issue #259 automaticly get tables to db hash computation
The old database during the update got injected hash computation algorithm from the newer version. The same algorithm is used for old and new versions of Hive Fork Manager. This way in case of a bug in the old version it is possible to repair hash computation with the new hfm version.
Hash is computed for all tables in the hafd schema (except state providers tables), not only for those included in the list forgotten by devs.
Only state providers used in db (registered in contexts) impact the whole db hash.
hash db functions without schema name parameter There is no sense to pass schema parameter to hash db computation, at the end, there was a list of tables to take into computation, and all of they belong to the hafd schema.
Remove hafd.table_schema table, the hash is used only to check if an update is possible, it is not a database hash because some elements are omitted during its computations and its value cannot be used to determine if schemas of two databases with hfm are similar. hafd.table_schema table blocked the total separation of hive_updata schema elements from the rest of the hive fork manager
Merge request reports
Activity
assigned to @Ickiewicz
added 1 commit
- ae1724ea - issue #259 (closed): compute database hash base on all tables on schema hafd
added 2 commits
- de08a2fe - hash db functions without schema name parameter
- 2cdc8286 - issue #259 (closed): compute database hash base on all tables on schema hafd
added 1 commit
- c02b24a7 - during update old and new version of db use the same hash computation
added 1 commit
- b86d63ad - use drop function if exists instead create or replace fior hash functions
added 1 commit
- 2d55bbc2 - all update functions in one file in schema hive_update
added 1 commit
- c524e404 - all update functions in one file in schema hive_update
added 1 commit
- a09ebdc5 - all update functions in one file in schema hive_update
added 1 commit
- 4110a96e - all update functions in one file in schema hive_update
