f



Cambridge Pascal Compiler on newer machines?

Hallo,

don't want to start a discussion or war ;-) about the optimal 
programming-language...
.... but I'm still using the Cambridge Pascal Compiler V1.00 from 1989 
and have written many applications I'm regulary using 'til today with 
it.
On my BBxM (RISCOS 5.23) the programs are only running under Aemulor, 
well and that's my problem: it's not possible to run RISCOS with the 
newer ZPP-ROMs.
So my question is, is it possible to make this (yes I know, very 
old...) compiler running "native" on 32-bit and ZPP? There is an 
module CPCLib for runtime and PCompile for the compiler.
A longer time ago I saw this compiler in the public domain (or at 
least to download free) but can't remember, where it was.

It's not my aim to convert those (mostly very special) pascal programs 
to another programming language, they are several thousands of 
codelines I developed over years and time is rare ;-) ...

The requirements to do that are far behind my capabilities, so is 
there any chance to convert the compiler or has anybody even done 
that? Any Pascaleros out there ;-) ?

Thanks in advance,
Greetings,
J�rgen
0
freak67
10/14/2016 8:36:32 PM
comp.sys.acorn.programmer 2499 articles. 0 followers. Post Follow

9 Replies
169 Views

Similar Articles

[PageSpeed] 2

On 14/10/2016 21:36, freak67@freakmail.de wrote:
> ... but I'm still using the Cambridge Pascal Compiler V1.00 from 1989
> and have written many applications
[Snip]
> So my question is, is it possible to make this (yes I know, very
> old...) compiler running "native" on 32-bit and ZPP? There is an
> module CPCLib for runtime and PCompile for the compiler.

Even if it was converted to run on 32 bit systems (big job without the 
source), it would still produce 26 bit output, and changing that would 
be an even bigger job, even with the source.

Emulation is the only practical way with such relics.

---druck

0
druck
10/14/2016 9:20:17 PM
On 14/10/2016 22:20, druck wrote:
> On 14/10/2016 21:36, freak67@freakmail.de wrote:
>> ... but I'm still using the Cambridge Pascal Compiler V1.00 from 1989
>> and have written many applications
> [Snip]
>> So my question is, is it possible to make this (yes I know, very
>> old...) compiler running "native" on 32-bit and ZPP? There is an
>> module CPCLib for runtime and PCompile for the compiler.
>
> Even if it was converted to run on 32 bit systems (big job without the
> source), it would still produce 26 bit output, and changing that would
> be an even bigger job, even with the source.
>
> Emulation is the only practical way with such relics.
>
> ---druck
>
I seem to remember there were some fairly major bugs in it that were an 
absolute pain to work around (not being able to close files correctly 
comes to mind, along with the fact that all the source had to be in the 
correct order with no libraries etc), and when I asked Oak Solutions 
about it (back in 1993/4) got a very shirty answer back about it not 
being properly supported even then, let alone 22 years later.
0
Someone
10/15/2016 8:40:19 AM
In Nachricht <ntri61$8ku$1@dont-email.me>
          druck <news@druck.org.uk> haben Sie geschrieben:

> Even if it was converted to run on 32 bit systems (big job without the
> source), it would still produce 26 bit output, and changing that would
> be an even bigger job, even with the source.

> Emulation is the only practical way with such relics.

> ---druck

Thanks for your answer, druck.
I'm not sure, if this compiler produces ARM-code, it's more probable 
that it's a kind of P-Code which is interpreted. Otherwise yes, a 
rewrite is more or less impossible.

Greetings,
J�rgen
0
JF
10/15/2016 12:49:53 PM
In Nachricht <ntsq1h$1nlb$1@gioia.aioe.org>
          Someone Somewhere <nntpac@gmail.com> haben Sie geschrieben:

> On 14/10/2016 22:20, druck wrote:
>> On 14/10/2016 21:36, freak67@freakmail.de wrote:
>>> ... but I'm still using the Cambridge Pascal Compiler V1.00 from 1989
>>> and have written many applications
>> [Snip]
>>> So my question is, is it possible to make this (yes I know, very
>>> old...) compiler running "native" on 32-bit and ZPP? There is an
>>> module CPCLib for runtime and PCompile for the compiler.
>>
>> Even if it was converted to run on 32 bit systems (big job without the
>> source), it would still produce 26 bit output, and changing that would
>> be an even bigger job, even with the source.
>>
>> Emulation is the only practical way with such relics.
>>
>> ---druck
>>
> I seem to remember there were some fairly major bugs in it that were an
> absolute pain to work around (not being able to close files correctly
> comes to mind, along with the fact that all the source had to be in the
> correct order with no libraries etc), and when I asked Oak Solutions
> about it (back in 1993/4) got a very shirty answer back about it not
> being properly supported even then, let alone 22 years later.

You are right, some bugs are very annoying. The left-open files was 
one of them, but it's possible to close them with an SWI-call.

And this Pascal-compiler worked relativly strict after the "real" 
pascal-standard at that time (absolutly not comparable to the other 
"pseudo" or better said "de facto"-standards, like Turbo-Pascal etc. 
which influenced most of the later compilers on other platforms).

But it was fine to program WIMP-applications with it ;-).

