lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ian Ribas (JIRA)" <>
Subject [jira] [Commented] (LUCENE-6582) SynonymFilter should generate a correct (or, at least, better) graph
Date Mon, 22 Jun 2015 22:30:01 GMT


Ian Ribas commented on LUCENE-6582:

Ok, I think I have an idea on how to do this, but it will definitely need some thinking, so
it will probably take a while.

bq. I think that's fine, I think better correctness trumps the added buffering cost.

I completely agree.

bq. I think if we fix syn filter here to produce the correct graph, we should also insert
a "sausagizer" phase that turns this graph back into a sausage for indexing?

I think Robert also commented something on these lines in his answer to my email. I think
I understand the general idea of what that means, but I would certainly appreciate some guidance,
when the time comes. I'll focus on producing a correct graph first.

This also means that maybe I'll need changes on the test validations, since we might run into
conditions that are considered wrong now. Specially regarding offsets (start and end) and
their relation to position lengths. But I'll see what I can do about that too.

bq. However, if you apply syn filter at search time, we could fix query parsers to possibly
"do the right thing" here

I was planning taking a shot at that too, once this part is finished. To make the solution
more complete. And, again, I'll certainly appreciate ideas when the time comes.

> SynonymFilter should generate a correct (or, at least, better) graph
> --------------------------------------------------------------------
>                 Key: LUCENE-6582
>                 URL:
>             Project: Lucene - Core
>          Issue Type: Bug
>            Reporter: Ian Ribas
>         Attachments: LUCENE-6582.patch, LUCENE-6582.patch, after.png, after2.png, after3.png,
> Some time ago, I had a problem with synonyms and phrase type queries (actually, it was
elasticsearch and I was using a match query with multiple terms and the "and" operator, as
better explained here:
> That issue led to some work on Lucene: LUCENE-6400 (where I helped a little with tests)
and  LUCENE-6401. This issue is also related to LUCENE-3843.
> Starting from the discussion on LUCENE-6400, I'm attempting to implement a solution.
Here is a patch with a first step - the implementation to fix "SynFilter to be able to 'make
positions'" (as was mentioned on the [issue|]).
In this way, the synonym filter generates a correct (or, at least, better) graph.
> As the synonym matching is greedy, I only had to worry about fixing the position length
of the rules of the current match, no future or past synonyms would "span" over this match
(please correct me if I'm wrong!). It did require more buffering, twice as much.
> The new behavior I added is not active by default, a new parameter has to be passed in
a new constructor for {{SynonymFilter}}. The changes I made do change the token stream generated
by the synonym filter, and I thought it would be better to let that be a voluntary decision
for now.
> I did some refactoring on the code, but mostly on what I had to change for may implementation,
so that the patch was not too hard to read. I created specific unit tests for the new implementation
({{TestMultiWordSynonymFilter}}) that should show how things will be with the new behavior.

This message was sent by Atlassian JIRA

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message