From: Paolo Bonzini <pbonzini@redhat.com>
To: Markus Armbruster <armbru@redhat.com>,
Daniel Berrange <berrange@redhat.com>,
qemu-devel <qemu-devel@nongnu.org>
Subject: Re: [PATCH 02/11] keyval: introduce keyval_merge
Date: Wed, 16 Jun 2021 12:49:17 +0200 [thread overview]
Message-ID: <4e2ae7a5-0549-7851-88f2-124d8c324416@redhat.com> (raw)
In-Reply-To: <87a6nqkxvz.fsf@dusky.pond.sub.org>
On 16/06/21 10:38, Markus Armbruster wrote:
>
> ... and both values are the same other QType (QTYPE_QNULL, QTYPE_QNUM,
> QTYPE_QSTRING, QTYPE_QBOOL): overwrite.
>
> Why is overwrite restricted to same QType? Is there no need for
> overwriting say a string with a number? Hmm, I guess it's okay, because
> keyval_parse() only ever produces QTYPE_QSTRING scalars. May be worth a
> comment, preferably in a function contract.
Good point, I can add an assert that the only scalars are QTYPE_QSTRING.
> However, some users look at*all* pairs. This provides "repeated keys
> build up a list" semantics.
>
> Example 4: repeated option key builds a list
>
> $ qemu-system-x86_64 -S -spice tls-port=12345,tls-channel=main,tls-channel=display
>
> This fails for me because I don't have Spice set up (and know basically
> nothing about it), but a working variation of it should configure*two*
> channels, not one: the second tls-channel= does*not* overwrite the
> first one, it configures another channel.
>
> Since -spice is a merge_lists option, this should be the case for
> multiple -spice, too. Merging the two options with keyval_merge() would
> not preserve that behavior, I'm afraid.
>
> QemuOpts is... complicated.
Right, QemuOpts is complicated and keyval a bit less so, which makes
keyval not applicable to every case: see -object where we had to resort
to the OptsVisitor. For those cases where repeated keys build up a
list, it will not be possible to switch them to keyval.
The other remarks are mostly cosmetic, so there's not much to say on them.
Paolo
next prev parent reply other threads:[~2021-06-16 10:50 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-10 13:35 [PATCH 00/11] vl: compound properties for machines Paolo Bonzini
2021-06-10 13:35 ` [PATCH 01/11] qom: export more functions for use with non-UserCreatable objects Paolo Bonzini
2021-06-10 13:49 ` Daniel P. Berrangé
2021-06-10 13:35 ` [PATCH 02/11] keyval: introduce keyval_merge Paolo Bonzini
2021-06-16 8:38 ` Markus Armbruster
2021-06-16 10:49 ` Paolo Bonzini [this message]
2021-06-10 13:35 ` [PATCH 03/11] keyval: introduce keyval_parse_into Paolo Bonzini
2021-06-10 13:53 ` Daniel P. Berrangé
2021-06-16 14:09 ` Markus Armbruster
2021-06-16 18:05 ` Paolo Bonzini
2021-06-10 13:35 ` [PATCH 04/11] vl: switch -M parsing to keyval Paolo Bonzini
2021-06-10 13:35 ` [PATCH 05/11] qemu-option: remove now-dead code Paolo Bonzini
2021-06-10 13:54 ` Daniel P. Berrangé
2021-06-10 13:35 ` [PATCH 06/11] machine: move dies from X86MachineState to CpuTopology Paolo Bonzini
2021-06-10 13:55 ` Daniel P. Berrangé
2021-06-10 13:35 ` [PATCH 07/11] machine: move common smp_parse code to caller Paolo Bonzini
2021-06-10 14:34 ` Daniel P. Berrangé
2021-06-10 13:35 ` [PATCH 08/11] machine: add error propagation to mc->smp_parse Paolo Bonzini
2021-06-10 14:35 ` Daniel P. Berrangé
2021-06-10 13:35 ` [PATCH 09/11] machine: pass QAPI struct " Paolo Bonzini
2021-06-10 14:37 ` Daniel P. Berrangé
2021-06-10 13:35 ` [PATCH 10/11] machine: reject -smp dies!=1 for non-PC machines Paolo Bonzini
2021-06-10 14:37 ` Daniel P. Berrangé
2021-06-10 13:35 ` [PATCH 11/11] machine: add smp compound property Paolo Bonzini
2021-06-10 14:41 ` Daniel P. Berrangé
2021-06-14 22:57 ` [PATCH 00/11] vl: compound properties for machines no-reply
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=4e2ae7a5-0549-7851-88f2-124d8c324416@redhat.com \
--to=pbonzini@redhat.com \
--cc=armbru@redhat.com \
--cc=berrange@redhat.com \
--cc=qemu-devel@nongnu.org \
/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).