[
https://issues.apache.org/jira/browse/TIKA-2963?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16949772#comment-16949772
]
Tim Allison commented on TIKA-2963:
-----------------------------------
The xlsx parser uses SAX for the sheet parsing but relies on DOM/beans for the
SharedStrings table, which can be a, um, problem. There’s been work on POI, not
yet merged, to store that table in H2 for a lower memory footprint.
If you unzip the file, how big is it? How big is its shared strings table.
Obviously, there could be other areas for improvement...
> Tika在抽取.xlsx类型的大文件时出现OOM错误
> --------------------------
>
> Key: TIKA-2963
> URL: https://issues.apache.org/jira/browse/TIKA-2963
> Project: Tika
> Issue Type: Improvement
> Components: core
> Affects Versions: 1.20
> Reporter: Feng Jiao Jiang
> Priority: Major
> Attachments: demo.java
>
>
> 对于docx和pptx类型的文件，Tika可配置SAX解析器来提高抽取性能。但是Tika在抽取.xlsx类型的大文件时仍会出现OOM错误，我暂时没有从官方找到解决方案，下面附上自己的代码，也是基于SAX解析器的解决方案，代码可根据实际情况进行参数调优，多有不足之处，大家批评指正，谢谢
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
