[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[lmi-commits] [lmi] master 50c1d57 1/6: Don't write solve trace to statu

From: Greg Chicares
Subject: [lmi-commits] [lmi] master 50c1d57 1/6: Don't write solve trace to statusbar
Date: Thu, 5 Aug 2021 17:09:59 -0400 (EDT)

branch: master
commit 50c1d57eef09bf37230fc6a14396c30cd100a692
Author: Gregory W. Chicares <gchicares@sbcglobal.net>
Commit: Gregory W. Chicares <gchicares@sbcglobal.net>

    Don't write solve trace to statusbar
    The calculation time reported on the statusbar, for a particular cell
    with a solve that takes dozens of iterations, is:
     - 156 msec (previously) with each iteration written to the statusbar
     -  80 msec (now) without writing details of each iteration
    so that indicator of progress took as much time as the operation it was
    reporting on. Twenty years ago, each iteration was so slow that it was
    at least entertaining to watch the iterates dance, but now they'd become
    nothing but a costly blur.
 ihs_avsolve.cpp | 3 ++-
 solve.cpp       | 1 -
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/ihs_avsolve.cpp b/ihs_avsolve.cpp
index 6ec6b7d..f1abc27 100644
--- a/ihs_avsolve.cpp
+++ b/ihs_avsolve.cpp
@@ -39,6 +39,7 @@
 #include "ledger_variant.hpp"
 #include "mc_enum_types_aux.hpp"        // set_run_basis_from_cloven_bases()
 #include "miscellany.hpp"               // ios_out_app_binary()
+#include "null_stream.hpp"
 #include "outlay.hpp"
 #include "zero.hpp"
@@ -456,7 +457,7 @@ currency AccountValue::Solve
     LMI_ASSERT(nullptr != solve_set_fn);
-    std::ostream os_trace(status().rdbuf());
+    std::ostream os_trace(&null_streambuf());
     std::ofstream ofs_trace;
         (  global_settings::instance().regression_testing()
diff --git a/solve.cpp b/solve.cpp
index 741e6dc..5d31aaf 100644
--- a/solve.cpp
+++ b/solve.cpp
@@ -329,7 +329,6 @@ currency AccountValue::Solve()
-        ,status()
     currency const solution_cents = round_to_cents.c(solution.root);

reply via email to

[Prev in Thread] Current Thread [Next in Thread]