On 2021-01-09 02:46, Michael Paquier wrote:
> +/* LCOV_EXCL_START */
> Does it really make sense to add those markers here? It seems to me
> that we would ignore any new coverage if regression tests based on
> older versions are added (we may really want to have such tests for
> more in-core extensions to be able to verify the portability of an
> extension, but that's not the job of this patch of course).
If we had a way to do such testing then we wouldn't need these markers.
But AFAICT, we don't.
> - elog(ERROR, "block 0 is a meta page");
> + ereport(ERROR,
> + (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
> + errmsg("block 0 is a meta page")));
> [...]
> + errmsg("block number %llu is out of range for relation \"%s\"",
> This does not follow the usual style for error reports that should not
> be written as full sentences? Maybe something like "invalid block
> number %u referring to meta page" and "block number out of range for
> relation %s: %llu"?
There are many error messages that say "[something] is out of range". I
don't think banning that would serve any purpose.