Program-source in ISAM-files?

Décade III, Duodi de Prairial de l'Année 216 de la Révolution
13:09:00 +0200

"Peter Flass" wrote:

> ...

> ... I'm paranoid about backups.

So am I. I am the only one in my company, as
far as I know, ho does cryptographic hashes on
Backups. Same thing at home.

Herwig

Décade III, Duodi de Prairial de l'Année 216 de la Révolution
13:13:13 +0200

"Peter Flass" wrote:

> ...

> ... By the way, what language is your compiler written in, if that's not
> "confidential"?

It cannot be confidental because it's the way to do it: PL1
of course, and a little bit of Assembler. More Assembler in
the Runtime-System, of course. The compiler is fully
bootstrapped.

A tale of history - also not a secret because there is public
money in it: I was in the Ada-Compiler project - Ada 83,
on BS2000, a joint venture between Siemens, Honeywell
Bull and Alsys. That was in the eigthies.

This Ada-compiler was written in Ada, and someone had
written an Ada to PL1 translator. With that, we developed
until the first bootstrap.

If I remember righly: producing the Ada-compiler via
PL1 took all night.

Well, you know, the hardware in those times ...

Herwig

"Herwig Huener AQSR" <news20061103@Herwig-Huener.de> wrote in message
news:g2dgjj$c6u$01$1@news.t-online.com...
20 Prairial an 216 de la Révolution
10:25:00 +0200

>robin wrote:

>> Who made the BS2000 ?

>Siemens. And BS2000 is running Europe.

>Herwig

Thanks.

"Herwig Huener" <Herwig.Huener@fujitsu-siemens.com> wrote in message
news:g2j39r$grs$1@nntp.fujitsu-siemens.com...
> Décade III, Duodi de Prairial de l'Année 216 de la Révolution
> 13:13:13 +0200
>
> "Peter Flass" wrote:
>
> > ... By the way, what language is your compiler written in, if that's not
> > "confidential"?
>
> It cannot be confidental because it's the way to do it: PL1
> of course, and a little bit of Assembler. More Assembler in
> the Runtime-System, of course. The compiler is fully
> bootstrapped.

That makes at least five written in PL/I :
IBM's PL/I for the PC and mainframe compilers,
Multix, and Peter Flass's OS/2 PL/I compiler.

24 Prairial an 216 de la Révolution
23:03:20 +0200

robin wrote:

> ...

>>It cannot be confidental because it's the way to do it: PL1
>>of course, and a little bit of Assembler. More Assembler in
>>the Runtime-System, of course. The compiler is fully
>>bootstrapped.

> That makes at least five written in PL/I :
> IBM's PL/I for the PC and mainframe compilers,
> Multix, and Peter Flass's OS/2 PL/I compiler.

Writing a compiler in it's own language with the
intent of bootstrapping makes the compiler-writer
*very* interested in the quality of his work -
after all, he is the first to suffer or to benefit.

This has been done with languages which are not
very suitable for compiler-writing. MicroFocus,
for instance, did (does?) it with their
COBOL-Compiler products.

Herwig

--
**************************************************
* Herwig Huener http://www.quantenrente.de *
* Assistierender QuantenScharfRichter *
**************************************************

robin wrote:
> "Herwig Huener" <Herwig.Huener@fujitsu-siemens.com> wrote in message
> news:g2j39r$grs$1@nntp.fujitsu-siemens.com...
>
>>Décade III, Duodi de Prairial de l'Année 216 de la Révolution
>>13:13:13 +0200
>>
>>"Peter Flass" wrote:
>>
>>
>>>... By the way, what language is your compiler written in, if that's not
>>>"confidential"?
>>
>>It cannot be confidental because it's the way to do it: PL1
>>of course, and a little bit of Assembler. More Assembler in
>>the Runtime-System, of course. The compiler is fully
>>bootstrapped.
>
>
> That makes at least five written in PL/I :
> IBM's PL/I for the PC and mainframe compilers,
> Multix, and Peter Flass's OS/2 PL/I compiler.
>
>

Many more. I believe Freiburghouse's compiler was licensed several
times to different vendors and retargeted. Kednos (if I've gotten this
correct) is selling a descendent of this compiler for HP Vax and Alpha,
at least.

Décade I, Decadi de Messidor de l'Année 216 de la Révolution
16:32:48 +0200

"Shmuel (Seymour J.) Metz" wrote:

> ...

>>Second, comments in the code often refer to line numbers (ISAM-keys)
>>rather than functions or modules.
>
> Ouch!

> ...

> Source code in ISAM files may be common, but comments referring to line
> numbers are insane. Please tell me you're joking.

The joke continues.

I just stumbled over a line

/* ****** A C H T U N G : N E U N U M E R I E R U N G ! ! !****** */

in English, that means "Attention! Re-numbering!", and it refers to
a re-sequencing of ISAM-Keys. Obviously, they'd run out of
line-numbers in a few spots.

I have kept a copy of the old code-base in ISAM-files, just in case
that I need to understand a comment which refers to a specific line.
Now I learn that older such comments are worthless.

Why didn't they just enlengthen the ISAM-keys where necessary
I wonder? All source-files with their individual ISAM-key-length?

BRW: The word "NUMERIERUNG" (meaning "numbering")
is nowadays written "NUMMERIERUNG", because we
had a new orthography in Germany. Reading through code and
documentation of an old project means that one is
confrontated with the old orthography all the time.

There are many hardships in working on an old project.

Herwig

In <g42tl7$b7f$1@nntp.fujitsu-siemens.com>, on 06/27/2008
at 04:31 PM, "Herwig Huener" <Herwig.Huener@fujitsu-siemens.com> said:

>Why didn't they just enlengthen the ISAM-keys where necessary I wonder?

They didn't think that anybody would be dumb enough to refer to line
numbers in the comments?

Extended line numbers may be useful in integrating independent changes to
the same code body, but using them as labels in the comments is just plain
insane.

--
Shmuel (Seymour J.) Metz, SysProg and JOAT <http://patriot.net/~shmuel>

Unsolicited bulk E-mail subject to legal action. I reserve the
right to publicly post or ridicule any abusive E-mail. Reply to
domain Patriot dot net user shmuel+news to contact me. Do not
reply to spamtrap@library.lspace.org