* Kevin Wolf (kw...@redhat.com) wrote:
> Instead of going through the QemuOpts-based parser, go directly from the
> given option string to ChardevOptions. This doesn't only avoid legacy
> code, but it also simplifies the implementation.
>
> Signed-off-by: Kevin Wolf <kw...@redhat.com>
OK, from HMP I think
Acked-by: Dr. David Alan Gilbert <dgilb...@redhat.com>
I'm assuming there's no change in the escaping from you extracting it
from the qdict and then printfing it back to go throguh the parser?
Dave
> ---
> monitor/hmp-cmds.c | 27 +++++++++------------------
> 1 file changed, 9 insertions(+), 18 deletions(-)
>
> diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
> index a6a6684df1..0244068de8 100644
> --- a/monitor/hmp-cmds.c
> +++ b/monitor/hmp-cmds.c
> @@ -1793,34 +1793,25 @@ void hmp_chardev_add(Monitor *mon, const QDict *qdict)
> void hmp_chardev_change(Monitor *mon, const QDict *qdict)
> {
> const char *args = qdict_get_str(qdict, "args");
> - const char *id;
> + const char *id = qdict_get_str(qdict, "id");
> + char *optstr;
> Error *err = NULL;
> - ChardevBackend *backend = NULL;
> + ChardevOptions *options = NULL;
> ChardevReturn *ret = NULL;
> - QemuOpts *opts = qemu_opts_parse_noisily(qemu_find_opts("chardev"), args,
> - true);
> - if (!opts) {
> - error_setg(&err, "Parsing chardev args failed");
> - goto end;
> - }
>
> - id = qdict_get_str(qdict, "id");
> - if (qemu_opts_id(opts)) {
> - error_setg(&err, "Unexpected 'id' parameter");
> - goto end;
> - }
> + optstr = g_strdup_printf("%s,id=%s", args, id);
>
> - backend = qemu_chr_parse_opts(opts, &err);
> - if (!backend) {
> + options = qemu_chr_parse_cli_str(optstr, &err);
> + if (!options) {
> goto end;
> }
>
> - ret = qmp_chardev_change(id, backend, &err);
> + ret = qmp_chardev_change(options->id, options->backend, &err);
>
> end:
> + g_free(optstr);
> qapi_free_ChardevReturn(ret);
> - qapi_free_ChardevBackend(backend);
> - qemu_opts_del(opts);
> + qapi_free_ChardevOptions(options);
> hmp_handle_error(mon, err);
> }
>
> --
> 2.28.0
>
--
Dr. David Alan Gilbert / dgilb...@redhat.com / Manchester, UK
