various

[Matthias Andree]

On Tue, 22 Oct 2002, Clint Adams wrote:

> The change in -vv behavior has been causing me some consternation,
> especially when I am unfortunate enough to have to use a dialup.
> I think there is a fundamental problem in treating 'debug' the same way
> as 'verbose'.  I suggest keeping -v's for verbose and using another
> option for debug (too bad -d's taken).  Multiple -v's would, as now,
> increase verbosity, but the debug option would take the bitmask.

-D is still available, and -x is not too uncommon for debugging flags.

Other than that, it's hard to tell what someone considers "debug" output
-- someone else might see it as "verbose regular" output instead.

> Base64 parsing is a problem.  Not only will 'solitary' on a
> line be ignored, but a legitimate b64 line, such as
> 'c29saXRhcnkgd29yZAo=', will be tokenized as 'c29saxrhcnkgd29yzao'.
> Unfortunately, 'xrhcnk' on a line by itself is also legal base64.
> While this is permissible, I don't think it's common; I think we're
> more likely to find base64 lines to be longer than the maximum token
> length, except when they end in '='.  I haven't thought this through,
> but I think we may get better results if the base64 re was modified to
> only catch '='-terminated strings.

That won't work, this termination only happens when "padding" is
necessary, i. e. if the length is not divisible by 3, but leaves a
remainder of 1 or 2.

> I also wanted to gripe about the versioning system but I'll have to
> defer that to my next time online.

:-) Well, it's beta and we're tagging CVS now.

-- 
Matthias Andree