All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jean Delvare <khali@linux-fr.org>
To: lm-sensors@vger.kernel.org
Subject: Re: [lm-sensors] [PATCH 6/8] sensord: Refactoring of loadConfig()
Date: Mon, 20 Apr 2009 10:10:48 +0000	[thread overview]
Message-ID: <20090420121048.7e8f2b46@hyperion.delvare> (raw)
In-Reply-To: <20090406075745.GG4734@ubuntu>

On Mon, 6 Apr 2009 09:57:45 +0200, Andre Prendel wrote:
> This patch does some refactoring of the loadConfig()
> function.
> ---
> 
> * Simplifying the conditions makes code flow clearer and eliminates
> long lines (> 80 chars).
> * Removed useless stat() call.
> * Return -1 in error case, instead of several positiv values (never defined).

Spelling: positive.

If you make loadConfig() always return -1 on error then you also want
to clean up the error message that is printed when reloadLib() fails,
to no longer include the (now pointless) error value.

This patch adds the following compilation warning:
prog/sensord/lib.c: In function ‘loadConfig’:
prog/sensord/lib.c:72: warning: ‘ret’ may be used uninitialized in this function
Please fix.

> 
>  lib.c |   76 +++++++++++++++++++++++++++++++++++-------------------------------
>  1 file changed, 41 insertions(+), 35 deletions(-)
> 
> --- quilt-sensors.orig/prog/sensord/lib.c	2009-04-03 09:31:57.000000000 +0200
> +++ quilt-sensors/prog/sensord/lib.c	2009-04-04 18:30:46.000000000 +0200
> @@ -20,7 +20,7 @@
>   * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
>   * MA 02110-1301 USA.
>   */
> -
> +#include <errno.h>

I do like a blank line before the first include ;)

>  #include <stdio.h>
>  #include <stdlib.h>
>  #include <string.h>
> @@ -34,46 +34,59 @@
>  
>  static int loadConfig(const char *cfgPath, int reload)
>  {
> -	struct stat stats;
> -	FILE *cfg = NULL;
> -	int ret = 0;
> -
> -	if (cfgPath && !strcmp(cfgPath, "-")) {
> -		if (!reload) {
> -			if ((ret = sensors_init(stdin))) {
> -				sensorLog(LOG_ERR,
> -					  "Error loading sensors configuration file <stdin>: %s",
> -					  sensors_strerror(ret));
> -				ret = 12;
> -			}
> -		}
> -	} else if (cfgPath && stat(cfgPath, &stats) < 0) {
> -		sensorLog(LOG_ERR,
> -			  "Error stating sensors configuration file: %s",
> -			  cfgPath);
> -		ret = 10;
> -	} else {
> -		if (reload) {
> -			sensorLog(LOG_INFO, "configuration reloading");
> -			sensors_cleanup();
> -		}
> -		if (cfgPath && !(cfg = fopen(cfgPath, "r"))) {
> -			sensorLog(LOG_ERR,
> -				  "Error opening sensors configuration file: %s",
> -				  cfgPath);
> -			ret = 11;
> -		} else if ((ret = sensors_init(cfg))) {
> -			sensorLog(LOG_ERR,
> -				  "Error loading sensors configuration file %s: %s",
> -				  cfgPath ? cfgPath : "(default)",
> -				  sensors_strerror(ret));
> -			ret = 11;
> -		}
> -		if (cfg)
> -			fclose(cfg);
> -	}
> +	int ret;
> + 	FILE *fp;
>  
> -	return ret;
> + 	/* Load default configuratinon. */

Typo: configuration.

> + 	if (!cfgPath) {
> + 		if (reload)
> +  			sensors_cleanup();

The original code was logging that the configuration file was being
reloaded. This seemed a useful feature to me, why did you remove it?

> +
> + 		ret = sensors_init(NULL);
> + 		if (ret) {
> + 			sensorLog(LOG_ERR, "Error while loading default"

I suggest dropping the "while" for consistency.

> + 				  " configuration file: %s",
> + 				  sensors_strerror(ret));
> + 			return -1;
> +  		}
> + 		return 0;
> + 	}
> +
> + 	/* Read config from stdin. */
> + 	if (!strcmp(cfgPath, "-")) {

As a side note, reading the configuration from stdin doesn't strike me
as something terribly useful. I wouldn't mind dropping support for
this, to make the code more simple.

> + 		if (reload)
> + 			return 0;
> +
> + 		ret = sensors_init(stdin);
> + 		if (ret) {
> + 			sensorLog(LOG_ERR, "Error loading sensors"
> + 				  " configuration file <stdin>: %s",
> +  				  sensors_strerror(ret));
> + 			return -1;
> +  		}
> + 		return 0;
> +  	}
> +
> + 	fp = fopen(cfgPath, "r");
> + 	if (!fp) {
> + 		sensorLog(LOG_ERR, "Error opening config file %s: %s",
> + 			  sensors_strerror(ret));

This call to sensors_strerror() is clearly incorrect, fopen() isn't
part of the libsensors API ;)

> + 		return -1;
> + 	}
> +
> + 	ret = sensors_init(fp);
> + 	if (ret) {
> + 		if (reload)
> + 			sensors_cleanup();

The logic looks all wrong to me. In reload mode, you have to call
sensors_cleanup() _before_ sensors_init()!

> +
> + 		sensorLog(LOG_ERR, "Error loading sensors config file %s: %s",
> + 			  cfgPath, sensors_strerror(ret));

It would also be nice to consistently use "config file" or
"configuration file" in all messages. I think the latter has my
preference.

> + 		fclose(fp);
> + 		return -1;
> + 	}
> + 	fclose(fp);
> +
> + 	return 0;
>  }
>  
>  int loadLib(const char *cfgPath)


-- 
Jean Delvare

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors


  reply	other threads:[~2009-04-20 10:10 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-04-06  7:57 [lm-sensors] [PATCH 6/8] sensord: Refactoring of loadConfig() Andre Prendel
2009-04-20 10:10 ` Jean Delvare [this message]
2009-04-22  7:34 ` Andre Prendel

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=20090420121048.7e8f2b46@hyperion.delvare \
    --to=khali@linux-fr.org \
    --cc=lm-sensors@vger.kernel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.