[COCOON-267] Lucene index/search broken, wrong contentType each doc so ignored

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 2.0.5-dev (Current CVS)
• Component/s: - Components: Sitemap
• Environment: Operating System: All; Platform: All
• Bugzilla Id: 8932
• Assignee: Unassigned
• Reporter: David Crossley
• Created: 09/May/02 13:22
• Updated: 16/Feb/11 13:58
• Resolved: 11/Oct/06 18:16

Description

Christian Zoffoli wrote:
>
> I have just downloaded the last CVS ...and I have found that lucene
> cant do any search on the indexed pages. The indexing process seems to
> work fine but "Index Statistics " says Total Count of Documents 0.

I looked at my core.log and found many entries like this ...
-----
DEBUG (2002-05-04) 17:28.33:193 [core.search.lucene](/cocoon/search/create)
HttpProcessor[8080][3]/SimpleLuceneXMLIndexerImpl: Ignoring
http://localhost:8080/cocoon/documents/mail-archives.html?cocoon-view=content
(text/html)
-----

I see from the code that the "Ignoring" message arises when
the contentType is null or not an allowed contentType. The code
builds a small list of allowedContentType (xml, xhtml). CVS shows
that the code and the samples pages have not changed for a while.

So why would the contentType returned by cocoon-view have suddenly
changed to be html? According to the sample/search/sitemap.xmap
the cocoon-view=content does in fact serialise to xml. So why would
the contentType be reporting as html? Perhaps there has been a change
in that arena.

--David