You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@opennlp.apache.org by "Joern Kottmann (JIRA)" <ji...@apache.org> on 2014/04/24 12:01:24 UTC

[jira] [Commented] (OPENNLP-675) Absence of logging and usage of System.out

    [ https://issues.apache.org/jira/browse/OPENNLP-675?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13979530#comment-13979530 ] 

Joern Kottmann commented on OPENNLP-675:
----------------------------------------

The usage of System.out and Sysem.err is only intentional in code which is part of the command line interface. In other areas stdout or stderr should not be used. The mentioned GISTrainer is an example of that. The training code typically writes log information to stdout.

There is a jira issues which tries to deal with it and suggest to implement an API to capture log events produced by the trainer. I suggest we have a second thought on that design and see how it could be implemented.

I believe it would be valuable to store the training log file inside the produced model.

> Absence of logging and usage of System.out
> ------------------------------------------
>
>                 Key: OPENNLP-675
>                 URL: https://issues.apache.org/jira/browse/OPENNLP-675
>             Project: OpenNLP
>          Issue Type: New Feature
>          Components: Sentence Detector, Tokenizer
>            Reporter: Eugene Prystupa
>
> There seems to be no concept of logging used by the libraries. Instead System.out.println is hard-coded in many places where debug information using a logging framework would do it.
> This makes awkward to use the modules integrated into a different application (as it spams our logs or console). 
> Is the usage of System.out in core classes (like GISTrainer) by choice? Or is it simply a technical debt? I am happy to work on it and provide a patch if this is a technical debt.



--
This message was sent by Atlassian JIRA
(v6.2#6252)