tibia/querymanager/src/database_sqlite.cc, branch masterTibia 7.7 compatible query managerhttps://kaplar.net/tibia/querymanager/atom?h=master2026-02-09T17:25:21Zfix small issue with TransferHouses query2026-02-09T17:25:21Zfusion32marcopuzziello@gmail.com2026-02-09T17:25:21Zurn:sha1:edea08d11cc306955d8d732164ec383d37ea1f62change how guild data is stored2025-10-23T07:26:11Zfusion32marcopuzziello@gmail.com2025-10-23T07:26:11Zurn:sha1:27a0df43b33b712462ab5be65462162b9e4bcd29properly handle NULL values with postgres result helpers2025-10-18T06:40:40Zfusion32marcopuzziello@gmail.com2025-10-18T06:40:40Zurn:sha1:ebf536a7910f6e8d5bff569d2d717287ad3ba716modify how automatic SQLite patches/upgrades works + overall cleanup2025-10-18T05:35:24Zfusion32marcopuzziello@gmail.com2025-10-18T05:32:59Zurn:sha1:601ce32ee8d0cee32b66c87b0bfefa7424a5b8c8world startup/shutdown timestamps + adjust a few column names2025-10-17T19:12:31Zfusion32marcopuzziello@gmail.com2025-10-17T19:12:31Zurn:sha1:d92f642810dc3125515d64f953d2bee840de1cc1properly handle text encoding differences with QueryManager clients2025-10-17T02:04:00Zfusion32marcopuzziello@gmail.com2025-10-17T01:51:58Zurn:sha1:63705942a366d39308251549a018451808c20ac8move ResolveHostName outside of GetCharacterEndpoints2025-10-16T07:23:52Zfusion32marcopuzziello@gmail.com2025-10-16T07:23:52Zurn:sha1:01a6e8dbe42fb6571ad846bee5e1d8cc7c9fba5eprevent a few SQLite queries to fail when they affected no rows2025-10-16T05:00:22Zfusion32marcopuzziello@gmail.com2025-10-16T05:00:22Zurn:sha1:9825f08f2b46b147faa48e4d127c93694522e6d6
This was done with the old design in mind, but with the query
thread offloading, returning false is meant to signal a failure
in completing it. A connection lost with PostgreSQL or BUSY error
with SQLite, for example.
lower monotonic uptime resolution from MILLISECONDS to SECONDS2025-10-16T04:46:10Zfusion32marcopuzziello@gmail.com2025-10-16T04:35:10Zurn:sha1:95022ca95b1d2792bae447d90694dcc0d67b9de2
The monotonic uptime was used exclusively with caches and having
a resolution of SECONDS allows it to be stored as an int without
risk of wrapping (~68 years). Using MILLISECONDS meant that it
would wrap after ~25 days which is totally possible and EXPECTED.
Just as an example, the small test server I ran for about 1 month
had ZERO downtime on the QueryManager except for when I manually
restarted it. It was probably very close to wrapping when I took
it down.
implement all postgres queries2025-10-15T23:27:29Zfusion32marcopuzziello@gmail.com2025-10-15T23:27:29Zurn:sha1:e89144eb2d13532dd8802509a48c27e9934113b7