Mathpix文字处理功能中新增了美化cases环境的功能

工具v2/database_tools.py

@@ -986,6 +986,7 @@ def RefineMathpix(raw_string): # 进一步修改mathpix得到的字符串
     string = re.sub(r"\(([^\(\)]*(?:\\in |=|\\ge|\\le|>|<)[^\(\)]*)\)\$",lambda matchobj: "$($" + matchobj.group(1) + "$)",string) #公式最后的范围陈述的括号放到公式环境外
     string = SplitMathComma(string) #判断数学环境中的","是否需要用$ $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
 
     return string
@@ -1054,13 +1055,49 @@ def MergeMathComma(string): #判断非数学环境中的","是否需要合并在
             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]+1:]+rmatter
+            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
+    return string #返回处理后的字符串
 
 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)]