tree 91ebd192986b29f9038e74b102e2ed4dd9670f56
parent f6e6af7c2d286135cfafea330aaa61839c149cce
author Rafsun Masud <rafsun82@gmail.com> 1700611055 -0800
committer GitHub <noreply@github.com> 1700611055 -0800
gpgsig -----BEGIN PGP SIGNATURE-----
 
 wsBcBAABCAAQBQJlXUPwCRBK7hj4Ov3rIwAAtgUIABXpxM7GcxjR8ry2AP3ioltl
 GSX20I/1W2KrXke7cqAmPVlIaYmeP3fwPIxIGa4GTD2gPbSKfEXp57H6IpGfTZKV
 L8CRkYKga9gk/vaYBSA8wLlJTF/aTW8xlE5x96qwGbxmtFe5+ZyK+wKeJU00TDSH
 jaKfl5IeWBBpkjJFxaxhD9obDnK5SBz4hPNID9vXAtWjBXHif8Gx3BHpIeB6rS3F
 IOwV7EUU7V3+5tbWBM/8yuERZpaI1tOAI1SW2ak6Is2s39BjiNU7SscPqDVw0vjd
 rCvfi4rm8WliZ65dtN7Kf5iDiarbRnA9kNz7EeM9AN/tz16zOfNxUYcdnDATiNY=
 =jHei
 -----END PGP SIGNATURE-----
 

Fix issue #1398 - SET followed by DELETE does not delete (#1412)

In the DELETE executor, tuples were being locked\deleted using the command ID
returned by GetCurrentCommandId(). However, the original tuples were retrieved
using a command ID stored in the estate.

To fix it- before the retrieval of a tuple, estate command IDs are set using
GetCurrentCommandId() so it matches if the tuple is to be deleted later.

Additional changes:
-------------------
Fixed an incorrect cypher_delete test. The query
`MATCH (n1)-[e]->() DELETE n1, e RETURN n1` should be able to delete n1 and e
without requiring DETACH DELETE.

TODO:
-----
It may be a good idea to audit the executors for any inconsistent use of
command IDs.