This commit is contained in:
2020-05-20 16:48:40 +00:00
parent 42803bb0d8
commit 463084e99e
3 changed files with 28 additions and 13 deletions

2
.vscode/launch.json vendored
View File

@@ -51,7 +51,7 @@
"AYANOVA_FOLDER_BACKUP_FILES": "c:\\temp\\RavenTestData\\backupfiles",
"AYANOVA_METRICS_USE_INFLUXDB": "false",
"AYANOVA_SERVER_TEST_MODE":"true",
"AYANOVA_SERVER_TEST_MODE_SEEDLEVEL":"small",
"AYANOVA_SERVER_TEST_MODE_SEEDLEVEL":"huge",
"AYANOVA_SERVER_TEST_MODE_TZ_OFFSET":"-7",
"AYANOVA_BACKUP_PG_DUMP_PATH":"C:\\data\\code\\PostgreSQLPortable_12.0\\App\\PgSQL\\bin\\"

View File

@@ -6,19 +6,7 @@
}
working on: BACKUP finish routes.
Add integration tests as much as possible
maybe need whole other category of integration tests for ops that can be run seprately?
Then test with actual attachments to ensure it's working for that part as well
Upload backup file so can restore?
only works if we have automated restore which is doubtful at any point
todo: OPS routes (SERVER AND CLIENT)
- Backup, restore https://rockfish.ayanova.com/default.htm#!/rfcaseEdit/3369
- These need to be done fairly early on in order to have shit to play with for testing etc
todo: Test backup on Linux
- going to need to run in vm I think maybe
todo: Search indexing performance improvement and exception avoidance (Search.cs 828)
ON CONFLICT IDEA
@@ -60,3 +48,7 @@ todo: onboarding and default manager account password
- see what other programs do, like our forum software
todo: API docs, make separate page for datalists and remove from api-response-format.md doc but put a reference link to it there.
DEPLOY AND TEST
todo: Test backup on Linux
- going to need to run in vm I think maybe

View File

@@ -827,6 +827,29 @@ ON CONFLICT IDEA
{
if (!ExistingKeywordMatches.ContainsValue(KeyWord))
{
/*
NEW IDEA
todo: Search indexing performance improvement and exception avoidance (Search.cs 828)
ON CONFLICT IDEA
https://www.postgresql.org/docs/current/sql-insert.html#SQL-ON-CONFLICT
Idea: do the insert manually with the clause "on conflict do nothing"
if detect it hasn't inserted (conflict) trigger a fetch instead
like what is being done now but won't have the exception to deal with!!
ON CONFLICT IDEA
https://www.postgresql.org/docs/current/sql-insert.html#SQL-ON-CONFLICT
Idea: do the insert manually with the clause "on conflict do nothing"
if detect it hasn't inserted (conflict) trigger a fetch instead
like what is being done now but won't have the exception to deal with!!
//NEED TO BASELINE THIS SHIT
*/
//algorithm: Attempt to add it to the db and get the id, if it fails with the expected exception for a duplicate word insertion attempt, then immediately read back that word and handle it
//ATTEMPT TO ADD THE WORD TO THE SEARCHDICTIONARY