tlm@pur-phy (Timothy Lee Meisenheimer) (04/06/88)
[ 'ello meeester line muncher.......]
I was unsure about which group this went to but since not everyone
gets both.........
As a follow up to some previous questions - It seems that the manx
make does indeed look at SYS1: before doing any and everything. Once
I "assign SYS1: RAM:" all those bothering requestors GO AWAY and make
will go it's merry way! What a cludge!! Why the special emphasis on
SYS1:???? I don't use the distribution floppies to boot(or their copies etc)
and only load up the "c" environment when I'm 'programming'. I don't shut
my machine off much and would rather not reboot either. Oh well, maybe
that will get fixed as I also hope the '%g' argument to printf(and friends)
will get fixed. For example printf("%10.1g",f) - (f is a float) - when
f=100 gets you "1" and not "100.0"!!! Then '%10.2g' gets you "10" etc.,
(I think you'll see the sequence). I don't mean to slash manx I really
like it a lot but it gets a little aggrevating as '%g' is something I'd
like to use is the "SC" spreadsheet so I can use it for both money type
math and scientific computing math (note '%g' does not work right for
numbers < 1 either and '%f' and '%e' work fine!).
On a different line with this spreadsheet I've noticed a problem
when catching ^C (so to speak). If I disable the ^C checking and let sc
handle it - i.e. just treat it as another character everything works fine, since
it just quits like hitting the 'q' key. But if I enable checking and
Check_abort() sees the ^C this calls _abort and some other things and
then does an exit(). With exit() all the memory gotten be malloc() etc.
is given back as well as stdio files. But I've got some DOS funtions (namely
an Open() on a RAW: which is not taken care of :-( ) so I have resources
which need to be cleaned up. Do I write my own _abort() routine? Or
should it be exit()? Or is it ill advised to mix manx and DOS routines like
this (somebody else wrote it and I'm trying to fix it)? If I substitute
my own routines what do I need to look out for?
Will be interested in anyone's advice.
tlm@newton.physics.purdue.edu