的结尾

.
And don't be surprised to find out the divs are not matching exactly.
If you have code like
while(counter > 0) {
....
}
You may loop forever in production, just keep those in mind.

【在 g**********y 的大作中提到】
: 那些library太heavy了，对于简单问题，我现在越来越不想用别人的library, 冷不丁
: 地升级一下，有可能引进一些莫名其妙的bug.
: 我倒是该去读读他们的source code, 他们的解法可能有启发。

under some
condition. That's how the problem comes.

【在 g*****g 的大作中提到】
: If you are sure your html string is well formed, the way you propose
: (search by regex and use a counter) is pretty efficient.
: It's when you try to ignore/not ignore certain errors you'll find out
: 3rd party library is very helpful.
: Html is not well formed, don't be surprised to see something like

.
: And don't be surprised to find out the divs are not matching exactly.
: If you have code like
: while(counter > 0) {
: ....
: }

id是1的div先进站，id 2 div进站，div/进站，发现栈底的div成对，于是取出。
/div id 2进站，和这个时候的栈底div id 2成对， 于是取出。 最后栈里还剩一个，
于是不成对。

【在 g**********y 的大作中提到】
: 给一个HTML String, 怎么找

的结尾

?
: 直觉办法是：counter=1, search ", counter++, 如果是: div>, counter--, counter归零时就是结尾。
: 有更简单有效的办法吗？或者说用Java自己的library, 有更简洁的coding吗？