|
Coconut Pete posted:This is not exactly a programming question, but it is related to programming: I would think using a framework as a starting point would lessen the bill on their end, wouldn't it? Why reinvent? There will still be customizations and general work you can bill for, I don't see why they would have a problem with it. I don't think you can profit from reselling open source, so I would be careful how you implement and charge (and read the small print of any source package you use). What I would have a problem with, as a client, is not being presented with a functional specification outlining what you are going to build the framework on (frameworks, languages, etc) prior to you starting your work.
|
# ? Jul 3, 2008 15:02 |
|
|
# ? May 14, 2024 19:21 |
|
I have a general MVC framework question. I have a HTML/JSP/Javascript page which renders a page and provides some basic interface, interaction and control. I then have a servlet which is executed from a button on the page described above. The servlet sends all information to the core model and service code. The core stuff executes and sends the response back to the servlet which sends it back to the HTML/JSP/Javascript page. So my question is, do I have this right?: HTML/JSP/Javascript page = view Servlet = controller Model/Service = model
|
# ? Jul 3, 2008 15:09 |
|
Coconut Pete posted:This is not exactly a programming question, but it is related to programming: Hell no - the client's paying you for the end result, anything you can do to lessen the work on your end is 100% ok. (Where the "end result" includes maintainability, so don't think I'm advocating taking stupid shortcuts and hardcoding things.) You're not selling every bit of code as being your own, you're selling the entire package. You do have to make sure the license terms on all the components you assemble allow you to resell them, but frameworks should all be ok with that, because that's what they're for.
|
# ? Jul 3, 2008 18:32 |
|
Triple Tech posted:Why again is emulation for video games slow? Is it because the programmers lack the EXACT documentation for the original architecture? Basically is it a matter of accuracy or is it really processing power? What emus are you talking about? I've only tried emus up to Playstation 1 and they all worked well. Emus of modern consoles are most likely limited by processing power. In the past, emulators were simple interpreters of the game code (fetch instruction, simulate it on x86 hardware, loop forever). Unless the modern console emus try to get some JIT compilation going I don't see a lot of hope to get them to emulate games properly on current systems.
|
# ? Jul 3, 2008 19:10 |
|
Triple Tech posted:Why again is emulation for video games slow? Is it because the programmers lack the EXACT documentation for the original architecture? Basically is it a matter of accuracy or is it really processing power? I believe your guesses are right - we pretty much just react to the game instructions instead of really understanding them. I know there are some systems we "understand," like the NES, that we can perfectly emulate. Other than that things like the Emotion Engine and Mode 7 chips provide specialized speed for games that utilize it in the same way that a PhysX card does for computer games - we can emulate it, but it uses up speed on something else.
|
# ? Jul 3, 2008 20:16 |
|
Skychrono posted:I believe your guesses are right - we pretty much just react to the game instructions instead of really understanding them. I know there are some systems we "understand," like the NES, that we can perfectly emulate. Other than that things like the Emotion Engine and Mode 7 chips provide specialized speed for games that utilize it in the same way that a PhysX card does for computer games - we can emulate it, but it uses up speed on something else. It's really more because game console hardware is a hell of a lot more complex than it was 20 years ago. Also, the operating systems and libraries on the consoles (those exist now) have to be duplicated almost exactly, and the only documentation for them (that you can get legally, at least) is how games use them. And, you don't have the source. It'd be like handing someone a copy of Office and the x86 spec and saying "emulate Windows running on a PC, but make it run on MIPS." It's theoretically possible, but it'd be a monumental effort.
|
# ? Jul 4, 2008 00:25 |
|
more falafel please posted:It's really more because game console hardware is a hell of a lot more complex than it was 20 years ago. Also, the operating systems and libraries on the consoles (those exist now) have to be duplicated almost exactly, and the only documentation for them (that you can get legally, at least) is how games use them. And, you don't have the source. It'd be like handing someone a copy of Office and the x86 spec and saying "emulate Windows running on a PC, but make it run on MIPS." It's theoretically possible, but it'd be a monumental effort. why can't some brave soul bust into the xbox labs and take off with the specs? A boy could only dream, i guess.
