import json,re,os,Levenshtein,fitz,time,sys,subprocess import pandas as pd import numpy as np import pyperclip import mysql.connector BuildFullScheme = { "输出路径": "临时文件", "教师版": True, "字段显示设置": { "题后空间": True, "课时目标": True, "题目标签": True, "答案": True, "解答与提示": True, "使用记录": [3,-1], "使用记录说明": "[a,b]表示显示最好的a个和最差b个, 有-2表示不显示, 无-2但有-1表示全部显示", "来源": True, "备注": True, "届别": [] } } def get_git_username(): command = "git config --global user.name" stream = os.popen(command) username = stream.read().strip() return username def connect(hostname,port,username,pwd,db): mydb = mysql.connector.connect(host = hostname, port = port, user = username, password = pwd, database = db) return mydb def GetDate(): #获得当前日期 currentdate = str(time.localtime().tm_year)+str(time.localtime().tm_mon).zfill(2)+str(time.localtime().tm_mday).zfill(2) return currentdate #返回当前日期yyyymmdd def GetTime(): #获得当前时间 t = time.localtime() currenttime = f"{t.tm_hour:0>2}{t.tm_min:0>2}{t.tm_sec:0>2}" return currenttime #返回当前时间hhmmss def datestrtotimestamp(date_str): # 将yyyymmdd的时间转为时间戳timestamp(以分钟为单位的整数) date_obj = time.strptime(date_str, "%Y%m%d") timestamp = int(time.mktime(date_obj)) return timestamp def ReadTextFile(filepath): #读取文本格式的文件 with open(filepath,"r",encoding="u8") as f: data = f.read() return data #返回文本格式文件的内容 def SaveTextFile(data,filepath): #写入文本格式的文件 pathname = os.path.dirname(filepath) if not os.path.exists(pathname): makedir(pathname) with open(filepath,"w",encoding="u8") as f: f.write(data) return filepath #返回文件名 def AppendTextFile(string,filepath): #在文本文件后添加内容 try: with open(filepath,"a",encoding = "u8") as f: f.write(string.strip()+"\n\n") except: makedir(os.path.split(filepath)[0]) with open(filepath,"w",encoding = "u8") as f: f.write(string.strip()+"\n\n") return filepath #返回文件名 def SortDict(adict): #按字典项顺序排序字典 return dict(sorted(adict.items())) #返回排序后的字典 #读取存储json数据库相关(不限于题号数据库) def load_dict(filename): #根据filename读取json数据库并转化为python字典 with open(filename,"r",encoding = "u8") as f: adict = json.loads(f.read()) return adict #返回python字典 def save_dict(adict,filepath): #将adict字典转化为json文件并保存至filename文件中 try: jsondata = json.dumps(adict,indent=4,ensure_ascii=False) SaveTextFile(jsondata,filepath) return filepath # 成功则返回文件路径 except: print("保存 %s 文件失败"%filepath) return 1 # 失败则返回1 def pre_treating(string): #删除字符串中对比较无用的字符, 以供比较 string = re.sub(r"\\begin\{center\}[\s\S]*?\\end\{center\}","",string) string = re.sub(r"(bracket\{\d+\})|(blank\{\d+\})|(fourch)|(twoch)|(onech)","",string) string = re.sub(r"[\s\\\{\}\$\(\)\[\]]","",string) string = re.sub(r"[\n\t]","",string) string = re.sub(r"(displaystyle)|(overrightarrow)|(overline)","",string) string = re.sub(r"[,\.:;?]","",string) return string #返回处理后的字符串 def findsru(id,alist): #在same_list中寻找与id相同的题号 set1 = set([pair[1] for pair in alist if pair[0] == id]) set2 = set([pair[0] for pair in alist if pair[1] == id]) return sorted(list(set1 | set2)) def findsameinDB(id): #在数据库中寻找与id相同的题号 mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() sql = "SELECT SAME_ID FROM same WHERE ID = (%s);" val = (id,) mycursor.execute(sql,val) set1 = set([ret[0] for ret in mycursor.fetchall()]) sql = "SELECT ID FROM same WHERE SAME_ID = (%s);" mycursor.execute(sql,val) set2 = set([ret[0] for ret in mycursor.fetchall()]) return sorted(list(set1 | set2)) def findrelatedinDB(id): #在数据库中寻找与id关联的题号 mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() sql = "SELECT RELATED_ID FROM related WHERE ID = (%s);" val = (id,) mycursor.execute(sql,val) set1 = set([ret[0] for ret in mycursor.fetchall()]) sql = "SELECT ID FROM related WHERE RELATED_ID = (%s);" mycursor.execute(sql,val) set2 = set([ret[0] for ret in mycursor.fetchall()]) return sorted(list(set1 | set2)) def get_unit_tags(id): mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() sql = "SELECT tagname FROM tagcorresp WHERE ID = (%s);" val = (id,) mycursor.execute(sql,val) unittags = [ret[0] for ret in mycursor.fetchall() if "单元" in ret[0]] mydb.close() return unittags def get_problem_content(id): mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() sql = "SELECT content FROM problems WHERE ID = (%s);" val = (id,) mycursor.execute(sql,val) content = mycursor.fetchall()[0][0] mydb.close() return content def generate_same_list(): mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() mycursor.execute("SELECT ID,SAME_ID FROM same;") same_list = [(ret[0],ret[1]) for ret in mycursor.fetchall()] mydb.close() return same_list.copy() def generate_related_list(): mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() mycursor.execute("SELECT ID,RELATED_ID FROM related;") related_list = [(ret[0],ret[1]) for ret in mycursor.fetchall()] mydb.close() return related_list.copy() def treat_dict(): #对整个题库字典中的内容部分进行预处理，删除无用字符 treated_dict = {} mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() mycursor.execute("SELECT ID,content FROM problems;") p_dict = {id:content for id,content in mycursor.fetchall()} # mycursor.execute("SELECT ID,SAME_ID FROM same;") # same_list = [(ret[0],ret[1]) for ret in mycursor.fetchall()] for id in p_dict: treated_dict[id] = {} treated_dict[id] = pre_treating(p_dict[id]) mydb.close() return treated_dict #返回处理后的字典, 含内容字段及相同题目字段 def detectmaxsim(currentid,excludelist,adict): #检测与已知题目关联程度最大的题目（除外列表之外的部分） maxsim = -1 argmaxsim = "000000" for id in adict: if not id in excludelist and not "OBSOLETE" in adict[id]["content"]: simrate = Levenshtein.jaro(adict[id]["content"],adict[currentid]["content"]) if simrate > maxsim: maxsim = simrate argmaxsim = id return (maxsim,argmaxsim) #返回最大关联系数与关联程度最大的题号 def stringmaxsim(string,adict,listlength): # maxsim = -1 # argmaxsim = "000000" maxsimlist = [] string = pre_treating(string) for id in adict: if not "OBS" in adict[id]: simrate = Levenshtein.jaro(adict[id],string) # if simrate > maxsim: # maxsim = simrate # argmaxsim = id if len(maxsimlist) < listlength: maxsimlist.append((id,simrate)) elif simrate > maxsimlist[-1][1]: maxsimlist = maxsimlist[:-1] maxsimlist.append((id,simrate)) maxsimlist = sorted(maxsimlist,key = lambda x: x[1], reverse = True) return maxsimlist #返回最大关联的listlength个题号及关联系数 def generate_sim_group(startingids,prodict,max_size,threshold): #生成与已知题块startingids(冒号和逗号连接的字符串)关联程度超过threshold的题目组, 超过max_size即停止 id_list = generate_number_set(startingids) output_list = [] treated_dict = treat_dict(prodict) continue_flag = True while continue_flag: appending_id_list = [] for oldid in id_list: for newid in [id for id in prodict if not id in id_list and not id in output_list and not "OBSOLETE" in treated_dict[id]["content"]]: simrate = Levenshtein.jaro(treated_dict[oldid]["content"],treated_dict[newid]["content"]) if simrate >= threshold: appending_id_list.append(newid) output_list = output_list + id_list.copy() id_list = appending_id_list.copy() if len(appending_id_list) == 0 or len(output_list)>max_size: continue_flag = False return generate_exp(sorted(output_list)) #返回冒号和逗号连接的题目组字符串 def generate_problem_series(startingid,length,adict): #在adict字典里返回从startingid开始的一系列题号, 每一题都是与上一题的关联程度最大的 excludelist = [startingid] currentid = startingid for i in range(length): maxsim,currentid = detectmaxsim(currentid,excludelist,adict) excludelist.append(currentid) return ",".join(excludelist) #返回按顺序的题号列表 def generate_number_set(string,*thedict): #根据可能含有":"和","的题号字符串生成一个用逗号分隔的六位题号列表, 例如"1:3,5"会生成["000001","000002","000003","000005"] #可变参数*dict如果存在, 将只生成dict的keys中包含的题号列表 string = re.sub(r"[\n\s]","",string).strip() while not string[-1] in "0123456789": string = string[:-1] string = RefinePunctuations(string) string_list = string.split(",") numbers_list = [] for s in string_list: if not ":" in s: numbers_list.append(s.zfill(6)) else: start,end = s.split(":") for ind in range(int(start),int(end)+1): numbers_list.append(str(ind).zfill(6)) if len(thedict) == 0: return numbers_list #返回六位题号列表 elif len(thedict) == 1 and type(thedict[0]) == dict: numbers_list = [id for id in numbers_list if id in thedict[0]] return numbers_list #返回字典中存在的六位题号列表 else: return "输入参数有误" def generate_classid(string): #返回班级列表 string = re.sub(r"[\n\s]","",string).strip() while not string[-1] in "0123456789": string = string[:-1] string = RefinePunctuations(string) string_list = string.split(",") numbers_list = [] for s in string_list: if not ":" in s: numbers_list.append(s.zfill(2)+"班") else: start,end = s.split(":") for ind in range(int(start),int(end)+1): numbers_list.append(str(ind).zfill(2)+"班") return numbers_list #返回2位班级号列表 def generate_id_set(string,*thedict): #除了生成题号列表外, 还能根据首字母生成基础知识编号列表或课时目标列表 string = RefinePunctuations(string) if re.findall(r"[BXK]",string) == []: if thedict == (): return generate_number_set(string) else: return generate_number_set(string,thedict) else: if string[0] == "B": string = re.sub("B","",string) return ["B"+i[-5:] for i in generate_number_set(string)] elif string[0] == "K": suffix = string.strip()[-1] string = re.sub("[KBX]","",string) return ["K"+i.zfill(7)+suffix for i in generate_number_set(string)] def generate_exp(id_list): #根据题号列表生成字符串式的含":"和","的题号字符串, 例如["000001","000002","000003","000005"]生成"000001:000003,000005", 若列表为空则生成"无有效题号" if not len(id_list) == 0: exp_list = [] start = id_list[0] current = start end = start for id in id_list[1:]: # print(id,current) if int(id)-1 == int(current): current = id end = id else: if not start == end: exp_list.append('"'+start+":"+end+'"') else: exp_list.append('"'+start+'"') start = id current = id end = id if not start == end: exp_list.append('"'+start+":"+end+'"') else: exp_list.append('"'+start+'"') exp_str = ",".join(exp_list).replace('"',"") else: exp_str = "无有效题号" return exp_str #返回含有":"或","的题号字符串 def parsePDF(filePath): #提取pdf文件中的字符 with fitz.open(filePath) as doc: text = "" for page in doc.pages(): text += page.get_text() + "\n" return text def extractIDs(filePath): #提取.txt,.tex或.pdf文件中的题号, 返回含有":"或","的题号字符串 if filePath[-4:] == ".txt" or filePath[-4:] == ".tex": with open(filePath,"r",encoding = "u8") as f: data = f.read() elif filePath[-4:] == ".pdf": data = parsePDF(filePath) else: return "格式不正确" ids = re.findall(r"\((\d{6})\)",data) return generate_exp(ids) def spareIDs(): #返回空闲题号, 已更新为适合mariadb的版本 mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() mycursor.execute("SELECT ID FROM problems;") idlist = [ret[0] for ret in mycursor.fetchall()] used_str = generate_exp(idlist) used_list = used_str.split(",") output = "" for group in range(len(used_list)-1): output += "首个空闲id: %s, 直至: %s"%(str(int(used_list[group][-6:])+1).zfill(6),str(int(used_list[group+1][:6])-1).zfill(6)) + "\n" output += "首个空闲id: %s, 直至: %s"%(str(int(used_list[-1][-6:])+1).zfill(6),"999999") mydb.close() return output #返回的是一个多行的字符串, 每一行中含有一个空闲题号的闭区间 def NextSpareID(num): #返回adict中下一个空闲的题号 mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() mycursor.execute("SELECT ID FROM problems;") idlist = [ret[0] for ret in mycursor.fetchall()] num = int(num) while str(num).zfill(6) in idlist: num += 1 mydb.close() return num def NextSpareIDBlock(num): #返回adict中下一个空闲的题号块 mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() mycursor.execute("SELECT ID FROM problems;") idlist = [ret[0] for ret in mycursor.fetchall()] start = NextSpareID(num) larger_ID_list = [int(id) for id in idlist if int(id)>start] if larger_ID_list == []: end = 999999 else: end = min(larger_ID_list) - 1 mydb.close() return str(start)+":"+str(end) def GenerateProblemListFromString(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 try: data = re.findall(r"\\begin\{document\}([\s\S]*?)