[lucy-user] Possible highlighter bug when searching for duplicate [wordX wordX]?

[goran kent <go...@gmail.com>]

Hi,

I stumbled onto this one when searching for [business to business].

Source <TITLE>: ...Companies, Products, Trade Leads, Business Marketplace

'to' is a stopword which is ignored - no problem.
So the query then becomes [business business].  The highlighter then produces:

...Companies, Products, Trade Leads, <strong>Business</strong>
<strong>Marketp</strong>lace

I then spent some time chasing my tail trying to reduce things down to
a small reproducible unit, and finally decided to try searching for
any duplicate [wordX wordX], and sure enough it's reproducible with
all my indexes.

Is this a bug or a problem on my side?

regards

Re: [lucy-user] Possible highlighter bug when searching for duplicate [wordX wordX]?

[Marvin Humphrey <ma...@rectangular.com>]

On Tue, Oct 04, 2011 at 02:24:19PM +0200, goran kent wrote:
> ...Companies, Products, Trade Leads, <strong>Business</strong>
> <strong>Marketp</strong>lace

This has now been fixed both on both the main line and the 0.2.x branch.  It
will be in our next release.

Marvin Humphrey

Re: [lucy-user] Possible highlighter bug when searching for duplicate [wordX wordX]?

[Marvin Humphrey <ma...@rectangular.com>]

On Tue, Oct 04, 2011 at 02:24:19PM +0200, goran kent wrote:

> ...Companies, Products, Trade Leads, <strong>Business</strong>
> <strong>Marketp</strong>lace

> Is this a bug or a problem on my side?

Definitely a bug in the loop of Highlighter_highlight_excerpt() in
Highlighter.c.  Please open a ticket.

Thanks for the clear explanation.

Marvin Humphrey