Re: [PATCH v5 15/24] monitor: introduce monitor_cur_is_hmp() helper

Daniel P. Berrangé <[email protected]> writes:

> Note that this is not simply the inverse of monitor_cur_is_qmp(),
> as both helpers require that monitor_cur() is first non-NULL.
>
> Reviewed-by: Richard Henderson <[email protected]>
> Signed-off-by: Daniel P. Berrangé <[email protected]>
> ---
>  include/monitor/monitor.h      |  1 +
>  monitor/monitor.c              | 10 ++++++++++
>  stubs/monitor-core.c           |  6 ++++++
>  tests/unit/test-util-sockets.c |  1 +
>  4 files changed, 18 insertions(+)
>
> diff --git a/include/monitor/monitor.h b/include/monitor/monitor.h
> index 296690e1f1..9c71e6cf3c 100644
> --- a/include/monitor/monitor.h
> +++ b/include/monitor/monitor.h
> @@ -16,6 +16,7 @@ extern QemuOptsList qemu_mon_opts;
>  Monitor *monitor_cur(void);
>  Monitor *monitor_set_cur(Coroutine *co, Monitor *mon);
>  bool monitor_cur_is_qmp(void);
> +bool monitor_cur_is_hmp(void);
>  
>  void monitor_init_globals(void);
>  void monitor_init_globals_core(void);
> diff --git a/monitor/monitor.c b/monitor/monitor.c
> index 6dc5a7016d..b81cc7d2ed 100644
> --- a/monitor/monitor.c
> +++ b/monitor/monitor.c
> @@ -116,6 +116,16 @@ bool monitor_cur_is_qmp(void)
>      return cur_mon && monitor_is_qmp(cur_mon);
>  }
>  
> +/**
> + * Is the current monitor, if any, a HMP monitor?

"an HMP"?  Not a native speaker...

> + */
> +bool monitor_cur_is_hmp(void)
> +{
> +    Monitor *cur_mon = monitor_cur();
> +
> +    return cur_mon && !monitor_is_qmp(cur_mon);
> +}
> +

The only use at the end of this series looks like this:

    Monitor *cur = NULL;

    if (monitor_cur_is_hmp()) {
        cur = monitor_cur();
    } else {
        // some stderr code
    }

Meh.  Could do

    Monitor *cur = monitor_cur();

    if (monitor_cur_is_qmp(cur)) {
        cur = NULL;
        // some stderr code
    }

Or with a helper returning the HMP monitor, we could have something like

    cur = monitor_cur_hmp();

    if (!cur) {
        // some stderr code
    }

Doesn't mirror monitor_cur_is_qmp() then.  But I just pointed out in
reply to PATCH 11 that we don't actually need monitor_cur_is_qmp()>

Ideas, not demands.

>  /**
>   * Is @mon is using readline?
>   * Note: not all HMP monitors use readline, e.g., gdbserver has a
> diff --git a/stubs/monitor-core.c b/stubs/monitor-core.c
> index a7c32297c9..674211f48f 100644
> --- a/stubs/monitor-core.c
> +++ b/stubs/monitor-core.c
> @@ -7,6 +7,12 @@ Monitor *monitor_cur(void)
>      return NULL;
>  }
>  
> +bool monitor_cur_is_hmp(void)
> +{
> +    /* since monitor_cur() above returns NULL, this can't be true */

Maybe

       /* We can't have a monitor, let alone an HMP monitor */

Up to you.

> +    return false;
> +}
> +
>  Monitor *monitor_set_cur(Coroutine *co, Monitor *mon)
>  {
>      return NULL;
> diff --git a/tests/unit/test-util-sockets.c b/tests/unit/test-util-sockets.c
> index ee66d727c3..4b7f408902 100644
> --- a/tests/unit/test-util-sockets.c
> +++ b/tests/unit/test-util-sockets.c
> @@ -74,6 +74,7 @@ int monitor_get_fd(Monitor *mon, const char *fdname, Error **errp)
>  Monitor *monitor_cur(void) { return cur_mon; }
>  Monitor *monitor_set_cur(Coroutine *co, Monitor *mon) { abort(); }
>  int monitor_vprintf(Monitor *mon, const char *fmt, va_list ap) { abort(); }
> +bool monitor_cur_is_hmp(void) { return false; }

Fine.  abort() would also be fine.  Matter of taste.

>  
>  #ifndef _WIN32
>  static void test_socket_fd_pass_name_good(void)