[lmi-commits] [lmi] master f10c1147 3/3: Augment documentation

[Greg Chicares]

branch: master
commit f10c114784411f2bf2857e732f54e710b543337c
Author: Gregory W. Chicares <gchicares@sbcglobal.net>
Commit: Gregory W. Chicares <gchicares@sbcglobal.net>

    Augment documentation
    
    Solving for premium to yield a target NAAR, and simultaneously
    determining specified amount as a function of premium, has been seen
    not to work well without a deep redesign of the solve code. Can such
    specified-amount strategies therefore be grayed out when solving for
    premium? No, not without redesigning the input-sequence classes, which
    would take too much work for too little benefit.
---
 input_harmonization.cpp | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/input_harmonization.cpp b/input_harmonization.cpp
index 300104b4..3d8bf107 100644
--- a/input_harmonization.cpp
+++ b/input_harmonization.cpp
@@ -504,6 +504,14 @@ void Input::DoHarmonize()
     // strategy, then toggle back and forth between gpt and cvat to
     // see what difference that makes. TAXATION !! Rethink that.
 
+    // Conditionally restricting SpecifiedAmountStrategyFromIssue has
+    // no visible effect because that scalar variable is obsolete,
+    // having been supplanted by
+    //   specamt_sequence SpecifiedAmount
+    // which has no knowledge of class Input. This is a general issue:
+    // input-sequence controls are unaware of the context that would
+    // be required for selective enablement.
+
     SpecifiedAmountStrategyFromIssue.allow(mce_sa_input_scalar, !specamt_solve 
&& !specamt_from_term_proportion);
     SpecifiedAmountStrategyFromIssue.allow(mce_sa_salary      , !specamt_solve 
&& !specamt_from_term_proportion);
     SpecifiedAmountStrategyFromIssue.allow(mce_sa_maximum     , 
!inhibit_premium_based_strategies);