[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[lmi-commits] [5933] Improve documentation
From: |
Greg Chicares |
Subject: |
[lmi-commits] [5933] Improve documentation |
Date: |
Wed, 06 Aug 2014 13:12:43 +0000 |
Revision: 5933
http://svn.sv.gnu.org/viewvc/?view=rev&root=lmi&revision=5933
Author: chicares
Date: 2014-08-06 13:12:42 +0000 (Wed, 06 Aug 2014)
Log Message:
-----------
Improve documentation
Modified Paths:
--------------
lmi/trunk/loads.cpp
Modified: lmi/trunk/loads.cpp
===================================================================
--- lmi/trunk/loads.cpp 2014-07-20 14:34:13 UTC (rev 5932)
+++ lmi/trunk/loads.cpp 2014-08-06 13:12:42 UTC (rev 5933)
@@ -186,14 +186,19 @@
)
);
- // ET !! As for rounding, we do want an expression-template library
- // to apply a scalar function like rounding to all elements of a
- // matrix, with some natural syntax like
- // separate_account_load_ =
details.round_interest_rate_(separate_account_load_);
for(int j = mce_gen_curr; j != mc_n_gen_bases; j++)
{
#if 0
-// ET !! This crashes:
+// ET !! As for rounding, we do want an expression-template library
+// to apply a scalar function like rounding to all elements of a
+// matrix, ideally with some natural syntax like
+// separate_account_load_ =
details.round_interest_rate_(separate_account_load_);
+// However, this throws an "Erroneous rounding function" exception,
+// because apply_unary() and apply_binary() use a default-constructed
+// function object:
+//
http://svn.savannah.nongnu.org/viewvc/lmi/trunk/tools/pete-2.1.1/PETE/Tools/PeteOps.in?root=lmi&r1=4151&r2=4150&pathrev=4151
+// but round_to<> is stateful, and deliberately throws that exception
+// when a default-constructed object is used.
assign
(separate_account_load_[j]
,apply_unary
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- [lmi-commits] [5933] Improve documentation,
Greg Chicares <=