Differences From Artifact [713fa3d927]:
- Executable file
mttroot/mtt/bin/trans/ese2rdae_r
— part of check-in
[4767414321]
at
2001-07-23 23:31:17
on branch origin/master
— Added -cr option to load CRs first - avoids alg. loops with R
implementation of mutiports. (user: gawthrop@users.sourceforge.net, size: 5905) [annotate] [blame] [check-ins using] [more...]
To Artifact [78718944fc]:
- Executable file
mttroot/mtt/bin/trans/ese2rdae_r
— part of check-in
[ea9834becc]
at
2002-06-05 11:14:51
on branch origin/optimise-algebraic-equations
— ae.r now generated using def2write_r like cse?.r
fix_c.r called at ese2rdae stage so that pow gets fixed in ae.r.These changes produce the desired result (optimised algebraic equations) but
have highlighted a problem; when optimisation fails, Reduce does not write
a result. For complicated systems, this can lead to missing assignments in
the resultant code. (user: geraint@users.sourceforge.net, size: 6188) [annotate] [blame] [check-ins using] [more...]
| ︙ | ︙ | |||
10 11 12 13 14 15 16 17 18 19 20 21 22 23 | # Copyright (c) P.J.Gawthrop, 1991, 1994, 1996 ############################################################### ## Version control history ############################################################### ## $Id$ ## $Log$ ## Revision 1.11 2001/07/06 00:46:50 gawthrop ## Added -cr option -- forces cr to be loaded before the ese.r file ## This avoids causality problems when using multi-port Rs to represent ## arbitary equations ## ## Revision 1.10 2000/11/30 15:23:16 peterg ## Taken out all subsystem stuff - now done in mtt using | > > > > | 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 | # Copyright (c) P.J.Gawthrop, 1991, 1994, 1996 ############################################################### ## Version control history ############################################################### ## $Id$ ## $Log$ ## Revision 1.12 2001/07/23 23:31:17 gawthrop ## Added -cr option to load CRs first - avoids alg. loops with R ## implementation of mutiports. ## ## Revision 1.11 2001/07/06 00:46:50 gawthrop ## Added -cr option -- forces cr to be loaded before the ese.r file ## This avoids causality problems when using multi-port Rs to represent ## arbitary equations ## ## Revision 1.10 2000/11/30 15:23:16 peterg ## Taken out all subsystem stuff - now done in mtt using |
| ︙ | ︙ | |||
96 97 98 99 100 101 102 103 104 105 106 107 108 109 |
case $1 in
-I )
info=info;;
-cr )
load_cr=yes;
blurb2=' using cr and subs first';
;;
-partition )
partition=yes;
blurb='with partitioning';
;;
*)
echo "$1 is an invalid argument - ignoring";
exit ;;
| > > > > | 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 |
case $1 in
-I )
info=info;;
-cr )
load_cr=yes;
blurb2=' using cr and subs first';
;;
-fixcc )
include=`echo 'in "'$MTT_LIB'/reduce/fix_c.r";'`
blurb3='fixing c and cc code';
;;
-partition )
partition=yes;
blurb='with partitioning';
;;
*)
echo "$1 is an invalid argument - ignoring";
exit ;;
|
| ︙ | ︙ | |||
132 133 134 135 136 137 138 |
if [ -n "${load_cr}" ]; then
load_cr_comm="in \"${crname}\";"
load_subs_comm="in \"${subsname}\";"
fi
# Inform user
| | > > > | 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 |
if [ -n "${load_cr}" ]; then
load_cr_comm="in \"${crname}\";"
load_subs_comm="in \"${subsname}\";"
fi
# Inform user
echo Creating $daename $blurb $blurb2 $blurb3
# Remove the old log file
rm -f $logname
# Use symbolic algebra to accomplish the transformation
$SYMBOLIC >$logname << EOF
%Read in the cr and sub when -cr is set
$load_cr_comm
$load_subs_comm
%Read the formatting function
in "$MTTPATH/trans/reduce_matrix.r";
% Definitions
in "$defname";
% Elementary system equations
in "$esename";
% Fix c code if required
$include
OFF Echo;
OFF Nat;
%Create the output file
OUT "$daename";
|
| ︙ | ︙ |