|
# ? Jul 4, 2008 00:34 |
|
Entheogen posted:why can't some brave soul bust into the xbox labs and take off with the specs? A boy could only dream, i guess. Probably because they can tell which people have access to that level of specs and it wouldn't be worth jeopardizing their awesome careers so some kids on the Internet can play games for free.
|
# ? Jul 5, 2008 11:39 |
|
Coconut Pete posted:This is not exactly a programming question, but it is related to programming: Do you write your own language and libraries when you contract? Of course it's not cheating.
|
# ? Jul 8, 2008 18:52 |
|
I have a small problem right now, I want to capture the data of a field on a .dbf but i dont know how to capture chars by chars on a string. Its kinda difficult to explain but i want for example, change this field (i kid you not, this is how its formatted) "NAME/NAME2/MORE NAMES" to 3 different fields where everything is separated. Sorry for the bad english I guess i should be doing a SQL script to change all the data, and i am not really a pro at Visual FoxPro :I
|
# ? Jul 8, 2008 19:07 |
|
BAD DOUBLE POSTING BAD!
|
# ? Jul 8, 2008 19:07 |
|
Why does the command line version of md5 on my OS X laptop give me different results than the Javascript implementation at http://pajhome.org.uk/crypt/md5/ ? Since there's an RFC for md5, shouldn't they be the same?
|
# ? Jul 9, 2008 03:49 |
|
ashgromnies posted:Why does the command line version of md5 on my OS X laptop give me different results than the Javascript implementation at http://pajhome.org.uk/crypt/md5/ ? Since there's an RFC for md5, shouldn't they be the same? Is one including the newline at the end of the line and one isn't?
|
# ? Jul 9, 2008 03:59 |
|
JoeNotCharles posted:Is one including the newline at the end of the line and one isn't? Durrrr. `echo -n` fixed it. Thanks!
|
# ? Jul 9, 2008 04:11 |
|
A book I'm reading has all its examples in C++, and in all its examples, every class function is implemented outside the class declaration instead of inside it. Why do they do this? My best guess is to keep their focus on class relationships instead of the implementation. Example: code:
|
# ? Jul 9, 2008 18:12 |
|
ColdSnickersBar posted:A book I'm reading has all its examples in C++, and in all its examples, every class function is implemented outside the class declaration instead of inside it. Why do they do this? My best guess is to keep their focus on class relationships instead of the implementation. Normally you'd put the declarations in a header file (maze.h) and the defintions/implementation in the .cpp/.c file. Then, when something else wants to use your class, it can #include your maze.h and, by only looking at what's exposed there, knows exactly what it can use the class for, without requiring any knowledge of implementation.
|
# ? Jul 9, 2008 18:16 |
|
Ugg boots posted:Normally you'd put the declarations in a header file (maze.h) and the defintions/implementation in the .cpp/.c file. Then, when something else wants to use your class, it can #include your maze.h and, by only looking at what's exposed there, knows exactly what it can use the class for, without requiring any knowledge of implementation. Oh that's awesome. I wonder why it didn't just say that? Especially since re-usability and separating design from implementation is the main theme of the book. Thanks! EDIT: So, then, wouldn't declaring arguments in the function assume too much about the implementation? EDIT2: Nevermind. I get it. ColdSnickersBar fucked around with this message at 18:37 on Jul 9, 2008 |
# ? Jul 9, 2008 18:20 |
|
ColdSnickersBar posted:A book I'm reading has all its examples in C++, and in all its examples, every class function is implemented outside the class declaration instead of inside it. Why do they do this? My best guess is to keep their focus on class relationships instead of the implementation. That's one of the reasons, along with it being much easier for a programmer to comprehend since they most-often just needs to know how to interface with a type rather than know its implementation, but there's also a lot more to it. If you put the function definitions in the class definition it means that every translation unit that simply needs the class definition, for instance, just to interface with an instance of the class, now has to go through all of the function definitions as well. This can increase compile times, especially if you have a lot of units, and now whenever you change the implementation, you are going to have to recompile all the units that simply interface with your class because you put the implementation in the headers that they all include. If you had the implementation in its own unit, then changing the implementation would not imply a recompile for everything else as long as the interface stayed the same. As well, by putting the implementation in the header that people need to include in order to interface with your class, you are pulling in any dependencies that the implementation has which the interface may not. Again, this can increase your compile times by a very large amount, especially if you are doing something like including <windows.h>, heh. Putting the function definition inside of the class definition also implicitly makes your functions as though you used the keyword "inline," though I wouldn't worry about that since most compilers are smart enough when considering the inline hint to disregard it if you have a complex function definition anyway. Finally a more subtle issue is that it's not always possible to put the function definition inside of the class definition because of conflicting dependencies which can only be resolved by putting includes between the class definition and the function definitions.
