nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

As we transition to the new Hyland Community, some post links may be temporarily disrupted as we finalize redirects. We apologize for the inconvenience and are working to resolve this as quickly as possible. Details.

Show search results with excerpts (like Google)

nicolasraoul

I am dreaming of a search UI that would show a text excerpt for each search result.

Does this already exist in Alfresco?
Or do I have to develop a specific Dialog that would call Lucene?
Or maybe it is not even possible to do this with Lucene?

Thanks a lot!
Nicolas

Note:Because I am searching only text files, showing each matching line would be fine too.

Find more posts tagged with

Comments

zaizi

Yes this is doable. We've implemented this for a customer project. You can use Lucene for selecting best text excerpts and display it in Alfresco results.

DM me for details.
Ainga

nicolasraoul

I am actually asking this question because someone asked on the Japanese forum, and I did not know the answer so I asked on the English forum. Here is the original message: http://forums.alfresco.com/jp/viewtopic.php?f=4&t=530

Does it require development or is it just configuration/Spring?

Thanks a lot!
Nicolas

zaizi

Yes, it requires development. You can utilise Lucene to provide the hit highlighting functionality. See http://www.cocooncenter.org/articles/lucene.html.

However to enable to Lucene to do that you'll need to extend Alfresco through Java coding. For one Alfresco's lucene index does not store the complete text just the index. The text needs to be stored / generated so they can be displayed in the search results. This would depend on your specific requirements.

Ainga

nicolasraoul

Thanks a lot! 🙂

Nicolas

aymens

Hello,

I implemented it for Alfresco CE 3.4.d, wrote something about it and provided an svn patch for that.

If you're interested check this: http://aymen-s.blogspot.com/2011/03/implementing-search-hits-highlight-in.html

Regards,

nicolasraoul

Thanks aymens, that looks great!
Nicolas

zaizi

We tried this approach first and then changed it because the Lucene index was growing quite large in production. Instead we created a child association to the content and stored the text version in the repository. Similar to how renditions are stored.

Ainga

aymens

Hi Ainga,

That would require querying node service for each node to get content at LuceneResultSet level.

That may slow down search response time significantly, but if index size becomes too big to put up with, one should choose your solution for content storage.

Thanks,

aymens

Thanks aymens, that looks great!
Nicolas

You're welcome