Georg Traar
Georg Traar
> Not important, as you like, if it's more clear with a separate issue, you could open it. I created a new issue: https://github.com/crate/crate/issues/13578 and close this
> Can you give some more context on concrete use cases? - Long-term storage / Archiving - for efficient long-term storage of data, while still keeping metadata information and being...
> Referencing the columns by ordinal is an alternative - although a bit less convenient: Which is even more error-prone when adjusting queries. But overall this is more about convenience...
This seems to be resolved with v0.4.4 ``` >>> sqlparse.split('''CREATE OR REPLACE FUNCTION public.bug() RETURNS text ... LANGUAGE sql ... AS $func$ ... select $string$ Run this query: SELECT 123;...
ello @c224600, Could you provide additional details to help us understand the issue better? I'm interested in knowing whether this problem occurs across all tables or if it's isolated to...
> Jmx Exporter has been updated to expose the new metrics with: [crate/jmx_exporter#68](https://github.com/crate/jmx_exporter/pull/68) Thanks, will plan to test it asap 👍
While the example query could probably be written as ... ```sql SELECT url, max(request_duration) request_duration FROM logs GROUP BY url ORDER BY request_duration DESC; ``` ... for more complex scenarios...
It seems like while on can define explicit casts for objects including sub-columns this information is dropped: ### Object Literal ```sql SELECT pg_typeof({x=1}['x']); -- undefined SELECT pg_typeof(({x=1}::OBJECT AS (x LONG))['x']);...
The described workaround (`CREATE TABLE / INSERT INTO`) has some flaws: - primary keys on the materialisation table are required for `INSERT INTO` - during updates, there are inconsistencies in...
a better approach might be to use a partitioned table with a view on top of it, which gets updated only after the insert into a new partition was successful....