Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Pedro Alves <palves@redhat.com>
To: Tom Tromey <tom@tromey.com>, gdb-patches@sourceware.org
Subject: Re: [RFA 08/11] Use string and unique_xmalloc_ptr in mi-main.c
Date: Thu, 28 Sep 2017 09:57:00 -0000	[thread overview]
Message-ID: <106d764a-be15-161a-355e-323e7e28a036@redhat.com> (raw)
In-Reply-To: <20170912185736.20436-9-tom@tromey.com>

On 09/12/2017 07:57 PM, Tom Tromey wrote:
> Change a couple of spots in mi-main.c to use std::string or
> unique_xmalloc_ptr.  unique_xmalloc_ptr is used here where the string
> is writeable; I generally prefer to pretend that string is read-only,
> perhaps not for a good resaon.

Yes, I think not for a good reason.  :-)  You can
pass &str[0] to execute_command get access to the underlying
modifiable raw string.  See b064640146bb for example.

> diff --git a/gdb/mi/mi-main.c b/gdb/mi/mi-main.c
> index c8c4a97..0147fb9 100644
> --- a/gdb/mi/mi-main.c
> +++ b/gdb/mi/mi-main.c
> @@ -2264,41 +2264,31 @@ mi_execute_cli_command (const char *cmd, int args_p, const char *args)
>  {
>    if (cmd != 0)
>      {
> -      struct cleanup *old_cleanups;
> -      char *run;
> +      gdb::unique_xmalloc_ptr<char> run;
>  
>        if (args_p)
> -	run = xstrprintf ("%s %s", cmd, args);
> +	run.reset (xstrprintf ("%s %s", cmd, args));
>        else
> -	run = xstrdup (cmd);
> +	run.reset (xstrdup (cmd));
>        if (mi_debug_p)
>  	/* FIXME: gdb_???? */
>  	fprintf_unfiltered (gdb_stdout, "cli=%s run=%s\n",
> -			    cmd, run);
> -      old_cleanups = make_cleanup (xfree, run);
> -      execute_command (run, 0 /* from_tty */ );
> -      do_cleanups (old_cleanups);
> -      return;
> +			    cmd, run.get ());
> +      execute_command (run.get (), 0 /* from_tty */ );
>      }
>  }
>  
>  void
>  mi_execute_async_cli_command (const char *cli_command, char **argv, int argc)
>  {
> -  struct cleanup *old_cleanups;
> -  char *run;
> +  gdb::unique_xmalloc_ptr<char> run;
>  
>    if (mi_async_p ())
> -    run = xstrprintf ("%s %s&", cli_command, argc ? *argv : "");
> +    run.reset (xstrprintf ("%s %s&", cli_command, argc ? *argv : ""));
>    else
> -    run = xstrprintf ("%s %s", cli_command, argc ? *argv : "");
> -  old_cleanups = make_cleanup (xfree, run);
> -
> -  execute_command (run, 0 /* from_tty */ );
> +    run.reset (xstrprintf ("%s %s", cli_command, argc ? *argv : ""));
>  
> -  /* Do this before doing any printing.  It would appear that some
> -     print code leaves garbage around in the buffer.  */
> -  do_cleanups (old_cleanups);
> +  execute_command (run.get (), 0 /* from_tty */ );
>  }
>  

I think the above would look better with std::string + string_printf.


>  	struct trace_state_variable *tsv;
> @@ -2822,10 +2808,8 @@ mi_cmd_trace_frame_collected (const char *command, char **argv, int argc)
>  
>  	if (tsv != NULL)
>  	  {
> -	    tsvname = (char *) xrealloc (tsvname, strlen (tsv->name) + 2);
> -	    tsvname[0] = '$';
> -	    strcpy (tsvname + 1, tsv->name);
> -	    uiout->field_string ("name", tsvname);
> +	    std::string tsvname = std::string ("$") + tsv->name;
> +	    uiout->field_string ("name", tsvname.c_str ());

How about replacing this string building + field_string with a single
call to:

	    uiout->field_fmt ("name", "$%s", tsv->name);

Thanks,
Pedro Alves


  reply	other threads:[~2017-09-28  9:57 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-09-12 18:57 more cleanup removal, particularly in MI Tom Tromey
2017-09-12 18:57 ` [RFA 06/11] Change some gdb_* functions to use a std::string out parameter Tom Tromey
2017-09-28  9:42   ` Pedro Alves
2017-09-28 19:58     ` Tom Tromey
2017-09-12 18:57 ` [RFA 07/11] Use gdb::byte_vector in mi_cmd_data_write_memory_bytes Tom Tromey
2017-09-28  9:46   ` Pedro Alves
2017-09-12 18:57 ` [RFA 08/11] Use string and unique_xmalloc_ptr in mi-main.c Tom Tromey
2017-09-28  9:57   ` Pedro Alves [this message]
2017-09-29  1:42     ` Tom Tromey
2017-09-29 10:23       ` Pedro Alves
2017-09-12 18:57 ` [RFA 11/11] Change captured_mi_execute_command to use scoped_restore Tom Tromey
2017-09-28 10:35   ` Pedro Alves
2017-09-12 18:57 ` [RFA 09/11] Use std::set in mi-main.c Tom Tromey
2017-09-28 10:10   ` Pedro Alves
2017-10-02 13:05     ` Simon Marchi
2017-10-03 11:21   ` Simon Marchi
2017-10-03 11:39     ` Tom Tromey
2017-09-12 18:57 ` [RFA 10/11] Use a std::vector for ada_exceptions_list Tom Tromey
2017-09-28 10:20   ` Pedro Alves
2017-09-12 18:57 ` [RFA 02/11] Remove cleanups from mi_cmd_break_insert_1 Tom Tromey
2017-09-28  9:24   ` Pedro Alves
2017-09-28 19:57     ` Tom Tromey
2017-09-29  1:40     ` Tom Tromey
2017-09-29 10:21       ` Pedro Alves
2017-09-12 18:57 ` [RFA 04/11] Don't copy a string in mi_cmd_disassemble Tom Tromey
2017-09-28  9:40   ` Pedro Alves
2017-09-12 18:57 ` [RFA 05/11] Remove unused declaration Tom Tromey
2017-09-28  9:40   ` Pedro Alves
2017-09-12 18:59 ` [RFA 01/11] Remove make_cleanup_defer_target_commit_resume Tom Tromey
2017-09-28  9:17   ` Pedro Alves
2017-09-12 19:03 ` [RFA 03/11] Remove cleanups from mi-cmd-var.c Tom Tromey
2017-09-28  9:36   ` Pedro Alves
2017-09-29  1:40     ` Tom Tromey
2017-09-29 10:22       ` Pedro Alves
2017-09-23 16:15 ` more cleanup removal, particularly in MI 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=106d764a-be15-161a-355e-323e7e28a036@redhat.com \
    --to=palves@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