareIDBlock(num,database): #返回adict中下一个空闲的题号块
     mydb = connect(hostname = db_host, port = db_port, username=db_user, pwd=db_pwd, db = database)
     mycursor = mydb.cursor()
@@ -433,6 +445,52 @@ 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 GenerateNewProblemDataFromString2024(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, 返回三元组(题目内容, 中缀, 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"\\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()
+        meta = {"same":[],"related":[],"unrelated":[]}
+        sim_problems = re.findall(r"\\definecolor[\s\S]*?\\end\{tcolorbox\}",content_raw)
+        for sim_p in sim_problems:
+            metainfo = re.findall(r"相似度:[\s]*\d\.\d{3}[\\\s]*?(\d{6})\(([\s\S]*?)\)",sim_p)
+            if len(metainfo) == 1:
+                id,metatype = metainfo[0]
+                metatype = metatype.upper()
+                if metatype.startswith("REP"):
+                    meta["rep"] = id.zfill(6)
+                elif metatype.startswith("S"):
+                    meta["same"].append(id.zfill(6))
+                elif metatype.startswith("R"):
+                    meta["related"].append(id.zfill(6))
+                elif metatype.startswith("U"):
+                    meta["unrelated"].append(id.zfill(6))
+            content_raw = content_raw.replace(sim_p,"")
+        content = content_raw.strip()
+        if "rep" in meta:
+            repid = meta["rep"]
+            meta = {"rep": id}
+        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))
+        # print(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

@@ -51,33 +51,24 @@ class MyWindow_bdsl(QWidget,Ui_Form):
         problems_raw = re.findall(r"\\begin\{enumerate\}([\s\S]*?)\\end\{enumerate\}",data_raw)
         data = "\n".join([item.strip() for item in problems_raw])
         problems = [(d.strip()) for d in data.split("\\item") if not d.strip() == ""]
-        output = "使用说明:\\\\\n rep??????表示使用已有题号??????,\\\\\n s??????表示和??????相同,\\\\\n r??????表示和??????相关\n\n\\begin{enumerate}\n\n"
+        output = "使用说明:\\\\\n ??????(rep)表示使用已有题号??????,\\\\\n ??????(s)表示和??????相同,\\\\\n ??????(r)表示和??????相关\n\n\\begin{enumerate}\n\n"
 
 
         for p in problems:
             p_treated = pre_treating(p)
             t = stringmaxsim(p_treated,treated_dict,5)
-            psrstring = ""
+            output += f"\\item {p}\n\n"
-            for id,simrate in t:
-                if simrate == 1:
-                    psrstring = "rep"+id+","
-                elif simrate > 0.5:
-                    psrstring += id + ","
-            if len(psrstring) > 0 and psrstring[-1] == ",":
-                psrstring = psrstring[:-1]
-            if len(psrstring) > 0:
-                output += f"\\item [{psrstring}] {p}\n\n"
-            else:
-                output += f"\\item {p}\n\n"
             count = 0
             for id,simrate in t:
                 if simrate > 0.5:
                     colors = get_color(simrate*2-1)
                     output += "\n\\definecolor{mycolor}{rgb}"+colors
-                    output += "\n\\begin{tcolorbox}"+f"[colback = mycolor, opacityback = 0.25, colframe = orange!10!white, breakable]\n"
+                    output += "\n\\begin{tcolorbox}"+f"[colback = mycolor, opacityback = 0.25, colframe = orange!10!white, breakable]\n\n"
-                    output += f"{simrate:.3f} \\ {id}\n\n"
+                    output += "\\begin{flushright}\\begin{Large}\n"
+                    output += f"相似度: {simrate:.3f} \\ {id}({'rep' if simrate == 1 else ''})\n"
+                    output += "\\end{Large}\\end{flushright}"
                     content = pro_dict[id]
-                    output += f"{content}\n"
+                    output += f"\n{content}\n"
                     output += "\\end{tcolorbox}\n"
                     count += 1
             output += "\n\n"
@@ -98,11 +89,12 @@ 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 = GenerateNewProblemDataFromString2024(data)
+        problem_count = len(problems)
         # pro_dict = load_dict("../题库0.3/Problems.json")
-        nextspareid = NextSpareID(starting_id,self.database_name)
         self.db = connect(hostname = db_host, port= db_port, username= db_user, pwd = db_pwd, db= self.database_name)
         self.cursor = self.db.cursor()
+        nextspareid = NextSpareIDseries(starting_id,self.db.cursor(),problem_count) #待修改（检测空白块数大小）
         rolled_back = False
         idlist,newproblemlist = AddProblemstoDict2024nocommit(nextspareid,raworigin,problems,editor,Indexed,self.cursor)
         for p in newproblemlist:
@@ -130,6 +122,8 @@ class MyWindow_bdsl(QWidget,Ui_Form):
         self.db.commit()
         self.db.close()
         print("已提交至数据库")
+        self.pushButton_commit.setDisabled(True)
+        self.problemstoimport = []