Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.java.programmer > #33632 > unrolled thread

Re: Checked Exceptions in Java

Back to article view | Back to comp.lang.java.programmer

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: Checked Exceptions in Java ruben <ceo@iran.gov> - 2016-02-06 20:37 +0000
    Re: Checked Exceptions in Java Lew <lewbloch@gmail.com> - 2016-02-06 13:45 -0800
      Re: Checked Exceptions in Java ruben <ceo@iran.gov> - 2016-02-08 01:31 +0000
        Re: Checked Exceptions in Java Lew <lewbloch@gmail.com> - 2016-02-07 21:47 -0800
          Re: Checked Exceptions in Java ruben <ceo@iran.gov> - 2016-02-15 11:39 +0000
            Re: Checked Exceptions in Java Eric Douglas <e.d.programmer@gmail.com> - 2016-02-15 05:03 -0800
          Re: Checked Exceptions in Java ruben <ceo@iran.gov> - 2016-02-15 11:42 +0000
        Re: Checked Exceptions in Java Eric Douglas <e.d.programmer@gmail.com> - 2016-02-08 05:34 -0800
          Re: Checked Exceptions in Java ruben <ceo@iran.gov> - 2016-02-15 11:41 +0000
          Re: Checked Exceptions in Java ruben <ceo@iran.gov> - 2016-02-15 11:44 +0000
      Re: Checked Exceptions in Java Eric Douglas <e.d.programmer@gmail.com> - 2016-02-08 05:33 -0800

#33632 — Re: Checked Exceptions in Java

On Sat, 06 Feb 2016 16:12:28 +0000, Stefan Ram wrote:


>   It is not always reasonable for library-grade code (as I call it),
>   that is, for code that resides in a library and might be called from
>   any client, or for professional code that is to be sold to other users
>   which are not programmers themselves.

Fuck that.  It is not the place for the language to tell the programmer 
how to program... period and the whole try catch throw library in Java is 
PIA and just sucks. 

[toc] | [next] | [standalone]

#33634

On Saturday, February 6, 2016 at 12:37:32 PM UTC-8, ruben wrote:
> On Sat, 06 Feb 2016 16:12:28 +0000, Stefan Ram wrote:
> 
> 
> >   It is not always reasonable for library-grade code (as I call it),
> >   that is, for code that resides in a library and might be called from
> >   any client, or for professional code that is to be sold to other users
> >   which are not programmers themselves.
> 
> Fuck that.  It is not the place for the language to tell the programmer 
> how to program... period and the whole try catch throw library in Java is 
> PIA and just sucks.

Now *that's* a solid, rational, engineeering argument!

Not.

The language doesn't tell the programmer how to program with Java any more than with any other programming language, so that premise is false

Checked exceptions are a *programmer's* decision to use or not. If an API writer included checked exceptions, they had a reason to. You might bitch and whine and moan and gripe and cry about it, but they did it for a reason. Just like they did for the argument and return types, and the name of the method.

If you don't like it, don't use that API.

The fact is that checked exceptions are very useful to indicate out-of-band conditions for a method that a client should, and in the mind of the author, must handle. Checked exceptions are the API's way of letting you, the client programmer, know what those out-of-band conditions. And thank goodness, or there'd be random, unpredictable, seen-first-in-production crashes when those predictable exceptions happen!

People who say "try..catch" "just sucks" are ignoring decades of advances in programming-language design. People who decry checked exceptions in Java have utterly failed to understand either their rationale and whom to blame for them.

-- 
Lew

[toc] | [prev] | [next] | [standalone]

#33659

On Sat, 06 Feb 2016 13:45:48 -0800, Lew wrote:


> The language doesn't tell the programmer how to program with Java any
> more than with any other programming language, so that premise is false



When you have to wrap an error in a check, then it DOES confine the 
coders and it sucks.  You have to get into this who deep psychological 
drama with the Java developers to try to figure out why this is checked 
and this is not checks.  It is such an obnoxious PIA.  Just stop the 
program and send my signal, or handle my condition as I wrote it... 
period and let me do my job.

Checked Errors and Exceptions is one of the stupidest ideas ever 
created.  

Ruben

[toc] | [prev] | [next] | [standalone]

#33663

On Sunday, February 7, 2016 at 5:31:24 PM UTC-8, ruben wrote:
> On Sat, 06 Feb 2016 13:45:48 -0800, Lew wrote:
> 
> 
> > The language doesn't tell the programmer how to program with Java any
> > more than with any other programming language, so that premise is false
> 
> When you have to wrap an error in a check, then it DOES confine the 

Shouting doesn't make it more true.

> coders and it sucks.  

Nice engineering argument there. The logic just flows from evidence to conclusion, doesn't it?

Or in this case, in a circle like Ouroboros eating its own tail.

> You have to get into this who deep psychological 
> drama with the Java developers to try to figure out why this is checked 

What the...?

NO, I don't. The only "deep psychological drama" is in your posts.

> and this is not checks.  It is such an obnoxious PIA.  Just stop the 

Again with the logic and evidence! Oh, my!

> program and send my signal, or handle my condition as I wrote it... 
> period and let me do my job.

I'm sorry, you say that every method that hits an out-of-band condition should stop the program?

