Adds support for templates index pattern

[jalvz]

See motivation in #64 (comment)

It will require Kibana support, basically if this new property is true, the index pattern generated for the templates should be <type>-<dataset>.*-*
I will open a ticket in Kibana if this is accepted.

[elasticmachine]

💚 Build Succeeded

the below badges are clickable and redirect to their specific view in the CI or DOCS

Expand to view the summary

Build stats

• Build Cause: Pull request #102 updated

• • Start Time: 2021-01-18T14:00:19.763+0000

• Duration: 3 min 51 sec

• Commit: 09954e6

[ruflin]

@jalvz Can you share a bit more details on how this will be used?

[jalvz]

@ruflin you have been too fast :)
Backlink in description now

[mtojek]

Please modify the "good" test package, so we can be sure that the change works as intended.

[ruflin]

I'm currently blocking this PR as we should first complete the discussion in #64 (comment)

I like the simplicity of this change but I worry it has many side effects. On the Stack side we loose the knowledge if the templates / mappings / index patterns for a data stream are all aligned or if a dev might have just introduced an invalide index pattern.

[jalvz]

Updated as per #64 (comment)

[ruflin]

@skh @jfsiii @jen-huang Could you chime in from the Kibana side if this is feasible?
@mtojek @ycombinator Not happy with the name yet (see conversation in the linked issue), so if you have ideas for better names ...
@jalvz LGTM but lets wait on the feedback from the Kibana side to not merge this too soon.

[jalvz]

Filed Kibana issue: elastic/kibana#88307

[skh]

Is this only about the field index_patterns in an ES index template (this is what I guessed from the discussions), or also about Kibana index patterns (this is how this line reads without context)?

[ruflin]

It is only about the ES index patterns. It should not affect the Kibana ones as these are based on the types and include all datasets anyways.

[skh]

Then this description should be changed accordingly.

[jalvz]

Hey all, are there any more objections to this?

[simitt]

LGTM

[mtojek]

LGTM, but let's leave the final decision to @ruflin .

EDIT:

please rebase the branch against master and update the changelog.

[ruflin]

LGTM

@mtojek @ycombinator Side discussion: Should this go into https://github.com/elastic/package-spec/blob/master/versions/1/changelog.yml or where is the right place for this?

[mtojek]

@mtojek @ycombinator Side discussion: Should this go into https://github.com/elastic/package-spec/blob/master/versions/1/changelog.yml or where is the right place for this?

Yes, this is the right place.