Re: Clear the Screen
From: nos (nos_at_nospam.com)
Date: 01/09/04
- Next message: Raymond DeCampo: "Re: Are primitives objects?"
- Previous message: Fred: "Re: Are primitives objects?"
- In reply to: Anthony Borla: "Re: Clear the Screen"
- Next in thread: Stewart Gordon: "Re: Clear the Screen"
- Reply: Stewart Gordon: "Re: Clear the Screen"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Fri, 09 Jan 2004 15:04:19 GMT
But if they can implement "beep", and they do,
they should be able to implement "cls" too.
"Anthony Borla" <ajborla@bigpond.com> wrote in message
news:g0yLb.2893$Wa.1516@news-server.bigpond.net.au...
>
> "Denz" <RUBBISH@RUBBISHhotmail.com> wrote in message
> news:KvlLb.1671$Wa.792@news-server.bigpond.net.au...
> >
> > Ive seen this question before- with no real solution.
> > Surely it wouldnt have been difficult to provide a clear
> > screen, and even basic screen cursor positioning?
> >
> > It would make text-mode java alot more useful.
> >
>
> Unfortunately these tasks are inherently operating system-specific, and
> cannot be implemented in any standard way across platforms. Some
platforms,
> in fact, don't even understand the concepts of 'screen', or 'cursor', so
> wouldn't even have any use for classes that implemented these
abstractions.
> Even the 'standard' C and C++ languages which, like Java, aim for platform
> independance, do not implement such functionality.
>
> System-specific tasks such as these may be performed via:
>
> * 'Runtime.getRuntime().exec(...)'
>
> * Writing Java Native Interface [JNI] routines which tap into
> the relevant operating system functionality
>
> The first option should really be seen as a 'quick and dirty' approach -
it
> allows you to perform certain system-specific tasks but with very little
> control. A program, to be considered robust, would make no more than
sparing
> use of this facility. If you find you are relying on it heavily you may
need
> to seriously question whether Java is the ideal development tool for this
> task [a scripting language (Perl, Python) might be more suitable ?], or
more
> of a development commitment needs be made by using JNI.
>
> There is nothing inherently wrong with JNI except that its use helps
defeat
> one of Java's most important qualities - platform independance - and it
> introduces 'weaknesses' into the application because program execution
> occurs outside the control of the JVM, and any failure here could be
> catastrophic. On the flipside, Java would be a very limited development
> environment without it since it allows for the tapping into
system-specific
> functionality in a controlled, and efficient way.
>
> So, to provide the facilities you seek, you need to implement suitable JNI
> routines. Again, options are available:
>
> * Tap into someone else's work. As another respondant has
> already meantioned, the JCurses package provides a
> reasonable suit of screen handling routines for the *NIX
> and Windows-familiy platforms
>
> * Write your own. This isn't as difficult as you might imagine,
> and might be the best approach if all you require is a
> few simple routines like:
>
> - Clearing the screen
> - Moving the cursor to specified positions
> - Accepting input [say a single keystroke] without pressing
> ENTER
>
> A good start would be the JNI tutorial at the Sun site, and,
> of course, perusal of the documentation for the system
> functions implementing this functionality
>
> Anyway, apologies for the longwindedness of the response. Hopefully,
though,
> some useful insights have been provided.
>
> Cheers,
>
> Anthony Borla
>
>
- Next message: Raymond DeCampo: "Re: Are primitives objects?"
- Previous message: Fred: "Re: Are primitives objects?"
- In reply to: Anthony Borla: "Re: Clear the Screen"
- Next in thread: Stewart Gordon: "Re: Clear the Screen"
- Reply: Stewart Gordon: "Re: Clear the Screen"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
|
