You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@opennlp.apache.org by "Richard Zowalla (Jira)" <ji...@apache.org> on 2023/03/09 14:17:00 UTC

[jira] [Closed] (OPENNLP-509) opennlp.tools.parser.Parse.getParent() returning incorrect object

     [ https://issues.apache.org/jira/browse/OPENNLP-509?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Richard Zowalla closed OPENNLP-509.
-----------------------------------
    Fix Version/s: 2.2.0
       Resolution: Fixed

> opennlp.tools.parser.Parse.getParent() returning incorrect object
> -----------------------------------------------------------------
>
>                 Key: OPENNLP-509
>                 URL: https://issues.apache.org/jira/browse/OPENNLP-509
>             Project: OpenNLP
>          Issue Type: Bug
>          Components: Parser
>    Affects Versions: tools-1.5.2-incubating
>         Environment: OpenNLP invoked from C# (.Net 4) via IKVM ikvm-7.0.4335.0
> OpenNLP library dll created using the command: ikvmc -target:library -assembly:opennlp %OPENNLP_LIB_PATH%\opennlp-maxent-3.0.2-incubating.jar %OPENNLP_LIB_PATH%\jwnl-1.3.3.jar %OPENNLP_LIB_PATH%\opennlp-tools-1.5.2-incubating.jar
>            Reporter: Ofer Tal
>            Assignee: Martin Wiesner
>            Priority: Major
>             Fix For: 2.2.0
>
>
> After parsing a sentence with opennlp.tools.parser.Parse.parse() some (many) Parse children do not have the correct parent set.
> Details:
> given a Parse node in the tree (let's assume it is in a variable named p)
> When iterating over the Parse[] returned by p.getChildren(), checking p.equals(children[i].getParent()) returns false in many, if not all of the nodes.
> More background --
> to create the parse tree, I used the code:
> {code}
> opennlp.tools.parser.Parse p = new opennlp.tools.parser.Parse(parseSentence, new opennlp.tools.util.Span(0, parseSentence.Length), opennlp.tools.parser.AbstractBottomUpParser.INC_NODE, 1, null);
>             // create a parse object for each token and add it to the parent
>             int start = 0;
>             foreach (string token in tokenizedSentence)
>             {
>                 {
>                     opennlp.tools.parser.Parse tokenParse = new opennlp.tools.parser.Parse(parseSentence,
>                                                             new opennlp.tools.util.Span(start, start + token.Length),
>                                                             opennlp.tools.parser.AbstractBottomUpParser.TOK_NODE,
>                                                             0,
>                                                             0);                    
>                     p.insert(tokenParse);
>                     start += token.Length + 1;
>                 }
>             }
>             // fetch 1 possible parse trees
>             opennlp.tools.parser.Parse[] parses = parser.parse(p, 1);
> {code}



--
This message was sent by Atlassian Jira
(v8.20.10#820010)