|
# ? Jul 9, 2008 18:30 |
|
I don't code in C++ often. Is this something that is usually only done by C++ developers? I never see this practice on the web in, for example, C#.
|
# ? Jul 9, 2008 18:42 |
|
That's because you're not able to do it in C#.
|
# ? Jul 9, 2008 18:44 |
|
Yes, a lot of this is because of the way that C and C++ code is compiled, which is very different from the systems that C#, Java, and many other languages use.
|
# ? Jul 9, 2008 18:47 |
|
Avenging Dentist posted:That's because you're not able to do it in C#. That Turkey Story posted:Yes, a lot of this is because of the way that C and C++ code is compiled, which is very different from the systems that C#, Java, and many other languages use. Oh. It's alright, I guess, since the book is really about concepts and not the code itself. Thanks for the help.
|
# ? Jul 9, 2008 18:52 |
|
I have a regular expression question - I need to split a string on spaces, except spaces between quotes. Basically what passing arguments into a program does. Couldn't find how to do it anywhere... So input: Hello Mr "potato head" output: [ Hello, Mr, potato head ] Also would be best if unclosed quotes would be ignored: Input: Hello "mr potato head Output: [ Hello, "mr, potato, head ] So, what regex do I pass into java String.split to get this? Thanks
|
# ? Jul 9, 2008 20:46 |
I'm working on a way to transfer data from an excel spreadsheet into an old NetTerm interface. If the data is in 27 cells across a row, I should be able to copy the entire row and then hit paste in NetTerm to transfer it all. Sometimes NetTerm takes all the data... other times it hiccups midway through and stops. I think the problem is that copying and pasting the data passes it through too quickly for the NetTerm system to respond properly. Is there a way to slow down the speed at which data is copied and pasted?
|
|
# ? Jul 9, 2008 21:23 |
|
Pinkied_Brain posted:I need to split a string on spaces, except spaces between quotes. You should take a look at java.util.regex.Matcher as it sounds like more of what you want. Basically, you just want to find matches that are either "characters separated by spaces" or "characters separated by quotes", so this is a regex which matches that: " ?([^ ]+|\".+\") ?". So your code would be similar to: code:
|
# ? Jul 9, 2008 21:38 |
|
Quick, probably embarrassing, only tangentially programming related question. I'm using X11 on a Mac to ssh to a server for work, and I need to get a .zip file from said server. I can find the file on that server just fine, but how can I copy it from there to my computer? I've tried using the cp command with a destination path on my computer, but that doesn't work and I didn't really expect it to... Any ideas?
|
# ? Jul 10, 2008 00:12 |
|
Emo.fm posted:Quick, probably embarrassing, only tangentially programming related question. I'm using X11 on a Mac to ssh to a server for work, and I need to get a .zip file from said server. I can find the file on that server just fine, but how can I copy it from there to my computer? I've tried using the cp command with a destination path on my computer, but that doesn't work and I didn't really expect it to... Any ideas? Have you tried using scp or sftp on your local machine? ie scp user@remote:/path/to/file localfilename
|
# ? Jul 10, 2008 00:15 |
|
Emo.fm posted:Quick, probably embarrassing, only tangentially programming related question. I'm using X11 on a Mac to ssh to a server for work, and I need to get a .zip file from said server. I can find the file on that server just fine, but how can I copy it from there to my computer? I've tried using the cp command with a destination path on my computer, but that doesn't work and I didn't really expect it to... Any ideas? I) This goes in SH/SC II) Why are you using X11 to ssh, OS X has a perfectly function ssh client built in III) To grab files from a remote server, use a normal FTP client like Transmit or Cyberduck and just use your ssh credentials to log in and browse like a normal ftp. This is called sftp and it is very useful.
