This is the mail archive of the
cygwin-patches
mailing list for the Cygwin project.
Re: [Patch] Encode invalid chars in /proc/registry entries
On Nov 16 20:25, Christian Franke wrote:
> Corinna Vinschen wrote:
>> Thanks for this patch. Apart from the missing ChangeLog I'm inclined
>> to apply it to the upcoming 1.5.25 release, but I don't like to have it
>> in HEAD as is.
>
> Thanks, I would appreciate to have this issue fixed in the bugfix release.
>
> Here is a new version of the patch and a ChangeLog.
>
> The names "." and ".." are now also encoded. Theses are also valid as
> Key/Value Names and ".." may result in infinite recursion.
Thanks, I've tested it on my machine and I've applied the patch to the
cr-0x5f1 branch.
>> So, for HEAD I'd like to ask you to allow arbitrary path lengths in your
>> code. Personally I could live with restricting registry paths to
>> PATH_MAX as well.
>
> Agree. Probably Cygwin should never descend paths that exceed PATH_MAX, as
> an application using PATH_MAX may have no buffer overflow check.
I agree.
>> While you're digging in registry code anyway... would you be interested
>> to convert the entire registry code to wide char and long path names?
>> I'd be glad for any help.
>
> I will have a look at it, but be patient. Is current HEAD a reasonable
> starting point or is there a better (more stable) snapshot?
Usually HEAD is the *only* valid starting point.
Thanks again for the patch,
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Project Co-Leader cygwin AT cygwin DOT com
Red Hat