lookup join docs #124531
lookup join docs #124531
Conversation
georgewallace
added
>docs
|
Pinging @elastic/es-docs (Team:Docs) |
|
@leemthompo @alex-spies somehow I accidently rebased some things and messed up my old branch. It was too much to go and determine the fix so I just moved stuff over to this new PR. Apologize for the confusion. I think I have addressed all the feedback so please take another look. |
There was a problem hiding this comment.
I think this is in a nice state now! I have mostly minor remaining comments.
As discussed, we probably want to follow this up and reference examples from .csv-spec files directly. We should probably also then add an example that shows how name conflicts between columns in the lookup index and the existing table are resolved.
| | 10094 | 4 | Quenya | null| | ||
| | 10095 | 5 | null | Atlantis| |
There was a problem hiding this comment.
Thank you for including the example from the docs. In a follow-up, maybe we want to construct a nicer example for the docs? (We used to place such examples in docs.csv-spec.) Then readers won't have to wonder why we include Atlantis or the elven language of Quenya :D
There was a problem hiding this comment.
If you think it would be cleaner, I just just replace those values with something more realistic @alex-spies
There was a problem hiding this comment.
👍 on having better sample data that ties to the existing sample data in Kibana (which can have its own separate section).
| | LOOKUP JOIN service_owners ON service_id | ||
| ``` | ||
|
|
||
| In case of name collisions, the newly created columns will override existing columns. |
There was a problem hiding this comment.
Also, we should provide an example here if possible.
There was a problem hiding this comment.
@alex-spies I was unable to find a proper example for that. Can you point me in the right direction?
There was a problem hiding this comment.
There are examples in the tests (we sometimes internally call the behavior "shadowing", so all tests with "shadowing" in their name deal with this behavior.)
However, these tests are super constructed to reveal the details of the behavior - I don't think they would flow very well in this document.
The same is true for the example that illustrates left-join behavior; that made sense as a test but may not be an intuitive example for esql-lookup-join.md.
I think we'd have to construct nicer examples for this, maybe in a follow up.
georgewallace
force-pushed
the
esql_lookup_join
branch
from
b5fe3c4 to
2ad2296
Compare
|
cc @craigtaverner's re how your auto-generation work will intersect with the syntax reference part of this |
Co-authored-by: Alexander Spies <alexander.spies@elastic.co>
This was initially part of elastic#124531 but got lost during the manual backport.
This was initially part of #124531 but got lost during the manual backport.
This was initially part of elastic#124531 but got lost during the manual backport.
No description provided.