Boost logo

Boost :

Subject: Re: [boost] boost::program_options question - enabling multipleusage of the same switch
From: Khaikin, Sergey (sergey.khaikin_at_[hidden])
Date: 2008-10-11 06:03:32


Hi Volodya,
  Semantics of multiple occurrences would be "ignore all the occurrences
but one and assume the switch value is true". So it's pretty much a
syntactic issue. Maybe if I describe the use case the need for this
syntax will become clearer. In my application, there is some external
tool that generates command line for my tool. In some cases, it may
generate Boolean switches more than one time and I need to support this
"silently" instead of throwing "multiple_occurences" exception. Do you
have any kind of workaround in mind ? (well, except using external
script to preprocess the command line which makes the flow cumbersome).

 Thanks,
Sergey

-----Original Message-----
From: boost-bounces_at_[hidden]
[mailto:boost-bounces_at_[hidden]] On Behalf Of Vladimir Prus
Sent: Tuesday, October 07, 2008 5:06 PM
To: boost_at_[hidden]
Subject: Re: [boost] boost::program_options question - enabling
multipleusage of the same switch

Khaikin, Sergey wrote:

> Hi,
>
> This question was originally brought up here:
>
> http://lists.boost.org/boost-users/2006/08/21633.php
>
> but unfortunately was not answered. What I basically want to do is to
> enable something like running
>
>
>
> $ myprog -v -v -v
>
>
>
> Where -v is some Boolean switch.

Boolean? What would be the semantics of multiple occurences? If you want
integer variable that is incremented by 1 for each occurrence, then I
think
this is entirely syntax matter, and probably is best handled by taking
vector<option> returned by parse_command_line, finding all -v
occurences,
and merging them into a single option, with a value equal to the number
of occurrences. There is no built-in way to do this.

- Volodya

_______________________________________________
Unsubscribe & other changes:
http://lists.boost.org/mailman/listinfo.cgi/boost
---------------------------------------------------------------------
Intel Israel (74) Limited

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.


Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk