Bug 47731

Summary: Word Extractor considers text copied from some website as an embedded object
Product: POI Reporter: Gitu <gi.bijlani>
Component: HWPFAssignee: POI Developers List <dev>
Severity: enhancement CC: gi.bijlani, yegor
Priority: P2    
Version: 3.2-FINAL   
Target Milestone: ---   
Hardware: PC   
OS: Windows Server 2003   
Attachments: This attachment contains text copied from a web page

Description Gitu 2009-08-24 22:50:21 UTC

I have copied some text from some web page and pasted that in a word document.
Now, when I use WordExtractor to extract the content of that document, then complete content gets extracted but the summary information comes multiple times.

After investigating I came to know that each part in that document is considered as an embedded object and hence for each embedded object, summary is getting extracted ie. same value is coming those many times.

I also wanted to know if considering an HTML content as an Embedded object is a valid behaviour.

I have attached a document which can reproduce the scenario.

Many thanks in advance,
Comment 1 Yegor Kozlov 2009-08-31 10:05:14 UTC
You seem to forget to attach the file. Please re-attach.

Comment 2 Gitu 2009-08-31 22:04:02 UTC
Created attachment 24197 [details]
This attachment contains text copied from a web page

Attached the document!!

Comment 3 Sergey Vladimirov 2011-07-24 18:55:07 UTC
Text extractor does extract all text from document, but not from included OLE objects. Those objects can be actually other Word documents and/or Excel stylesheet and/or vector images.

There is can be an enchancement to TextExtractor to allow extracting text from OLE objects, but surely current behaviour not a bug. Changing importance to "enchancement".
Comment 4 Sergey Vladimirov 2011-08-09 12:43:12 UTC
Fixed/improved in r1155337, will be part of 3.8-beta4 release.