GNU bug report logs - #52440
28.0.50; [PATCH] Quis custodiet ipsos custodes (sqlite3)

Previous Next

Full log

View this message in rfc822 format

From: Eli Zaretskii <eliz <at> gnu.org>
To: dick <dick.r.chiang <at> gmail.com>
Cc: 52440 <at> debbugs.gnu.org, larsi <at> gnus.org
Subject: bug#52440: 28.0.50; [PATCH] Quis custodiet ipsos custodes (sqlite3)
Date: Sun, 12 Dec 2021 14:01:23 +0200

> From: dick <dick.r.chiang <at> gmail.com>
> Cc: Lars Ingebrigtsen <larsi <at> gnus.org>,  52440 <at> debbugs.gnu.org
> Date: Sun, 12 Dec 2021 06:32:27 -0500
> 
> EZ> The amount of sqlite stuff in lisp.h is quite small...  So I tend
> EZ> to think a separate header is overkill.
> 
> Thinking about this in terms of number of bytes added to lisp.h is the
> wrong way to think about it.

If that's the only aspect, perhaps.  But it isn't the only aspect:
there are others, which I mentioned, and you elided.  Together with
them, the size is relevant to decide between the alternatives.

> I agree that Lisp_Process, while important, doesn't belong in lisp.h.
> That something as ancillary, third-party, and controversial as
> Lisp_Sqlite does get into lisp.h is a testimony about its author.

We are not dogmatic about this stuff.  This is C, not C++, so what
goes into what header is not subject to dogma.  And this isn't the
case that just files dealing with sqlite will include this header:
every source file that deals with Lisp objects will need it, so at
least conceptually it belongs to lisp.h.  Triggering unnecessary
rebuilds is an annoyance during development, so it's no less important
than purity of the headers.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.