Amber Beerends

Greenhorn
+ Follow
since Jun 11, 2008
Merit badge: grant badges
For More
Cows and Likes
Cows
Total received
0
In last 30 days
0
Total given
0
Likes
Total received
0
Received in last 30 days
0
Total given
0
Given in last 30 days
0
Forums and Threads
Scavenger Hunt
expand Ranch Hand Scavenger Hunt
expand Greenhorn Scavenger Hunt

Recent posts by Amber Beerends

The original post here is almost 2 years old. Is there any indication of when the Java EE 6 exam will be released? I'm leery of investing time on an exam still based on 1.4...
rinke - That solution doesn't work. If the server writes X and Y to the file, the client can still overwrite the entire file so that it just contains A and B.

I think what Jean is saying makes more sense. I had been delaying saving the configuration until the application was shut-up down, but if I do it after a successful connection is made with the submitted data, this is less likely to happen. Although, it could be possible if someone opens both, then starts the server, then starts the client. :S

- amber
My exam specifies that some configuration must be saved between runs in a specific file in the current working directory and must be persistent between runs. I've done this and its all well and good UNLESS I'm running the server and networked client on the same machine from the same working directory. Then the saves from one has the potential to wipe out changes made by the other. I'm torn about handling this. One one hand, its totally illogical in a real setting to be running the networked client and server on the same machine so this scenario is really just a developer testing problem. On the other hand, I'm technically not meeting the requirement in this (albeit unrealistic) scenario.

Thoughts??
Sorry, Andrew, I wasn't really trying to ask that as it's own question. It was more of a rhetorical question. My point was more my second statement:

I guess what I'm wondering if the book's example is planting seeds of non-existent requirements simply because the book's author choose to do something a certain way.



So, regarding the previous poster saying that your PostiveIntegerField code could be copied directly, I was trying to point out there is nothing in the assignment saying that the checking done by PositiveIntegerField is necessary which supports your point about book's code being more difficult than the Sun assignment.

IMO, in that case, the book solution is going above and beyond the call of duty, but I don't think it qualifies as giving "too much information" since the information is not even really required! I do wonder, though, if those kinds of more intricate details will trip up the more inexperienced developer and make them think that is required even if it isn't!

FWIW, I've been coding Java for 11 years but most of my experience, but I haven't done anything regarding RMI or sockets for 10 years so I found the chapters covering those topics to be really useful. The rest of the design ideas closely mirrored what I had already planned to do based on my own reading of the assignment. I think that if someone was new to Java, they would have a harder time understanding all of the code and explanations in the book and would, at the least, take a much longer time to complete the project. Also, I'm sure most of the information in the book could be gleamed from other sources. It would just take longer to find it all that way so the book becomes "one stop shopping" for the relevant info.
Maybe I'm oversimplifying, but my assignment clearly says you don't get extra credit for exceeding the requirements and there is no statement that says I "must" validate the port number. Would it really be considered wrong to just let the server app fail to connect (crash?!) if the port number is invalid?

I guess what I'm wondering if the book's example is planting seeds of non-existent requirements simply because the book's author choose to do something a certain way.