Greetings,
J�rgen
0
JF
10/15/2016 1:04:58 PM
On 15/10/2016 14:04, JF wrote:
> In Nachricht <ntsq1h$1nlb$1@gioia.aioe.org>
>           Someone Somewhere <nntpac@gmail.com> haben Sie geschrieben:
>
>>>
>> I seem to remember there were some fairly major bugs in it that were an
>> absolute pain to work around (not being able to close files correctly
>> comes to mind, along with the fact that all the source had to be in the
>> correct order with no libraries etc), and when I asked Oak Solutions
>> about it (back in 1993/4) got a very shirty answer back about it not
>> being properly supported even then, let alone 22 years later.
>
> You are right, some bugs are very annoying. The left-open files was
> one of them, but it's possible to close them with an SWI-call.
>
> And this Pascal-compiler worked relativly strict after the "real"
> pascal-standard at that time (absolutly not comparable to the other
> "pseudo" or better said "de facto"-standards, like Turbo-Pascal etc.
> which influenced most of the later compilers on other platforms).
>
> But it was fine to program WIMP-applications with it ;-).
>

Indeed - I managed to write a fairly capable spreadsheet and charting 
app using it, the only problem I remember is that the compile time ended 
up being north of 20 minutes or so for every change (although that may 
be 22 years of exaggeration playing a part!

0
Someone
10/16/2016 7:59:05 PM
In message <7b3c70cf55.JF@freak67>
          JF <jfortkamp@freenet.de> wrote:

> In Nachricht <ntri61$8ku$1@dont-email.me>
>           druck <news@druck.org.uk> haben Sie geschrieben:

>> Even if it was converted to run on 32 bit systems (big job without the
>> source), it would still produce 26 bit output, and changing that would
>> be an even bigger job, even with the source.

>> Emulation is the only practical way with such relics.

>> ---druck

> Thanks for your answer, druck.
> I'm not sure, if this compiler produces ARM-code, it's more probable
> that it's a kind of P-Code which is interpreted. Otherwise yes, a
> rewrite is more or less impossible.

Have you tried the Author - There is info about EasyC/C++ - here in 
the Acorn Programmer group.
Author :- Terry Moffitt BT Belfast Engineering Centre +44 1232894436
(might be worth a letter)

-- 
Colin Ferris Cornwall UK
0
cferris
10/19/2016 12:47:30 PM
In message <805c7fd155.cferris@cferris.freeuk.com>
          cferris@freeRemoveuk.com.invalid wrote:

> In message <7b3c70cf55.JF@freak67>
>           JF <jfortkamp@freenet.de> wrote:

>> In Nachricht <ntri61$8ku$1@dont-email.me>
>>           druck <news@druck.org.uk> haben Sie geschrieben:

>>> Even if it was converted to run on 32 bit systems (big job without the
>>> source), it would still produce 26 bit output, and changing that would
>>> be an even bigger job, even with the source.

>>> Emulation is the only practical way with such relics.

>>> ---druck

>> Thanks for your answer, druck.
>> I'm not sure, if this compiler produces ARM-code, it's more probable
>> that it's a kind of P-Code which is interpreted. Otherwise yes, a
>> rewrite is more or less impossible.

> Have you tried the Author - There is info about EasyC/C++ - here in
> the Acorn Programmer group.
> Author :- Terry Moffitt BT Belfast Engineering Centre +44 1232894436
> (might be worth a letter)

Sorry - bit of mix up - thats Beebugs EasyC not Pascal.


-- 
Colin Ferris Cornwall UK
0
cferris
10/19/2016 1:02:21 PM
cferris@freeremoveuk.com.invalid wrote:
> In message <805c7fd155.cferris@cferris.freeuk.com>
>           cferris@freeRemoveuk.com.invalid wrote:
> > Have you tried the Author - There is info about EasyC/C++ - here in
> > the Acorn Programmer group.
> 
> Sorry - bit of mix up - thats Beebugs EasyC not Pascal.

Discussions have already been under way (mentioning Pascal as a side issue
in the context of other things) - I'm not party to them, and I'm sure those
who have anything to report will do so as and when they have some news.

Theo
0
Theo
10/19/2016 6:13:16 PM
In Nachricht <8Go*V3Eaw@news.chiark.greenend.org.uk>
          Theo Markettos <theom+news@chiark.greenend.org.uk> haben Sie 
geschrieben:

> cferris@freeremoveuk.com.invalid wrote:
>> In message <805c7fd155.cferris@cferris.freeuk.com>
>>           cferris@freeRemoveuk.com.invalid wrote:
>>> Have you tried the Author - There is info about EasyC/C++ - here in
>>> the Acorn Programmer group.
>> 
>> Sorry - bit of mix up - thats Beebugs EasyC not Pascal.

> Discussions have already been under way (mentioning Pascal as a side issue
> in the context of other things) - I'm not party to them, and I'm sure those
> who have anything to report will do so as and when they have some news.

> Theo

Hallo,

thanks to all who replied and tried to help.

I've already seen the discussion on ROO here (thanks to Colin):
https://www.riscosopen.org/forum/forums/2/topics/6653
but I'm not registered at ROO, so could only follow it passiv.

Not sure about it, but perhaps a hint: the ABCLibrary-module also must 
be present. Could it be, that the compiler was written in BASIC and 
has been compiled? Who wrote the compiler?

Greetings,
J�rgen
0
freak67
10/19/2016 9:52:10 PM
Reply: