Re: [PATCH] jvyaml sanity: printStackTrace for client object creation exceptions

Wouldn't it be a better idea to store backtrace and forward it higher,
along with exception, so it can be used when needed, but wouldn't take
log-space otherwise?

I strongly believe, that printing is a task for application, not for a
library (with an exceptional case of a printing library)

On 12/20/07, Zenaan Harkness <[email protected]> wrote:
> Hello, here's patch #2 :)
>
> This one adds a simple printStackTrace, around object creation time.
>
> Without this, when a ctor or a newly-created object (I think), and a
> setXXX method of such an object (I have tested - faced this problem),
> throws an exception, eg a NullPointerException, the name of the
> exception (eg "NullPointerException") as well as the stack trace for my
> own code (the client code of jvYaml) which would normally pinpoint the
> exception, is completely consumed and is not displayed to the user when
> running their (my) code.
>
> So, dumping the stack at the specified point in SafeConstructorImpl
> solves this problem.
>
> Both this and the previous patch I am using on a daily basis in my
> running code.
>
> Regards
> Zen
>
>
> ---
>  src/org/jvyaml/SafeConstructorImpl.java |    2 ++
>  1 files changed, 2 insertions(+), 0 deletions(-)
>
> diff --git a/src/org/jvyaml/SafeConstructorImpl.java b/src/org/jvyaml/SafeConstructorImpl.java
> index 33da8f5..c3e29e2 100644
> --- a/src/org/jvyaml/SafeConstructorImpl.java
> +++ b/src/org/jvyaml/SafeConstructorImpl.java
> @@ -366,6 +366,8 @@ public class SafeConstructorImpl extends BaseConstructorImpl {
>                  }
>              }
>          } catch(final Exception e) {
> +            // to make debugging for YAML client code sane, we must dump stack:
> +            e.printStackTrace();
>              throw new YAMLException("Can't construct a java object from class " + pref + ": " + e.toString());
>          }
>          return outp;
> --
> 1.5.3.7.1066.g7a4a2e
>
>
> --
> Homepage: www.SoulSound.net -- Free Australia: www.UPMART.org
> Please respect the confidentiality of this email as sensibly warranted.
>
> -------------------------------------------------------------------------
> SF.Net email is sponsored by:
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services
> for just about anything Open Source.
> http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
> _______________________________________________
> Yaml-core mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/yaml-core
>


-- 
Alexey Zakhlestin
http://blog.milkfarmsoft.com/

-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services
for just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace