Unserialize This issue was found by a emlog forum user in the use of the Times wrong
The performance of the problem is as follows:
The Emlog cache is saved by serializing the data Object (array) of PHP (serialize) and storing it as a file, and then deserializing (unserialize) the cache string to read the data when the cache is read.
The cached serialized data that I fetch from the user is:
A:1:{s:8: "Kl_album"; A:4:{s:5: "title"; S:12: "Wonderful Moment"; s:3: "url"; s:41: "Http://www.kaisay.cn/?plugin=kl_album"; S:8: " Is_blank "; s:7:" _parent "; s:4:" Hide "; s:1:" N ";}}
I look at the people who understand serialization will say, this data is very normal Ah, no problem. But the string is unserialize directly, the value returned is a false;
Code
Var_dump (Unserialize (' A:1:{s:8: "Kl_album"; A:4:{s:5: "title"; S:12: "Wonderful Moment"; s:3: "url"; s:41: "http://www.kaisay.cn/? Plugin=kl_album "; s:8:" Is_blank "; s:7:" _parent "; s:4:" Hide "; s:1:" N ";}} ');
Run results
What is the problem? The answer is s:41: "Http://www.kaisay.cn/?plugin=kl_album"
The serialized string http://www.kaisay.cn/?plugin=kl_album the length of the string is 41, but we have only 37 characters to count. This is the problem that causes PHP to deserialize the string invalidation.
If you change the length of the string to 37, the program will be successfully deserialized
Code:
Var_dump (Unserialize (' A:1:{s:8: "Kl_album"; A:4:{s:5: "title"; S:12: "Wonderful Moment"; s:3: "url"; s:37: "http://www.kaisay.cn/? Plugin=kl_album "; s:8:" Is_blank "; s:7:" _parent "; s:4:" Hide "; s:1:" N ";}} ');
After Google only found that the problem abroad has been a lot of netizens encountered, in the official manual unserialize function page comments There are many netizens in the discussion and study the solution of the problem.
There are many possible reasons for this, and the biggest possibility is that the encoding and deserialization at the time of serializing the data is not the same as causing the length of the string to deviate. For example, database encoding Latin1 and UTF-8 character lengths are not the same.
Summary: Solution:
Situation One: UTF-8
function mb_unserialize ($serial _str) { $serial _str= preg_replace ('!s: (\d+): "(. *?)";! Se ', ' s: '. strlen (' $ "). ': \ ' $2\ "; '", $serial _str); $serial _str= str_replace ("\ R", "", $serial _str); Return Unserialize ($serial _str);}
Scenario Two: ASC
function asc_unserialize ($serial _str) { $serial _str = preg_replace ('!s: (\d+): "(. *?)";! Se ', ' "s:". strlen ("$"). ": \" $2\ ";" ', $serial _str); $serial _str= str_replace ("\ R", "", $serial _str); Return Unserialize ($serial _str);}
PHP function Unserialize Data return false problem analysis