\\end\{document\}",data)[0] except: pass 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 = data[pos[0]:pos[1]].strip() content = re.sub(r"\n\%[\s\S]*$","",content) #题目内容 content = re.sub(r"\\\\$","",content) # 删去题目最后一行处可能存在的\\ 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)) 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 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] except: pass 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() 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 CreateEmptyProblem(problem): # 根据已有的题目创建新的空题目 NewProblem = problem.copy() for field in NewProblem: if type(NewProblem[field]) == str: NewProblem[field] = "" elif type(NewProblem[field]) == list: NewProblem[field] = [] elif type(NewProblem[field]) == int or type(NewProblem[field]) == float: NewProblem[field] = -1 return NewProblem #返回一个空题目的字典, ID和内容待赋值 def CreateNewProblem(id,content,origin,dict,editor): # 构建一道新题目的字典 NewProblem = CreateEmptyProblem(dict["000001"]) NewProblem["id"] = str(id).zfill(6) NewProblem["content"] = content NewProblem["origin"] = origin NewProblem["edit"] = [editor] return NewProblem # 返回一道新题目的字典, 已赋新的ID, 内容, 来源和编辑者 def AddRelatedProblemToDB(id,content,oid,editor): mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() id = str(id).zfill(6) content = content.strip() if "blank" in content: genre = "填空题" space = "" elif "bracket" in content: genre = "选择题" space = "" else: genre = "解答题" space = "4em" origin_json = {"来源": "改编题目", "前序": str(oid).zfill(6)} origin = json.dumps(origin_json,ensure_ascii=False) sql = "INSERT INTO problems (ID,content,genre,origin,space) VALUE (%s,%s,%s,%s,%s);" val = (id,content,genre,origin,space) mycursor.execute(sql,val) sql = "INSERT INTO edit_history (ID,date,editor) VALUE (%s,%s,%s);" val = (id,GetDate(),editor.strip()) mycursor.execute(sql,val) sql = "INSERT INTO related (ID,RELATED_ID) VALUE (%s,%s);" val = (id, oid) if id > oid: val = (oid, id) mycursor.execute(sql,val) sql = "INSERT INTO logs (DATE,TIME,username,action,id,db_content) VALUE (%s,%s,%s,%s,%s,%s);" val = (GetDate(),GetTime(),get_git_username(),"加关联题",id,content) mycursor.execute(sql,val) mydb.commit() mydb.close() return id def AddProblemstoDict2024(startingid,raworigin,problems,editor,indexed): #将来自GenerateProblemListFromString的列表中的题目添加到thedict字典, 返回题号列表(包括用老题号替代的题目) mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() idlist = [] id = int(startingid) currentsuffix = problems[0][1] problemindex = 0 for p_and_suffix_and_meta in problems: p, suffix, meta = p_and_suffix_and_meta pid = str(id).zfill(6) if suffix == currentsuffix: problemindex += 1 else: problemindex = 1 currentsuffix = suffix if indexed: origin = {"来源": raworigin + suffix, "题号": problemindex} else: origin = {"来源": raworigin + suffix} if not "rep" in meta: if "blank" in p: genre = "填空题" space = "" elif "bracket" in p: genre = "选择题" space = "" else: genre = "解答题" space = "4em" sql = "INSERT INTO problems (ID,content,genre,origin,space) VALUE (%s,%s,%s,%s,%s);" val = (pid,p.strip(),genre,json.dumps(origin,ensure_ascii=False),space) mycursor.execute(sql,val) sql = "INSERT INTO edit_history (ID,date,editor) VALUE (%s,%s,%s);" val = (pid,GetDate(),editor.strip()) mycursor.execute(sql,val) sql = "INSERT INTO logs (DATE,TIME,username,action,id,db_content) VALUE (%s,%s,%s,%s,%s,%s);" val = (GetDate(),GetTime(),get_git_username(),"新增题目",pid,p.strip()) mycursor.execute(sql,val) if "same" in meta: sql = "INSERT INTO same (ID,SAME_ID) VALUE (%s,%s);" for sid in meta["same"]: val = (pid, sid) if pid > sid: val = (sid, pid) mycursor.execute(sql,val) if "related" in meta: sql = "INSERT INTO related (ID,RELATED_ID) VALUE (%s,%s);" for rid in meta["related"]: val = (pid, rid) if pid > rid: val = (rid, pid) mycursor.execute(sql,val) if "unrelated" in meta: sql = "INSERT INTO unrelated (ID,UNRELATED_ID) VALUE (%s,%s);" for uid in meta["unrelated"]: val = (pid, uid) if pid > uid: val = (uid, pid) mycursor.execute(sql,val) print(f"已收录题号: {pid}, 题目类型: {genre}, 题目来源: {origin['来源'] + ('试题'+str(origin['题号'])) if '题号' in origin else ''}, 题目内容: {p.strip()}") id += 1 idlist.append(pid) else: idlist.append(meta["rep"]) print(f"该题 {idlist[-1]} {p} 已在题库中, 不必收录.") mydb.commit() mydb.close() return idlist def CreateIDLinks(old_id_list,new_id_list): #建立已有id和新id之间的联系, thedict为可选, 选中的话即为当前字典, 会从new_id_list中排除当前字典中有的项 if len(old_id_list)>len(new_id_list): return "新ID个数不足." else: id_links = [] for i in range(len(old_id_list)): id_links.append((old_id_list[i],new_id_list[i])) return id_links # 返回id联系, 每个元组表示一对id, 前者是旧id, 后者是新id def CreateRelatedProblems(links): # 根据links关联生成待编辑的新题目字典, 等待编辑修改 output = "\\begin{enumerate}\n" mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() sql = "SELECT content FROM problems WHERE ID = %s;" for id,rid in links: val = (id,) mycursor.execute(sql,val) ret = mycursor.fetchone()[0] output += f"\\item ({id}->{rid}) {ret}\n\n" output = f"{output}\n\\end{{enumerate}}\n" mydb.close() return output #正常返回0 def ImportRelatedProblems(new_json,main_json): # 导入编辑过的关联题目json文件到主数据库 pro_dict = load_dict(main_json) new_dict = load_dict(new_json) edited = [] for id in new_dict: new_id = new_dict[id]["id"].replace("待替换","") #新题号后需要跟"待替换"字样 if new_id in pro_dict: print("题号有重复") return 1 #异常返回1 else: edited.append(new_id) pro_dict[new_id] = new_dict[id].copy() pro_dict[new_id]["id"] = new_id pro_dict[id]["related"] += [new_id] pro_dict[new_id]["related"] += [id] p = pro_dict[new_id]["content"] if "blank" in p: pro_dict[new_id]["genre"] = "填空题" pro_dict[new_id]["genre"] = "" elif "bracket" in p: pro_dict[new_id]["genre"] = "选择题" pro_dict[new_id]["genre"] = "" else: pro_dict[new_id]["genre"] = "解答题" pro_dict[new_id]["genre"] = "4em" print("导入关联题目 %s -> %s 信息成功."%(id,new_id)) save_dict(SortDict(pro_dict),main_json) #保存至目标pro_dict文件 return generate_exp(edited) #正常返回新题号list def strip_suffix(originalString, suf_words_list): # 字符串去除指定后缀 for sw in suf_words_list: output = re.sub(sw+r"[\S]*$","",originalString) return(output) # 返回原字符串中截去suf_words_list及之后字符的部分 def get_striped_origin(pro_dict,id,suf_words_list): # 题目来源去除指定后缀 return strip_suffix(pro_dict[id]["origin"],suf_words_list) # 返回去除指定后缀后的题目来源 def SeperateFirstLine(string): # 切割一个含有换行的字符串 thelist = string.split("\n") firstline = thelist[0].strip() contents = "\n".join([lines.strip() for lines in thelist[1:]]) return (firstline,contents) # 返回第一行, 其余行形成的二元组, 每一行前后的空格都被删去 def ObtainDatatoModify(metadatafilepath,fields_dict): #从metadata文件中获取需要修改的综合信息, metadata文件用双回车区分项, 单行的项表示此后的字段, 不小于两行的项中第一行是题目id, 第二行起是要修改的内容. #fieldsdictpath是字段信息数据库文件路径 data = ReadTextFile(metadatafilepath) datalines = data.split("\n") data = "" for l in datalines: if l.strip() in fields_dict: data += l + "\n\n" else: data += l + "\n" data = re.sub(r"\n[\s\n]*\n","\n\n",data.strip()) #去除一些无意义的空格和多余的回车 datalist = data.split("\n\n") to_modify_list = [] currentfield = "NotAField" for line in datalist: if line.strip() in fields_dict: currentfield = line.strip() elif not "\n" in line: currentfield = "NotAField" else: id,content = SeperateFirstLine(line) to_modify_list.append((currentfield,str(id).zfill(6),content)) return to_modify_list #返回一个列表, 每一项是一个三元组, 三项依次为字段, 题号, 要修改的内容 def FloatToInt(string): #从字符串返回浮点数，如果值非常接近整数则返回该整数 f = float(string) if abs(f-round(f))<0.01: f = round(f) return f #返回浮点数或整数 def OverwriteData(prodict,fieldsdict,field_id_and_content): #用覆盖方式修改题目某字段信息 field,id,content = field_id_and_content fieldType = fieldsdict[field]["FieldType"] if not id in prodict: print("题号 %s 并不在数据库中"%id) else: if fieldType == "str": prodict[id][field] = content elif fieldType == "int" or fieldType == "float": content = FloatToInt(content) prodict[id][field] = content print("已覆盖 %s 的 %s 字段, 内容为 %s"%(id,field,content)) return (id,field,content) #返回三元组: 题号, 字段, 覆盖内容 def AppendData(prodict,fieldsdict,field_id_and_content): #用添加方式修改题目某字段信息 field,id,content = field_id_and_content fieldType = fieldsdict[field]["FieldType"] if not id in prodict: print("题号 %s 并不在数据库中"%id) else: if fieldType == "str": if content.strip() in prodict[id][field]: print("题号 %s 的 %s 字段, 内容 %s 已存在"%(id,field,content)) elif prodict[id][field].strip() == "": prodict[id][field] = content.strip() print("已于 %s 的 %s 字段执行添加, 内容为 %s"%(id,field,content)) else: prodict[id][field] = (prodict[id][field].strip() + "\\\\\n" + content).strip() print("已于 %s 的 %s 字段执行添加, 内容为 %s"%(id,field,content)) elif fieldType == "list": lines = [line.strip() for line in content.split("\n")] for line in lines: if line in prodict[id][field]: print("题号 %s 的 %s 字段, 内容 %s 已存在"%(id,field,line)) else: prodict[id][field] = prodict[id][field].copy() + [line] print("已于 %s 的 %s 字段执行添加, 内容为 %s"%(id,field,line)) return (id,field,content) #返回三元组: 题号, 字段, 内容 def AppendMutualData(prodict,field_id_and_content): #添加两个id之间的same, related, unrelated关联 field,id,content = field_id_and_content lines = [str(line).zfill(6) for line in content.split("\n")] for id2 in lines: if not id in prodict: print("题号 %s 并不在数据库中"%id) elif not id2 in prodict: print("题号 %s 并不在数据库中"%id2) else: if id2 in prodict[id][field]: print("题号 %s 的 %s 字段, 内容 %s 已存在"%(id,field,id2)) else: prodict[id][field] = prodict[id][field] + [id2] print("已于 %s 的 %s 字段执行添加, 内容为 %s"%(id,field,id2)) if id in prodict[id2][field]: print("题号 %s 的 %s 字段, 内容 %s 已存在"%(id2,field,id)) else: prodict[id2][field] = prodict[id2][field] + [id] print("已于 %s 的 %s 字段执行添加, 内容为 %s"%(id2,field,id)) return (id,field,content) #返回三元组: 题号, 字段, 内容 def AppendObjData(prodict,objdict,field_id_and_content): #添加目标编号数据 field,id,content = field_id_and_content if not id in prodict: print("题号 %s 并不在数据库中"%id) else: lines = [line.strip() for line in content.split("\n")] for objid in lines: if not objid in objdict: print("目标编号 %s 并不在数据库中"%objid) elif objid in prodict[id][field]: print("题号 %s 的 %s 字段, 内容 %s 已存在"%(id,field,objid)) else: prodict[id][field] = prodict[id][field] + [objid] print("已于 %s 的 %s 字段执行添加, 编号为 %s, 内容为 %s"%(id,field,objid,objdict[objid]["content"])) return (id,field,content) #返回三元组: 题号, 字段, 内容 def AppendUsageData(prodict,field_id_and_content): #添加使用记录数据 field,id,content = field_id_and_content lines = [re.sub(r"\s+",r"\t",line.strip()) for line in content.strip().split("\n")] pending_list = [] for line in lines: if not "FORCE" in line.upper(): time_stripped = re.sub(r"^\d{4,}\t","",line) #读取去除时间的数据 if time_stripped in "\n".join(prodict[id][field]): print("题号 %s 的 %s 字段, 内容 %s 已存在"%(id,field,line)) else: classid = [info for info in line.split("\t") if len(re.findall(r"[班高一二三]",info))>0][0] # 读取班级号 if classid in "\n".join(prodict[id][field]): print("班级 %s 在题号为 %s 的题目处已有使用记录, 在pending list中记录"%(classid,id)) oldinfo = [usage for usage in prodict[id][field] if classid in usage] pending_list = pending_list + [(id,line,oldinfo)] else: prodict[id][field].append(line) print("已于 %s 的 %s 字段执行添加, 内容为 %s"%(id,field,line)) else: line = re.sub(r"\s+",r"\t",re.sub(r"FORCE","",line.upper())).strip() prodict[id][field].append(line) print("已于 %s 的 %s 字段执行强制添加, 内容为 %s"%(id,field,line)) output = "" for item in pending_list: id,new,oldlist = item output += id + "\n" output += new+"\tFORCE\n" output += "\n".join(oldlist)+"\n\n" return (field,id,content,output) #返回四元组: 题号, 字段, 内容, 待确定是否要添加的字符串（不含FORCE字样的行为旧结果，含FORCE字样的行为新结果，FORCE是运行后强制添加） def AppendUsageData2024(prodict,field_id_and_content): field,id,content = field_id_and_content lines = [re.sub(r"\s+",r"\t",line.strip()) for line in content.strip().split("\n")] pending_list = [] for line in lines: if not "FORCE" in line.upper(): time_stripped = re.sub(r"^\d{4,}\t","",line) #读取去除时间的数据 if time_stripped in "\n".join(prodict[id][field]): print(f"******题号 {id} 的 {field} 字段, 内容 {line} 已存在") else: usage = parseUsage(line) oldusages = prodict[id]["usages"] importflag = True for u in oldusages: oldusage = parseUsage(u) if usage["classid"] == oldusage["classid"] and abs(datestrtotimestamp(usage["date"])-datestrtotimestamp(oldusage["date"])) < 7*86400 and usagelistdifference(usage["difficulties"],oldusage["difficulties"])<0.05: print(f"######班级 {usage['classid']} 在题号为 {id} 的题目处已有非常类似的使用记录, 不作记录") importflag = False break if importflag: for u in oldusages: oldusage = parseUsage(u) if usage["classid"] == oldusage["classid"]: print(f"!!!!!!