Why don't you create a synonym for + that expands to your customers
product name that includes the plus? You can even have your FE do
this sort of replacement BEFORE submitting to Solr.
Thanks,
Matt Weber
On Sep 14, 2009, at 11:42 AM, AHMET ARSLAN wrote:
Thanks Ahmet,
Thats excellent, thanks :) I may have to increase the
gramsize to take into account other possible uses but i can
now read around these filters to make the adjustments.
With regard to WordDelimiterFilterFactory. Is there a way
to place a delimiter on this filter to still get most of its
functionality without it absorbing the + signs?
Yes you are right, preserveOriginal="1" will causes the original
token to be indexed without modifications.
Will i loose a lot of 'good' functionality by removing it?
It depends of your input data. It is used to break one token into
subwords.
Like: "Wi-Fi" -> "Wi", "Fi" and "PowerShot" -> "Power", "Shot"
If you input data set contains such words, you may need it.
But I think just to make last character searchable, using
NGramFilter(s) is not an optimal solution. I don't know what type of
dataset you have but, I think using separate two fields (with
different types) for that is more suitable. One field will contain
actual data itself. The other will hold only the last character(s).
You can achieve this by a copyField or programatically during
indexing. The type of the field lastCharsField will be using
EdgeNGramFilter so that only last character of token(s) will pass
that filter.
During searching you will search those two fields:
originalField:\+ OR lastCharsField:\+
The query lastCharsField:\+ will return you all the products ending
with +.
Hope this helps.
