不过VB.NET确实有许多VB6没有的新功能，代码的自动排版就是一项，这也正是我们今天要实现的功能——VB代码格式化。
先看下实现的效果：
格式化前：
复制代码 代码如下:
For i = 0 To WebBrowser1.Document.All.length - 1
If WebBrowser1.Document.All(i).tagName = "HTML" Then
strContent = strContent & WebBrowser1.Document.All(i).innerHTML
Exit For
End If
Next
格式化后：
复制代码 代码如下:
For i = 0 To WebBrowser1.Document.All.length - 1
If WebBrowser1.Document.All(i).tagName = "HTML" Then
strContent = strContent & WebBrowser1.Document.All(i).innerHTML
Exit For
End If
Next
C++水平一直很烂，所以选择了C++作为练手语言写了这个简陋的VB代码格式化工具。代码不长，才200多行，标准C++实现。另外，为了彻底打消某些人继续使用VC6的念头，使用了auto关键字嘿嘿。好吧，废话少说，直接上代码：
复制代码 代码如下:
#include<iostream>
#include<string>
#include<vector>
#include<algorithm>
#include<fstream>
using namespace std;
//判断是否为空格
bool isSpace(const char chr)
{
return chr == ' ';
}
//去除左侧空白
void leftTrim(string &str)
{
string::iterator p=find_if(str.begin(),str.end(),not1(ptr_fun(isSpace)));
str.erase(str.begin(),p);
}
//去除右侧空白
void rightTrim(string& str)
{
string::reverse_iterator p=find_if(str.rbegin(),str.rend(),not1(ptr_fun(isSpace)));
str.erase(p.base(),str.end());
}
//去除两侧空白
string trim(const string& str)
{
string strRet(str);
leftTrim(strRet);
rightTrim(strRet);
return strRet;
}
//转换成小写
string toLower(const string& str)
{
string strRet(str);
transform(strRet.begin(),strRet.end(),strRet.begin(),(int (*)(int))tolower);
return strRet;
}
//判断是否以给定关键字开头
bool startWith(const vector<string>& vecCodeKeywords,const string& strCodeLine)
{
string line(toLower(strCodeLine));
for(auto keyword=vecCodeKeywords.begin(); keyword!=vecCodeKeywords.end(); keyword++)
if(line.find(*keyword + " ")==0 || line== *keyword)
return true;
return false;
}
//IF...Then...特殊检查
bool checkForIfThen(const string& strCodeLine)
{
vector<string> vecIf;
vecIf.push_back("if");
if(!startWith(vecIf,strCodeLine))
return false;
if(toLower(strCodeLine).find("then")==string::npos)
return false;
string line(trim(toLower(strCodeLine)));
if(line.length()<7)
return false;
return !(line.substr(line.length()-4,4) == "then");
}
//格式化给定行并标记相关信息
int formatAndMarkLine(string& strCodeLine)
{
//起始关键字 "if","for","[Private | Friend | Public] [Static] [Sub | Function | Property | Type]","with","do","select"
vector<string> vecStartKeywords;
vecStartKeywords.push_back("if");
vecStartKeywords.push_back("for");
vecStartKeywords.push_back("with");
vecStartKeywords.push_back("do");
vecStartKeywords.push_back("select");
string _pfp[] = {"private","friend","public"}; //可空
string _s[] = {"static"}; //可空
string _sfpt[] = {"sub","function","property","type"};
//_pfp _s 都为空
for(auto i=0; i<4; i++)
vecStartKeywords.push_back(_sfpt[i]);
//_pfp 为空
for(auto i=0; i<4; i++)
vecStartKeywords.push_back(_s[0] + " " + _sfpt[i]);
//_s 为空
for(auto i=0; i<4; i++)
for(auto j=0; j<3; j++)
vecStartKeywords.push_back(_pfp[j] + " " + _sfpt[i]);
//_pfp _s 都不为空
for(auto i=0; i<4; i++)
for(auto j=0; j<3; j++)
vecStartKeywords.push_back(_pfp[j] + " " + _s[0] + " " + _sfpt[i]);
//终止关键字 "end if","next","End [Sub | Function | Property | Type]","end with","loop","end select"
vector<string> vecEndKeywords;
vecEndKeywords.push_back("end if");
vecEndKeywords.push_back("next");
vecEndKeywords.push_back("end with");