班级 {usage['classid']} 在题号为 {id} 的题目处已有使用记录, 在pending list中记录") oldinfo = [v for v in prodict[id][field] if usage['classid'] in v] pending_list = pending_list + [(id,line,oldinfo)] importflag = False break if importflag: prodict[id][field].append(line) print(f"已于 {id} 的 {field} 字段执行添加, 内容为 {line}") else: line = re.sub(r"\s+",r"\t",re.sub(r"FORCE","",line.upper())).strip() prodict[id][field].append(line) print(f"&&&&&&&&&&&& 已于 {id} 的 {field} 字段执行强制添加, 内容为 {line} &&&&&&") output = "" for item in pending_list: id,new,oldlist = item output += id + "\n" output += new+"\tFORCE\n" output += "\n".join(oldlist)+"\n\n" return (field,id,content,output) #返回四元组: 题号, 字段, 内容, 待确定是否要添加的字符串（不含FORCE字样的行为旧结果，含FORCE字样的行为新结果，FORCE是运行后强制添加） def ImportMetadata(prodict,objdict,fieldsdict,metadatafilepath,pendingdatafilepath): #metadata自动修改, 根据字段自适应修改, 参数为题库字典, 目标字典, 字段字典, metadata文本文件路径, 待确定是否替换的内容的存放路径 data_to_modify = ObtainDatatoModify(metadatafilepath,fieldsdict) outputstring = "usages\n\n" for item in data_to_modify: field = item[0] if field == "NotAField": print("字段名有误") return 1 else: method = fieldsdict[field]["Method"] if method == "overwrite": feedback = OverwriteData(prodict,fieldsdict,item) elif method == "append": feedback = AppendData(prodict,fieldsdict,item) elif method == "mutualappend": feedback = AppendMutualData(prodict,item) elif method == "objappend": feedback = AppendObjData(prodict,objdict,item) elif method == "usageappend": feedback = AppendUsageData2024(prodict,item) outputstring += feedback[3] elif method == "fixed": print("字段 %s 不可按此方式修改"%field) SaveTextFile(outputstring,pendingdatafilepath) return outputstring # 已在数据库中修改, 之后需要将数据库写入一次, 返回1表示字段名有误, 返回其他表示成功进行了修改 def parseUsage(usagestring): #对单行usage信息进行分词 usagedict = {} datalist = re.sub(r"\s+",r"\t",usagestring.strip()).split("\t") if re.findall(r"20[\d]{2,6}",datalist[0]) != [] and re.findall(r"[\u4e00-\u9fa5]",datalist[0]) == []: date = datalist.pop(0) else: date = "" usagedict["date"] = date if re.findall(r"[高一二三班]",datalist[0]) != []: classid = datalist.pop(0) else: classid = "" usagedict["classid"] = classid usagedict["subproblems"] = len(datalist) usagedict["difficulties"] = [float(u) for u in datalist] usagedict["glossdiff"] = round(sum(usagedict["difficulties"])/usagedict["subproblems"],3) return usagedict #返回词典, 含date日期, classid班级标识, subproblems小题数目, difficulties得分率, glossdiff小题平均得分率 def StringSubstitute(regex,template,stringtuple): # 用stringtuple里的字符串逐个替换template中的符合某种regex规则的字符串 toSub = re.findall(regex,template) if not len(toSub) == len(stringtuple): print("长度不符.") return 1 #若长度不符则显示“长度不符”并返回1 else: output = template for i in range(len(toSub)): output = output.replace(toSub[i],stringtuple[i]) return output #若长度符合则返回替换后的字符串 def GenerateTexDataforEditing(id_string,prodict,templatefilepath,editor): # 根据id_string的题号列表在prodict中生成一个可修改的tex文件, 包含题目内容，答案和解答，editor表示编辑者 id_list = generate_number_set(id_string,prodict) output = "编辑者: " + editor + "\n\n\\begin{enumerate}\n\n" for id in id_list: content = prodict[id]["content"] answer = prodict[id]["ans"] solution = prodict[id]["solution"] remark = prodict[id]["remark"] output += "\\item (%s) "%str(id).zfill(6) + content + "\n\n" + "答案: " + answer + "\n\n" + "解答与提示: " + solution + "\n\n" + "备注: " + remark + "\n\n" output += "\\end{enumerate}" template = ReadTextFile(templatefilepath) texdata = StringSubstitute(r"<<[\s\S]*?待替换[\s\S]*?>>",template,[output]) return texdata # 返回Tex文件的字符串, 待保存至tex文件 def GetEditedProblems(string): #从.tex文件所读取的字符串中取出正文内容, 并切割成以题号为key的字典, 内容为[题目内容, 答案, 解答与提示, 备注]四元数组 bodydata = re.findall(r"\\begin\{enumerate\}([\s\S]*)\\end\{enumerate\}",string)[0]+r"\item" problems = re.findall(r"\((\d{6})\)([\s\S]*?)\\item",bodydata) edited_dict = {} for problem in problems: id, pro_string = problem edited_dict[id] = parseProblem(pro_string) return edited_dict # 返回以题号为key, 内容为[题目内容, 答案, 解答与提示, 备注]四元数组的字典 def parseProblem(string): # 对以不小于两个回车切分的四段式字符串进行分词(通常第二段有"答案:", 第三段有"解答与提示:", 第四段有"备注:") data = string.strip() data = re.sub(r"\n{2,}","\n\n",data) content,ans,solution,remark = data.split("\n\n") content = content.strip() ans = re.sub("答案:","",ans).strip() solution = re.sub("解答与提示:","",solution).strip() remark = re.sub("备注:","",remark).strip() return (content,ans,solution,remark) # 返回四元组(题目内容, 答案, 解答与提示, 备注) def ModifyProblembyTeX(id_string,prodict,toeditfilepath,editor): # vscode打开.tex文件修改题目的内容, 答案与解答 # 读取题号列表并生成待编辑的.tex文件 texdata = GenerateTexDataforEditing(id_string,prodict,"模板文件/题目编辑.txt",editor) SaveTextFile(texdata,toeditfilepath) # 打开待编辑的.tex文件 print("编辑完成后保存文件, 关闭文件继续...") os.system("code -w -g "+toeditfilepath) # 生成修改后的题号与内容, 答案, 解答与提示的对应 editor = GetDate() + "\t" + editor editedtexdata = ReadTextFile(toeditfilepath) edited_dict = GetEditedProblems(editedtexdata) editedIDList = [] # 将.tex文件中的修改反映到原题库字典, 并保存 for id in edited_dict: content, ans, solution, remark = edited_dict[id] if not (content == prodict[id]["content"] and ans == prodict[id]["ans"] and solution == prodict[id]["solution"] and remark == prodict[id]["remark"]): prodict[id]["content"] = content prodict[id]["ans"] = ans prodict[id]["solution"] = solution prodict[id]["remark"] = remark prodict[id]["edit"] = prodict[id]["edit"] + [editor] editedIDList.append(id) save_dict(prodict,"../题库0.3/Problems.json") return generate_exp(editedIDList) # 返回编辑过的字典 def RemoveMutualLink(metadata,prodict): # 删除双向关联id(same,related,unrelated), 输入为一个字符串, 每行表示一对题号, 用空格或制表符隔开 lines = [re.sub(r"[\s]+",r"\t",line).strip() for line in metadata.split("\n")] for line in lines: if not line == []: id1,id2 = line.split("\t") id1 = id1.zfill(6) id2 = id2.zfill(6) for field in ["same","related","unrelated"]: if id1 in prodict[id2][field]: prodict[id2][field].remove(id1) if id2 in prodict[id1][field]: prodict[id1][field].remove(id2) return 0 # 返回0 def jsonEditProblemMetadata(id_string,prodict,editor): #用vscode在json模式下编辑题目综合信息 jsontoeditpath = "临时文件/problem_edit.json" idlist = generate_number_set(id_string,prodict) edit_dict = {} for id in idlist: edit_dict[id] = prodict[id].copy() save_dict(edit_dict,jsontoeditpath) #打开待编辑的json文件 os.system("code -w -g "+jsontoeditpath) #编辑后关闭文件窗口自动执行下面步骤 editeddict = load_dict(jsontoeditpath) editlist = [] for id in editeddict: if not prodict[id] == editeddict[id]: prodict[id] = editeddict[id].copy() prodict[id]["edit"] = prodict[id]["edit"] + [GetDate() + "\t" + editor] editlist.append(id) return(generate_exp(editlist)) # 返回编辑过的题号字符串 def GetSamePairs(prodict): #获取已标注的相同题目组 same_groups = [] for id in prodict: same = prodict[id]["same"] if len(same) > 0 and not len(prodict[id]["usages"]) == 0: same_groups.append([id]+same) return(same_groups) #返回相同题目组, 每组由一些相同的题目构成 def ShareSameUsagesinDB(id1,id2,db): #有问题， 待修改 mycursor = db.cursor() sql = "SELECT date,classname,diff FROM usages WHERE ID = (%s);" val = (id1,) mycursor.execute(sql,val) id1_usages_list = mycursor.fetchall() val = (id2,) mycursor.execute(sql,val) id2_usages_list = mycursor.fetchall() for item in set(id1_usages_list)-set(id2_usages_list): print(f"{id1} -> {id2}: {item}") sql = "INSERT INTO usages (ID,date,classname,diff) VALUE (%s,%s,%s,%s);" val = (id2,item[0],item[1],item[2]) mycursor.execute(sql,val) sql = "INSERT INTO logs (DATE,TIME,username,action,db_content) VALUE (%s,%s,%s,%s,%s);" val = (GetDate(),GetTime(),get_git_username(),"分享使用数据",f"{id1} -> {id2}: {item}") mycursor.execute(sql,val) for item in set(id2_usages_list)-set(id1_usages_list): print(f"{id2} -> {id1}: {item}") sql = "INSERT INTO usages (ID,date,classname,diff) VALUE (%s,%s,%s,%s);" val = (id1,item[0],item[1],item[2]) mycursor.execute(sql,val) sql = "INSERT INTO logs (DATE,TIME,username,action,db_content) VALUE (%s,%s,%s,%s,%s);" val = (GetDate(),GetTime(),get_git_username(),"分享使用数据",f"{id2} -> {id1}: {item}") mycursor.execute(sql,val) return 0 def ShareSameUsages(prodict,same_group): #对same_group中的所有题号共享使用记录 current_usages = [] for id in same_group: for usage in prodict[id]["usages"]: if not usage in current_usages: current_usages = current_usages + [usage] for id in same_group: prodict[id]["usages"] = current_usages.copy() return((same_group,current_usages)) #返回same_group中的题号列表 及 所有题号记录 组成的二元组 def SortUsages(prodict): #对使用记录按字符串进行排序 for id in prodict: usages = prodict[id]["usages"].copy() usages.sort() prodict[id]["usages"] = usages.copy() return 0 #返回0 def SortUsagesbyAverage(theusages): #根据使用记录每一条的平均值进行排序, 越高的在越前面 glossdifflist = [] for i in range(len(theusages)): glossdiff = np.mean([float(d) for d in theusages[i][2]]) if glossdiff <= 0.999: #去除全对的使用记录 glossdifflist = glossdifflist + [(i,glossdiff)] sortedglossdifflist = sorted(glossdifflist, key = lambda x:x[1], reverse = True) newusages = [theusages[i[0]] for i in sortedglossdifflist] return(newusages) # 返回排序后的list def StripSuffix(string, suf_words): #除去字符串前后的空格及suf_words中每一个符合regex的字符串及其之后的部分 string = string.strip() for sw in suf_words: string = re.sub(sw+r"[\S]*$","",string) return(string) # 返回处理以后的字符串 def generate_origin(origin_dict): data = origin_dict["来源"] if "题号" in origin_dict: data = f"{data}试题{origin_dict['题号']}" if "前序" in origin_dict: data = f"{data}-改编自{origin_dict['前序']}" return data def MatchConditioninMariaDB(condition_list): corr_dict = { "content": ('problems','content'), "objs": ('objcorresp','obj_ID'), "tags": ('tagcorresp','tagname'), "origin": ('problems','origin'), "genre": ('problems','genre'), "ans": ('problems','ans'), "solution": ('problems','solution'), "remark": ('remarks','remark_content'), "same": ('same','ID','SAME_ID'), "related": ('related','ID','RELATED_ID'), "usages": ('usages','date','classname','diff') } mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() mycursor.execute("SELECT ID FROM problems WHERE NOT content REGEXP 'OBS';") match = set([u[0] for u in mycursor.fetchall()]) for field,reverse,regexp in condition_list: if not field in ["same","related","usages"]: table,column = corr_dict[field] sql = f"SELECT ID FROM {table} WHERE {column} REGEXP '{regexp}';" mycursor.execute(sql) newmatch = set([u[0] for u in mycursor.fetchall()]) elif field in ["same","related"]: table,col0,col1 = corr_dict[field] sql1 = f"SELECT {col1} FROM {table} WHERE {col0} REGEXP '{regexp}';" mycursor.execute(sql1) newmatch = set([u[0] for u in mycursor.fetchall()]) sql2 = f"SELECT {col0} FROM {table} WHERE {col1} REGEXP '{regexp}';" mycursor.execute(sql2) newmatch = newmatch | set([u[0] for u in mycursor.fetchall()]) elif field == "usages": table,cols = corr_dict[field][0],corr_dict[field][1:] newmatch = set([]) for i in range(len(cols)): sql = f"SELECT ID FROM {table} WHERE {cols[i]} REGEXP '{regexp}';" mycursor.execute(sql) newmatch = newmatch | set([u[0] for u in mycursor.fetchall()]) if reverse: match = match - newmatch else: match = match & newmatch mydb.close() return match def MatchCondition2024(problem,condition_list): #判断problem这一字典是否符合condition_list中的所有筛选条件 match = True #初始设定符合条件 for field, flag_not, matchexp in condition_list: exps = [i.strip() for i in matchexp.split(",")] if type(problem[field]) == list: data = "\n".join(problem[field]) elif field == "origin": data = generate_origin(problem[field]) else: data = str(problem[field]) #至此将每个字段中的内容都转为string if flag_not == False: #表示肯定的筛选 if all([re.findall(exp,data) == [] for exp in exps]): #如果有一个条件中的每个关键字都找不到, 就返回"不符合条件"(False) return False if flag_not == True: if all([re.findall(exp,data) != [] for exp in exps]): #如果有一个条件中的每个关键字都能找到, 