The pgrminclude problem

From: Peter Geoghegan <peter(at)2ndquadrant(dot)com>
To: Bruce Momjian <bruce(at)momjian(dot)us>
Cc: PG Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: The pgrminclude problem
Date: 2012-08-16 15:48:34
Message-ID: CAEYLb_UWaRy+XsLY3+9S7doxf3sWdD2Qewa1zCC1Kuw_9zTTuw@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

I found a tool that Google authored that attempts to solve the same
problems as pgrminclude does in our codebase. It's called "include
what you use", and is based on Clang. The project is hosted here:

http://code.google.com/p/include-what-you-use/

I'm not suggesting that we should start using this tool instead of
pgrminclude, because it has enough caveats of its own, and is mostly
written with Google's C++ codebase in mind. However, it's worth being
aware of. There is a useful analysis of the general problem in the
README - "Why IWYU is Difficult" (you can probably just skip the
extensive analysis of C++ templates as they relate to the problem
there).

The tool is authored by Craig Silverstein, Google's director of
technology. If he believes that IWYU is a difficult problem, well, it
probably is.

--
Peter Geoghegan http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training and Services

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Bruce Momjian 2012-08-16 15:56:19 Re: The pgrminclude problem
Previous Message Bruce Momjian 2012-08-16 15:45:31 Re: [COMMITTERS] pgsql: In docs, change a few cases of "not important" to "unimportant".