Path: csiph.com!usenet.pasdenom.info!weretis.net!feeder4.news.weretis.net!feeds.phibee-telecom.net!newsfeed.xs4all.nl!newsfeed5.news.xs4all.nl!xs4all!post.news.xs4all.nl!not-for-mail
To: python-list@python.org
From: Dennis Lee Bieber <wlfraed@ix.netcom.com>
Subject: Re: =?ISO-8859-1?Q?the_meaning_of_r=EF=BE=95?= =?ISO-8859-1?Q?.......=EF=BE?=
Date: Mon, 23 Jul 2012 14:10:02 -0400
Organization: > Bestiaria Support Staff <
References: <CAEMsKDun53k2Kyk9TFT1kBxFB2zgp7rhA3W0LLPuS74+nbwiPA@mail.gmail.com> <mailman.2455.1343030385.4697.python-list@python.org> <500d0632$0$1504$c3e8da3$76491128@news.astraweb.com> <roy-D44797.08552123072012@news.panix.com> <jujja3$dfn$1@speranza.aioe.org> <mailman.2467.1343050549.4697.python-list@python.org> <jujkut$hr3$1@speranza.aioe.org> <mailman.2473.1343053204.4697.python-list@python.org> <jujnt8$osp$2@speranza.aioe.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Precedence: list
Newsgroups: comp.lang.python
Message-ID: <mailman.2489.1343067003.4697.python-list@python.org>
Lines: 56
NNTP-Posting-Host: 2001:888:2000:d::a6
Xref: csiph.com comp.lang.python:25909

On Mon, 23 Jul 2012 16:42:51 +0200, Henrik Faber <hfaber@invalid.net>
declaimed the following in gmane.comp.python.general:

> 
> If that was written by my coworkers, I'd strangle them.
>
	My first real assignment, 31 years ago, was porting an application
to CDC MP-60 FORTRAN (what I called "FORTRAN MINUS TWO"). This was a
minimal FORTRAN implementation in which one could not do things like:

	ix = 20
	call xyz(ix, ix+2, ix-2)

forcing us to produce such abominations as

	ix = 20

	jinx = ix + 2
	minx = ix - 2

	call xyz(ix, jinx, minx)

	Now, when you take into consideration that this application,
purportedly a requirements traceability system, was only a few
functions* shy of being a (pre-SQL) relational database using a form of
relational algebra, and did lots of record<>disk-block^ packing, then
there were many calls to subprograms that needed things like block #,
record #, start&end bytes within record/blocks, with both source and
destination information needed.

* It needed a dynamic "project" operation; records contained a "format
ID" (formats were themselves stored in the "database") which identified
the layout of the data in the record. A dynamic "project" operation
would have to build a format from the specifications of the desired
fields. It also would have needed a simple join operation (also
dynamically creating a format for the result set).

^ The MP-60 did not make that easy either... The F-2 did not have direct
access I/O, so one had to use OS services to position to "block #" of a
file, and then perform the FORTRAN I/O operation. BUT: disk blocks were
512 bytes, but only around 488 bytes were available to the user -- the
OS stored something in the rest of the block. That meant we had to
change the program to work in 488 byte blocks, but to address the file
in 512 byte blocks.

	My second major assignment was to maintain an application that had
been originally written (by a subcontractor) using preprocessor to
convert "C" style block statements into FORTRAN-IV (the only valid use
I've ever seen for the ASSIGNed GOTO -- emulating the return from
"subroutine" calls). I got to maintain the /output/ of the preprocessor
-- where programs were 3000-30000 lines of code in a single file, all
variables were global...
-- 
	Wulfraed                 Dennis Lee Bieber         AF6VN
        wlfraed@ix.netcom.com    HTTP://wlfraed.home.netcom.com/