IMPORT INTO with default expressions and computed columns #8764
Conversation
|
This change is |
|
@pbardea - friendly ping to PTAL early next week :) I'd like to address reviews and merge this PR this upcoming week! |
v20.2/import-into.md
Outdated
| - `transaction_timestamp()` | ||
|
|
||
| - `random()` | ||
| - `gen_random_uuid` |
There was a problem hiding this comment.
nit: parens here to match the rest?
v20.2/import-into.md
Outdated
| - `unique_rowid()` | ||
|
|
||
| {{site.data.alerts.callout_info}} | ||
| Non-targeted columns with constant default expressions are not required to be nullable. |
There was a problem hiding this comment.
I'm not sure that we define what "targeted" columns are in the docs? Do we have a way of referring to the columns that are specified in an IMPORT INTO command?
Thinking about this more, do we usually have callouts for things that are not required? I think this behaviour would be the "expected" case, rather than an exceptional one.
There was a problem hiding this comment.
TFTR!
Reviewable status:
complete! 0 of 0 LGTMs obtained (waiting on @mwang1026 and @pbardea)
v20.2/import-into.md, line 98 at r2 (raw file):
Previously, pbardea (Paul Bardea) wrote…
nit: parens here to match the rest?
Done.
v20.2/import-into.md, line 102 at r2 (raw file):
Previously, pbardea (Paul Bardea) wrote…
I'm not sure that we define what "targeted" columns are in the docs? Do we have a way of referring to the columns that are specified in an IMPORT INTO command?
Thinking about this more, do we usually have callouts for things that are not required? I think this behaviour would be the "expected" case, rather than an exceptional one.
Yeah, that's a good point. I don't think we need the callout. Removed.
There was a problem hiding this comment.
Reviewable status:
Closes #8585.
Closes #8291.
Closes #8105.
Closes #8289.
Closes #8106.
Closes #7863.
Closes #8419.