就返回"不符合条件"(False) return False return match #返回是否符合条件 def MatchCondition(problem,condition_dict): #判断problem这一字典是否符合condition_dict中的所有筛选条件 match = True #初始设定符合条件 for fieldraw in [c for c in condition_dict if not "_not" in c and not condition_dict[c] == [""]]: #选出正向的条件([""]表示该条件不起作用) cond_list = condition_dict[fieldraw] if type(cond_list) == str: cond_list = [cond_list] field = re.sub(r"\d","",fieldraw) if type(problem[field]) == list: #将题库字典中的相应字段转化成字符串string string = "\n".join((problem[field])) else: string = str(problem[field]) current_match = False for cond in cond_list: #有一个条件被满足, 就认为当前字段满足条件了 if len(re.findall(cond,string)) > 0: current_match = True if current_match == False: match = False # 如果某个正向条件未满足, 那么match就赋值为False; 如果每一个正向条件都被满足, 那么match仍为True for fieldraw in [c for c in condition_dict if "_not" in c and not condition_dict[c] == [""]]: #选出反向的条件([""]表示该条件不起作用) cond_list = condition_dict[fieldraw] fieldraw = fieldraw.replace("_not","") field = re.sub(r"\d","",fieldraw) if type(problem[field]) == list: #将题库字典中的相应字段转化成字符串string string = "\n".join((problem[field])) else: string = str(problem[field]) current_match = True for cond in cond_list: #有一个条件被满足, 就认为当前字段不被满足了 if len(re.findall(cond,string)) > 0: current_match = False if current_match == False: match = False return match #返回是否符合条件 def get_color(value): # 根据得分率获得rgb颜色代码 value = float(value) if value>=0.5: (r,g)=(1,2-2*value) else: (r,g)=(2*value,1) return "{%.3f,%.3f,0}"%(r,g) #返回用大括号包围的rgb数值 def GenerateValueColorCode(matchobj): # 生成三位小数代表的颜色方框(黑色边框, 难度对应的底色)的LaTeX代码 value = matchobj.group(1) return "\t\\fcolorbox[rgb]{0,0,0}%s{%s}"%(get_color(value),value) #返回代码 def StudentsGetAfterContent(id,prodict,answered,spaceflag): #生成学生版讲义后的答案及空格， answered表示有无答案, spaceflag表示有无空格 string = "" if "ans" in prodict[id]: if answered: string += "答案: \\textcolor{%s}{%s}\n\n"%(("red",prodict[id]["ans"]) if prodict[id]["ans"] != "" else ("blue","暂无答案")) if spaceflag: if "space" in prodict[id]: if prodict[id]["space"] != "": string += "\\vspace*{%s}\n\n"%prodict[id]["space"] return string #生成学生讲义后的答案及空格 def XeLaTeXCompile(filedir,filename,times = 2): #在filedir目录中用XeLaTeX编译filename文件两次（能编译出正确的引用和页码） flagsuc = True for i in range(times): if os.system("xelatex -interaction=batchmode -output-directory=%s %s"%(filedir,filename)) == 0: print(f"第 {i+1} 次编译 {filename} 成功.") else: flagsuc = False return flagsuc # 若第二次编译成功则返回True, 否则返回False def XeLaTeXTest(editedid,adict,objdict,misc,templatepath,outdir,outfile): #对adict中的题号字符串editedid进行编译测试, 返回成功True或失败False latex_raw = ReadTextFile(templatepath) if sys.platform == "darwin": #非win系统用默认字体 latex_raw = re.sub(r"fontset[\s]*=[\s]*none","fontset = fandol",latex_raw) latex_raw = re.sub(r"\\setCJKmainfont",r"% \\setCJKmainfont",latex_raw) if misc["教师版"] == True: latex_raw = latex_raw.replace(r"学号\blank{50} \ 姓名\blank{80}","上海市控江中学") bodystring = "\\begin{enumerate}\n\n" idlist = generate_number_set(editedid) for id in idlist: bodystring += generateLaTeXBodyContent(id,adict,objdict,misc) bodystring += "\n\n\\end{enumerate}\n\n" latex_data = StringSubstitute(r"<<[\s\S]*?待替换[\s\S]*?>>",latex_raw,("测试",bodystring)) #替换标题和bodystring SaveTextFile(latex_data,os.path.join(outdir,outfile)) succ = XeLaTeXCompile(outdir,outfile) return succ def GenerateSectionBodyStringfromMariaDB(cursor,problems,sectiontitles,misc,consecutivenumbering= True): #生成学生版的.tex文件的主体内容 bodystring = "" count = 0 for i in range(len(problems)): if problems[i][0] in "0123456789": idlist = generate_number_set(problems[i]) sectionstring = f"\\section{{{sectiontitles[i]}}}\n\\begin{{enumerate}}\n\\setcounter{{enumi}}{{{count if consecutivenumbering else 0}}}\n\n" for id in idlist: count += 1 sectionstring += generateLaTeXBodyContentfromMariaDB(cursor,id,misc) sectionstring += "\\end{enumerate}\n\n" bodystring += sectionstring if problems[i][0] == "K": idlist = problems[i].split(",") sectionstring = f"\\section{{{sectiontitles[i]}}}\n\\begin{{enumerate}}\n\n" for objid in idlist: sectionstring += generateLaTeXobjsfromMariaDB(cursor,objid) sectionstring += "\\end{enumerate}\n\n" bodystring += sectionstring if problems[i][0] == "B": idlist = problems[i].split(",") sectionstring = f"\\section{{{sectiontitles[i]}}}\n\\begin{{enumerate}}\n\n" for bnid in idlist: sectionstring += generateLaTeXbnsfromMariaDB(cursor,bnid) sectionstring += "\\end{enumerate}\n\n" bodystring += sectionstring return bodystring #返回主题内容字符串 # if not objlist == []: # output += "\\section{课时目标}\n\n" # output += "\\begin{enumerate}\n\n" # for objid in objlist: # output += "\\item %s \\ %s \n\n"%(objid,objdict[objid]["content"]) # output += "\\end{enumerate}\n\n" #生成学习目标 # if not bnlist == []: # output += "\\section{双基梳理}\n\n" # basic_body = "" # for bnid in bnlist: # basic_body += "\\item %s\n\n"%basicknowledgedict[bnid]["content"] # output += "\\begin{enumerate}\n\n %s\n\n\\end{enumerate}\n\n"%basic_body #生成基础知识梳理 def GenerateStudentBodyString(problems,sectiontitles,pro_dict,consecutivenumbering= True,answered= True,spaceflag=True): #生成学生版的.tex文件的主体内容 bodystring = "" if len(problems) == len(sectiontitles): count = 0 for i in range(len(problems)): idlist = generate_number_set(problems[i],pro_dict) sectionstring = "\\section{%s}\n\\begin{enumerate}\n\\setcounter{enumi}{%d}\n\n"%(sectiontitles[i],count if consecutivenumbering else 0) for id in idlist: count += 1 aftercontent = StudentsGetAfterContent(id,pro_dict,answered,spaceflag) sectionstring += "\\item {\\tiny(%s)} %s\n\n%s"%(id,pro_dict[id]["content"],aftercontent) sectionstring += "\\end{enumerate}\n\n" bodystring += sectionstring else: idstring = ",".join(problems) idlist = generate_number_set(idstring,pro_dict) sectionstring = "\\begin{enumerate}\n\n" for id in idlist: aftercontent = StudentsGetAfterContent(id,pro_dict,answered,spaceflag) sectionstring += "\\item {\\tiny(%s)} %s\n\n%s"%(id,pro_dict[id]["content"],aftercontent) sectionstring += "\\end{enumerate}\n\n" bodystring += sectionstring return bodystring #返回主题内容字符串 def TeachersGetAfterContentColored(id,prodict,objdict,topandbottomusagestuple = (3,3), showobjs = True, showtags = True, showans = True, showsolution = True, showusages = True, showorigin = True, showremark = True): #生成教师版讲义后的答案及空格， topandbottomusagestuple表示保留得分率最高的使用记录与最低的使用记录的个数, 有负数表示不排列, 彩色背景 string = "" objs = GenerateObjTexCode(id,prodict,objdict) if showobjs else "" tags = ("\\begin{tcolorbox}[colback = orange!10!white, colframe = orange!10!white, breakable]\n标签: %s\n\n\\end{tcolorbox}\n"%("; ".join(prodict[id]["tags"]))) if showtags else "" ans = ("\\begin{tcolorbox}[colback = red!10!white, colframe = red!10!white, breakable]\n答案: %s\n\n\\end{tcolorbox}\n"%(prodict[id]["ans"] if prodict[id]["ans"] != "" else "暂无答案")) if showans else "" solution = ("\\begin{tcolorbox}[colback = green!10!white, colframe = green!10!white, breakable]\n解答或提示: %s\n\n\\end{tcolorbox}\n"%(prodict[id]["solution"] if prodict[id]["solution"] != "" else "暂无解答")) if showsolution else "" # solution = ("解答或提示: \\textcolor{magenta}{%s}\n\n"%(prodict[id]["solution"] if prodict[id]["solution"] != "" else "暂无解答")) if showsolution else "" origin = ("\\begin{tcolorbox}[colback = yellow!30!white, colframe = yellow!30!white, breakable]\n来源: %s\n\n\\end{tcolorbox}\n"%prodict[id]["origin"]) if showorigin else "" remark = ("\\begin{tcolorbox}[colback = cyan!30!white, colframe = cyan!30!white, breakable]\n备注: %s\n\n\\end{tcolorbox}\n"%(prodict[id]["remark"] if prodict[id]["remark"] != "" else "暂无备注")) if showremark else "" usages = ("\\begin{tcolorbox}[colback = lime!10!white, colframe = lime!10!white, breakable]\n使用记录(%s):\n\n%s\n\n\\end{tcolorbox}\n"%(str(topandbottomusagestuple),GenerateUsageTexCode(id,prodict,topandbottomusagestuple))) if showusages else "" string += objs + tags + ans + solution + usages + origin + remark return string #生成教师版讲义后的答案及空格 def TeachersGetAfterContentPlain(id,prodict,objdict,topandbottomusagestuple = (3,3), showobjs = True, showtags = True, showans = True, showsolution = True, showusages = True, showorigin = True, showremark = True): #生成教师版讲义后的答案及空格， topandbottomusagestuple表示保留得分率最高的使用记录与最低的使用记录的个数, 有负数表示不排列 string = "" objs = ("目标:\n\n%s\n\n"%GenerateObjTexCode(id,prodict,objdict)) if showobjs else "" tags = ("标签: \\textcolor[rgb]{0.5,0.6,0.8}{%s}\n\n"%("; ".join(prodict[id]["tags"])) if not prodict[id]["tags"] == [] else "标签: \n\n") if showtags else "" ans = ("答案: \\textcolor{%s}{%s}\n\n"%(("red",prodict[id]["ans"]) if prodict[id]["ans"] != "" else ("blue","暂无答案"))) if showans else "" solution = ("解答或提示: \\textcolor{magenta}{%s}\n\n"%(prodict[id]["solution"] if prodict[id]["solution"] != "" else "暂无解答")) if showsolution else "" origin = ("来源: %s\n\n"%prodict[id]["origin"]) if showorigin else "" remark = ("备注: \\textcolor[rgb]{0,0.5,0.2}{%s}\n\n"%(prodict[id]["remark"] if prodict[id]["remark"] != "" else "暂无备注")) if showremark else "" usages = ("使用记录:\n\n%s\n\n"%GenerateUsageTexCode(id,prodict,topandbottomusagestuple)) if showusages else "" string += objs + tags + ans + solution + usages + origin + remark return string #生成教师版讲义后的答案及空格 def GenerateObjTexCode(id,prodict,objdict): #生成目标代号对应的学习目标字符串（含蓝色编码） # string = "" # if prodict[id]["objs"] != []: string = "\n\\begin{tcolorbox}[colback = blue!10!white, colframe = blue!10!white, breakable]" for objid in prodict[id]["objs"]: if objid.upper() == "KNONE": string += "\n\n%s\t%s"%(objid,"无当前有效关联目标") else: string += "\n\n%s\t%s"%(objid,objdict[objid]["content"]) if prodict[id]["objs"] == []: string += "暂未关联\n" string += "\n\n\\end{tcolorbox}\n" return string #返回目标代码字符串 def ChooseUsage(usages,topandbottomusagestuple): #生成题号对应的题目的使用记录, topandbottomusagestuple表示保留得分率最高的使用记录与最低的使用记录的个数, 有负数表示不排列, 两数之和大于记录数则从高到低排列后全部展示 top,bottom = topandbottomusagestuple if top < 0 or bottom < 0: return usages # 返回原本的全部使用记录 elif top + bottom > len(usages): return SortUsagesbyAverage(usages) # 返回排列后的使用记录 else: return (SortUsagesbyAverage(usages)[:top] + SortUsagesbyAverage(usages)[(len(usages)-bottom):]) # 返回排序后的最前面top个和最后面bottom个 def GenerateUsageTexCode(usage_list_raw,misc): #根据topandbottomusagestuple的要求生成题号为id的题目的缩减版的使用记录列表， topandbottomusagestuple表示保留得分率最高的使用记录与最低的使用记录的个数, 有负数表示不排列, 两数之和大于记录数则从高到低排列后全部展示 grade = misc["字段显示设置"]["届别"] topandbottomusagestuple = misc["字段显示设置"]["使用记录"] usage_list = [] if len(grade) == 0: usage_list = usage_list_raw.copy() else: for u in usage_list_raw: for g in grade: if re.findall(g,u[1]) != []: usage_list.append(u) break usages = ChooseUsage(usage_list,topandbottomusagestuple) usages_str = ["\t".join((u[0],u[1],("\t".join(u[2])))) for u in usages] usagecode = re.sub("\\t([\d]\.