|
# ? Jul 10, 2008 01:09 |
|
6174 posted:Have you tried using scp or sftp on your local machine? ie Night Chaos posted:I) This goes in SH/SC 2) I am pretty inexperienced, just took a few comp sci classes in college, and my professor was very adamant that X11 is superior to OS X's Terminal. I don't know what his reasoning is, but I've always just used X11 as a result. 3) Interesting, thanks.
|
# ? Jul 10, 2008 01:40 |
|
Your compsci professor was either misinformed or just out of date -- the Terminal.app that ships with OS X has been quite decent compared to average xterm since around 10.3/10.4, and is even better now in 10.5. Given that the X11 terminal looks and behaves like complete rear end, there's no reason to use it unless you're just using it to launch a GTK app like e.g. Wireshark.
|
# ? Jul 10, 2008 01:58 |
|
bitprophet posted:Given that the X11 terminal looks and behaves like complete rear end, there's no reason to use it unless you're just using it to launch a GTK app like e.g. Wireshark. alias wireshark='open -a /Applications/Wireshark.app' There is no reason to use xterm/iterm/anydamnedterm on OS X.
|
# ? Jul 10, 2008 02:22 |
|
Plastic Jesus posted:alias wireshark='open -a /Applications/Wireshark.app' Except for, you know, using terminal applications.
|
# ? Jul 10, 2008 03:33 |
|
Plastic Jesus posted:alias wireshark='open -a /Applications/Wireshark.app' Sorry, I meant the wireshark that installs via MacPorts I forget why I installed it that way and not via whatever gives you a nice shiny .app, but that's how it's set up vv it's a GTK app and has to be launched in X11. And, yea, it's rather difficult to avoid the terminal entirely, at least if you're a non-Cocoa developer or sysadmin.
|
# ? Jul 10, 2008 04:39 |
|
GT_Onizuka posted:Except for, you know, using terminal applications. I meant that there's no reason to use anything other than Terminal.app
|
# ? Jul 10, 2008 05:43 |
|
Plastic Jesus posted:I meant that there's no reason to use anything other than Terminal.app Terminal.app doesn't do mouse reporting, which is definitely a huge reason not to use it. It's also pretty spartan feature-wise. The only thing it has going over the only other native terminal emulator, iTerm, is that it's much faster and less buggy. It really is a matter of preference. Terminal.app isn't leagues ahead of everything else or anything like that.
|
# ? Jul 10, 2008 06:05 |
|
How is rxtv-unicode on OS X?
|
# ? Jul 10, 2008 06:17 |
|
Plastic Jesus posted:alias wireshark='open -a /Applications/Wireshark.app' iTerm is perfectly acceptable, if that's what you're into. The only reason I used to like it was tabs, and Leopard's Terminal.app has tabs now. But it's certainly not inferior to Terminal.app.
|
# ? Jul 10, 2008 06:35 |
|
Terminal.app seems to screw up the basic 16 colors. Turning off bold and using "Use bright colors for bold text" is still not good enough. It also doesn't let you tweak the exact colors you use (except for foreground and background). I can't seem to get rid of the scrollbar. This is totally a nitpick but it's annoying. Even tuning scrollback lines down to zero doesn't hide it. I use screen so I never actually scroll up in the terminal itself. The scrollbar is ugly wasted space. The block cursor does not draw whatever character is under it. Playing with its color's opacity does help, but I shouldn't have to do that. I don't like the other cursor displays because it's harder to find the cursor. iTerm gets these all of these right, so I use it. Unfortunately, it crashes practically every day. But I always run everything in screen, so it's no more than a few seconds of annoyance. Filburt Shellbach fucked around with this message at 13:14 on Jul 10, 2008 |
# ? Jul 10, 2008 13:11 |
|
You can use TerminalColours to customize the color scheme. If you want mouse reporting, I'm working on MouseTerm which implements it for Terminal.app. So far it only does mouse wheel scrolling, but it does simulated scrolling in fullscreen applications as well (like gnome-terminal).
|
# ? Jul 10, 2008 16:08 |
|
|
# ? May 14, 2024 19:21 |
Counter Punch posted:I think the problem is that copying and pasting the data passes it through too quickly for the NetTerm system to respond properly. Any ideas? Am I asking this in the wrong place? Thanks...
|
|
# ? Jul 10, 2008 18:55 |