Обсуждение: Re: [BUGS] about test_parser installation failure problem(PostgreSQL in 9.5.0)?
"=?utf-8?B?6Zas6Zas44Kk44G1?=" <paaple@foxmail.com> writes: > test_parser install is ok (postgresql 9.2.4) > but at (postgresql 9.5.0) failure? Yes, we moved test_parser and some other only-useful-for-testing modules from contrib to src/test/modules, which means they won't get installed in standard installations. regards, tom lane
Re: [BUGS] about test_parser installation failure problem(PostgreSQL in 9.5.0)?
От
Michael Paquier
Дата:
On Sat, Jan 16, 2016 at 2:03 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote: > "=?utf-8?B?6Zas6Zas44Kk44G1?=" <paaple@foxmail.com> writes: >> test_parser install is ok (postgresql 9.2.4) >> but at (postgresql 9.5.0) failure? > > Yes, we moved test_parser and some other only-useful-for-testing modules > from contrib to src/test/modules, which means they won't get installed in > standard installations. Additional note: on Windows when code is compiled with MSVC, they are installed. -- Michael
Michael Paquier <michael.paquier@gmail.com> writes: > On Sat, Jan 16, 2016 at 2:03 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote: >> Yes, we moved test_parser and some other only-useful-for-testing modules >> from contrib to src/test/modules, which means they won't get installed in >> standard installations. > Additional note: on Windows when code is compiled with MSVC, they are installed. Really? Surely that's a bug. regards, tom lane
Re: [GENERAL] Re: [BUGS] about test_parser installation failure problem(PostgreSQL in 9.5.0)?
От
Michael Paquier
Дата:
On Sun, Jan 17, 2016 at 12:28 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote: > Michael Paquier <michael.paquier@gmail.com> writes: >> On Sat, Jan 16, 2016 at 2:03 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote: >>> Yes, we moved test_parser and some other only-useful-for-testing modules >>> from contrib to src/test/modules, which means they won't get installed in >>> standard installations. > >> Additional note: on Windows when code is compiled with MSVC, they are installed. > > Really? Surely that's a bug. We are doing so to simplify the work of vcregress.pl where an instance needs to be usually in place before running the tests. -- Michael