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.

Lucene analyser - How it works and how to configure it

sergio

Hi all.

Some questions about the lucene analyzer:
1- What kind of files does Lucene analyze during uploading?;
2- Is there any way to configure the analyzer in order to avoid content indexing for some kinds of file?
3- Is it possible to modify the rules for tokenization? For example, is it possible to avoid lucene tokenizing the string "hello word" into two different tokens "hello" and "word" in order to consider the "hello word" as a unique token?

Many thanks to anyone who will be pleasured do reply some useful exaplanations.

All the best,

Sergio

Find more posts tagged with

Comments

andy

Hi

1) All that can be converted to text in some way
2) Make sure there is not conversion to text
3) You can set properties to be untokenized in the DD - they are then indexed "as is"

Andy

sergio

Hi.

That's all clear, but with an exception, the following.

As I can understand, Lucene tries to extract text in any way from files. As a consequence, from an XML file probably it will extract a huge amount of text (but I don't know in which way it extract useful text) that will be indexed in a second time. This is a natural behaviour for a CMS.

Unfortunately, I have some files (.CEL, .DAT, .XML) I would like not to be indexed at all by Lucene. Is there any way to control in Alfresco this behaviour? My problem is not only to limit tokenization to entire phases, but also to avoid that Lucene tries to extract text and index it.

Any possibility of performing this task?

Thanks for your further explanations.

All the best,

Sergio