Re[2]: [lmi] wxmsw-2.9.0 regression: messagebox doubling

[Vadim Zeitlin]

On Thu, 26 Feb 2009 06:04:18 +0000 Greg Chicares <address@hidden> wrote:

GC> Does "lmi_wx.exe!MortalityRates..." mean you're using msvc?

 Yes, sorry for not making it clear. I use MSVC in order to be able to
debug it conveniently.

GC> If so, then does it work with gcc?

 I haven't tested yet, will do it a.s.a.p.

GC> My fear is that we might have discovered some area where lmi
GC> unintentionally relies on implementation-defined behavior of gcc.

 I don't see how could 5-argument transform() work with input sequences of
different sizes with any standard library implementation. Or do you mean
some difference upstream which is responsible for CurrentCoiMultiplier_
being empty? This is possible, of course.

 BTW just to be totally sure: neither CCoiMultiplier_ nor
CurrentCoiMultiplier_ are supposed to be empty, aren't they?

GC> Can you run
GC>   make cli_tests
GC> successfully? That would exercise the code that's failing here, in a
GC> way that doesn't use wx at all.

 I don't have MSVC projects for all tests yet but I'll create them and do
this a.s.a.p. too.

GC> 'install_msw.sh' contains all the steps and is well tested. It takes
GC> an hour or so to run, and builds its own libraries, which I know you
GC> would rather avoid, but it does work for me.

 The problem is that the last time I tried I spent almost a day on it and
gave up. This is certainly due to the unusual setup of my system (tiny C:
drive, LMI and wxWidgets on different drives, ...) but the trouble is that
it's difficult for me to change this. I probably ought to set up a VM just
for building LMI but then it's more likely to take 3-4 hours to build and
not 1...

GC> It is interesting that similar failures are observed in classes
GC> InterestRates and MortalityRates, under similar circumstances.
GC> Below is an experimental patch that tests the data at various
GC> points for the MortalityRates case, which fails first.

 Just in case it can help, here is the sequence of messages I get after
applying this patch:

        ---------------------------
        55 is CurrentCoiMultiplierRealized_.size()
        55 is years_to_maturity()
        45 is issue_age        ()
        65 is retirement_age   ()
        0 is inforce_year     ()
        2009 is effective_year   ()

        [file .\input_realization.cpp, line 440]
        ---------------------------

        0 is z.CurrentCoiMultiplierRealized_.size()
        0 is CurrentCoiMultiplier.size()

        [file .\yare_input.cpp, line 176]
        ---------------------------
        0 is basic_values.yare_input_.CurrentCoiMultiplier.size()
        0 is CurrentCoiMultiplier_.size()

        [file .\mortality_rates_fetch.cpp, line 82]
        ---------------------------

Do I understand correctly that z.CurrentCoiMultiplierRealized_.size()
should be 55 too in the second message?

 Thanks,
VZ