Re: [Nmh-workers] corruption when attaching all-null files

[Paul Fox]

lyndon wrote:
 > 
 > On May 7, 2014, at 12:49 PM, Paul Fox <address@hidden> wrote:
 > 
 > > and i think i understand
 > > the containsnuls issue
 > 
 > There should only be one way to declare if the content is 7bit.
 > 
 > If you have 'is7bit' and 'containsnul' + 'contains8bit', there are
 > two possible decision paths:
 > 
 > 1) 'is7bit' is set, or
 > 2) 'containsnul' and 'contains8bit' are both unset.
 > 
 > If a chunk of code, e.g., sets 'contains8bit' but forgets to clear
 > 'is7bit', who wins when you try to decide if the code is 7bit safe?

got it.

aside from the logic aspect:  i take it that text with a NUL must
be treated differently than text with an 8th bit set.  i assume it
must be encoded with q-p?  (or base64, but q-p would be sufficient?)

paul
----------------------
 paul fox, address@hidden (arlington, ma, where it's 59.0 degrees)