Here is what I have decided about the inchar, outchar, lineechar
controversy:
In a concession to the "I don't want to type too many letters" camp,
we'll make the defaults %i#, %o# and %t# for input, output and
inTermediate expressions. It bothers me that %i# and %i look similar,
but are completely unrelated but I'll get over it.
I will write up a customization tutorial.
I have modified all the info files to reflect the new defaults. I
haven't carefully proofread the entire manual, but that needs to be done
after we change cases, anyway. I did (quickly) go over each line that
changed. I caught a couple of cases where my script* was too zealous. I
do not think the number of errors this introduced into the manual is a
significant increase in the total number of errors in the manual.
It is interesting to note that the file Symmetries.tex makes extensive
use of c#, d# and e# for variables.
I would like to include the new defaults in the first beta for 5.9.1,
which will appear soon. If the change introduces unforeseen problems, we
can take it out before the release.
Lastly, as I previously mentioned, there is a code freeze in effect. I
am therefore offering up my patch for consideration before comitting:
--------------------------------------------------------------------
Index: suprv1.lisp
===================================================================
RCS file: /cvsroot/maxima/maxima/src/suprv1.lisp,v
retrieving revision 1.13
diff -r1.13 suprv1.lisp
164,165c164,165
< (DEFMVAR $% '$% "The last D-line computed, corresponds to lisp *"
NO-RESET)
< (DEFMVAR $INCHAR '$C
---
> (DEFMVAR $% '$% "The last out-line computed, corresponds to lisp *"
NO-RESET)
> (DEFMVAR $INCHAR '|$%i|
167c167
< (DEFMVAR $OUTCHAR '$D
---
> (DEFMVAR $OUTCHAR '|$%o|
169c169
< (DEFMVAR $LINECHAR '$E
---
> (DEFMVAR $LINECHAR '|$%t|
--------------------------------------------------------------------
The changes in the info directory are a little too large to include in
this message. I have included the script I used to generate them at the
end.
--Jim
*Here is the script I used:
for i in *.texi
do
sed 's/\([^A-Za-z]\)[Cc]\([0-9]\+\)/\1%i\2/g' $i | \
sed 's/\([^A-Za-z]\)[Dd]\([0-9]\+\)/\1%o\2/g' | \
sed 's/\([^A-Za-z]\)[Ee]\([0-9]\+\)/\1%t\2/g' > tmp
mv tmp $i
done
The script erroneously changed numbers like 1.0E20. There were three
instances I fixed by hand.