Lua 字符串实践 -- 过滤词库算法

note目录

• 过滤词字典结构
• 构造过滤词树结构
• 测试example

---

1: 过滤词字典结构

WordMap.lua文件

WordMap = luaclass("WordMap")

function WordMap:Init()
    self.isLast = false
    self.map = {}
    return self
end

return WordMap

过滤词的单个结构包括2个元素：
1：isLast是否是为节点
2：子节点-是一张表table{}，用单个字符做key存放WordMap的表

---

2: 构造过滤词树结构

假设过滤词库的里面有如下过滤词:

local configFilter = {}
configFilter[1] = 
{
   filterWord = "李泽东,李泽西,李泽北,李泽南,李克勤,周树人,周就,周佳佳"
}

构造的过滤词库的树结构：

完整的过滤词库代码：
WordFilterManager.lua

WordFilterManager = {}
local WordMap = require("WordMap")

--根据过滤词库来初始化过滤词库群树结构
function WordFilterManager.InitWordFilter()
    -- 初始化一张过滤词树空结构
    WordFilterManager.wordMap = WordMap():Init()
    local str = configFilter[1].filterWord
    local strArr = StringUtil.Split(str,",")
    for i = 1,#strArr do
        local wordItem = strArr[i]
        if wordItem ~= "" then
            WordFilterManager.AddWordsChar(WordFilterManager.wordMap , wordItem , 1)
        end
    end    
end

--添加单个字符到过滤词树的结构中
function WordFilterManager.AddWordsChar(wordMap , word , charIdIndex)
    local map = wordMap.map
    local singleChar = StringUtil.utf8sub(word,charIdIndex , 1)
    if map[singleChar] == nil then
        map[singleChar] = WordMap():Init()
    end

    local subWordMap = map[singleChar]
    --单个词条是最后一个字符，则是叶子节点
    if charIdIndex == StringUtil.utf8len(word) then
        subWordMap.isLast = true
    end

    if charIdIndex < StringUtil.utf8len(word) then
        WordFilterManager.AddWordsChar(subWordMap,word,charIdIndex + 1)
    end

end

--判断一个字符串是否有过滤词存在，并同时返回过滤词的字符个数
function WordFilterManager.CheckWord(wordMap , word , beginIndex , filterCounts)
    local map = wordMap.map
    local singleChar = StringUtil.utf8sub(word,beginIndex,1)
    if map[singleChar] ~= nil then 
       filterCount = filterCount + 1
       if map[singleChar].isLast then
           return true,filterCount
       else
           if beginIndex > StringUtil.utf8len(word) then
               return false , 0 
           else
               WordFilterManager.CheckWord(map[singleChar],word,beginIndex + 1,filterCount)
           end
        end
    end
    rturn false , 0
end

-- 判断一个字符串是否有过滤词存在
function WordFilterManager.IsContains(str)
    local len = StringUtile.utf8len(str)
    for i = 1,len do
        if WordFilterManager.CheckWord(WordFilterManager.wordMap,str,i,0) then
            rturn true
        end    
    end
    rturn false
end

--将给过来的字符串中的所有的过滤词改为"***"
function WordFilterManager.Filter(str)
    local len = StringUtil.utf8len(str)
    for i = 1,len then
        local isFilter,filterCount =                         WordFilterManager.CheckWord(WordFilterManager.wordMap,str,i,0)
        if isFilter then
            local filterWord = ""
            for i = 1,filterCount do
                filterWord = filterWord .. "*"
            end
            str = string.sub(str,StringUtil.utf8sub(str,i,filterCount),filterWord)
            i = i + filterCount - 1
    end    
    return str
end

---

3: 测试example

local str = "aaa我是李泽西啊aaa"
local isContainsFilterWord = WordFilterManager.IsContains(str)
local beforeFilterWord = WordFilterManager.Filter(str)

print(tostring(isContainsFilterWord))        ---> output: true
print(beforeFilterWord)                      ---> output:aaa我是***啊aaa