[\d]{0,10})",GenerateValueColorCode,"\n\n".join(usages_str)) return usagecode #返回缩减后的使用记录列表 def generateLaTeXBodyContentfromMariaDB(cursor,id,misc): #根据id,读取的json内容adict,和字典misc来生成讲义 #misc 样例 #{ # "教师版": True, #如果设置为True则除了 题后空间 之外都进行判断并处理, 否则只处理 题后空间 和 答案 # "字段显示设置": { # "题后空间": True, # "课时目标": True, # "题目标签": True, # "答案": True, # "解答与提示": True, # "使用记录": (3,-1), # "使用记录说明": "(a,b)表示显示最好的a个和最差b个, 有-2表示不显示, 无-2但有-1表示全部显示", # "来源": True, # "备注": True, # "届别": [] # } # } id = str(id).zfill(6) sql = "SELECT content,ans,solution,origin,space FROM problems WHERE ID = %s;" val = (id,) cursor.execute(sql,val) ret_list = cursor.fetchall() if len(ret_list) != 0: content,ans,solution,raw_origin,space = ret_list[0] if ans is None: ans = "" if solution is None: solution = "" origin = json.loads(raw_origin) #以下生成obj_list sql = "SELECT obj_ID FROM objcorresp WHERE ID = %s;" cursor.execute(sql,val) ret_list = cursor.fetchall() obj_list = [] for id_raw in ret_list: obj_id = id_raw[0] sql = "SELECT obj_content FROM lessonobj WHERE objid = %s;" if not obj_id.upper() == "KNONE": objval = (obj_id,) cursor.execute(sql,objval) obj_list.append(f"{obj_id}\t{cursor.fetchall()[0][0]}") else: cursor.fetchall() obj_list.append(f"{obj_id}\t暂无目标对应") sql = "SELECT tagname FROM tagcorresp WHERE ID = %s;" cursor.execute(sql,val) tag_list = [t[0] for t in cursor.fetchall()] sql = "SELECT date,remark_content FROM remarks WHERE ID = %s;" cursor.execute(sql,val) remark_list = sorted([f"{t[0]}\t{t[1]}" for t in cursor.fetchall()]) sql = "SELECT date,classname,diff FROM usages WHERE ID = %s;" cursor.execute(sql,val) usages_fetched = [(u[0] if not u[0] is None else "",u[1],json.loads(u[2])) for u in cursor.fetchall()] usages_raw = sorted(usages_fetched) if not "教师版" in misc or misc["教师版"] == False: output = f"\n\\item {{\\tiny ({id})}} {content}" if "字段显示设置" in misc and "答案" in misc["字段显示设置"] and misc["字段显示设置"]["答案"] == True: ans = ans if len(ans) > 0 else "暂无答案" output += f"\n\n答案: \\textcolor{{{('red' if ans != '暂无答案' else 'blue')}}}{{{ans}}}\n\n" if "字段显示设置" in misc and "题后空间" in misc["字段显示设置"] and misc["字段显示设置"]["题后空间"] == True: space = f"\n\n\\vspace*{{{space}}}\n\n" if len(space) > 0 else "" output += space else: output = f"\n\\item ({id}) {content}" if "备注" in misc["字段显示设置"] and misc["字段显示设置"]["备注"] == True: remark = r"\\".join(remark_list) if len(remark_list) > 0 else "暂无备注" output += f"\n\n备注: \\textcolor[rgb]{{0,0.5,0.2}}{{{remark}}}\n\n" if "课时目标" in misc["字段显示设置"] and misc["字段显示设置"]["课时目标"] == True: if len(obj_list) == 0: obj_string = "暂无目标" else: obj_string = '\n\n'.join(obj_list) objs = f"\n\n目标:\n\n{obj_string}\n\n" output += objs if "题目标签" in misc["字段显示设置"] and misc["字段显示设置"]["题目标签"] == True: if len(tag_list) == 0: tags = "暂无标签" else: tags = '; '.join(tag_list) output += f"\n\n标签: \\textcolor[rgb]{{0.5,0.6,0.8}}{{{tags}}}\n\n" if "答案" in misc["字段显示设置"] and misc["字段显示设置"]["答案"] == True: ans = ans if len(ans) > 0 else "暂无答案" output += f"\n\n答案: \\textcolor{{{('red' if ans != '暂无答案' else 'blue')}}}{{{ans}}}\n\n" if "解答与提示" in misc["字段显示设置"] and misc["字段显示设置"]["解答与提示"] == True: solution = solution if len(solution) > 0 else "暂无解答或提示" output += f"\n\n解答或提示: \\textcolor{{magenta}}{{{solution}}}\n\n" if "使用记录" in misc["字段显示设置"] and type(misc["字段显示设置"]["使用记录"]) in (list,tuple) and not -2 in misc["字段显示设置"]["使用记录"]: usages = f"\n\n使用记录:\n\n{GenerateUsageTexCode(usages_raw,misc)}\n\n" output += usages if "来源" in misc["字段显示设置"] and misc["字段显示设置"]["来源"] == True: origin = generate_origin(origin) if len(origin) > 0 else "未记录来源" output += f"\n\n来源: {origin}\n\n" else: output = f"\\item ({id}) 题号有误!!!" return output def generateLaTeXobjsfromMariaDB(cursor,objid): objid = str(objid).upper().strip() if objid == "KNONE": return "" else: sql = "SELECT obj_content FROM lessonobj WHERE objid = %s;" val = (objid,) cursor.execute(sql,val) obj_content = cursor.fetchall()[0][0] output = f"\\item {objid} {obj_content}\n\n" return output def generateLaTeXbnsfromMariaDB(cursor,bnid): bnid = str(bnid).upper().strip() sql = "SELECT bn_content FROM basic_knowledges WHERE bn_id = %s;" val = (bnid,) cursor.execute(sql,val) try: bn_content = cursor.fetchall()[0][0] except: bn_content = "基础知识目标编号有误" output = f"\\item {bn_content}\n\n" return output def GenerateTeacherBodyString(problems,sectiontitles,prodict,objdict,consecutivenumbering = True,topandbottomusagestuple = (3,3),sectionname = "section", showobjs = True, showtags = True, showans = True, showsolution = True, showusages = True, showorigin = True, showremark = True, colored = False): #生成教师版的.tex文件的主体内容, 各项是否显示为可选 bodystring = "" GetAfterContent = TeachersGetAfterContentPlain if not colored else TeachersGetAfterContentColored if len(problems) == len(sectiontitles): count = 0 for i in range(len(problems)): idlist = generate_number_set(problems[i],prodict) sectionstring = "\\%s{%s}\n\\begin{enumerate}\n\\setcounter{enumi}{%d}\n\n"%(sectionname,sectiontitles[i],count if consecutivenumbering else 0) for id in idlist: count += 1 aftercontent = GetAfterContent(id,prodict,objdict,topandbottomusagestuple = topandbottomusagestuple, showobjs = showobjs, showtags = showtags, showans = showans, showsolution = showsolution, showusages = showusages, showorigin = showorigin, showremark = showremark) sectionstring += "\\item (%s) %s\n\n%s"%(id,prodict[id]["content"],aftercontent) sectionstring += "\\end{enumerate}\n\n" bodystring += sectionstring else: idstring = ",".join(problems) idlist = generate_number_set(idstring,prodict) sectionstring = "\\begin{enumerate}\n\n" for id in idlist: aftercontent = GetAfterContent(id,prodict,objdict,topandbottomusagestuple = topandbottomusagestuple, showobjs = showobjs, showtags = showtags, showans = showans, showsolution = showsolution, showusages = showusages, showorigin = showorigin, showremark = showremark) sectionstring += "\\item (%s) %s\n\n%s"%(id,prodict[id]["content"],aftercontent) sectionstring += "\\end{enumerate}\n\n" bodystring += sectionstring return bodystring #返回主体内容字符串 def GetValidTexFiles(path): #获取 有题号的.tex文件列表 及 有题号的.tex文件所在的路径列表 texlist = [] pathlist = [] for root,folders,files in os.walk(path): for file in files: if file[-4:] == ".tex": texdata = ReadTextFile(os.path.join(root,file)) ids = re.findall(r"\((\d{6})\)",texdata) if len(ids) > 0: pathlist = pathlist + [root] texlist = texlist + [os.path.join(root,file)] texlist = sorted(list(set(texlist))) pathlist = sorted(list(set(pathlist))) return (texlist,pathlist) # 返回 有题号的.tex文件列表 与 所在路径列表 组成的二元组 def generate_lessonid_list(lessonidstr,lessonsdict): #根据lessonidstr的条件(开头字母与数字)返回课时代码列表 raw_criterion_list = lessonidstr.split(",") criterion_list = [] for cr in raw_criterion_list: #生成满足要求的课时代码的前若干位 if not ":" in cr: criterion_list.append(cr) else: criterion_list = criterion_list + ["K"+c[-4:] for c in generate_number_set(cr.upper().replace("K",""))] lessons_list = [] for id in lessonsdict: #对每一课时作比对, 生成课时代码列表 for cr in criterion_list: if id.startswith(cr): lessons_list.append(id) break return lessons_list # 返回课时代码列表 def GenerateLessonPreparationDraft(notetitle, outputdir, adict, prodict, objdict, lessonsdict, topandbottomusagestuple = (3,3), showobjs = True, showtags = True, showans = True, showsolution = True, showusages = True, showorigin = True, showremark = True): #根据adict中课时与题号的对应生成每一课时的备课草稿 output = "\\tableofcontents\n\\newpage\n\n" outputfilepath = os.path.join(outputdir,notetitle+".tex") for lid in adict: output += "\\section{%s \ %s}\n\n"%(lid,lessonsdict[lid]["name"]) output += "\\subsection{课时目标}\n\n" output += "\\begin{enumerate}\n\n" for objid in objdict: if objid.startswith(lid): output += "\\item %s \\ %s \n\n"%(objid,objdict[objid]["content"]) output += "\\end{enumerate}\n\n" output += GenerateTeacherBodyString([adict[lid]],["参考题目资源"],prodict,objdict,sectionname = "subsection",topandbottomusagestuple= topandbottomusagestuple, showobjs = showobjs, showtags = showtags, showans = showans, showsolution = showsolution, showusages = showusages, showorigin = showorigin, showremark = showremark) output += "\\newpage\n\n" latex_raw = ReadTextFile("模板文件/讲义模板.txt") latex_raw = latex_raw.replace(r"学号\blank{50} \ 姓名\blank{80}","上海市控江中学") #替换掉模板中的姓名学号 if sys.platform == "darwin": #非win系统用默认字体 latex_raw = re.sub(r"fontset[\s]*=[\s]*none","fontset = fandol",latex_raw) latex_raw = re.sub(r"\\setCJKmainfont",r"% \\setCJKmainfont",latex_raw) latex_data = StringSubstitute(r"<<[\s\S]*?待替换[\s\S]*?>>",latex_raw,(notetitle,output)) #替换标题和bodystring SaveTextFile(latex_data,outputfilepath) #保存.tex文件 if XeLaTeXCompile(outputdir,notetitle+".tex"): print("编译成功") return 0 # 返回0 else: print("编译失败") return latex_data # 返回有错误的latex源代码 def GenerateLessonPreparation(notetitle, outputdir, adict, prodict, objdict, basicknowledgedict, homeworkspaces = False): #根据adict生成每一课时的教案 output = "\\tableofcontents\n\\newpage\n\n" outputfilepath = os.path.join(outputdir,notetitle+".tex") for key in adict: lessonid = key lessonname = adict[key]["name"] objlist = adict[key]["objects"] bnlist = adict[key]["basicknowledges"] eblist = adict[key]["examples_basic"] ealist = adict[key]["examples_adv"] worklist = adict[key]["work_inclass"] homeworklist = adict[key]["homework"] remarks = adict[key]["remarks"] if not "复习" in lessonname: output += "\\section{%s \ %s}\n\n"%(lessonid,lessonname) if not objlist == []: output += "\\subsection{课时目标}\n\n" output += "\\begin{enumerate}\n\n" for objid in objlist: output += "\\item %s \\ %s \n\n"%(objid,objdict[objid]["content"]) output += "\\end{enumerate}\n\n" #生成学习目标 if not bnlist == []: output += "\\subsection{双基梳理}\n\n" basic_body = "" for bnid in bnlist: basic_body += "\\item %s\n\n"%basicknowledgedict[bnid]["content"] output += "\\begin{enumerate}\n\n %s\n\n\\end{enumerate}\n\n"%basic_body #生成基础知识梳理 output += "\\subsection{知识体验}\n\n" output += "\\subsubsection{必讲例题}\n\n" output += GenerateStudentBodyString(eblist,[],prodict,consecutivenumbering=False,answered=False) output += "\\subsubsection{选讲例题}\n\n" output += GenerateStudentBodyString(ealist,[],prodict,consecutivenumbering=False,answered=False) output += "\\subsection{巩固新知}\n\n" output += GenerateStudentBodyString(worklist,[],prodict,consecutivenumbering=False,answered=False) if not remarks == "": output += "\\subsection{备注}\n\n" output += remarks + "\n\n" if homeworkspaces: output += "\\newpage\n\n" output += "\\subsection{课后作业}\n\n" output += GenerateStudentBodyString(homeworklist,[],prodict,consecutivenumbering=False,answered=False,spaceflag=homeworkspaces) output += "\\newpage\n\n" else: output += "\\section{%s \ %s}\n\n"%(lessonid,lessonname) output += "\\subsection{复习题}\n\n" output += GenerateStudentBodyString(homeworklist,[],prodict,consecutivenumbering=False,answered = False,spaceflag=homeworkspaces) output += "\\newpage\n\n" latex_raw = ReadTextFile("模板文件/讲义模板.txt") latex_raw = latex_raw.replace(r"学号\blank{50} \ 姓名\blank{80}","上海市控江中学") #替换掉模板中的姓名学号 if sys.platform == "darwin": #非win系统用默认字体 latex_raw = re.sub(r"fontset[\s]*=[\s]*none","fontset = fandol",latex_raw) latex_raw = re.sub(r"\\setCJKmainfont",r"% \\setCJKmainfont",latex_raw) latex_data = StringSubstitute(r"<<[\s\S]*?待替换[\s\S]*?>>",latex_raw,(notetitle,output)) #替换标题和bodystring SaveTextFile(latex_data,outputfilepath) #保存.tex文件 if XeLaTeXCompile(outputdir,notetitle+".tex"): print("编译成功") return 0 # 返回0 else: print("编译失败") return latex_data # 返回有错误的latex源代码 def GenerateSingleLessonPreparation(lessonid, outputdir, adict, prodict, objdict, basicknowledgedict, homeworkspaces = False, filename = "default"): #根据adict生成每一课时的教案 output = "" notetitle = lessonid + r" \ " + adict[lessonid]["name"] if filename == "default": filename = lessonid + adict[lessonid]["name"]+".tex" else: filename = filename+".tex" outputfilepath = os.path.join(outputdir,filename) lessonname = adict[lessonid]["name"] objlist = adict[lessonid]["objects"] bnlist = adict[lessonid]["basicknowledges"] eblist = adict[lessonid]["examples_basic"] ealist = adict[lessonid]["examples_adv"] worklist = adict[lessonid]["work_inclass"] homeworklist = adict[lessonid]["homework"] remarks = adict[lessonid]["remarks"] if not "复习" in lessonname: if not objlist == []: output += "\\section{课时目标}\n\n" output += "\\begin{enumerate}\n\n" for objid in objlist: output += "\\item %s \\ %s \n\n"%(objid,objdict[objid]["content"]) output += "\\end{enumerate}\n\n" #生成学习目标 if not bnlist == []: output += "\\section{双基梳理}\n\n" basic_body = "" for bnid in bnlist: basic_body += "\\item %s\n\n"%basicknowledgedict[bnid]["content"] output += "\\begin{enumerate}\n\n %s\n\n\\end{enumerate}\n\n"%basic_body #生成基础知识梳理 output += "\\section{知识体验}\n\n" # output += "\\subsection{必讲例题}\n\n" output += GenerateStudentBodyString(eblist,[],prodict,consecutivenumbering=False,answered=False) # output += "\\subsection{选讲例题}\n\n" # output += GenerateStudentBodyString(ealist,[],prodict,consecutivenumbering=False,answered=False) output += "\\section{巩固新知}\n\n" output += GenerateStudentBodyString(worklist,[],prodict,consecutivenumbering=False,answered=False) if not remarks == "": output += "\\section{备注}\n\n" output += remarks + "\n\n" if homeworkspaces: output += "\\newpage\n\n" output += "\\section{课后作业}\n\n" output += GenerateStudentBodyString(homeworklist,[],prodict,consecutivenumbering=False,answered=False,spaceflag=homeworkspaces) else: output += "\\section{复习题}\n\n" output += GenerateStudentBodyString(homeworklist,[],prodict,consecutivenumbering=False,answered = False,spaceflag=homeworkspaces) latex_raw = ReadTextFile("模板文件/讲义模板.txt") latex_raw = latex_raw.replace(r"学号\blank{50} \ 姓名\blank{80}","上海市控江中学") #替换掉模板中的姓名学号 if sys.platform == "darwin": #非win系统用默认字体 latex_raw = re.sub(r"fontset[\s]*=[\s]*none","fontset = fandol",latex_raw) latex_raw = re.sub(r"\\setCJKmainfont",r"% \\setCJKmainfont",latex_raw) latex_data = StringSubstitute(r"<<[\s\S]*?