[ https://issues.apache.org/jira/browse/OAK-7251?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16356632#comment-16356632 ]
Chetan Mehrotra commented on OAK-7251: -------------------------------------- This convention was a carry over of [Jackrabbit |https://github.com/apache/jackrabbit/blob/407bd5c7803e8518531559d3bf6fa480197e6341/jackrabbit-core/src/main/java/org/apache/jackrabbit/core/query/lucene/LazyTextExtractorField.java#L185] way of doing things!. +1 to change that now > BinaryTextExtractor should not ignore parse exception - they should at least > be logged at DEBUG in all cases > ------------------------------------------------------------------------------------------------------------ > > Key: OAK-7251 > URL: https://issues.apache.org/jira/browse/OAK-7251 > Project: Jackrabbit Oak > Issue Type: Improvement > Components: lucene > Reporter: Vikas Saurabh > Assignee: Vikas Saurabh > Priority: Major > Fix For: 1.10 > > > BinaryTextExtractor ignores missing library error like: > {noformat} > } catch (LinkageError e) { > // Capture and ignore errors caused by extraction libraries > // not being present. This is equivalent to disabling > // selected media types in configuration, so we can simply > // ignore these errors. > {noformat} > or > {noformat} > // Capture and report any other full text extraction problems. > // The special STOP exception is used for normal termination. > if (!handler.isWriteLimitReached(t)) { > {noformat} > We should at not skip these errors - some information should at least be > available at DEBUG. -- This message was sent by Atlassian JIRA (v7.6.3#76005)