Re: Making C function declaration parameter names consistent with corresponding definition names - Mailing list pgsql-hackers

From Peter Geoghegan
Subject Re: Making C function declaration parameter names consistent with corresponding definition names
Date
Msg-id CAH2-Wzk7VatgOM0m9tpqfq_jtaYuYFK_RHoN+s=aM1Hf=xC9nQ@mail.gmail.com
Whole thread Raw
In response to Re: Making C function declaration parameter names consistent with corresponding definition names  (Peter Geoghegan <pg@bowt.ie>)
Responses Re: Making C function declaration parameter names consistent with corresponding definition names
List pgsql-hackers
On Mon, Sep 19, 2022 at 11:36 PM Peter Geoghegan <pg@bowt.ie> wrote:
> Attached revision v4 fixes those pg_dump patch items.
>
> It also breaks out the ecpg changes into their own patch.

I pushed much of this just now. All that remains to bring the entire
codebase into compliance is the ecpg patch and the pg_dump patch.
Those two areas are relatively tricky. But it's now unlikely that I'll
need to push a commit that makes relatively many CF patches stop
applying against HEAD -- that part is over.

Once we're done with ecpg and pg_dump, we can talk about the actual
practicalities of formally adopting a project policy on consistent
parameter names. I mostly use clang-tidy via my editor's support for
the clangd language server -- clang-tidy is primarily a linter, so it
isn't necessarily run in bulk all that often. I'll need to come up
with instructions for running clang-tidy from the command line that
are easy to follow.

I've found that the run_clang_tidy script (AKA run-clang-tidy.py)
works, but the whole experience feels hobbled together. I think that
we really need something like a build target for this -- something
comparable to what we do to support GCOV. That would also allow us to
use additional clang-tidy checks, which might be useful. We might even
find it useful to come up with some novel check of our own. Apparently
it's not all that difficult to write one from scratch, to implement
custom rules. There are already custom rules for big open source
projects such as the Linux Kernel, Chromium, and LLVM itself.

-- 
Peter Geoghegan



pgsql-hackers by date:

Previous
From: Mark Dilger
Date:
Subject: Re: why can't a table be part of the same publication as its schema
Next
From: James Coleman
Date:
Subject: Re: Auto explain after query timeout