On Sun, Nov 27, 2011 at 11:54 AM, Rocky Bernstein <[email protected]
> wrote:

>
>
> On Sun, Nov 27, 2011 at 6:42 AM, Bastiaan Timmer <[email protected]>wrote:
>
>> Ok, I created a small example yesterday. I don't know if I can attach
>> files here or send it to your personal email addresses so I've pasted it
>> here: http://paste2.org/p/1799532
>
>
> This works. I have a copy of the program now and run it and see it
> failing. Thanks.
>
> It should be more straightforward now to check that the return value
> should indeed be a DRIVER_OP_BAD_POINTER instead of BAD_PARAMETER.
>

I've looked now at the code and understand why you get this from the
library's point of view: that is the error message that comes back from the
MMC command. That said, since passing a NULL is possibly a common problem,
I changed mmc_read_cd() to look for this special case and return
DRIVER_OP_BAD_POINTER here.


>
>
>>
>>
>> Just let me know if I should get it to you some other way. The file is
>> just the example_paranoia.c program, with the wav-writing removed and a
>> custom read_audio and callback added.
>>
>> The reason the read_audio fails is in line 38, where 0 is passed instead
>> of the allocated buffer (and it returns BAD_PARAMETER). The reason
>> read_paranoia doesn't return is in line 131, where the paranoia mode is set
>> to PARANOIA_MODE_DISABLE.
>>
>> To come back to Robert's message about the paranoia mode, it seems that
>> when I set the mode to PARANOIA_MODE_FULL^PARANOIA_MODE_NEVERSKIP, it does
>> indeed allow skipping and so, it doesn't hang. I would have thought
>> PARANOIA_MODE_DISABLE would also allow skipping, but that doesn't seem to
>> be the case. I've also tried PARANOIA_MODE_DISABLE^PARANOIA_MODE_NEVERSKIP,
>> but that doesn't work either. So I guess that might be a bug? Or is it
>> intended that skipping is not allowed when paranoia is disabled?
>>
>
>
> Again, this is paranoia's API not libcdio's. So although you may get
> thought and opinions here,I would imagine the best place to ask for a
> change is in the paranoia mailing list if it is deemed that this is an API
> bug.
>
> Also, if I were doing that I'd probably rewrite the program so that it
> doesn't use libcdio at all. I am pretty sure that you'd get the same
> results but they'll probably want you to make sure.
>
>
>>
>> thanks,
>> Bas
>>
>> _______________________________________________
>> Libcdio-help mailing list
>> [email protected]
>> https://lists.gnu.org/mailman/listinfo/libcdio-help
>>
>>
>
_______________________________________________
Libcdio-help mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/libcdio-help

Reply via email to