Re: svn commit: r1145673 - in /incubator/opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/cmdline: chunker/ChunkerEvaluatorTool.java namefind/TokenNameFinderEvaluatorTool.java postag/POSTaggerEvaluatorTool.java

[Jörn Kottmann <ko...@gmail.com>]

On 7/12/11 6:53 PM, colen@apache.org wrote:
> -    File testData = new File(CmdLineUtil.getParameter("-data", args));
> -    CmdLineUtil.checkInputFile("Test data", testData);
> +    BasicEvaluationParameters params = ArgumentParser.parse(args,
> +        BasicEvaluationParameters.class);
>
> -    Charset encoding = CmdLineUtil.getEncodingParameter(args);
> +    File testData = params.getData();

Shouldn't we keep the check for the input file?

Maybe we should later replace it with an annotation, then the Argument
Parser can check it for us. e.g. an InputFile/OutputFile Annotation, or
just one with a parameter.

Jörn