Wireshark · Ethereal-dev: Re: [Ethereal-dev] Re: [Ethereal-users] error opening C:\Program Files\Ethereal-0.9.16\asn1/default.

[Ulf Lamping <ulf.lamping@xxxxxx>]

Guy Harris wrote:

> On Dec 15, 2003, at 5:59 PM, Ian Schorr wrote:
>
> > I could be wrong, but it looks like the way the code's written, it 
> > also will error if the pref that it read in differs from default 
> > value (current ethereal dir + "asn1/default.tt").  The code actually 
> > gets the same error either way (file does not exist), but doesn't 
> > report the error unless the Ethereal dir in the pref is different 
> > than the dir you're running Ethereal out of.
>
>
> The goal was to arrange that if you explicitly specified a file, you'd 
> get an error if the file didn't exist, but if you left it as the 
> default, it'd just drive on silently.
>
> Unfortunately, the only way it has to determine whether you explicitly 
> specified a file is to see whether the value from the preference file 
> is the default.  The preferences code doesn't indicate whether a given 
> preference was actually set in the preferences file or not.
>
> Arguably, it should, *and* that indication should be set if the user 
> explicitly changes a preference, *and* any preference the user hasn't 
> explicitly set should *not* be written to the preferences file.
>
> Unfortunately, that doesn't provide any mechanism for explicitly 
> saying "I want this particular preference item to revert to the 
> default setting" other than "edit your preference file by hand and 
> remove the item".  We could have a "reset to default" button that 
> resets *all* settings, but that's a bit of a crude mechanism.
>
> Also, that raises the question of whether the default should be the 
> compiled-in defaults or the compiled-in defaults *as overridden by any 
> global preference setting*.
>
> _______________________________________________
> Ethereal-dev mailing list
> Ethereal-dev@xxxxxxxxxxxx
> http://www.ethereal.com/mailman/listinfo/ethereal-dev
Yesterday, I've changed the path seperator in "asn1/default.tt" to the 
GLIB define, so it will use the correct backslash on win32 systems.

This *seems* to fix the problem, at least on my machine.

Regards, ULFL