Re: Is a modern build system acceptable for older platforms

Поиск
Список
Период
Сортировка
От Yuriy Zhuravlev
Тема Re: Is a modern build system acceptable for older platforms
Дата
Msg-id CANiD2e-DvaK0ZpFWx4ki_gLywCq7L8-5VX+esS1ca6LfqUu8Rw@mail.gmail.com
обсуждение исходный текст
Ответ на Re: Is a modern build system acceptable for older platforms  (Craig Ringer <craig@2ndquadrant.com>)
Список pgsql-hackers
Дерево обсуждения
Which is nice

OK

 Again, nice
 
Yeah, that's nice

I already use CMake

I'd do it, personally

I suppose we have no one silver bullet reason to change autoconf+make to CMake but it's cumulative impression.

Also, I suppose this thread started to resolve at least one small question. It's should be something like voting, but I see here only a few people and your and Tom's
answers very strange, you can't say definitely yes or no and you more thinking about another people who not exist here. 

----
Some specific anwers:

Again, nice, but does that solve a real current problem?

This is main reason why PGXS on windows not exist, also it's solve problems with differents between MSVC versions and future releases.  

Yeah, that's nice, but again, what're the real world benefits?

it's just convinient, for easy work with Xcode for example.  

Also, most people working on PostgreSQL are probably less bothered by
the terminal.

It's another one reson why windows users and students don't want to hack Postgres. 

Yep. FWIW, I already use CMake for some PostgreSQL extensions because
of PGXS limitations and Windows support.

And now, it can be problem if your postgres build by mingw or llvm for example. (even under MSVC you can use clang now) 
Enviroment start change quickly and it's too much efforts to support all this by yourself. 

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Peter Geoghegan
Дата:
Сообщение: Re: Local partitioned indexes and pageinspect
Следующее
От: Yuriy Zhuravlev
Дата:
Сообщение: Re: Is a modern build system acceptable for older platforms