h**********c
发帖数: 4120 | 1 I have no chinese input. The online one is for tranditional.
The question, e.g. Java, arrayIndexOutofbound or nullpointer exceptions,
how are they handled?
The developers deal with them, or the QA guys,or both. | g*****g
发帖数: 34805 | 2 Test driven approach should help. High unit test coverage reveals
most of such problems before QA find them. And devs still need to
fix them no matter where they are being found.
【在 h**********c 的大作中提到】
: I have no chinese input. The online one is for tranditional.
: The question, e.g. Java, arrayIndexOutofbound or nullpointer exceptions,
: how are they handled?
: The developers deal with them, or the QA guys,or both.
| h**********c
发帖数: 4120 | 3 I once tried to do something with JUnit.Still not that auto-mated.
Is there something can do bulk tests of many classes?
Some tools to debug multithreading ?
【在 g*****g 的大作中提到】
: Test driven approach should help. High unit test coverage reveals
: most of such problems before QA find them. And devs still need to
: fix them no matter where they are being found.
| g*****g
发帖数: 34805 | 4 You can run all JUnit testcases as part of build. You can use
Maven to manage your build and Jerkins to do continous integration.
It's certainly automatic. For multithreading debuging, simple logging
and mock classes may be more effective IMHO.
【在 h**********c 的大作中提到】
: I once tried to do something with JUnit.Still not that auto-mated.
: Is there something can do bulk tests of many classes?
: Some tools to debug multithreading ?
| h**********c
发帖数: 4120 | 5 These things are useful quite placebo.
Fundamental I believe is still design.
Let's talk about pass by ref in Java. You encapsulate something in a class.
This class should take care of this member variable. Ok, they just pass by
ref, get the ref, modify change them even new a new one in another class and
set it back.
Cross call function. lets say a.fun1() call b.fun2(), .... finally z.fun26()
call a.fun1(). They dont' even know what is called dependency graph.In fact
, they know, they just don't bother one second to think about it.In this
aspect, I don't think even unit test will help.
Another horrible thing is handle. handle may cause the cyclic call just
mentioned.
【在 g*****g 的大作中提到】
: You can run all JUnit testcases as part of build. You can use
: Maven to manage your build and Jerkins to do continous integration.
: It's certainly automatic. For multithreading debuging, simple logging
: and mock classes may be more effective IMHO.
| g*****g
发帖数: 34805 | 6 Unit test is not necessarily within a class. You can pick granuity wherever
you like. And it doesn't matter if a function ends up calling itself. It's
like a blackbox, you mock the interface,you set some input, you expect some
output. You can even use JMock to verify every exact step if you like.
.
and
()
fact
【在 h**********c 的大作中提到】
: These things are useful quite placebo.
: Fundamental I believe is still design.
: Let's talk about pass by ref in Java. You encapsulate something in a class.
: This class should take care of this member variable. Ok, they just pass by
: ref, get the ref, modify change them even new a new one in another class and
: set it back.
: Cross call function. lets say a.fun1() call b.fun2(), .... finally z.fun26()
: call a.fun1(). They dont' even know what is called dependency graph.In fact
: , they know, they just don't bother one second to think about it.In this
: aspect, I don't think even unit test will help.
| h**********c
发帖数: 4120 | 7 A funct ca'aslf ain't good.
May infinite'lpoo, de facto, bad practice in sw eng.
Do you think MySQL is industrially robust?
I would prefer to read't guando open la'ource.
wherever
some
【在 g*****g 的大作中提到】
: Unit test is not necessarily within a class. You can pick granuity wherever
: you like. And it doesn't matter if a function ends up calling itself. It's
: like a blackbox, you mock the interface,you set some input, you expect some
: output. You can even use JMock to verify every exact step if you like.
:
: .
: and
: ()
: fact
| g*****g
发帖数: 34805 | 8 I don't care if a function calls itself. That's not what unit test
is about. And that's not where bugs are coming from.
【在 h**********c 的大作中提到】
: A funct ca'aslf ain't good.
: May infinite'lpoo, de facto, bad practice in sw eng.
: Do you think MySQL is industrially robust?
: I would prefer to read't guando open la'ource.
:
: wherever
: some
| h**********c
发帖数: 4120 | 9 That is what I ask?
How to detect stack overflow, index out of bound, null pointer exception etc?
The old c/cpp problem still viable.
【在 g*****g 的大作中提到】
: I don't care if a function calls itself. That's not what unit test
: is about. And that's not where bugs are coming from.
| t****t
发帖数: 6806 | 10 我很好奇你是哪里人, 为什么我看你写的中文英文都很吃力?
【在 h**********c 的大作中提到】
: A funct ca'aslf ain't good.
: May infinite'lpoo, de facto, bad practice in sw eng.
: Do you think MySQL is industrially robust?
: I would prefer to read't guando open la'ource.
:
: wherever
: some
| | | h**********c
发帖数: 4120 | 11 Sorry for your in-convenience.
Your question is difficult to answer. It is totally non-technical. OK?
【在 t****t 的大作中提到】
: 我很好奇你是哪里人, 为什么我看你写的中文英文都很吃力?
| t****t
发帖数: 6806 | 12 不方便回答当然就算了, 我说了我是好奇. 就算techinical的问题, 也不一定要回答的.
【在 h**********c 的大作中提到】
: Sorry for your in-convenience.
: Your question is difficult to answer. It is totally non-technical. OK?
| C***y
发帖数: 2546 | 13 一般人都写da niu 或者big cow
big niu的写法比较独特
的.
【在 t****t 的大作中提到】
: 不方便回答当然就算了, 我说了我是好奇. 就算techinical的问题, 也不一定要回答的.
| t****t
发帖数: 6806 | 14 倒不是这个, 他最后那句read't 啥啥的我愣是没看懂, 照理我这么老的网虫不该有这
种事, 我想难道他是墨墨? 同作者了一下, 发现第二篇我就看得不是很明白, 而且以前
就没看明白, 比如说"我也比较直率一点表白，应该从哪里gogo." 这个我倒是能猜到意
思, 但是这语法很怪啊, 好象是老外说的中文, 或者是机器翻译的结果...
http://www.mitbbs.com/article_t/Programming/31203303.html
【在 C***y 的大作中提到】
: 一般人都写da niu 或者big cow
: big niu的写法比较独特
:
: 的.
| b***i
发帖数: 3043 | 15 I don't think he is chinese.
he just google translated to traditional chinese and pasted the question
here.
【在 t****t 的大作中提到】
: 倒不是这个, 他最后那句read't 啥啥的我愣是没看懂, 照理我这么老的网虫不该有这
: 种事, 我想难道他是墨墨? 同作者了一下, 发现第二篇我就看得不是很明白, 而且以前
: 就没看明白, 比如说"我也比较直率一点表白，应该从哪里gogo." 这个我倒是能猜到意
: 思, 但是这语法很怪啊, 好象是老外说的中文, 或者是机器翻译的结果...
: http://www.mitbbs.com/article_t/Programming/31203303.html
|
|
