merge top-level of tap and regress test specifications, shouldn't be two separate arrays with the same information at the top #31
Description
Right now regression, isolation, tap tests are specified separately as array elements of regress_tests, isolation_tests, tap_tests. Sometimes one part of the code uses all three types of tests, see e.g.: https://github.com/anarazel/postgres/blob/meson/contrib/test_decoding/meson.build#L11
The same boilerplate is repeated for each test type:
regress_tests += {
'name': 'test_decoding',
'sd': meson.current_source_dir(),
'bd': meson.current_build_dir(),
...
isolation_tests += {
'name': 'test_decoding',
'sd': meson.current_source_dir(),
'bd': meson.current_build_dir(),
...
tap_tests += {
'name': 'test_decoding',
'sd': meson.current_source_dir(),
'bd': meson.current_build_dir(),Instead we should have one tests array, and then sub-elements below for each type of test. Perhaps something like
tests += {
'name': 'test_decoding',
'sd': meson.current_source_dir(),
'bd': meson.current_build_dir(),
regress: { sql : [ 'ddl', ...], ... },
isolation: { specs: [...]},
...Metadata
Metadata
Assignees
Labels
No labels