待替换[\s\S]*?>>",latex_raw,(notetitle,output)) #替换标题和bodystring SaveTextFile(latex_data,outputfilepath) #保存.tex文件 if XeLaTeXCompile(outputdir,filename): print("编译成功") return 0 # 返回0 else: print("编译失败") return latex_data # 返回有错误的latex源代码 def getlineindex(string,filepath): #返回字符串在文本文件中第一次出现的行数, 若未出现则返回1 with open(filepath,"r",encoding = "u8") as f: lines = f.readlines() for l in lines: if string in l: return lines.index(l)+1 return 1 def getCopy(): # 获取剪切板内容 t = pyperclip.paste().replace("\r","") return t def setCopy(string): # 写入剪切板内容 pyperclip.copy(string) # wc.OpenClipboard() # wc.EmptyClipboard() # wc.SetClipboardData(win32con.CF_UNICODETEXT, string) # wc.CloseClipboard() def itemizeProblems(string): #将题号替换为\item string_list = string.split("\n") itemed_list = [] for line in string_list: if not "&" in line or "tabular" in line or "array" in line or "cases" in line: itemed_list.append(re.sub(r"(?:(?:^|\n)+[例]*[\s]*[0-9]+[\s]*[\.、\s]+|\[[\s]*例[\s]*[0-9]*[\s]*\][\s]*)","\\n\\\\item ",line)) else: itemed_list.append(line) string = "\n".join(itemed_list) return string def RefinePunctuations(raw_string): puctuationsfulltosemi = {"　": " ","。": ". ","．": ". ","，": ", ","：": ": ","；": "; ","（": "(","）": ")","？": "? ","“": "``","”": "''", "【": "[", "】": "]", "！": "!"} string = raw_string.strip() for s in puctuationsfulltosemi: string = re.sub(s,puctuationsfulltosemi[s],string) #将部分全角标记替换为半角 return string def RefineMathpix(raw_string): # 进一步修改mathpix得到的字符串 puctuationsfulltosemi = {"　": " ","。": ". ","．": ". ","，": ", ","：": ": ","；": "; ","（": "(","）": ")","？": "? ","“": "``","”": "''", "【": "[", "】": "]"} replacestrings = {r"\\overparen": r"\\overset\\frown", "eqslant": "eq", r"\\vec": r"\\overrightarrow ", r"\\bar": r"\\overline", r"\\lim": r"\\displaystyle\\lim", r"\\sum":r"\\displaystyle\\sum", r"\\prod":r"\\displaystyle\\prod", r"\\mid":"|", r"\^\{\\prime\}":"'",r"e\^":r"\\mathrm{e}^",r"/\s*/":r"\\parallel "} wrongrecog = {"雉":"锥","[粗秿]圆":"椭圆","投郑":"投掷","抛郑":"抛掷","范目":"范围","揷":"插","末见":"未见","末成":"未成","针角":"钝角","幕函数":"幂函数","末知":"未知","阀值":"阈值","祖[桓晅]":"祖暅","图象":"图像","末使用":"未使用","末来":"未来","竟赛":"竞赛"} string = raw_string string = re.sub(r"\\left([\.\(\[|])",lambda matchobj: "" if matchobj.group(1) == "." else matchobj.group(1),string) #删去括号前的\left标记 string = re.sub(r"\\right([\.\)\]|])",lambda matchobj: "" if matchobj.group(1) == "." else matchobj.group(1),string) #删去括号前的\right标记\ string = re.sub(r"\\left\\\{","\\{",string) #删去大括号前的\left标记 string = re.sub(r"\\right\\\}","\\}",string) #删去大括号前的\right标记 string = string.replace("\\left\\langle","\\langle").replace("\\right\\rangle","\\rangle") for s in puctuationsfulltosemi: string = re.sub(s,puctuationsfulltosemi[s],string) #将部分全角标记替换为半角 for s in replacestrings: string = re.sub(s,replacestrings[s],string) #修改部分LaTeX命令成为惯用的 for s in wrongrecog: string = re.sub(s,wrongrecog[s],string) #修改mathpix识别的一些常见错别字 string = RipTestPaperDesc(string) string = re.sub(r"[\s]*(``|''|\}|\{) *",lambda matchobj: matchobj.group(1),string) #去除符号前后的空格 string = itemizeProblems(string) #将题号替换为\item string = re.sub(r"\$\$","$",string) #行间公式替换为行内公式 string = re.sub(r"\$\s+\$"," ",string) #删除多余的$符号 string = re.sub(r"([,.:;?!])\$",lambda x:x.group(1)+" $",string) #标点和$符号分开 string = re.sub(r"\\left\\lvert\\,","|",string) #替换|的错误代码 string = re.sub(r"\\frac",r"\\dfrac",string) #替换frac为dfrac string = re.sub(r"\n(?:A\.|\(A\))([\s\S]*?)(?:B\.|\(B\))([\s\S]*?)(?:C\.|\(C\))([\s\S]*?)(?:D\.|\(D\))([\s\S]*?)\n",lambda matchobj: "\n\\fourch{%s}{%s}{%s}{%s}\n"%(matchobj.group(1).strip(),matchobj.group(2).strip(),matchobj.group(3).strip(),matchobj.group(4).strip()),string+"\n\n") # 选择题的选择支处理 string = re.sub(r"[\.;](\}\{|\}\n)",lambda matchobj: matchobj.group(1),string) #去除选择题选项最末尾的句号或分号 string = re.sub(r"\n\s+","\n",string) #删除多余的回车 string = re.sub(r"\\q+uad","",string) #删除\quad,\qquad等 string = re.sub(r"\$\\q+uad\$","",string)#删除两边有$的\quad,\qquad等 string = re.sub(r"~","",string) #删除~ string = re.sub(r"\s*\([\s]{,10}\)",r"\\bracket{20}",string) string = re.sub(r"\s*\\bracket\{20\}\s*\n",r"\\bracket{20}.\n",string)#行末无内容括号的处理 for i in range(2): string = re.sub(r"(\^|_)\{([0-9a-zA-Z])\}",lambda matchobj: matchobj.group(1)+matchobj.group(2),string) #删除一些无用的大括号 for i in range(2): string = re.sub(r"([0-9A-Z])\s+([0-9A-Z])",lambda matchobj: matchobj.group(1)+matchobj.group(2),string) #合并一些公式中的无效空格 for i in range(2): string = re.sub(r"([\u4e00-\u9fa5])\s+([\u4e00-\u9fa5])",lambda matchobj: matchobj.group(1)+matchobj.group(2),string) #合并一些文本中的无效空格 string = re.sub(r"([CP])(_[^_\^]{,5}\^)",lambda x:r"\mathrm{"+x.group(1)+"}"+x.group(2),string) #处理排列数和组合数字母的正体 string = re.sub(r"([\^_])\{([-]{0,1})\\dfrac",lambda matchobj: matchobj.group(1) + "{" + matchobj.group(2) + "\\frac",string) string = re.sub(r"ldots",r"cdots",string) #将ldots替换为cdots string = re.sub(r"\\text\s*\{,\s*当\s*\}",", ",string) string = re.sub(r"[\\{]*\\(begin|end)\{array\}(?:\{[rcl]*\}){0,1}",lambda matchobj: "\\" + matchobj.group(1) + "{cases}",string) #将分段函数的array环境改为cases环境 string = re.sub(r"([\u4e00-\u9fa5\$])[\s]*\n\\item",lambda matchobj: matchobj.group(1)+"\\blank{50}.\n\\item",string) #给中文或公式结尾的题目最后一行加上填空的空格. string = re.sub(r"(是|为|(?:=\$))\s*([,.;\n])",lambda matchobj: matchobj.group(1) + "\\blank{50}" + ("." if matchobj.group(2) == "\n" else "") + matchobj.group(2),string) #给行中的题目需要的地方加上空格 string = re.sub(r"(有|为|(?:确定))([种条个])",lambda matchobj: matchobj.group(1) + "\\blank{50}" + matchobj.group(2),string) #给行中的题目需要的地方加上空格 if not "\\bracket" in string: string = re.sub(r"([\u4e00-\u9fa5\$])(?:\\bracket\{20\})*[\.]*[\s]*\n\\fourch",lambda matchobj: matchobj.group(1)+"\\bracket{20}.\n\\fourch",string) #给中文或公式结尾的题目最后一行加上选择题的括号. string = re.sub(r"\\bracket\{(\d*)\}\$",lambda matchobj: "$"+"\\bracket{"+matchobj.group(1)+"}",string)#交换出现在$之前的括号与$的位置 string = re.sub(r"(%[^\n]*)\\blank\{50\}\.",lambda matchobj:matchobj.group(1),string) #注释行不加\blank{50} string = re.sub(r"\\blank\{50\}\.\s*\n\\fourch",r"\\bracket{20}."+"\n"+r"\\fourch",string) #选择题前的空格改为括号 string = re.sub(r"[\\\\]*\n(\(\d{1,2}\))(?:(?!\n)\s)*",lambda matchobj: "\\\\\n"+matchobj.group(1)+" ",string) #新一行的小题号回车前加上换行符 string = re.sub(r"\(([^\(\)]*(?:\\in|=|\\ge|\\le|\\ne|>|<|\\parallel|\\perp)[^\(\)]*)\)\$",lambda matchobj: "$($" + matchobj.group(1) + "$)",string) #公式最后的范围陈述的括号放到公式环境外 string = re.sub(r"\$\$\(","(",string) #删去上一步造成的多余双$ string = re.sub(r"\(\s\$","($",string) #删去首个括号内公式前的空格 string = re.sub(r"\(\\begin\{cases\}",r"\\begin{pmatrix}",string) #修改错误的\begin{cases}) string = re.sub(r"\\end\{cases\}\)",r"\\end{pmatrix}",string) #修改错误的(\end{cases} string = SplitMathComma(string) #判断数学环境中的","是否需要用$ $分离, 如果需要则执行分离 string = MergeMathComma(string) #判断非数学环境中的","是否需要合并在一个数学环境中, 如果需要则执行合并 string = RefineCasesEnv(string) #美化cases环境 string = RefineChineseComma(string) #改顿号 string = RefineInterval(string) #改错误的dollars符号和括号的顺序 string = string.replace("$$"," ") string = re.sub(r"\s+\\blank\{50\}",r"\\blank{50}",string) string = re.sub(r"\s+\\bracket\{20\}",r"\\bracket{20}",string) return string def RefineInterval(string): newstring = string matches = re.finditer(r"\+\\infty\$\)",newstring) #先处理右括号 poslist = [(match.start(),match.end()) for match in matches] poslist.reverse() #从后往前逐一更改 for s,e in poslist: leftdollar = newstring[:s].rfind("$") leftbracket = leftdollar-1 lb = newstring[leftbracket] if lb in ["[","("]: #如果上一个$前面是括号, 就交换这一对相应的$符号和括号 print("$"+lb+newstring[leftbracket+2:e-2] + ")$") newstring = newstring[:leftbracket] + "$"+lb+newstring[leftbracket+2:e-2] + ")$" + newstring[e:] matches = re.finditer(r"\(\$-\\infty",newstring) poslist = [(match.start(),match.end()) for match in matches] poslist.reverse() #从后往前逐一更改 for s,e in poslist: # print(s,e) rightdollar = newstring[e:].find("$")+e rightbracket = rightdollar+1 rb = newstring[rightbracket] if rb in ["]",")"]: print("$("+newstring[s+2:rightbracket-1]+rb+"$") newstring = newstring[:s]+"$("+newstring[s+2:rightbracket-1]+rb+"$"+newstring[rightbracket+1:] newstring = newstring.replace("($-\\infty$, $+\\infty$)","$(-\\infty, +\\infty)$") return newstring def RefineChineseComma(string): #顿号如果在数学环境中, 则在两侧加上$符号 CommaPositions = [match.start() for match in re.finditer("、",string)] CommaPositions.reverse() for pos in CommaPositions: if string[:pos].count("$") % 2 == 1: string = string[:pos].rstrip() + "$、$" + string[(pos+1):].lstrip() return string def SplitMathComma(string): #判断数学环境中的","是否需要用$ $分离, 如果需要则执行分离 UnsplittedPairs = [(r"[\(\[]",r"[\)\]]"),(r"\{",r"\}"),(r"\\begin\{cases\}",r"\\end\{cases\}")] lmatter,rmatter = (string,"") while "," in lmatter: pos = lmatter.rfind(",") if lmatter[:pos].count("$") % 2 == 1: stringtemp = lmatter+rmatter start = stringtemp[:pos].rfind("$") end = stringtemp[pos:].find("$")+pos #寻找,两侧的数学环境 stringtemp[start:end+1] locallmatter = stringtemp[start+1:pos] localrmatter = stringtemp[pos+1:end] tosplit = True for p1,p2 in UnsplittedPairs: p1l = SubstringOccurence(p1,locallmatter) p1r = SubstringOccurence(p1,localrmatter) p2l = SubstringOccurence(p2,locallmatter) p2r = SubstringOccurence(p2,localrmatter) if len(p1l)>0 and len(p2l) == 0: tosplit = False if len(p2r)>0 and len(p1r) == 0: tosplit = False if len(p1l)*len(p2l)>0: if p1l[-1]>p2l[-1]: tosplit = False if len(p1r)*len(p2r)>0: if p1r[0]>p2r[0]: tosplit = False if len(SubstringOccurence(r"(?:=|\\ge|\\le|\\ne|\\in|>|<|\\parallel|\\perp)",locallmatter))*len(SubstringOccurence(r"(?:=|\\ge|\\le|\\ne|\\in|>|<|\\parallel|\\perp)",localrmatter)) == 0: tosplit = False if tosplit: rmatter = ", $"+lmatter[pos+1:].lstrip()+rmatter.lstrip() lmatter = lmatter[:pos]+"$" else: rmatter = lmatter[pos-1:]+rmatter lmatter = lmatter[:pos-1] else: rmatter = lmatter[pos-1:]+rmatter lmatter = lmatter[:pos-1] return lmatter+rmatter def MergeMathComma(string): #判断非数学环境中的","是否需要合并在一个数学环境中, 如果需要则执行合并 UnsplittedPairs = [(r"[\(\[]",r"[\)\]]"),(r"\{",r"\}"),(r"\\begin\{cases\}",r"\\end\{cases\}")] lmatter,rmatter = (string,"") while not [item for item in re.finditer(r"\$,\s*\$",lmatter)] == []: CommaPos = [(item.start(),item.end()) for item in re.finditer(r"\$,\s*\$",lmatter)] ThePos = CommaPos[-1] tempstring = lmatter+rmatter lpos = tempstring[:ThePos[0]].rfind("$") rpos = tempstring[ThePos[1]+1:].find("$") + ThePos[1] + 1 locallmatter = tempstring[lpos+1:ThePos[0]] localrmatter = tempstring[ThePos[1]:rpos] tomerge = False for p1,p2 in UnsplittedPairs: if len(SubstringOccurence(p1,locallmatter))-len(SubstringOccurence(p2,locallmatter)) == 1 and len(SubstringOccurence(p2,localrmatter))-len(SubstringOccurence(p1,localrmatter)) == 1: tomerge = True if tomerge: rmatter = ", "+lmatter[ThePos[1]:]+rmatter lmatter = lmatter[:ThePos[0]] else: rmatter = lmatter[ThePos[0]:]+rmatter lmatter = lmatter[:ThePos[0]] return lmatter+rmatter def ReverseMatchBrackets(string):#从最后开始寻找第一组匹配的括号, 返回左括号位置与右括号位置组成的数对 if not ")" in string: return -1 else: endpos = string.rfind(")") lmatter = string[:endpos] startpos = -1 for i in range(len(lmatter)-1,-1,-1): templmatter = lmatter[i:] if templmatter.count("(")-templmatter.count(")") == 1: startpos = i break return (startpos,endpos) def RefineCasesEnv(string): # 美化cases环境 CasesPos = [(item.start(1),item.end(1)) for item in re.finditer(r"\\begin\{cases\}(.