[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[lmi-commits] [6305] Demote, rewrite, and move a not-so-important defect
From: |
Greg Chicares |
Subject: |
[lmi-commits] [6305] Demote, rewrite, and move a not-so-important defect marker |
Date: |
Sat, 26 Sep 2015 01:46:18 +0000 |
Revision: 6305
http://svn.sv.gnu.org/viewvc/?view=rev&root=lmi&revision=6305
Author: chicares
Date: 2015-09-26 01:46:18 +0000 (Sat, 26 Sep 2015)
Log Message:
-----------
Demote, rewrite, and move a not-so-important defect marker
Modified Paths:
--------------
lmi/trunk/group_values.cpp
lmi/trunk/ledger.hpp
Modified: lmi/trunk/group_values.cpp
===================================================================
--- lmi/trunk/group_values.cpp 2015-09-26 00:50:35 UTC (rev 6304)
+++ lmi/trunk/group_values.cpp 2015-09-26 01:46:18 UTC (rev 6305)
@@ -686,6 +686,19 @@
{
census_run_result result;
+ // Use 100 as the presumptive maximum composite length, assuming
+ // that issue age 0 and endowment age 100 is the worst case. Class
+ // Ledger takes length as a constructor argument, but the true
+ // length is inconvenient to determine here--in particular, when
+ // running life by life, individual cells have not yet been
+ // initialized. The minimum age could be determined by inspecting
+ // all cells' input parameters, but the age at endowment would
+ // require a database lookup for each cell, which would probably
+ // cost more than the overhead of carrying superfluous durations
+ // in the composite ledger. OTOH, doing that (perhaps in class
+ // multiple_cell_document) might not be noticeably slow; and it
+ // may be the best way if coverage beyond age 100 is to be
+ // illustrated someday without any presumptively fixed maximum.
composite_.reset
(new Ledger
(100
Modified: lmi/trunk/ledger.hpp
===================================================================
--- lmi/trunk/ledger.hpp 2015-09-26 00:50:35 UTC (rev 6304)
+++ lmi/trunk/ledger.hpp 2015-09-26 01:46:18 UTC (rev 6305)
@@ -65,13 +65,6 @@
class LMI_SO Ledger
{
public:
- // TODO ?? It's pretty weak to use 100 as a default max length.
- // But when running a composite, life by life, we don't know the max
- // length until we've initialized all lives. We could get the min age
- // by inspecting the input parms for all lives, but to get the age at
- // endt requires initializing each life and doing a database lookup
- // for endt age. Yet 100 won't work for issue age 0 if coverage
- // beyond age 100 is to be shown.
explicit Ledger
(int length
,mcenum_ledger_type ledger_type
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- [lmi-commits] [6305] Demote, rewrite, and move a not-so-important defect marker,
Greg Chicares <=