qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Pierrick Bouvier <pierrick.bouvier@linaro.org>
To: "Daniel P. Berrangé" <berrange@redhat.com>
Cc: qemu-devel@nongnu.org, alex.bennee@linaro.org,
	stefanha@redhat.com, peter.maydell@linaro.org,
	Markus Armbruster <armbru@redhat.com>,
	richard.henderson@linaro.org, pbonzini@redhat.com,
	jsnow@redhat.com, philmd@linaro.org, thuth@redhat.com,
	Michael Roth <michael.roth@amd.com>
Subject: Re: [PATCH 03/13] qobject/qlit: allow to hide dict or list entries
Date: Thu, 8 May 2025 13:25:10 -0700	[thread overview]
Message-ID: <49dc014b-bfd1-43de-8d0c-c916e178432a@linaro.org> (raw)
In-Reply-To: <aBy94N6JSwDj53mb@redhat.com>

On 5/8/25 7:21 AM, Daniel P. Berrangé wrote:
> On Wed, May 07, 2025 at 04:14:33PM -0700, Pierrick Bouvier wrote:
>> We add a new .hidden field to qlit entries, which gets ignored when
>> creating the associated QObject.
>> By default .hidden is 0, so it means the entry is visible. This way,
>> only potentially hidden elements need to be assigned.
> 
> IMHO this feels like a somewhat dubious concept to have in the
> qobject code, as it is quite specialized to a single use case.
> A more general purpose approach would be to have some mechanism
> for cloning while applying a data filter, though I admit that
> may be more tedious to actually use.
>

It's the least worse solution I found, for conditionally defining the 
schema.
Trying to dynamically create the qlit stuff with individual .push() was 
a nightmare in code generator, so I dropped this idea.

>>
>> Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
>> ---
>>   include/qobject/qlit.h | 12 ++++++++++++
>>   qobject/qlit.c         | 10 ++++++++--
>>   2 files changed, 20 insertions(+), 2 deletions(-)
>>
>> diff --git a/include/qobject/qlit.h b/include/qobject/qlit.h
>> index c0676d5daf2..3b66c22013c 100644
>> --- a/include/qobject/qlit.h
>> +++ b/include/qobject/qlit.h
>> @@ -28,25 +28,37 @@ struct QLitObject {
>>           QLitDictEntry *qdict;
>>           QLitObject *qlist;
>>       } value;
>> +    bool hidden;
>>   };
>>   
>>   struct QLitDictEntry {
>>       const char *key;
>>       QLitObject value;
>> +    bool hidden;
>>   };
>>   
>>   #define QLIT_QNULL \
>>       { .type = QTYPE_QNULL }
>>   #define QLIT_QBOOL(val) \
>>       { .type = QTYPE_QBOOL, .value.qbool = (val) }
>> +#define QLIT_QBOOL_HIDDEN(val, cond) \
>> +    { .type = QTYPE_QBOOL, .value.qbool = (val), .hidden = (cond) }
>>   #define QLIT_QNUM(val) \
>>       { .type = QTYPE_QNUM, .value.qnum = (val) }
>> +#define QLIT_QNUM_HIDDEN(val, cond) \
>> +    { .type = QTYPE_QNUM, .value.qnum = (val), .hidden = (cond) }
>>   #define QLIT_QSTR(val) \
>>       { .type = QTYPE_QSTRING, .value.qstr = (val) }
>> +#define QLIT_QSTR_HIDDEN(val, cond) \
>> +    { .type = QTYPE_QSTRING, .value.qstr = (val), .hidden = (cond) }
>>   #define QLIT_QDICT(val) \
>>       { .type = QTYPE_QDICT, .value.qdict = (val) }
>> +#define QLIT_QDICT_HIDDEN(val, cond) \
>> +    { .type = QTYPE_QDICT, .value.qdict = (val), .hidden = (cond) }
>>   #define QLIT_QLIST(val) \
>>       { .type = QTYPE_QLIST, .value.qlist = (val) }
>> +#define QLIT_QLIST_HIDDEN(val, cond) \
>> +    { .type = QTYPE_QLIST, .value.qlist = (val), .hidden = (cond) }
>>   
>>   bool qlit_equal_qobject(const QLitObject *lhs, const QObject *rhs);
>>   
>> diff --git a/qobject/qlit.c b/qobject/qlit.c
>> index a44f47eaa57..7b372c5ebaa 100644
>> --- a/qobject/qlit.c
>> +++ b/qobject/qlit.c
>> @@ -90,6 +90,8 @@ bool qlit_equal_qobject(const QLitObject *lhs, const QObject *rhs)
>>   
>>   QObject *qobject_from_qlit(const QLitObject *qlit)
>>   {
>> +    g_assert(!qlit->hidden);
>> +
>>       switch (qlit->type) {
>>       case QTYPE_QNULL:
>>           return QOBJECT(qnull());
>> @@ -102,7 +104,9 @@ QObject *qobject_from_qlit(const QLitObject *qlit)
>>           QLitDictEntry *e;
>>   
>>           for (e = qlit->value.qdict; e->key; e++) {
>> -            qdict_put_obj(qdict, e->key, qobject_from_qlit(&e->value));
>> +            if (!e->hidden) {
>> +                qdict_put_obj(qdict, e->key, qobject_from_qlit(&e->value));
>> +            }
>>           }
>>           return QOBJECT(qdict);
>>       }
>> @@ -111,7 +115,9 @@ QObject *qobject_from_qlit(const QLitObject *qlit)
>>           QLitObject *e;
>>   
>>           for (e = qlit->value.qlist; e->type != QTYPE_NONE; e++) {
>> -            qlist_append_obj(qlist, qobject_from_qlit(e));
>> +            if (!e->hidden) {
>> +                qlist_append_obj(qlist, qobject_from_qlit(e));
>> +            }
>>           }
>>           return QOBJECT(qlist);
>>       }
>> -- 
>> 2.47.2
>>
> 
> With regards,
> Daniel



