SUMMARY

this is a library which collect all kinds of useful source met in work. I am the coder for most of them, and some, very great code block, are coded by others-colleague, friends etc. which will be marked in the code.

CONTENT

util:

bit_array.h : bit map, a great number of count but only two value in one position

arrary_lock.h : amount of lock with bit flag

lifo.h && lifo.hpp : a stack for thread safety under read and write

autolock.h : a lock with RAII mechanism based on type

singleton.h : process and thread singleton, especially thread singleton can manage, generate, get in thread field gobally and release thread data automatically

auto_release.h : resouce release with RAII

file_watcher.h : check whether is file updated

hash_fun.h : fast hash function from facebook

lock_free_atomic_ops.h : automic operations

path.h : path join and split

timer.h : time monitor

monitor:

cpu_switch_monitor.h : monitor a cpu back time ine one thread with sleep, which can describe the cpu load to some degree

algorithm: computer algorithm test

container:

chain_hash.h : a chain hash

circle_cache.h : circle buffer cache

dict_container.h : two buffer dict container which can manage all kinds of dict

pchain_map.h : a thread-safety hashmap

pqueue.h : a light-weight thread-safety high performance queue based link

resource_stack.h : a thread-safety stack

sequence_recorder.h : record event and do sume simple summary

framework:

hining_pool.h : a input queue and output queu with threads to process

dict_manager.* : manager all kinds of dict

ml_util: machine learning util

algorighm:

tree.h:

迭代方式打印二叉树核心思想：
1）每一次循环是对二叉树根，左右三个节点处理切面
2）确认满足需求打印节点思路，并在三个节点处理上体现这个思路
3）思路包括：处理顺序，打印，入栈顺序 这几个核心处理要素
4）充分利用栈后进先出特性和遍历顺序特性，这两个顺序组成需求顺序