diff options
| author | Peter Eisentraut <peter@eisentraut.org> | 2021-04-07 07:49:27 +0200 |
|---|---|---|
| committer | Peter Eisentraut <peter@eisentraut.org> | 2021-04-07 08:05:55 +0200 |
| commit | dd13ad9d39a1ba41cf329b6fe408b49be57c7b88 (patch) | |
| tree | e75e26cae720ad2f4a0e1d8ef071ae197924c45a /src/backend/access/gist/gistbuild.c | |
| parent | 0b5e8245283eef67e88fb5380836cdc2c743d848 (diff) | |
| download | postgresql-dd13ad9d39a1ba41cf329b6fe408b49be57c7b88.tar.gz postgresql-dd13ad9d39a1ba41cf329b6fe408b49be57c7b88.zip | |
Fix use of cursor sensitivity terminology
Documentation and comments in code and tests have been using the terms
sensitive/insensitive cursor incorrectly relative to the SQL standard.
(Cursor sensitivity is only relevant for changes made in the same
transaction as the cursor, not for concurrent changes in other
sessions.) Moreover, some of the behavior of PostgreSQL is incorrect
according to the SQL standard, confusing the issue further. (WHERE
CURRENT OF changes are not visible in insensitive cursors, but they
should be.)
This change corrects the terminology and removes the claim that
sensitive cursors are supported. It also adds a test case that checks
the insensitive behavior in a "correct" way, using a change command
not using WHERE CURRENT OF. Finally, it adds the ASENSITIVE cursor
option to select the default asensitive behavior, per SQL standard.
There are no changes to cursor behavior in this patch.
Discussion: https://www.postgresql.org/message-id/flat/96ee8b30-9889-9e1b-b053-90e10c050e85%40enterprisedb.com
Diffstat (limited to 'src/backend/access/gist/gistbuild.c')
0 files changed, 0 insertions, 0 deletions