This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH 00/12] remove some cleanups using a cleanup function
- From: Joel Brobecker <brobecker at adacore dot com>
- To: Tom Tromey <tom at tromey dot com>
- Cc: gdb-patches at sourceware dot org
- Date: Fri, 11 Jan 2019 10:56:41 +0400
- Subject: Re: [PATCH 00/12] remove some cleanups using a cleanup function
- References: <20190109033426.16062-1-tom@tromey.com>
> This is the series to introduce a new cleanup_function class, that can
> be used in situations where a simple function call is needed to do
> some cleanup.
>
> As I mentioned in the other thread, I was (and still am) uncertain as
> to whether this is a good idea. My worry is just that this will be
> used in ways leading to unclear code. (You can judge for yourself
> whether this series has already entered that territory...)
>
> A couple of rules I used when deciding when to use the new class:
>
> * There should not be too many uses of the existing cleanup (one or
> two). If there are more than several it should probably be a
> bespoke class.
>
> * Any lambdas should capture only a small number of things (one or
> maybe two), and only capture by value should be used.
>
> Hopefully I actually followed these.
>
> I wonder if the new class should take a std::function rather than a
> gdb::function_view. Some of the changes needs a bit of extra code to
> deal with the latter.
>
> Also I wonder if it should be possible to default-construct a
> cleanup_function, and then set it later. This would eliminate some
> uses of gdb::optional; and seemed maybe reasonable given that the
> class already has to maintain a (sort of-) validity flag.
>
> This series includes a couple of cleanup-related comment fixes for
> things I happened to notice along the way.
I don't know C++ well enough to have an opinion on the general
questions asked here, but FWIW, the patch series in itself seems
like a nice improvement already (compared to before).
--
Joel