From: Kevin Buettner <kevinb@redhat.com>
To: Tom Tromey <tom@tromey.com>
Cc: gdb-patches@sourceware.org
Subject: Re: [RFC 2/5] Introduce the "critical" testsuite command
Date: Sat, 15 Apr 2023 16:35:45 -0700 [thread overview]
Message-ID: <20230415162429.4cf48de9@f37-zws-nv> (raw)
In-Reply-To: <20230415151336.4104455-3-tom@tromey.com>
Hi Tom,
Overall, I like this idea, but I think I found an easily fixable problem
in the implementation of the "critical" command - see below.
On Sat, 15 Apr 2023 09:13:33 -0600
Tom Tromey <tom@tromey.com> wrote:
> This patch introduces a new "critical" command for the test suite.
> Unlike require, "critical" can handle multiple arguments and a variety
> of return value conventions. It is implemented as a Tcl namespace
> ensemble for this reason.
>
> This version of the patch provides a few such subcommands. Currently
> these are all implemented via a helper proc, but there's no
> requirement that this always be true.
> ---
> gdb/testsuite/lib/gdb.exp | 28 ++++++++++++++++++++++++++++
> 1 file changed, 28 insertions(+)
>
> diff --git a/gdb/testsuite/lib/gdb.exp b/gdb/testsuite/lib/gdb.exp
> index a1808c699e9..0ae40bceba2 100644
> --- a/gdb/testsuite/lib/gdb.exp
> +++ b/gdb/testsuite/lib/gdb.exp
> @@ -9287,6 +9287,34 @@ proc require { args } {
> }
> }
>
> +#
> +# The "critical" command is used to invoke some code that must
> +# succeed. If the code in question fails, "unsupported" is logged and
> +# the command's caller will return.
> +#
> +# This is implemented as an ensemble namespace so that individual
> +# sub-commands can handle varying return values correctly.
> +#
> +namespace eval critical {
> + proc _make_wrapper {name bad_result} {
> + proc $name {args} [format {
> + set result [uplevel [list %s {*}$args]]
> + if {$result == %s} {
> + unsupported "error adding gdb index"
This message makes sense for ensure_gdb_index, but I don't think
it makes sense for the others. Should a suitable message to print
here be passed to _make_wrapper?
> + return -code return 0
> + }
> + } $name $bad_result]
> + namespace export $name
> + }
> +
> + _make_wrapper ensure_gdb_index -1
> + _make_wrapper prepare_for_testing_full -1
> + _make_wrapper prepare_for_testing -1
> + _make_wrapper runto 0
> +
> + namespace ensemble create
> +}
> +
> # Wait up to ::TIMEOUT seconds for file PATH to exist on the target system.
> # Return 1 if it does exist, 0 otherwise.
>
> --
> 2.39.2
>
next prev parent reply other threads:[~2023-04-15 23:35 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-15 15:13 [RFC 0/5] Add new "critical" command to test suite Tom Tromey
2023-04-15 15:13 ` [RFC 1/5] Bug fix in prepare_for_testing Tom Tromey
2023-04-15 15:13 ` [RFC 2/5] Introduce the "critical" testsuite command Tom Tromey
2023-04-15 23:35 ` Kevin Buettner [this message]
2023-04-15 15:13 ` [RFC 3/5] Use "critical ensure_gdb_index" in gdb.rust Tom Tromey
2023-04-15 15:13 ` [RFC 4/5] Use "critical prepare_for_testing" " Tom Tromey
2023-04-15 15:13 ` [RFC 5/5] Use "critical runto" " Tom Tromey
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230415162429.4cf48de9@f37-zws-nv \
--to=kevinb@redhat.com \
--cc=gdb-patches@sourceware.org \
--cc=tom@tromey.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).