[Biojava-l] Exception handling

[mark.schreiber at group.novartis.com]

Hello -

Exceptions that extend RuntimeException don't have to be caught. This was 
a contraversial decision by the Java designers so they could make things 
like NullPointerException. If you think of all the times you could throw a 
null pointer exception you would need a hell of a lot of try catch 
statements (one for everytime you pass an object reference). Classes that 
extend RuntimeException are supposed to indicate failures due to bad 
programming practice. They shouldn't really occur in a well made program. 
You can catch them, you just have to know when they might be thrown. You 
should only try to catch them if you can recover sensibly though so most 
times you wouldn't bother unless you have a system that needs to run 24/7 
without crashing. BioRuntimeException is one of these but hardly used in 
the API.

The other type of Throwable is that doesn't need to be caught is Error 
(BioError is one of these, also rarely used). Errors are generally 
considered to be unrecoverable. You generally log them, shutdown as 
gracefully as possible and call tech support.

BioException is much more common. This extends good old Exception and must 
be declared and caught. It is used all over.

So I'm not sure what the problem is. If you call DNATools.createDNA and 
throw some rubbish at it you will probably get something like 
IllegalSymbolException (subclass of BioException). You can catch this.

Without seeing the program or stack trace all I can think of is that some 
bug is causing an Error to be thrown somewhere. The only other possibility 
is that you have two catch statements that conflict. Eg if you catch 
BioException and exit before you catch IllegalSymbolException then you 
won't reach this. Although your compiler should complain about this so it 
seems unlikely.

Can you send the source? or that exception stack?

- Mark





"Richard HOLLAND" <hollandr at gis.a-star.edu.sg>
Sent by: biojava-l-bounces at portal.open-bio.org
12/17/2004 12:42 PM

 
        To:     <biojava-l at biojava.org>
        cc:     (bcc: Mark Schreiber/GP/Novartis)
        Subject:        [Biojava-l] Exception handling


Hi all,

I'd like to be able to catch exceptions and handle them but it seems that 
many BioJava exceptions are sub-classes of the system kill everything no 
matter what exception, which makes them uncatchable. For instance if I try 
and load a junk sequence using DNATools, I get an exception quite rightly 
complaining about the alphabet, but even though I try-catch the code that 
parses this junk sequence (for Exception), the program quits completely 
without my exception handler or any further code being called.

Is this deliberate? Surely it would be nicer to allow these exceptions to 
be caught? I have noticed that many routines in the BioJava source code 
throw BioException (or a name similar to that, can't remember...), yet do 
not declare it in their method signatures. This could be why. It'd be much 
nicer to actually declare you are going to throw an exception, what type 
it is likely to be, then behave nicely instead of killing the whole app.

cheers,
Richard

Richard Holland
Bioinformatics Specialist
GIS extension 8199 
 
---------------------------------------------
This email is confidential and may be privileged. If you are not the 
intended recipient, please delete it and notify us immediately. Please do 
not copy or use it for any purpose, or disclose its content to any other 
person. Thank you.
---------------------------------------------


_______________________________________________
Biojava-l mailing list  -  Biojava-l at biojava.org
http://biojava.org/mailman/listinfo/biojava-l