Re: [evolution-patches] camel_folder_summary_decode_token len bug

[Jeffrey Stedfast <fejj ximian com>]

sure, looks good

On Mon, 2004-09-20 at 06:34, James Ascroft-Leigh wrote:
> As discussed on IRC, camel/camel-folder-summary.c's
> camel_folder_summary_decode_token is not consistent with whatever writes
> the summary files.  In particular reading an actual length of 0 (encoded
> as 32) is treated as an error despite apparently being written to the
> file for at least one of the messages in my large in box.
> 
> Symptoms:  Every time the summary file is read it is deemed to be
> corrupt and all of the summary information is re-fetched from the (in my
> case) IMAP server.
> 
> The attached patch, authored by NotZed, is a fix for this that works for
> me at least.
> 
> Thanks!
> 
> ______________________________________________________________________
> Index: camel/camel-folder-summary.c
> ===================================================================
> RCS file: /cvs/gnome/evolution/camel/camel-folder-summary.c,v
> retrieving revision 1.136
> diff -u -3 -r1.136 camel-folder-summary.c
> --- camel/camel-folder-summary.c	17 May 2004 03:44:27 -0000	1.136
> +++ camel/camel-folder-summary.c	20 Sep 2004 10:07:26 -0000
> @@ -1363,7 +1363,7 @@
>  	} else {
>  		len -= 32;
>  		ret = g_malloc(len+1);
> -		if (fread(ret, len, 1, in) != 1) {
> +		if (len > 0 && fread(ret, len, 1, in) != 1) {
>  			g_free(ret);
>  			*str = NULL;
>  			return -1;