Channy's blog

//Description: In a new project, teammates offer a base version of the fundamental classes. Heres are just some notes after reading theirs code.

//Create Date: 2019-07-26 09:24:57

//Author: channy

2019-07-26-Notes_Learning_Others_Code

Summaries generally use code blocks to macro definition #define

For example, most of my functions use try-catch block, then it can be written as

#define TRY_CATCH_BLOCK(func, msg) \
	try { \
		func;
	} \
	catch(std::exception &e) { \
		writeLog(#msg" catch an exception %s.", e.what()); \
	} \
	catch(...) { \
		writeLog(#msg" catch an unknow exception.");
	}

Then I can use it freely and simply…

把常用的代码段总结成宏定义

中文翻译好烦。。。请求外援。。。

If different inputs needs to run different processes, then can define them in config files (ini,xml,etc)

For example, if input is x, need to run A -> B -> C;

while if input is y, need to run A -> C -> B;

then I can define like this in config files:

"input":{
	list[{
	"type":"x",
	"process":["A","B","C"]
	},
	{
	"type":"y",
	"process":["A","C","B"]
	}]
}

Once demand is changed, I can only change config files but don’t need to change my code and release a new version. Of cause the config files can be encrypted to make our project more safe.

Class definition and declaration can also use macro definication

没想好怎么总结，想好再补。。。

Same as above, when loading others dynamic library, some common code blocks can also be written as macro definition

#define LOAD_DLLFUN(FNAME, FNAME_STR) FNAME=(TYPE_##FNAME) qLib.resolve(FNAME_STR); \
        if (!FNAME){writeLog(QString().sprintf("load %s failed", FNAME_STR));return UNKNOW_ERROR_CODE;}

back