doudiandi6967 2012-05-25 18:30
浏览 33
已采纳

字典搜索词 - PHP和MySQL中的最佳性能

I would like to implement a dictionary for searching words and their inflection. Dictionary size is about 60MB and it is stored in txt file.

Currently I'm passing a whole dictionary (60MB) to an variable and then searching for interesting words. I can't resist an impression that is not properer way (to allocate 60MB in memory just for dictionary)

Can anyone recommend other solution? Maybe there are some other ways to search for interesting phrases within a file without storing whole content of it into memory? or maybe putting whole dictionary into MySQL db would help (I could slice that dictionary and put every word into db with additional id to recognize inflection between different word that are coming form the same root)

  • 写回答

2条回答 默认 最新

  • duankeng2026 2012-05-25 18:41
    关注

    Of course putting dictionary inside a DB would help. Database works MUCH faster due to internal indexing and residing in system memory as a daemon; Searching inflections would also be simpler, faster and more efficient with queries:

    SELECT * FROM dictionary WHERE word LIKE "devel%"

    would get you "developing", "developer", "develop" and so on. Also there's a lot of other possibilities. The only drawback is that you must have database server installed. On the other hand, loading an entire txt file in memory would consume more then 1GB of memory if 20 users connect at the same time, since script is started anew each time.

    本回答被题主选为最佳回答 , 对您是否有帮助呢?
    评论
查看更多回答(1条)

报告相同问题?

悬赏问题

  • ¥15 ogg dd trandata 报错
  • ¥15 高缺失率数据如何选择填充方式
  • ¥50 potsgresql15备份问题
  • ¥15 Mac系统vs code使用phpstudy如何配置debug来调试php
  • ¥15 目前主流的音乐软件,像网易云音乐,QQ音乐他们的前端和后台部分是用的什么技术实现的?求解!
  • ¥60 pb数据库修改与连接
  • ¥15 spss统计中二分类变量和有序变量的相关性分析可以用kendall相关分析吗?
  • ¥15 拟通过pc下指令到安卓系统,如果追求响应速度,尽可能无延迟,是不是用安卓模拟器会优于实体的安卓手机?如果是,可以快多少毫秒?
  • ¥20 神经网络Sequential name=sequential, built=False
  • ¥16 Qphython 用xlrd读取excel报错