Lua语言的屏蔽词过滤

参考了DFA算法的简单说明与案例实现

构造过滤器

local FilterData = {}local FilterTree = {}--构建时候的临时变量local IsEnd = "IsEnd"function FilterData.Load()--构建过滤器for _, v in ipairs(exd["GT_".."FilterText"]) do --这个循环是已经读取的屏蔽字表local FilterStr = v.MaskWordlocal CurMap = FilterTree--这个函数是循环遍历字符串ForEachString(FilterStr,function(Str)if CurMap[Str] == nil thenCurMap[Str] = {}CurMap[Str][IsEnd] = falseCurMap =  CurMap[Str]elseCurMap =  CurMap[Str]endend,true)CurMap[IsEnd] = trueend
end

查找

--基础的屏蔽词表匹配过滤
function FilterData.MaskWordFilter(Str)local BeginIndex = 1local MatchingTable = {}while BeginIndex < #Str dolocal NowMap = FilterTree--local MatchingNum = 0local CurStr = string.sub(Str, BeginIndex, #Str)local CurIndex = BeginIndexlocal LeftMatchIndex = 0local RightMatchIndex = 0local i = 1local AllowSpecialSymbols = falsewhile true dolocal char = string.sub(CurStr,i,i)local Ascll = string.byte(char)if Ascll > 128 thenchar = string.sub(CurStr,i,i+2)i = i + 3if NowMap[char] ~=nil thenif LeftMatchIndex == 0 thenLeftMatchIndex = CurIndexendif NowMap[char][IsEnd] == true thenRightMatchIndex = CurIndex+2BeginIndex = CurIndex+3endNowMap = NowMap[char]else--判断是否已经有一个字匹配if LeftMatchIndex ~= 0 and RightMatchIndex ~= 0 thenMatchingTable[LeftMatchIndex] = RightMatchIndexbreakelseif LeftMatchIndex ~= 0 thenBeginIndex =  BeginIndex+3breakelseBeginIndex =  BeginIndex+3endendCurIndex = CurIndex +3elsei = i + 1if 32 < Ascll thenif NowMap[char] ~=nil thenif LeftMatchIndex == 0 thenLeftMatchIndex = CurIndexendif NowMap[char][IsEnd] == true thenRightMatchIndex = CurIndexBeginIndex = CurIndex+1endNowMap = NowMap[char]elseif LeftMatchIndex ~= 0 and RightMatchIndex ~= 0 thenMatchingTable[LeftMatchIndex] = RightMatchIndexbreakelseif LeftMatchIndex ~= 0 thenBeginIndex =  BeginIndex+1breakelseBeginIndex =  BeginIndex+1endendelseif AllowSpecialSymbols thenendCurIndex = CurIndex +1endif i > #CurStr thenif LeftMatchIndex ~= 0 and RightMatchIndex ~= 0 thenMatchingTable[LeftMatchIndex] = RightMatchIndexelseBeginIndex =  BeginIndex + #CurStrendbreakendendendlocal res = trueif next(MatchingTable) ~= nil thenres = falseendreturn res, MatchingTable
end--调用前面的，把字符串先提取成中文和英文再分别过滤，替换成**
function FilterData.FineMaskWordFilter(Str)local Chi = {}local Eng = {}local k = 1while k < #Str dolocal c = string.byte(Str,k)if not c then break end--英文字母 (c>=48 and c<=57) orif  (c>= 65 and c<=90) or (c>=97 and c<=122) thenk = k + 1Eng[#Eng+1] = string.char(c)elseif c>=228 and c<=233 then--中文local c1 = string.byte(Str,k+1)local c2 = string.byte(Str,k+2)if c1 and c2 thenlocal a1,a2,a3,a4 = 128,191,128,191if c == 228 then a1 = 184elseif c == 233 then a2,a4 = 190,c1 ~= 190 and 191 or 165endif c1>=a1 and c1<=a2 and c2>=a3 and c2<=a4 thenk = k + 3Chi[#Chi+1] = string.char(c,c1,c2)endendelse--都不符合时候不加入表k = k + 1endendlocal ChiStr = table.concat(Chi)local EngStr = table.concat(Eng)local _,r =FilterData.MaskWordFilter(ChiStr)for key,v in pairs(r) dolocal tmp =string.sub(ChiStr,key,v)ForEachString(tmp, function(SubStr)  Str = string.gsub(Str,SubStr,"*") end)endlocal _,q = FilterData.MaskWordFilter(EngStr)for key,v in pairs(q) dolocal tmp =string.sub(EngStr,key,v)ForEachString(tmp, function(SubStr)  Str = string.gsub(Str,SubStr,"*") end)endreturn Str
end

