|
dojohansen wrote: I was debating whether or not *generally* speaking it is smart to develop unit tests BEFORE or AFTER a refactoring of code. Unfortunately not one word you input related to this subject
Perhaps it's because you replied to my post that said:
led mike wrote: Yes. It's hard to imagine someone is thinking about design and architecture and refactoring but they haven't yet thought about the fact they should have unit tests. As hard as it might be to imagine that someone, no doubt they exist.
What I was discussing in that post as well as my last one to you was the importance of Unit Tests during INITIAL DEVELOPMENT. The original post used terms like "old architecture" which I took to mean in production.
dojohansen wrote: It is my OPINION that it is not ALWAYS or AUTOMATICALLY best to develop unit tests PRIOR TO REFACTORING CODE
It's my opinion that it's insane to be in production without unit tests. And no this is not some new belief based on TDD, I learned to use Unit Tests long before I ever heard the term Test Driven Development and long before any frameworks existed, AFAIK. Therefore I was never discussing the merits of developing unit tests based on a need to refactor in post production since the Unit Tests would already exist prior to production. My comments were only to the point of developing unit tests as a standard practice of software development, period.
So you can get as upset as you want with what I said, it doesn't change any of that.
led mike
|
|
|
|
|
Somebody's been really hammering you with 1 votes on this board Mike. I've corrected the balance on some of them, and will continue to do so as time allows.
|
|
|
|
|
Pete O'Hanlon wrote: Somebody'
 Yeah, probably the same guy that was giving you the 2's. He explained that was an accident though. Apparently he didn't like my response to his argumentative reply in this thread about the importance of Unit Tests. Like I give a crap what he thinks.
led mike
|
|
|
|
|
Ahhh. Yup - I was reading this thread through with interest, especially because it highlights the differences in opinion in whether or not UT is important. It's a shame that people miss the point that good up front UT is a great way to force yourself to think more about the architecture of the system, and to look at putting good practices into place.
|
|
|
|
|
Pete O'Hanlon wrote: It's a shame that people miss the point
Sure, it's one in long list of points people miss in this field regarding good practices. witnessed here every single day. "here" having multiple usages:
1) Code Project
2) My peers and *gulp* managers or where I work
3) Me myself and I
Well maybe I don't miss them as much as I don't know them. I remember when I started out in University we worked with Unix and DOS. This was pre Windows and PCs were starting to move out so DOS became the hot bed. My internship, which became a job (long story or not), was DOS POS, yes Point of Sale and Piece of Shite. Anyway back then my vision of the future was I would be one of these guys that new almost everything about PC development. Then came (for me) OO, Windows and the Internet, etc., and I soon realized I would ever only scratch the surface or not, depends on your point of view. Most days it feels like not.
Anyway people that don't have a clue and don't even try, those that ignore people like Cunningham, Fowler, Beck, Booch and well there are just so many, as though they think they know better than they do I find annoying, at best, at worst they seem disgraceful. For my part I just hope I can even understand their points, the idea that I could judge them is not something I would even consider.
led mike
|
|
|
|
|
led mike wrote: I soon realized I would ever only scratch the surface
Yup - the more I learn, the more it becomes apparent that there's more to learn. For me, the forums are a great place to learn from people I respect. Even if I don't often ask questions on the forums, it's amazing how much I've picked up from the better answers.
|
|
|
|
|
Pete O'Hanlon wrote: Even if I don't often ask questions on the forums, it's amazing how much I've picked up from the better answers.
Exactly, that's the reason I am in them really. Replying is only a secondary consideration.
led mike
|
|
|
|
|
Well it wasn't YOUR thread. Just because you go off on a tangent doesn't mean everyone has to. I tried to write a post relevant to the thread, which was about existing code and refactoring. While it may be "insane" to get into production without unit tests, nevermind that 99.9% of all systems ever developed in the history of the universe has done exactly this, this is irrelevant.
You don't get any points for deliberately misunderstanding.
|
|
|
|
|
dojohansen wrote: Well it wasn't YOUR thread. Just because you go off on a tangent doesn't mean everyone has to.
