Everytime I visit I see certain questions asked again and again. This
certainly doesn't bother me; it just shows me that new folks are
showing an interest all the time. A favorite question of mine is,
"Where can I get a copy of QBasic 4.5 with the compiler?" Which is
answered that you can't anymore unless you are lucky, etc.
Here's my question. Why are there any sites out there where you can
e-mail/upload your "myprog.bas" file and get e-mailed back a
"myprog.exe" file. In other words, I'm surprised that there aren't
several groups of volunteers willing to compile files for people.
Nothing illlegal about that, is there.
Anyway, keep up the good work.
Sam
Many people here, myself included, will cheerfully compile source code for
anyone who cares to e-mail it to us. In my case, I can compile using PDS,
QB, PB (DOS or CC), TrueBASIC (DOS, Bronze or Silver).
Tom Lake
If you have the compiler, you can probably debug
the program.
If you don't have the compiler and program it in a
different variant
of basic, there's significant chance that it won't
compile.
I've never been able to get anything past "hello
world" to run
in another variant without some tweeking.
Most people, me included, would be glad to compile
"A" program.
Problem is that it always turns into a never
ending debug/enhance cycle.
If you'd post a link to the compiler, everyone
would download it and
be happy.
mike
--
Bunch of stuff For Sale and Wanted at the link
below.
Tek 2465, TFT projector, ham radio, 30pS pulser
Tektronix Concept Books...
http://www.geocities.com/SiliconValley/Monitor/4710/
One reason is that most BASIC-language compiler for the PC do not output a
complete listing of the source code compiled with errors indicated. Most of
them (Microsoft's BASIC compilers, PowerBASIC for sure) work via some kind
of interaction with the IDE, and it would be very awkward to report the
simple typing errors which inevitably occur when writing code.
That said, I know you can get source code output files from MS-QB 4.5, but I
cannot recall if you get the BASIC language error message when you do
something dumb. (Yes, that is "message", singular; both MS-DOS BASIC and
PowerBASIC compilers stop the compilation on the first error). (The more I
think about it, I think you *do* get the source code listing up to the point
of the error with Microsoft compilers).
Another reason is that it would take some kind of doing to allow compilation
of programs which have multiple source or object code modules - $INCLUDE
files (MS, PB), multiple modules (MS) and multiple precompiled comonents OBJ
files or PB "unit" files).
I can see in my head how to design a 'compiler service' for either
compiler.. maybe the user would have to submit some kind of ZIP file with a
control record indicating which compiler options are needed... kind of like
the IBM mainframe COBOL compiler lets you submit a 'control' record (or put
compiler options in-line). For MS-BASICs, you set up a command-line compile
with the source code option; for PowerBASIC, same thing, except you'd have
to generate the source code listing yourself,and report back the compiler
output from a file. I think the PB command line compiler sends its output to
STDOUT, meaning it could be redirected.( Gee, I just fired up the PB/DOS IDE
and cannot find the doc for the command-line compiler to see if there is a
log file option. That must be in the book and not in the help file.).
I have no interest in actually doing this, but if anyone is interested in
tapping my thoughts drop me a note.
--
Michael Mattias
Tal Systems, Inc.
Racine WI
mmat...@talsystems.com
http://www.electronic-projects.net/software.shtml
Not necessarily. My first programs were written without a compiler. It
worked like this
1) Carefully plan your program.
2) Write out the (Fortran) program on coding forms with a pencil.
3) Check what you've done.
4) Send to remote computer centre by (Royal) mail.
5) Wait for four days.
6) Receive results
7) Repeat until the program works
Programming like this makes you think pretty hard about what you are asking
the computer to do. You *really* want to avoid step 7, so you spend a lot
of time on steps 1 and step 3. As a result you take a lot more care about
the coding than you do when you can "ask the compiler" to find your mistakes
and thus you are much less likely to get into an endless debug/enhance
cycle.
I think that the OP's idea for an online compiler service is quite a
reasonable one.
Cheers
Derek
Derek Ross wrote:
cnip
> > Problem is that it always turns into a never ending debug/enhance cycle.
>
> Not necessarily. My first programs were written without a compiler. It
> worked like this
>
> 1) Carefully plan your program.
> 2) Write out the (Fortran) program on coding forms with a pencil.
> 3) Check what you've done.
> 4) Send to remote computer centre by (Royal) mail.
> 5) Wait for four days.
> 6) Receive results
> 7) Repeat until the program works
>
> Programming like this makes you think pretty hard about what you are asking
> the computer to do. You *really* want to avoid step 7, so you spend a lot
> of time on steps 1 and step 3. As a result you take a lot more care about
> the coding than you do when you can "ask the compiler" to find your mistakes
> and thus you are much less likely to get into an endless debug/enhance
> cycle.
>
> I think that the OP's idea for an online compiler service is quite a
> reasonable one.
>
> Cheers
>
> Derek
Online basic compiler might make a fun toy
project, but won't succeed as a commercial
venture.
Expereienced people have the compiler. Only
people who would need an online basic
compiler would be very inexperienced. They would
need support...the deal breaker.
There is a business model for custom IC layout
where the tool costs hundreds of killobucks.
But it was supported by the IC vendor. Don't know
if they ever made any money on it.
Your results are not typical. Software is poorly
documented and often doesn't work
the way one might interpret the documentation. If
you know exactly what you're doin'
maybe. But It's gotta be perfect. Like the song
says, "99 and a half just won't do.."
I don't think I've ever done anything that worked
exactly the way I wanted it to on the
first try...may have worked the way I told it to,
but not the way I eventually wanted...
and I been practicing for over half a century.
Well, sure, no disagreements there, but nobody said that it had to be
commercial. I was just thinking, freely accessible, no support, take it or
leave it. For QBASIC programmers who would develop with the interpreter on
their own computers and then send off the debugged code for compilation, it
could work.
Cheers
Derek