[parted-devel] [PATCH] parted: Display details of partition alignment failure (#726856)

Brian C. Lane bcl at redhat.com
Fri Feb 12 00:52:35 UTC 2016


On Wed, Feb 10, 2016 at 10:47:09AM -0500, Phil Susi wrote:
> First, sorry for the late reply; I seem to have fallen off the mailing
> list some time ago.  Also please take your misguided mail client's
> reply-to-list function and drown it in the bathtub.  Reply-to-all is the
> correct way to reply to public discussions.  Had you done this I still
> would have seen this message long ago.

I don't see how you falling off the list is my problem, and this really
isn't the place to discuss email etiquette.

> 
> On 12/16/2015 5:52 PM, Brian C. Lane wrote:
> > I took another look at this, and still disagree. do_mkpart throws
> > exceptions in a couple of places based on results from the attempt to
> > make the partition. There is also another use of partition_align_check
> 
> So?
> 
> > in do_align_check where it doesn't throw an exception. Moving these
> > exceptions down into a function doesn't make sense.
> 
> Hrm... it seems that if you want to get the reason in do_mkpart, you
> would also want to get it in do_align_check.  That is; this function
> should also be printing the reason rather than passing NULL.  Proper
> exception practice still says the exception should be pushed down to
> where the actual error is detected, but on the other hand,
> do_align_check wants to just print the result rather than process it as
> an exception and so it would have to catch it and work harder to print
> it as a regular message instead of as an error due to the wonky nature
> of libparted's makeshift C exceptions rather than having proper C++ or
> SEH exception handling.
> 
> Due to that and the fact that none of these functions are part of
> libparted, I suppose the string argument is ok, but it seems that it
> should also be added to do_align_check, don't you think?
> 

I can't actually remember why I didn't do it that way in the first
place so I'll extend the "not aligned" string to include the reason.


-- 
Brian C. Lane | Anaconda Team | IRC: bcl #anaconda | Port Orchard, WA (PST8PDT)



More information about the parted-devel mailing list