YOu replied to me you f***ing moron. If you didn't want to talk to me or my point then you could have replied to someone else. What f***ing idiot.
dojohansen wrote: You don't get any points for deliberately misunderstanding.
I have not deliberately misunderstood anything. Sure there is a lot of legacy code that has no unit tests, should you create them? There is no cook book answer to that. Each project is different and therefore must be considered based on it's own unique criteria, period.
That said, what's legacy code? In this context, what it means to me is anything I was not an original developer of because I had no control over the creation of unit tests or the design of it, therefore to me it's legacy. If I am developing new code, which includes new modules, classes, libraries whatever for use in legacy systems, I will develop unit tests, period, that's what I have done for the 18 years I have been working in this field and I am not about to stop now just because you come along thinking you know something substantial. You don't, anymore than I do. I didn't develop or discover the concept of unit tests, I was taught to use them by people far more skilled in this field then you or I will ever be.
led mike
|
|
|
|
|
led mike wrote: YOu replied to me you f***ing moron.
We're seeing more and more of your true colors I guess.
But you see, I was NOT replying just to you. I was replying to anyone involved in or just reading the thread, but since my starting point was a desire to take issue with dogmatic recommandations, not to mention your arrogant "how can there POSSIBLY be ANYONE out there who don't do things your way" attitude I chose to reply to your post.
led mike wrote: ... and I am not about to stop now just because you come along thinking you know something substantial. You don't, anymore than I do.
I'll let those who read decide for themselves. In any case that isn't the issue. And you continue to debate AS IF I have uttered anything even remotely suggesting that one should not use unit tests.
I assumed you were DELIBERATELY misunderstanding but I am beginning to doubt if you can control it. Have a nice weekend.
|
|
|
|
|
dojohansen wrote: AS IF I have uttered anything even remotely suggesting that one should not use unit tests.
I assumed you were DELIBERATELY misunderstanding but I am beginning to doubt if you can control it.
No, but I did lose the thread as it has been a long one so, yes, I was wrong, you are correct, you never suggested that, my mistake. We were basically talking about two different things in regards to unit tests and it seems we might actually have similar views on both of them. Therefore it is at least partly my fault, if not completely, that we ended up in a confrontation rather than a discussion.
dojohansen wrote: Have a nice weekend.
thanks, same to you.
led mike
|
|
|
|
|
Where can I find some online guidance regarding the creating the ultimate user experience, with regards to exception handling? Jeff Atwood's catch-all solution[^] is very attractive, but I'd like to read up on the finer points, in a design aspect, of handling those unpleasant little situations where the user gets told something bad.
|
|
|
|
|
My [self-serving] opinion is: just tell the users that a problem occurred but that tech support (or whoever) has already been notified. Then when you get the notification, call the user.
there is no point in telling the user that "someclass encountered a null blah blah exception in method whatever(arg,arg,arg,arg) at line 42075 of module someclass.cs": in general they have no context to understand what it means, they cannot do anything about it, and frankly it scares quite a few of them.
telling them instead "oops, my bad" and then following up in a timely fashion [i.e. before they forget what they were doing] is far better - it makes you look like you're on top of things, and it gives the user a warm fuzzy about the application and your commitment to supporting it
again, this is just my [self-serving] opinion. 
|
|
|
|
|
I'm looking for good books or online sources that describe techniques for using object composition. Specifically, I'm interested in the idea that you can have objects within objects, several layers deep and how this can be a beneficial approach to managing application complexity.
I'm already familiar with most of the books out there on OOP, e.g. Design Patterns, Refactoring, etc., so if you can recommend something that deals with this subject that's off the well-beaten path, I'd appreciate it.
|
|
|
|
|
I am not aware of any books. I don't know if this subject is on c2.com but there are certainly some that might qualify as off the well-beaten path.
led mike
|
|
|
|
|
I'm designing a group of tables where each table offers both natural keys, e.g. invoice number, or surrogate keys, e.g. InvoiceId or InvoiceGuid. The PK for each table is a domain independent integer identity column, and each table has an optional natural key field. <pre>
[Example Table] recordId int identity,
exampleId nvarchar(20)
</pre>
Where the 'example' prefix is specific to the Example table, and an invoice table will have an invoiceId natural key field. It strikes me that we are in an inheritance centric solution space, but that will have to wait for the Entity Framework. Are there any established patterns for going about maintaining both types of keys?