  reply	other threads:[~2025-05-08 20:25 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-07 23:14 [PATCH 00/13] single-binary: make QAPI generated files common Pierrick Bouvier
2025-05-07 23:14 ` [PATCH 01/13] qapi: introduce 'runtime_if' for QAPI json Pierrick Bouvier
2025-05-08  6:53   ` Philippe Mathieu-Daudé
2025-05-08 20:22     ` Pierrick Bouvier
2025-05-15  4:39   ` Markus Armbruster
2025-05-15 15:42     ` Pierrick Bouvier
2025-05-07 23:14 ` [PATCH 02/13] qapi/introspect: generate schema as a QObject directly Pierrick Bouvier
2025-05-07 23:14 ` [PATCH 03/13] qobject/qlit: allow to hide dict or list entries Pierrick Bouvier
2025-05-08 14:21   ` Daniel P. Berrangé
2025-05-08 20:25     ` Pierrick Bouvier [this message]
2025-05-07 23:14 ` [PATCH 04/13] qapi/introspect: hide fields in schema Pierrick Bouvier
2025-05-07 23:14 ` [PATCH 05/13] qapi/commands: register commands conditionally Pierrick Bouvier
2025-05-07 23:14 ` [PATCH 06/13] qapi/visit: hide fields in JSON marshalling Pierrick Bouvier
2025-05-07 23:14 ` [PATCH 07/13] qapi: add access to qemu/target-info.h Pierrick Bouvier
2025-05-08  6:57   ` Philippe Mathieu-Daudé
2025-05-07 23:14 ` [PATCH 08/13] qemu/target-info: implement missing helpers Pierrick Bouvier
2025-05-08  6:40   ` Philippe Mathieu-Daudé
2025-05-08 20:30     ` Pierrick Bouvier
2025-05-07 23:14 ` [PATCH 09/13] qapi: transform target specific 'if' in runtime checks Pierrick Bouvier
2025-05-08  6:44   ` Philippe Mathieu-Daudé
2025-05-08 14:40   ` Daniel P. Berrangé
2025-05-08 20:48     ` Pierrick Bouvier
2025-05-10  6:57     ` Markus Armbruster
2025-05-13  0:36       ` Pierrick Bouvier
2025-05-13  7:08         ` Markus Armbruster
2025-05-13 22:52           ` Pierrick Bouvier
2025-05-14  7:13             ` Markus Armbruster
2025-05-14 16:54               ` Pierrick Bouvier
2025-05-14 14:09   ` Markus Armbruster
2025-05-14 16:50     ` Pierrick Bouvier
2025-05-07 23:14 ` [PATCH 10/13] qapi: add weak stubs for target specific commands Pierrick Bouvier
2025-05-08  6:57   ` Philippe Mathieu-Daudé
2025-05-08 20:33     ` Pierrick Bouvier
2025-05-07 23:14 ` [PATCH 11/13] qapi: make all generated files common Pierrick Bouvier
2025-05-08  6:57   ` Philippe Mathieu-Daudé
2025-05-07 23:14 ` [PATCH 13/13] [ANNEX] build/qapi: after series Pierrick Bouvier
2025-05-07 23:33 ` [PATCH 00/13] single-binary: make QAPI generated files common Pierrick Bouvier

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=49dc014b-bfd1-43de-8d0c-c916e178432a@linaro.org \
    --to=pierrick.bouvier@linaro.org \
    --cc=alex.bennee@linaro.org \
    --cc=armbru@redhat.com \
    --cc=berrange@redhat.com \
    --cc=jsnow@redhat.com \
    --cc=michael.roth@amd.com \
    --cc=pbonzini@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=philmd@linaro.org \
    --cc=qemu-devel@nongnu.org \
    --cc=richard.henderson@linaro.org \
    --cc=stefanha@redhat.com \
    --cc=thuth@redhat.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 NNTP newsgroup(s).