FileUtil.extractText is not correctly working for Documents in docx format (Word 2007+) files

• Type: Bug
• Status: Closed
• Resolution: Fixed
• Affects Version/s: 6.2.X EE, 7.0.X, 7.1.X, Master
• Fix Version/s: 6.2.X EE, 7.0.0 DXP FP79, 7.0.10.11 DXP SP11, 7.0.X, 7.1.10 DXP FP10, 7.1.10.2 SP2, 7.1.3 CE GA4, 7.1.X, Master
• Component/s: Documents & Media
• Labels:
  • backend_test
  • ee-ts
  • liferay-dxp-70-sp11
  • liferay-dxp-71-sp2
  • liferay-fixpack-de-79-7010
  • liferay-fixpack-dxp-10-7110
  • liferay-fixpack-portal-171-6210
• Branch Version/s:
  7.1.x, 7.0.x, 6.2.x
• Backported to Branch:
  Committed
• Fix Priority: 3
• Sprint: 021 - Spearow
• Git Pull Request: https://github.com/brianchandotcom/liferay-portal/pull/69390
• Bug Type: Regression Bug

FileUtil.extractText is not correctly working for Documents in docx format (Word 2007+) files

It is handled as they were a zip file, unziping it and indexing internal xml files
 
Steps to reproduce

1. Create some test files with docx, odt, doc and rtf formats with content:

   This is a test

2. You can also use JIRA ticket attached files
3. Create a site and attach that files to document library
4. Get indexed data from elasticsearch, opening following URL: (note: set your groupId)

   http://localhost:9200/_search?q=%2BgroupId:<<YOUR_GROUPID>>+%2BentryClassName:com.liferay.document.library.kernel.model.DLFileEntry&pretty

1. Review "content_en_US" field of each returned document:
   • Expected behavior: all documents have "This is a test" text in "content_en_US" field
   • Wrong behavior: Test.docx document doesn't have "This is a test" text in "content_en_US" field and it have a different one instead. Test.odt, Test.doc and Test.rtf are fine