You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@tika.apache.org by "Hudson (Jira)" <ji...@apache.org> on 2021/06/28 21:29:00 UTC

[jira] [Commented] (TIKA-3456) LanguageDetector should try to respect hasEnoughText more intelligently

    [ https://issues.apache.org/jira/browse/TIKA-3456?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17370858#comment-17370858 ] 

Hudson commented on TIKA-3456:
------------------------------

UNSTABLE: Integrated in Jenkins build Tika ยป tika-branch1x-jdk8 #137 (See [https://ci-builds.apache.org/job/Tika/job/tika-branch1x-jdk8/137/])
TIKA-3456 -- LanguageDetector should chunk long strings and test for hasEnoughText. (tallison: [https://github.com/apache/tika/commit/4ba5fd7eb8b1a6ccc45fd773b73e6f809a652370])
* (edit) tika-core/src/main/java/org/apache/tika/language/detect/LanguageDetector.java


> LanguageDetector should try to respect hasEnoughText more intelligently
> -----------------------------------------------------------------------
>
>                 Key: TIKA-3456
>                 URL: https://issues.apache.org/jira/browse/TIKA-3456
>             Project: Tika
>          Issue Type: Task
>            Reporter: Tim Allison
>            Priority: Minor
>             Fix For: 1.27
>
>
> If a user calls LanguageDetector's detect(String txt) or addText(String txt), the full string is passed on to the subclasses and there is no check on "hasEnoughText()".  For large strings, LanguageDetector should break the string into smaller parts and check for hasEnoughText().



--
This message was sent by Atlassian Jira
(v8.3.4#803005)