Now that is not a good experience for the user.

Checked exceptions not only let you do your job, they help you do it better by handling the condition as the API writer wrote it, just as you suggest. So your argument actually supports the value of checked exceptions.

> Checked Errors and Exceptions is one of the stupidest ideas ever 
> created. 

Java doesn't have "checked Errors [sic]". Thus you prove you have absolutely no idea what you're talking about, and thus your illogically-presented notions have no merit.

You have provided nothing in the way of evidence, just strong emotional language and "deep psychological drama". Your words have no value, therefore. 

-- 
Lew

[toc] | [prev] | [next] | [standalone]

#33798

On Sun, 07 Feb 2016 21:47:36 -0800, Lew wrote:

> Nice engineering argument there.


It was, but you missed it.  Although considering how poorly Java is 
engineered from the ground up, I doubt that it make any difference.

But here is the funny part, there are no engineering problems, only human 
ones.

I guess you missed that class in into to engineering school ....

[toc] | [prev] | [next] | [standalone]

#33805

On Monday, February 15, 2016 at 6:39:46 AM UTC-5, ruben wrote:
> On Sun, 07 Feb 2016 21:47:36 -0800, Lew wrote:
> 
> > Nice engineering argument there.
> 
> 
> It was, but you missed it.  Although considering how poorly Java is 
> engineered from the ground up, I doubt that it make any difference.
> 
> But here is the funny part, there are no engineering problems, only human 
> ones.
> 
> I guess you missed that class in into to engineering school ....

Poorly engineered?  Why are there so many trolls in this forum just stopping in to say they dislike Java entirely and would prefer a different language?  Find your other language, hang out in their forum.

Checked exceptions are always better than unchecked exceptions when handled properly.  Poor handling by any particular developer(s) is not a reason to dislike checked exceptions.  I have had a job with nightly support, and I would much rather the program know to expect errors and know something it can do about them than get that call at 2am to say someone's program stopped working.  Every language with the power to do anything useful is going to have programmers writing poor code, and even the largest corporations let inexperienced developers put that code into production without a senior developer reviewing it first.

[toc] | [prev] | [next] | [standalone]

#33800

On Sun, 07 Feb 2016 21:47:36 -0800, Lew wrote:

>> When you have to wrap an error in a check, then it DOES confine the
> 
> Shouting doesn't make it more true.

that's not shouting.  THIS IS SHOUTING... god help us all.

Like

That wine was REALLY sweet...

[toc] | [prev] | [next] | [standalone]

#33673

On Sunday, February 7, 2016 at 8:31:24 PM UTC-5, ruben wrote:
> On Sat, 06 Feb 2016 13:45:48 -0800, Lew wrote:
> 
> 
> > The language doesn't tell the programmer how to program with Java any
> > more than with any other programming language, so that premise is false
> 
> 
> 
> When you have to wrap an error in a check, then it DOES confine the 
> coders and it sucks.  You have to get into this who deep psychological 
> drama with the Java developers to try to figure out why this is checked 
> and this is not checks.  It is such an obnoxious PIA.  Just stop the 
> program and send my signal, or handle my condition as I wrote it... 
> period and let me do my job.
> 
> Checked Errors and Exceptions is one of the stupidest ideas ever 
> created.  
> 
> Ruben

Have you ever had a job? with a large corporation? ever gotten the call at 2am because someone's program stopped?

Please stop posting here and stop using Java, it is not the language for you.

[toc] | [prev] | [next] | [standalone]

#33799

On Mon, 08 Feb 2016 05:34:28 -0800, Eric Douglas wrote:

> Have you ever had a job? with a large corporation? ever gotten the call
> at 2am because someone's program stopped?


you mean like a 1200 bed hospital.  Not since 9-11, no.

[toc] | [prev] | [next] | [standalone]

#33801

On Mon, 08 Feb 2016 05:34:28 -0800, Eric Douglas wrote:


> Have you ever had a job? with a large corporation? ever gotten the call
> at 2am because someone's program stopped?

because checked errors saved your program...  umm.  If you say so.

[toc] | [prev] | [next] | [standalone]

#33672

On Saturday, February 6, 2016 at 4:46:16 PM UTC-5, Lew wrote:
> Now *that's* a solid, rational, engineeering argument!
> 
> Not.
> 
> The language doesn't tell the programmer how to program with Java any more than with any other programming language, so that premise is false
> 
> Checked exceptions are a *programmer's* decision to use or not. If an API writer included checked exceptions, they had a reason to. You might bitch and whine and moan and gripe and cry about it, but they did it for a reason. Just like they did for the argument and return types, and the name of the method.
> 
> If you don't like it, don't use that API.
> 
> The fact is that checked exceptions are very useful to indicate out-of-band conditions for a method that a client should, and in the mind of the author, must handle. Checked exceptions are the API's way of letting you, the client programmer, know what those out-of-band conditions. And thank goodness, or there'd be random, unpredictable, seen-first-in-production crashes when those predictable exceptions happen!
> 
> People who say "try..catch" "just sucks" are ignoring decades of advances in programming-language design. People who decry checked exceptions in Java have utterly failed to understand either their rationale and whom to blame for them.
> 
> -- 
> Lew

+1

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.java.programmer

csiph-web