Adriano dos Santos Fernandes
Adriano dos Santos Fernandes
> The problem is that at some point merges become too hard. You should know it with your schemas branch. Of course, but this is software development. INTL refactor and...
> The "problem" is that the grammar for computed columns uses `non_array_type` > where it probably should be using `domain_or_non_array_type_name`. Not simple as that. The computed value/source is in RDB$FIELDS,...
Currently it's not accepted, so it's better to document it if not already.
> Doing so might potentially deoptimize other queries, though. Yes. DROP SCHEMA, for example. I can agree to add extra indexes with the order of the fields inversed, but to...
> Did you use AI to generate this post? Why are you question this?
> I wouldn't rush to implement operators not included in the SQL standard. Some DBMSs are now paying the price for initially poor decisions. For example, MySQL once made the...
What is the build commands that works? I tried this and it failled: ``` cmake -S . -B build-cmake ... -- Configuring done (9.2s) CMake Error at src/CMakeLists.txt:997 (add_executable): Cannot...
Suggested test below. Should create a v6 database, backup and restore into v5. Result should match v6. ``` set term !; create schema s1! create schema s2! create or alter...
Firebird 5.0.2.1613 Windows 32-bit "MERGE INTO" gives wrong returning value with the use of sequence
I think this is well know issue and more than just a MERGE thing. This example increments the sequence two times. ``` INSERT INTO TEST (TESTID, "YEAR", NAME) select *...
> NOTE: This seems to be a generic parser issue: for example `select 0alias from rdb$database` has the same problem. This should not be accepted either, but results in a...