ElasticSearch ID constraints

Question

For ElasticSearch document IDs, are there any character constraints or restrictions?

I am really interested to know if forward slash '/' would cause any issues here? I have some news feeds which I would like to index. The problem is that the database that contains this data that has UID set to the URL of the news feed. Don't ask me why it was designed this way because I haven't gotten a clue.

I want to use the same identifier(URL) for ElasticSearch document. I have successfully used GUIDs, alphanumeric and numeric characters without problems.

If I can't what would be the best workaround - should i encode the entire url?

Thanks

score 23 · Accepted Answer · answered Aug 30 '12 at 02:59

23

There are no constraints. Forward slashes can be used. But in order to use such id in the REST API, it has to be url encoded:

$ curl -XPUT "localhost:9200/id-test-index/rec/1+1%2F2" -d '{"field" : "one and a half"}'

answered Aug 30 '12 at 02:59

imotov

28,277
3
90
82

this is what i found out after way too much messing around .. thanks! – Gabbar Aug 30 '12 at 17:02
Actually in the earlier versions of Elasticsearch, around v1.7, some aggregations failed calculating correctly if you used colons or underscore in your _id – WoodyDRN Dec 20 '18 at 12:29

score 1 · Answer 2 · edited May 21 '20 at 01:05

1

~~Slash "/" URL encoding is broken: https://github.com/elasticsearch/elasticsearch/issues/2903~~

Slash "/" is no longer broken. This issue has been fixed.

edited May 21 '20 at 01:05

aioobe

413,195
112
811
826

answered May 15 '13 at 01:41

Ryan R.

2,478
5
27
48

11

for new visitors: this has been fixed a long time ago – glaed Apr 12 '16 at 15:33

ElasticSearch ID constraints

2 Answers2