vecEndKeywords.push_back("loop");
vecEndKeywords.push_back("end select");
for(auto i=0; i<4; i++)
vecEndKeywords.push_back("end " + _sfpt[i]);
//中间关键字 "else","elseif","case"
vector<string> vecMiddleKeywords;
vecMiddleKeywords.push_back("else");
vecMiddleKeywords.push_back("elseif");
vecMiddleKeywords.push_back("case");
auto mark = 0;
char c;
auto n=0;
string line;
auto quote = false; //双引号状态
/*
规则：
双引号内单引号不考虑，否则后面内容成为注释
*/
auto space = true; //空白符状态 false 表示未遇到任何空白符
while((c=strCodeLine[n++]))
{
switch(c)
{
case ' ':
case 't':
if(quote)
{
line += c;
}
else
{
if(!space)
{
line += c;
space = true;
}
}
break;
case '"':
space = false;
quote = !quote;
line += c;
break;
case ''':
space = false;
if(quote)
line += c;
else
{
line += " '"; //MsgBox("itianda") '单引号前有一个空格
while((c=strCodeLine[n++])) //直接附加单引号后面内容
line += c;
continue;
}
break;
case '_': //续行符
space = false;
line += c;
if(!quote && n==(int)strCodeLine.length() && n-2>=0 && strCodeLine[n-2]==' ')
mark |= 0x80; //10000000
break;
default:
space = false;
line += c;
}
}
strCodeLine = line;
if(startWith(vecStartKeywords,line) && !checkForIfThen(line))
mark += 1;
if(startWith(vecEndKeywords,line))
mark += 2;
if(startWith(vecMiddleKeywords,line))
mark += 3;
return mark;
}
//将代码按行分割
void splitToLines(const string& strCode, vector<string>& vecCodeLines)
{
vecCodeLines.clear();
char c;
auto n=0;
string line;
while((c=strCode[n++]))
{
if(c!='n')
line += c;
else
{
vecCodeLines.push_back(line);
line.clear();
}
}
if(line.length()) //最后一行为空则舍去
vecCodeLines.push_back(line);
}
//格式化给定代码
void formatCode(string& strCode,const string& strIndentString)
{
vector<string> vecLines; //所有代码行
splitToLines(strCode,vecLines); //获取所有代码行
if(vecLines.size()==0)
{
return;
}
auto indentLevel = 0; //缩进级别
auto incompleteLine = false; //是否是未结束行
for(auto line=vecLines.begin(); line!=vecLines.end(); line++)
{
auto indent = indentLevel;
auto mask = formatAndMarkLine(*line);
switch(mask & ~0x80)
{
case 0:
break;
case 1:
indentLevel++;
break;
case 2:
indent--;
indentLevel--;
break;
case 3:
indent--;
break;
}
if(incompleteLine)
indent++;
incompleteLine = mask & 0x80;
if(indent<0)
indent = 0;
if(indentLevel<0)
indentLevel = 0;
string strIndent;
for(auto i=0; i<indent; i++)
strIndent += strIndentString;
*line = strIndent + *line;
}
strCode.clear();
for(auto line=vecLines.begin(); line!=vecLines.end(); line++)
strCode+= trim(*line).length() ? "n" + *line : "";
}
int main()
{
string indentString = " ";
string code;
ifstream inputFile("in.txt");
string line;
while(getline(inputFile,line))
{
code += line + "n";
}
formatCode(code,indentString);
ofstream outputFile("out.txt");
outputFile<<"Your beautiful code:"<<endl<<
"-------------------------------------------------------------------"
<<endl<<code<<endl<<endl<<
"-------------------------------------------------------------------"
<<endl<<
" Formatted by itianda's PUVBFormatter"
<<endl<<
" http://www.programup.com/blog"
<<endl;
return 0;
}
看过代码应该知道这是多么基本的实现了吧，好多细节都没有去仔细处理，比如没有考虑冒号连接多行的情况，所以如果你希望使用此工具，请不要把多行语句写到一行哦！
最后提供一个我编译好的EXE下载：PUVBFormatter
更新：
增加select case…end select关键字，感谢jjww2999网友的反馈。
本文来自: itianda's blog