判断玩家名字是否只含有中英文数字和常用标点

function FilterData.PlayerNameFilter(Str)local ss = {}local k = 1while k < #Str dolocal c = string.byte(Str,k)if not c then break end--英文字母符号if (c>=48 and c<=57) or (c>= 65 and c<=90) or (c>=97 and c<=122) thenk = k + 1ss[#ss+1] = string.char(c)elseif c>=228 and c<=233 then--中文local c1 = string.byte(Str,k+1)local c2 = string.byte(Str,k+2)if c1 and c2 thenlocal a1,a2,a3,a4 = 128,191,128,191if c == 228 then a1 = 184elseif c == 233 then a2,a4 = 190,c1 ~= 190 and 191 or 165endif c1>=a1 and c1<=a2 and c2>=a3 and c2<=a4 thenk = k + 3ss[#ss+1] = string.char(c,c1,c2)endendelse--都不符合时候不加入表k = k + 1endendif #ss > 0 thenlocal res = table.concat(ss)local len = #resif #Str == len thenreturn true,reselsereturn false,resendelsereturn false , ""end
end

字符串遍历的代码

--[[字符串遍历
@Str: 被遍历的字符串
@Function：每个char需要调用的函数
@bAllowSpecialSymbols：是否允许空格遍历
@...：Function的其他参数
--]]
function ForEachString(Str , Function ,bAllowSpecialSymbols,...)if Str==nil or Function==nil ortype(Str)~= "string" or  type(Function) ~= "function"thenreturnendlocal i = 1local AllowSpecialSymbolsif bAllowSpecialSymbols then AllowSpecialSymbols = bAllowSpecialSymbolselseAllowSpecialSymbols = falseendwhile true dolocal char = string.sub(Str,i,i)local Ascll = string.byte(char)if Ascll > 128 thenchar = string.sub(Str,i,i+2)i = i + 3Function(char,...)elseif 32 < Ascll thenFunction(char,...)elseif AllowSpecialSymbols thenFunction(char,...)endi = i + 1endif i > #Str thenbreakendend
end

Lua语言的屏蔽词过滤相关推荐

游戏敏感词屏蔽词过滤
游戏和网站,都需要敏感词(屏蔽词)过滤,你懂的.而且敏感词库不断的增长,从十几年前的3000多词,已经增长到14000以上(各渠道获取的词库略有不同). 这1万多词的扫描处理,开销的性能就有点客观了. ...
python敏感词过滤replace_Serverless 实战：3 分钟实现文本敏感词过滤
敏感词过滤是随着互联网社区一起发展起来的一种阻止网络犯罪和网络暴力的技术手段,通过对可能存在犯罪或网络暴力的关键词进行有针对性的筛查和屏蔽,能够防患于未然,将后果严重的犯罪行为扼杀于萌芽之中. 随着各 ...
38个敏感词_Serverless 实战：3 分钟实现文本敏感词过滤
敏感词过滤是随着互联网社区一起发展起来的一种阻止网络犯罪和网络暴力的技术手段,通过对可能存在犯罪或网络暴力的关键词进行有针对性的筛查和屏蔽,能够防患于未然,将后果严重的犯罪行为扼杀于萌芽之中. 随着各 ...
dfa算法c语言,DFA跟trie字典树实现敏感词过滤（python和c语言）
DFA和trie字典树实现敏感词过滤(python和c语言) 现在做的项目都是用python开发,需要用做关键词检查,过滤关键词,之前用c语言做过这样的事情,用字典树,蛮高效的,内存小,检查快. 到了 ...
Android前端敏感词（屏蔽词）过滤
转载请标明出处:http://blog.csdn.net/donkor_/article/details/81232529 前言: 开发中经常要处理用户一些文字的提交,所以涉及到了敏感词过滤的功能,参 ...
java敏感词屏蔽器,“敏感词过滤”功能设置
"敏感词过滤"功能设置 "敏感词过滤"是指对互联网发布的言论和文章中含有的敏感词进行过滤.敏感词经过在互联网的传播和扩散会影响社会的稳定和用户的使用.云锁的&q ...
敏感词过滤算法为内容保驾护航 Java/.Net/C++/c/Python等语言是如何进行敏感词打码限制的高效防范违规内容
有人的地方,就有江湖,有输入框的地方,就有注入风险!有输入框的地方,就有敏感词!敏感词就像一个平台杀手,可能直接导致平台被封锁! 敏感词是一个APP.一个网站.一个内容平台的"杀手" ...
网站是怎么屏蔽脏话的呢：简单学会SpringBoot项目敏感词、违规词过滤方案
一个社区最重要的就是交流氛围与审查违规,而这两者都少不了对于敏感词进行过滤的自动维护措施.基于这样的措施,我们才能基本保证用户在使用社区的过程中,不至于被敏感违规词汇包围,才能够正常的进行发布帖子和评 ...
Java实现敏感词过滤 - IKAnalyzer中文分词工具
IKAnalyzer 是一个开源的,基于java语言开发的轻量级的中文分词工具包. 官网: https://code.google.com/archive/p/ik-analyzer/ 本用例借助 I ...

Lua语言的屏蔽词过滤

构造过滤器

查找

判断玩家名字是否只含有中英文数字和常用标点

字符串遍历的代码

Lua语言的屏蔽词过滤相关推荐

最新文章

热门文章