Developer forum

Hi Anna

I do not think that is a change - it has been like that all the time.

The search is based on Lucene - and it contains some built in features. One of them is a list of built in stop words that you would usually not index. "A" is not only a letter, but in english language also a common word like "a search engine" that in a regular search case would be messing up a lot of search results.

Take a look at this section: https://doc.dynamicweb.com/documentation-9/repositories/indexing/custom-fields#4786

You can create your own stopword list by creating a txt file in  /Files/System/Repositories/stopwords.txt based on what is in the default list of stopwords. And then remove the ones you do not want to be used as stopwords - maybe all of them.

BR Nicolai

Another thing.

The fields containing customer name in the index is probably marked as "Analyzed". This also means that & will not be indexed at all. This will also be the case when using another stopword list. Understanding "Analyzed".

When the name of the customer ("A & A sheet metal products, inc") is indexed, it is broken in to 'terms'. Also when you search, the search input, i.e. "a&a" is also broken into terms.

If the field is analyzed, and you are using the default stopwords list, these terms are going into the index:

• sheet
• metal
• products
• inc

And when searching for a&a, nothing is left when analyzed because a is a stopword and & is a non-word removed by the analyzer. So nothing is actually searched.

If you index "A & A sheet metal products, inc" without an analyzer, there will be only one term:

• A & A sheet metal products, inc

And then searching for "a&a" will also yield no results as it is not the same. If you search for "a & a" and you are using a startswith operator, the customer will be found.