database_tools增加一个函数ChooseIDsByUsageInterval选择一段时间内某年级平均正确率在区间内的题号

工具v2/database_tools.py

@@ -1,5 +1,6 @@
 import json,re,os,Levenshtein,fitz,time,sys,subprocess
 import pandas as pd
+import numpy as np
 import pyperclip
 
 
@@ -2251,5 +2252,35 @@ def RefineExclude(excludejson): #将excludejson的key变为6位题号
         newjson[key.zfill(6)] = excludejson[key]
     return newjson
 
+
+def ChooseIDsByUsageInterval(startdate,enddate,interval,grade,prodict): #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
+    used_problems = []
+    for id in prodict:
+        currentproblem = prodict[id]
+        currentusages = [parseUsage(u) for u in currentproblem["usages"] if startdate <= parseUsage(u)["date"] <= enddate and grade+"届" in parseUsage(u)["classid"]]
+        if not currentusages == []:
+            used_problems.append((id,currentusages))
+    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 item in used_problems:
+        id,usages = item
+        subproblems = [u["subproblems"] 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 = [u["difficulties"][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)
+    return (chosenproblems,cautionids,used_problems) #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
+
+
 if __name__ == "__main__":
     print("数据库工具, import用.")