From: Sven M. Hallberg (pesco_at_[hidden])
Date: 2001-05-10 11:38:15
On Thu, May 10, 2001 at 08:36:23AM -0700, Gary Powell wrote:
> > Is there interest in Boost for an option processor
> I'm interested.
> > Here's a synopsis of my processor: One creates an option_processor object
> > and
> > registers supported options with it, giving the following information:
> > - short/long ways for denominating the option on the command line
> On the short way, will you require this to be specified by the programmer,
> or will it automatically create a short flag from the long flags?
The programmer needs to specify both denominators denominators seperately.
Autogenerating the short form would be too much guesswork in my eyes.
> > - a description
> > - a handler function object
> You might also consider using the any_function stuff as well.
Yes, I have yet to check that library out, but had it marked. :)
> > Once that is done, the processor can be asked about an info table about
> > the
> > known options as frequently seen on programs' "--help" screens. Finally of
> > course it can be given argc/argv and it will parse it for options and call
> > the
> > appropriate handlers, throwing exceptions on errors (missing arguments,
> > etc.)
> This is very handy to have and if done so that "help" is required when you
> register the option it will remind programmers to fill in at least a token
> help statement.
Exactly, the description string is required.
Should I upload my current version to the vault already, or do more
boost-specific cleanup first? For example, I'm using POSIX regex functions, as
well as a custom smartpointer.
-- "Would the All-Seeing Eye please look in my direction?" [ KeyID........: 0xC297FEAB ] [ Fingerprint..: FEA5 0F93 7320 3F39 66A5 AED3 073F 2D5F C297 FEAB ]
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk