http://quantlib.414.s1.nabble.com/schedule-full-interface-tp16512p16522.html
yes I guess that would work as well for the specific case. I would
calendar. This can be tricky, since I think this very same calendar is
used to adjust the schedule's dates to get the payment dates. But
ok. Thank you for this.
> Hi All,
>
> I’m new to Quantlib so apologies if I’ve missed the point about
> your objective but for a swap with a nearly regular schedule, the only
> exception being that the first date is a holiday, could you not just add
> that holiday date to the Calendar object which is then passed into the
> Schedule object via the rules based constructor. The first date will then
> get adjusted based on the business day convention.
>
> I can see why you’d need a more comprehensive solution if the dates
> are at irregular intervals but the original case looks reachable with a
> regular schedule plus a holiday adjustment.
>
>
>
> Thanks
>
> Ed James
>
>
>
> From: Francois Botha [mailto:
[hidden email]]
> Sent: 26 April 2015 21:16
> To: Peter Caspers
> Cc: Luigi Ballabio; QuantLib users
> Subject: Re: [Quantlib-users] schedule / full interface
>
>
>
> Hi Peter,
>
>
>
> At a glance, it sounds good to me too. I'll be able to tell only later.
>
>
>
> thanks
>
>
> Francois Botha
>
>
>
> On 25 April 2015 at 18:45, Peter Caspers <
[hidden email]> wrote:
>
> I will try to send a PR which does that. Francois, you'd be happy with
> that kind of solution, too ?
> Peter
>
>
> On 25 April 2015 at 09:02, Luigi Ballabio <
[hidden email]> wrote:
>> Yes, this would work.
>>
>> Luigi
>>
>> On Apr 25, 2015 8:52 AM, "Peter Caspers" <
[hidden email]> wrote:
>>>
>>> ok thanks. Couldn't we do it just like that:
>>>
>>> Extend the already existing constructor (taking a vector of dates, a
>>> calendar and bdc) such that it accepts the other stuff (termination
>>> date convention, generation rule, tenor, end of month flag, a vector
>>> of booleans indicating regular periods, I guess that's it). Everything
>>> except the first vector of dates should be wrapped in boost::optional
>>> and defaulted to null.
>>>
>>> The inspector fullInterface() returns true if either the rule based
>>> constructor is used or the other one and _all_ meta information is not
>>> null.
>>>
>>> The single inspectors for the meta data return the meta data piece
>>> that is required, if not null, and throws otherwise (i.e. it does not
>>> check for the full interface given, only for the requested piece of
>>> information to be available).
>>>
>>> As far as I can see that would be a pragmatic solution giving full
>>> flexibility, wouldn't it ?
>>>
>>> Peter
>>>
>>>
>>> On 25 April 2015 at 00:38, Francois Botha <
[hidden email]> wrote:
>>> > I remember I had a similar problem:
>>> >
>>> >
>>> >
http://quantlib.10058.n7.nabble.com/End-of-month-bond-coupons-for-leap-years-td15311.html>>> >
>>> > It's been on my todo list to see if I can create a constructor for the
>>> > Schedule to accept an array of dates.
>>> >
>>> > Francois Botha
>>> >
>>> > On 24 April 2015 at 20:31, Peter Caspers <
[hidden email]>
>>> > wrote:
>>> >>
>>> >> Hello,
>>> >>
>>> >> I need to set up a swap with a nearly regular schedule. The only
>>> >> exception is the first date which is on a holiday. The clean way to
>>> >> generate this would be (?)
>>> >>
>>> >> - generate a preliminary schedule using the schedule's class date
>>> >> generation machinery
>>> >> - extract the dates, overwrite the first one with the correct value
>>> >> and then set up the final schedule based on this date vector
>>> >>
>>> >> However when feeding the second schedule into a VanillaSwap
>>> >> instrument, it complains that the schedule need to support the full
>>> >> interface, as far as I can see to extract the calendar and businessday
>>> >> convention which are used to get payment dates from the schedule
>>> >> dates.
>>> >>
>>> >> What can I do now ? As a workaround I just added a method
>>> >>
>>> >> std::vector<Date>& datesModifyable() { return dates_; }
>>> >>
>>> >> to the schedule class which allows to modify the generated dates, but
>>> >> retaining the full interface. But this is not meant like this, is it ?
>>> >>
>>> >> Can we somehow make the schedule class more flexible without breaking
>>> >> something ?
>>> >>
>>> >> Thanks
>>> >> Peter
>>> >>
>>> >>
>>> >>
>>> >>
>>> >> ------------------------------------------------------------------------------
>>> >> One dashboard for servers and applications across
>>> >> Physical-Virtual-Cloud
>>> >> Widest out-of-the-box monitoring support with 50+ applications
>>> >> Performance metrics, stats and reports that give you Actionable
>>> >> Insights
>>> >> Deep dive visibility with transaction tracing using APM Insight.
>>> >>
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y>>> >> _______________________________________________
>>> >> QuantLib-users mailing list
>>> >>
[hidden email]
>>> >>
https://lists.sourceforge.net/lists/listinfo/quantlib-users>>> >
>>> >
>>>
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> One dashboard for servers and applications across Physical-Virtual-Cloud
>>> Widest out-of-the-box monitoring support with 50+ applications
>>> Performance metrics, stats and reports that give you Actionable Insights
>>> Deep dive visibility with transaction tracing using APM Insight.
>>>
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y>>> _______________________________________________
>>> QuantLib-users mailing list
>>>
[hidden email]
>>>
https://lists.sourceforge.net/lists/listinfo/quantlib-users>
>