*?)\\end\{cases\}",string, re.DOTALL)] #找到所有的cases环境 CasesPos.reverse() for start,end in CasesPos: lmatter = string[:start] rmatter = string[end:] content = string[start:end] lines = [item.strip() for item in content.split(r"\\")] newlist = [] for line in lines: newline = line if line[-1] == ")": lbracket,rbracket = ReverseMatchBrackets(line) if re.findall(r"(?:=|<|>|\\ge|\\le|\\in)",line[lbracket:rbracket]) != []: newline = line[:lbracket] + ", " + line[lbracket+1:rbracket] + "," #用逗号代替表示范围的括号 newline = re.sub(r",",",&",newline) newlist.append(newline) newcontent = r"\\".join(newlist) newcontent = re.sub(r",[\s]*&[\s]*\\",r",\\",newcontent) newcontent = re.sub(r",[\s]*&[\s]*$",",",newcontent) #给逗号添加&分栏符, 并去除无效的& string = lmatter + newcontent + rmatter string = re.sub(r"&[,\s]&","& ",string) return string #返回处理后的字符串 def RipTestPaperDesc(string): string = re.sub(r"[一二三四五六][、\.\s]+(?:(?:填空)|(?:选择)|(?:解答))题","",string) #去除没有标分数的填空选择解答题描述 string = re.sub(r"\(本[大]*题满分[^\)]*\d+\s*分\)","",string) #去除带括号的分数描述 string = re.sub(r"本[大]*题[^\n]*?步骤\.{0,1}","",string) #去除解答题需要写出步骤提示 string = re.sub(r"\({0,1}本[大]*题共有[^\n]*\d+\s*分\.{0,1}\){0,1}","",string) #去除解答题单题分数提示 string = re.sub(r"每题有[^\n]*涂黑\.{0,1}","",string) #去除选择题提示 string = re.sub(r"考生[^\n]*结果\.*","",string) #去除作答提示 return string def SubstringOccurence(regex,string): #生成regex在string中出现的所有位置 poslist = [item.start() for item in re.finditer(regex,string)] return poslist def select_grade_from_pro_dict(prodict,grades): if len(grades) == 0: return prodict else: gradelist = [] for g in grades: if not "届" in g: gradelist.append(g+"届") else: gradelist.append(g) adict = prodict.copy() for id in prodict: raw_usages = prodict[id]["usages"].copy() new_usages = [] for u in raw_usages: for g in gradelist: if g in u: new_usages.append(u) adict[id]["usages"] = new_usages.copy() return adict def GenerateSingleLessonNotefromMariaDB(cursor,id,notesdict,templatepath,outputfilepath,misc,consecutivenumbering = False): #20240415版讲义生成 notetitle = id + r" \ " + notesdict["notes"][id]["name"] structure = notesdict["structures"][id[0].upper()]["structure"] note_contents = notesdict["notes"][id] output = "" sections_list = [] problems_list = [] for key in structure: if not len(note_contents[key]) == 0: sections_list.append(key) problems_list.append(",".join(note_contents[key])) rawoutput = GenerateSectionBodyStringfromMariaDB(cursor = cursor, problems=problems_list,sectiontitles=sections_list,misc=misc,consecutivenumbering= consecutivenumbering) paragraphs = [p for p in rawoutput.split("\\section") if not p.strip() == ""] for item in paragraphs: sectionkey, content = re.findall(r"\{([\S]*)\}\n([\S\s]*)$",item)[0] if not len(paragraphs) == 1: output += "\\section{" + structure[sectionkey]["name"] + "}\n\n" # if not structure[sectionkey]["spaceflag"] or answered: # content = re.sub(r"\\vspace[\*]?\{[\S]*\}","\n",content) output += content + "\n\n" latex_raw = ReadTextFile(templatepath) latex_raw = latex_raw.replace(r"学号\blank{50} \ 姓名\blank{80}","上海市控江中学") #替换掉模板中的姓名学号 if sys.platform == "darwin": #非win系统用默认字体 latex_raw = re.sub(r"fontset[\s]*=[\s]*none","fontset = fandol",latex_raw) latex_raw = re.sub(r"\\setCJKmainfont",r"% \\setCJKmainfont",latex_raw) latex_data = StringSubstitute(r"<<[\s\S]*?待替换[\s\S]*?>>",latex_raw,(notetitle,output)) #替换标题和bodystring SaveTextFile(latex_data,outputfilepath) #保存.tex文件 if misc["编译单个文件"] == True: outputdir,filename = os.path.split(outputfilepath) print(f"{filename}编译中...") if XeLaTeXCompile(outputdir,filename): print("编译成功") else: print("编译失败") return latex_data # 返回有错误的latex源代码 def GenerateSingleLessonNote(id,notesdict,metadict,templatepath,outputfilepath,consecutivenumbering = False, answered = False): #20231215版讲义生成 notetitle = id + r" \ " + notesdict["notes"][id]["name"] structure = notesdict["structures"][id[0].upper()]["structure"] note_contents = notesdict["notes"][id] output = "" sections_list = [] problems_list = [] for key in structure: if not len(note_contents[key]) == 0: sections_list.append(key) problems_list.append(",".join(note_contents[key])) rawoutput = GenerateStudentBodyString(problems=problems_list,sectiontitles=sections_list,pro_dict=metadict,consecutivenumbering= consecutivenumbering, answered= answered, spaceflag = True) paragraphs = [p for p in rawoutput.split("\\section") if not p.strip() == ""] for item in paragraphs: sectionkey, content = re.findall(r"\{([\S]*)\}\n([\S\s]*)$",item)[0] if not len(paragraphs) == 1: output += "\\section{" + structure[sectionkey]["name"] + "}\n\n" if not structure[sectionkey]["spaceflag"] or answered: content = re.sub(r"\\vspace[\*]?\{[\S]*\}","\n",content) output += content + "\n\n" latex_raw = ReadTextFile(templatepath) latex_raw = latex_raw.replace(r"学号\blank{50} \ 姓名\blank{80}","上海市控江中学") #替换掉模板中的姓名学号 if sys.platform == "darwin": #非win系统用默认字体 latex_raw = re.sub(r"fontset[\s]*=[\s]*none","fontset = fandol",latex_raw) latex_raw = re.sub(r"\\setCJKmainfont",r"% \\setCJKmainfont",latex_raw) latex_data = StringSubstitute(r"<<[\s\S]*?待替换[\s\S]*?>>",latex_raw,(notetitle,output)) #替换标题和bodystring SaveTextFile(latex_data,outputfilepath) #保存.tex文件 outputdir,filename = os.path.split(outputfilepath) print(f"{filename}编译中...") if XeLaTeXCompile(outputdir,filename): print("编译成功") return latex_data # 返回0 else: print("编译失败") return latex_data # 返回有错误的latex源代码 def getUnit(n): # 返回0-9的数字对应的单元名 unitlist = ["暂无对应","第一单元","第二单元","第三单元","第四单元","第五单元","第六单元","第七单元","第八单元","第九单元"] if 0<=n<=9: return unitlist[n] def getUnitNumber(string): unitlist = ["暂无对应","第一单元","第二单元","第三单元","第四单元","第五单元","第六单元","第七单元","第八单元","第九单元"] if string in unitlist[1:]: return unitlist.index(string) def ExtractProblemIDs(paperdict,pro_dict):#从备课组材料的每一张讲义的dict(paperdict)中提取题号 output = [] for key in paperdict.keys(): if type(paperdict[key]) == list: suslist = paperdict[key] flag = True for id in suslist: if not id in pro_dict: flag = False break if flag: for id in suslist: output.append(id) return(output) def ParseZipname(zipfilename): #小闲平台的zip文件中获得试卷编号, 返回试卷编号字符串 xiaoxianpid = re.findall(r"^[可选_]*(\d*?)_",os.path.split(zipfilename)[1]) return xiaoxianpid[0] def FindFile(dir,filename): #在指定目录及子目录下寻找特定文件名的文件, 返回文件所在的路径列表 pathlist = [] for path,m,filenames in os.walk(dir): if filename in filenames: pathlist.append(path) return pathlist def FindPaper(xiaoxianpid, answersheetpath): #根据小闲的试卷编号和答题纸对应json的根目录寻找题库的试卷编号,届别,题号, 返回(题库试卷编号,届别,题号列表), 如果未找到则返回False answersheetpathlist = FindFile(answersheetpath,"答题纸对应.json") foundpid = False for dir in answersheetpathlist: filepath = os.path.join(dir,"答题纸对应.json") anssheetjson = load_dict(filepath) if xiaoxianpid in anssheetjson: foundpid = True grade = "20"+re.findall(r"\d{2}届",dir)[0] pid = anssheetjson[xiaoxianpid]["id"] notesjson = load_dict(os.path.join(dir,"校本材料.json")) if not "idlist" in anssheetjson[xiaoxianpid]: idlist = [] for part in anssheetjson[xiaoxianpid]["parts"]: idlist += notesjson["notes"][pid][part].copy() else: idlist = anssheetjson[xiaoxianpid]["idlist"] if "marks" in anssheetjson[xiaoxianpid]: marks = anssheetjson[xiaoxianpid]["marks"] else: marks = [] if "exclude" in anssheetjson[xiaoxianpid]: excludejson = anssheetjson[xiaoxianpid]["exclude"] else: excludejson = {} break if foundpid: return(pid,grade,idlist,marks,excludejson) else: return False def CheckPaperType(filepath,filename): #根据filepath(通常是小闲的zip解压出的目录)和filename(通常是"小题分_按学号（数学）.xlsx")检测试卷类型, 未找到该文件则返回False, 找到文件且是日常试卷返回"日常卷", 找到文件且不是日常试卷返回"考试卷" statsfilepathlist = FindFile(filepath,filename) if statsfilepathlist == []: return False else: dir = statsfilepathlist[0] dfcurrent = pd.read_excel(os.path.join(dir,filename)) if re.findall(r"第\d*步",str(dfcurrent.loc[1,:])) == []: return "日常卷" else: return "考试卷" def generateColIndexandMarks(filepath,statsfilename,paperinfo): #根据filepath(是一个有statsfilename的文件夹列表)中第一个路径中的数据文件, statsfilename数据文件名, 及paperinfo(FindPaper返回的结果)寻找excel文件中有效的列的位置和相应的满分分数 dir = filepath[0] dfcurrent = pd.read_excel(os.path.join(dir,statsfilename)) validcols = [] for i in range(len(dfcurrent.columns)): colname = str(dfcurrent.iloc[1,i]) if ("单选" in colname or "填空" in colname or "主观" in colname or "步" in colname) and re.findall("[ABCD]",colname) == []: validcols.append(i) for col in range(len(validcols)-1,-1,-1): colname = str(dfcurrent.iloc[1,validcols[col]]) if "主观" in colname: colname_main = re.findall(r"^([\d\.]*)[\($]",colname[2:])[0] t = [dfcurrent.iloc[1,c] for c in validcols[col+1:]] t = str(t) if colname_main in t: validcols.pop(col) if paperinfo[3] == []: marks = [1] * len(validcols) else: marks = paperinfo[3] if len(marks) == len(validcols): return (validcols,marks) else: return False def CheckValidity(classpath,gradename,threshold): #根据文件夹classpath, 年级名gradename和提交比例threshold, 检测提交人数是否不小于threshold, 返回(班级名, 提交是否有效) classname_raw = re.findall(r"(高[一二三])(\d*?)班",classpath)[0] classname = gradename + classname_raw[0] + classname_raw[1].zfill(2) + "班" df = pd.read_excel(os.path.join(os.path.split(classpath)[0],"学科总体分析.xlsx")) totalstudents = df.loc[2,df.columns[1]] validstudents = df.loc[2,df.columns[2]] classvalidflag = False if threshold * totalstudents < validstudents: print(f"{classname} 有效, 共 {totalstudents} 人, 提交 {validstudents} 人") classvalidflag = True else: print(f"!!! {classname} 无效, 共 {totalstudents} 人, 提交 {validstudents} 人") return (classname,classvalidflag) def generateIDtoUsageCorrespondence(idlist,validcols,names): #根据idlist(题库ID列表), validcols(有效列位置列表), names(题目名称列表)自动生成一个字典, 键值为题库ID, 内容为该题对应的列位置列表 corr_dict = {} for i in range(len(idlist)): ind = i+1 collist = [] for j in range(len(validcols)): n = names[j] if not "步" in n: name = re.findall(r"^([^\(]*)",n)[0] else: name = re.findall(r"^([^第]*)",n)[0] if ind == getindex(name): collist.append(j) corr_dict[idlist[i]] = collist return corr_dict def CalculateUsages(statsfilepathlist,statsfilename,gradename,threshold,marks,correspondence_dict,validcols,date,exclude = {}): #根据统计数据所在的路径,文件名,年级,阈值,分数列表和题号列数(0-len(validcols))对应字典,以及原excel文件中的有效列位置validcols, 日期date, 生成usages的metadata.txt文件的内容, 如果有正确率大于1的则返回False output = "usages\n\n\n" validflag = True marks = [int(mark) for mark in marks] for dir in statsfilepathlist: classname, valid = CheckValidity(dir,gradename,threshold) if valid: dfcurrent = pd.read_excel(os.path.join(dir,statsfilename)) means = dfcurrent.iloc[2:-2,validcols].mean()/marks if max(means)>1: print("满分数据有误!!!") validflag = False else: means_out = [f"{t:.3f}" for t in means] for id in correspondence_dict: cols = correspondence_dict[id] if not len(cols) == 0 and not (id in exclude and classname[-3:] in generate_classid(exclude[id])): diffs = "\t".join([means_out[u] for u in cols]) usages = f"{date}\t{classname}\t{diffs}" output += f"{id}\n{usages}\n\n\n" if validflag: return output else: return False def getindex(string,pos = 2): para = string.split(".") return int(para[pos-1]) def generateListOfIDandContent(string): #根据标准的讲义LaTeX源码字符串(可能含答案)生成一个list, 每一项是一个(id,id后内容)的tuple return re.findall(r"\((\d{6})\)([\s\S]*?)(?:(?:\\item)|(?:\\end\{enumerate\}))",string) def CountEffectiveBraces(string): #在string中统计有效的(LaTeX的\{和\}不算作有效)的大括号的个数, 返回(左大括号个数, 右大括号个数) string_ref = re.sub(r"\\[\{\}]"," ",string) return (string_ref.count("{"),string_ref.count("}")) def generateAnswerTex(content,anspreamble = "答案: \\textcolor{red}{"): #在从anspreamble开始的字符串后找到答案字符串(anspreamble中的字符不算) startpos = content.index(anspreamble) + len(anspreamble) - 1 endpos = startpos + 1 l,r = CountEffectiveBraces(content[startpos:endpos]) while not l == r: endpos += 1 l,r = CountEffectiveBraces(content[startpos:endpos]) return content[startpos+1:endpos-1].strip() def generateAnswerList(string,anspreamble = "答案: \\textcolor{red}{"): #从LaTeX源代码string中分析题号与对应的答案 alist = generateListOfIDandContent(string) anslist = [] for a in alist: id,content = a if anspreamble in content: anslist.append((id,generateAnswerTex(content,anspreamble))) return anslist def unUnitted(idexp): #返回adict中未赋单元的id列表 idlist = generate_number_set(idexp) mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() mycursor.execute("SELECT ID FROM tagcorresp WHERE tagname REGEXP '第\\\\S*单元' or tagname = '暂无对应';") unittedids = set([ret[0] for ret in mycursor.fetchall()]) ununittedids = sorted(list(set(idlist)-unittedids)) return ununittedids def AutoAssignTagNotoLaTeX(newlist,allsamelist,allrelatedlist): output = "\\begin{enumerate}\n\n" tagrecord = "" print("正在生成简化题目的字典...") treateddict = treat_dict() print("简化题目字典生成完毕.") for id in newlist: samelist = findsru(id,allsamelist) relatedlist = findsru(id,allrelatedlist) if not samelist == []: unittags = get_unit_tags(samelist[0]) elif not relatedlist == []: unittags = get_unit_tags(relatedlist[0]) else: simgroup = stringmaxsim(treateddict[id],treateddict,10) unittags_raw = {} for g in simgroup: rid,sim = g if not rid == id and sim > 0.75: for t in get_unit_tags(rid): if t in unittags_raw: unittags_raw[t] += 1 else: unittags_raw[t] = 1 if len(unittags_raw) == 0: unittags = [] else: unittags = [max(unittags_raw, key = lambda x: unittags_raw[x])] tagnumbers = "" for u in unittags: tagnumbers += str(getUnitNumber(u)) output += f"\\item ({id})[{tagnumbers}] {get_problem_content(id)}\n\n" tagrecord += f"{id}\t{tagnumbers}\n" output += "\\end{enumerate}\n\n" return (output,tagrecord) def UnitRectoMetadataText(data): #把data中的单元对应变为metadata.txt中的单元对应文本 lines = [l for l in data.split("\n") if l.strip() != ""] output = "tags\n\n" for l in lines: id,unitno = l.split("\t") units = "" if not len(unitno.strip()) == 0: for n in unitno: units += f"{getUnit(int(n))}\n" output += f"{id.zfill(6)}\n" output += f"{units}\n\n" return output def gitdiff(commitid1,commitid2,fileinrepo="题库0.3/Problems.json",tempfilepath = "临时文件/temp.json"): # 根据两个git 的commit id 比较某一个json文件的不同, 返回一个tuple, 第一项是新增key的变化, 第二项是同一个key下哪些字段变化了 os.system(f"git show {commitid1}:{fileinrepo} > {tempfilepath}") json1 = load_dict("临时文件/temp.json") os.system(f"git show {commitid2}:{fileinrepo} > {tempfilepath}") json2 = load_dict("临时文件/temp.json") os.remove("临时文件/temp.json") diff_list = [] new_or_deleted_list = [id for id in json1.keys() if not id in json2.keys()] + [id for id in json2.keys() if not id in json1.keys()] for id in [i for i in json1.keys() if i in json2.keys()]: current_diff = [] for field in json1[id]: if not json1[id][field] == json2[id][field]: current_diff.append(field) if not current_diff == []: diff_list.append((id,current_diff.copy())) return (new_or_deleted_list,diff_list) def CheckUsagesValidity(data): #检查使用记录数据是否合理(没有超过1的), 若有则返回1, 合理则返回0 diffs = re.findall(r"\s(\d+\.\d{3})[\s\n$]",data) for d in diffs: if float(d)>1: print(f"有{d}, 数据有问题") return 1 return 0 def pairingbraces(string_raw,leftbracepos): #根据字符串中的左括号的位置寻找配对的右括号 types = {"(":"()","[":"[]","{":"{}"} braces = "".join([types[i] for i in types]) string = string_raw for i in braces: string = string.replace("\\"+i," ") #去除LaTeX的括号 if not string[leftbracepos] in ("(","[","{"): print("位置上的字符不是左括号") return -1 #位置上不是括号, 返回-1 else: braces = types[string[leftbracepos]] lbrace = braces[0] rbrace = braces[1] count = 1 currentpos = leftbracepos + 1 while not currentpos == len(string): if string[currentpos] == lbrace: count += 1 elif string[currentpos] == rbrace: count -= 1 if count == 0: return currentpos # 返回配对的右括号位置 currentpos += 1 return -1 #未找到配对括号 def InMathEnv(string,pos): #判断string的pos位置上的字符是否在math环境中 string = string.repalce("\$"," ") if not string[pos] == "$" and string[:pos].count("$") % 2 == 1: return True else: return False def MaskingMathEnv(string_raw): #把string_raw中数学环境中的符号都转为$后以新字符串的形式返回 string_raw = string_raw.replace(r"\$"," ") countdollars = 0 string = "" for i in range(len(string_raw)): if string_raw[i] == "$": countdollars = 1 - countdollars if countdollars == 0: string += string_raw[i] else: string += "$" return string def MultiplechoicetoBlankFilling(string_raw): #把多选题的题干和选项转为填空题 firstchoicepos = re.search(r"(?:(?:four)|(?:two)|(?:one))ch",string_raw).span()[1] headstring = string_raw[:firstchoicepos] headstring = re.sub(r"\\bracket\{[\d]*\}",r"\\blank{50}",headstring) headstring = re.sub(r"\n\\(?:(?:four)|(?:two)|(?:one))ch",r"\\\\\n",headstring) string = string_raw[firstchoicepos:] choices = [] while "{" in string: endpos = pairingbraces(string,0) choices.append(string[1:endpos]) string = string[endpos+1:] output = "" for i in range(len(choices)): output += f"\\textcircled{{{i+1}}} {choices[i]}; " output = headstring + output[:-2]+"." return output def ExtractIDList(filepath): #从文件获取题目序号和ID的对应, 返回一个列表, 列表中的每个tuple都是(文件中的题目序号(int),六位题号(string)) if filepath[-4:] == ".pdf": data = parsePDF(filepath) idlist = re.findall(r"(\d+)\.[\s\n]*\((\d{6})\)",data) else: data = ReadTextFile(filepath) enumerateBlocks = [item.strip() for item in re.findall(r"\\begin\{enumerate\}([\s\S]*?)\\end\{enumerate\}",data)] for i in range(len(enumerateBlocks)): if not enumerateBlocks[i].startswith("\\setcounter"): enumerateBlocks[i] = "\\setcounter{enumi}{0}\n\n" + enumerateBlocks[i] idlist = [] for item in enumerateBlocks: enumilist = [i.span() for i in re.finditer(r"\\setcounter\{enumi\}\{(\d+)\}",item)] enumilist.append((len(item),len(item))) for i in range(len(enumilist)-1): indstring = item[enumilist[i][0]:enumilist[i][1]] ind = int(re.findall(r"\d+",indstring)[0]) bodylist = re.findall(r"\((\d{6})\)",item[enumilist[i][1]:enumilist[i+1][0]]) for id in bodylist: ind += 1 idlist.append((str(ind),id)) idlist = [(int(id[0]),id[1]) for id in idlist] return idlist def ExportIDList(filepath): #从.tex或.pdf文件获取题目序号和ID的对应(每一行每个题号分别对应第1,2,3...题, 空缺的题号的ID用999999表示), 返回题号字符串 idlist = ExtractIDList(filepath) output = "" currentoutput = [] for i in range(len(idlist)): item = idlist[i] if i == 0: currentind = item[0] currentoutput +=["999999"]*(item[0]-1) currentoutput.append(item[1]) elif item[0] >= currentind + 1: currentoutput+=(["999999"]*(item[0]-currentind-1)) currentind = item[0] currentoutput.append(item[1]) else: output += ",".join(currentoutput)+"\n\n" currentoutput = [] currentind = item[0] currentoutput +=["999999"]*(item[0]-1) currentoutput.append(item[1]) output += ",".join(currentoutput)+"\n\n" return output def makedir(dirpath): #递归创建文件夹 dirpath = dirpath.strip().replace("\\","/") dirlist_raw = dirpath.split("/") if ":" in dirlist_raw[0]: dirlist_raw[1] = dirlist_raw[0] + "/" + dirlist_raw[1] dirlist = dirlist_raw[1:].copy() else: dirlist = dirlist_raw.copy() for i in range(len(dirlist)): if not i == 0: dirlist[i] = dirlist[i-1] + "/" + dirlist[i] for dpath in dirlist: try: os.mkdir(dpath) except: pass return dirlist def TitleIDStringtoTupleList(raw_string): #将类json的标题与题号对应变为(标题,题号字符串)的tuple之后组成list corresponding_list = [] raw_string = RefinePunctuations(raw_string).replace("{","").replace("}","") raw_list = [re.sub(r"(?:(?:\s)|(?:,\s*$))","",t) for t in raw_string.split("\n") if not t.strip() == ""] for raw_item in raw_list: item = raw_item.replace('"','').replace(";","") sep = item.find(":") title = item[:sep] content = item[sep+1:] corresponding_list.append((title,content)) return corresponding_list def TitleIDTupleListtoString(correspoinding_list): #将(标题,题号字符串)的tuple组成的list转化为更方便阅读的字符串, 前面有编号 output_string = "" count = 0 for corresp in correspoinding_list: count += 1 output_string += f"{count}. {corresp[0]}: {corresp[1]}\n" return output_string def startfile(filepath): #跨平台打开文件或文件夹 isfile = os.path.isfile(filepath) osname = sys.platform if osname == "win32": if not isfile: ret = subprocess.Popen(["explorer",filepath]) if isfile: ret = subprocess.Popen(["start",filepath], shell=True) elif osname == "darwin": ret = subprocess.Popen(["open",filepath]) elif osname == "linux": ret = subprocess.Popen(["xdg-open",filepath]) return ret def PaintRedAnswers(string,prodict): #将prodict中已有的答案标红色 output = string IDandContentList = generateListOfIDandContent(output) for id,content in IDandContentList: raw_string = id+")"+content # u = output.index(raw_string) if "textcolor{blue}" in raw_string: new_ans = generateAnswerTex(content,anspreamble="\\textcolor{blue}{") raw_ans = prodict[id]["ans"] if new_ans == raw_ans: sub_string = raw_string.replace("textcolor{blue}","textcolor{red}") output = output.replace(raw_string,sub_string) elif raw_ans == "" and new_ans != "暂无答案": print(f"{id} 已有答案, 请运行提取答案控件以获取新录入的答案") elif new_ans != "暂无答案": print(f"{id} 新答案 {new_ans} 和原答案 {raw_ans} 有所不同, 请仔细检查") return output def usagelistdifference(lista,listb): #比较两个usage数组的最大差别(绝对值), 数组长度不一致则返回1 if len(lista) != len(listb): return 1 else: maxdiff = 0 for i in range(len(lista)): maxdiff = max(abs(float(lista[i])-float(listb[i])),maxdiff) return maxdiff def RefineExclude(excludejson): #将excludejson的key变为6位题号 newjson = {} for key in excludejson: newjson[key.zfill(6)] = excludejson[key] return newjson def ChooseIDsByUsageInterval(startdate,enddate,interval,classregex): #返回根据条件选出的题号字典及对应小题, 需要留意的记录长度不一的题号, 以及所有使用过的题号 validusages = [] usedproblems = [] mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() sql = "SELECT ID,date,classname,diff FROM usages;" mycursor.execute(sql) usage_list = mycursor.fetchall() for id,date,classname,diff_raw in usage_list: if not date is None and startdate <= date <= enddate and re.findall(classregex,classname) != []: validusages.append((id,date,classname,json.loads(diff_raw))) if not id in usedproblems: usedproblems.append(id) cautionids = [] #使用记录长度不一的题目id chosenproblems = {} #使用记录的平均值在thresholds中的题目id(键值为题号id, 内容为小题号, 内容为空表示没有小题) for id in usedproblems: usages = [u[1:] for u in validusages if u[0] == id] subproblems = [len(u[2]) for u in usages] if max(subproblems) != min(subproblems): cautionids.append(id) print(f"!!! 题号 {id} 的使用记录中小题数目不全相同, 请检查") else: for i in range(subproblems[0]): difflist = [float(u[2][i]) for u in usages] diffmean = np.mean(difflist) if interval[0] <= diffmean <= interval[1]: if subproblems[0] == 1: chosenproblems[id] = [] elif not id in chosenproblems: chosenproblems[id] = [i+1] else: chosenproblems[id].append(i+1) mydb.close() return (chosenproblems,cautionids,usedproblems) #返回根据条件选出的题号字典及对应小题, 需要留意的记录长度不一的题号, 以及所有使用过的题号 def getAllIDsExp(obsincluded = True): #获取题库中所有题目的id(含:,)字符串, 不含 mydb = connect(hostname = "wwylss.synology.me", port = "13306", username="tikuuser", pwd="Kjmathds_2024", db = "tikutest") mycursor = mydb.cursor() if obsincluded: mycursor.execute("SELECT ID FROM problems;") else: mycursor.execute("SELECT ID FROM problems WHERE NOT content REGEXP 'OBS';") idlist = [ret[0] for ret in mycursor.fetchall()] return generate_exp(idlist) def parseRemark(string): #从一行remark字符串中分出日期和内容, 返回 日期,内容 元组 remark_raw = string.replace(r"\\","").strip() date_raw_list = re.findall(r"^\({0,1}\d{8}\){0,1}",remark_raw) if len(date_raw_list) > 0: date_raw = date_raw_list[0] date = re.sub(r"[\(\)]","",date_raw) remark = re.sub(date,"",remark_raw).strip().replace("()","") else: date = "00000000" remark = remark_raw return date,remark if __name__ == "__main__": print("数据库工具, import用.")