Possible bugs when using highlights in deep nested documents

GyllingSW · January 18, 2017, 12:16pm

Hi all,

I would like someone to help verify if I hit a bug or not, when using deep nested structures analyzed with the "danish" analyzer.

For steps to reproduce see this gist:

gist.github.com

https://gist.github.com/GyllingSW/0a7d85702916c2b5a19d7dcc3f7eca4c

gistfile1.txt

#
# Steps to reproduce using sence or kibana dev tools
#

PUT /test_index
{
   "settings": {
      "number_of_shards": 1,
      "number_of_replicas": 0
   },

This file has been truncated. show original

Basic idea is, that my client has a structure similar to this, and the highlight feature isn't working as expected, so to reproduce I created the gist above.

If I change for the "english" analyzer everything is working as expected. I have tested this using Elasticsearch versions 2.4.X and 5.1.X both with similar results.

GyllingSW · January 23, 2017, 11:12am

I discovered, that the issue mostly lies in the default danish stop words list, thus ES is actually doing exactly as expected. To solve the issue above - a re-index will have to happen ...

Consider this as solved

system · February 20, 2017, 11:12am

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.