Last modified: 123hrs 24mins after originally posted --
|
|
|
|
|
Are you asking a question?
Mike Lasseter
|
|
|
|
|
If you want to generate the invoice number as a natural key in code, I might suggest a singleton that would be responsible for generating it (possibly date + sequence based), however, you need to consider other factors that would get messy by storing the "Invoice Number" as a single column.
For instance, if the system started generating invoices today (080705-001) and shuts down unexpectedly, say at 10am, when it starts back up, it would have to reparse the invoice numbers (like '080705-%' to get the last sequence number...obviously, this would be far from ideal. Also, by storing the invoice number in the database, you are now "stuck" with the formating decision you made at design time. If you start generating more that 1000 invoices today, you will have duplicate invoice numbers (another bad situation). Many systems push back the ultimate problem by adding customer id to the invoice number...
Anyway, if you store the invoice date separate from the sequence number, you can retrieve the last sequence number quicker, you are not stuck with a format problem, and you can move the format issue to your business layer if you do have issues.
|
|
|
|
|
We are writing a server app for Linux in C++. In order to avoid reinventing the wheel we want to use some kind of library that already implements common architectural patterns and provides solutions to problems that arise during server development. The Java and C# community are well endowed in this regard. Is there anything similar available for C++??? We did find IBM's Reusable Server Kernel (RSK) but unfortunately that is meant solely for their VM systems. A similar truly cross-platform library would be just what we are looking for. Thanks in advance for your help.
|
|
|
|
|
variantseeker wrote: We are writing a server app for Linux in C++.
Did you check out the W3C and Apache libraries?
led mike
|
|
|
|
|
I posted a message to comp.object about this subject, but decided to post a modified version of the question here.
What I'm trying to grapple with is the cyclic nature of architectures like MVC that rely on the Observer pattern.
Say that you have a subject (or model in MVC terms) that has a set of observers attached to it. When the subject changes, it notifies its observers. Ok, good enough. But suppose that one of the observers is responsible for changing the subject. Being an observer it will receive a change notification from the subject after changing the subject. In turn, the observer may choose to update itself by querying the subject. However, this update is unnecessary as it was responsible for the change in the subject in the first place.
I was looking at MFC's Document/View architecture for inspiration. The CDocument class has an UpdateAllViews method. The method takes a CView object as a parameter. This object represents the view responsible for changing the document. The CDocument skips over it when updating its views so the specified view doesn't update itself unnecessarily. This approach requires that the View be the one responsible for triggering an update (maybe not in all cases, I'm not all that familiar with the Document/View architecture).
At any rate, this provides one way to break the cycle created by the Observer pattern when an observer can be responsible for changing its subject.
Thoughts?
|
|
|
|
|
Leslie Sanford wrote: Thoughts?
Yes this is a common problem. The solutions vary and discovering which one is best suited is likely dependent on specific aspects of each project.
See this article for example[^]
led mike
|
|
|
|
|
led mike wrote: Yes this is a common problem. The solutions vary and discovering which one is best suited is likely dependent on specific aspects of each project.
See this article for example[^]
Nice article! Thanks. I learned some new approaches.
However, what the article means by redundant notifications[^] isn't exactly what I described here. My concern is an observer receiving a notification when it is the one responsible for changing the subject. The Document/View's UpdateAllViews approach seems to be one way around this.
|
|
|
|
|
Leslie Sanford wrote: However, what the article means by redundant notifications[^] isn't exactly what I described here.
Yes that's what I meant by the solutions will vary and need to be specific to the aspects of any given project.
led mike
|
|
|
|
|
Perhaps try taking a look at how Datbinding works in .Net. It implements its updates exactly as you described.
This can be a PIA in some situations, especially if you require a field to be all caps and want to control it in your property like _someproperty = value.ToUpper(). In this case your ToUpper changes don't get pushed back out to the control.
Mike Lasseter
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
