Received: by 10.180.95.2 with SMTP id dg2mr905206wib.2.1348422748917;
        Sun, 23 Sep 2012 10:52:28 -0700 (PDT)
MIME-Version: 1.0
Path: ed8ni33250761wib.0!nntp.google.com!goblin1!goblin2!goblin.stu.neva.ru!aioe.org!.POSTED!not-for-mail
From: Kaz Kylheku <k...@kylheku.com>
Newsgroups: comp.lang.lisp,comp.emacs
Subject: Re: tricky text manipulation
Followup-To: comp.lang.lisp
Date: Sun, 23 Sep 2012 17:52:23 +0000 (UTC)
Organization: Aioe.org NNTP Server
Lines: 72
Message-ID: <20120923103407.343@kylheku.com>
References: <k3fojf$ri1$1@dont-email.me>
NNTP-Posting-Host: X+c6YNb3AaWMPA3YfA4opg.user.speranza.aioe.org
X-Complaints-To: abuse@aioe.org
User-Agent: slrn/pre1.0.0-18 (Linux)
X-Notice: Filtered by postfilter v. 0.8.2

On 2012-09-20, no.top.p...@gmail.com <no.top.p...@gmail.com> wrote:
> But now, the contents: A,B,C,D will contain common, 
> repeated/redundant text, which usually is near the
> page-beginning, but must be handled anywhere is the page.

The problem with your problem description is that it is described
in terms of these abstract hypotheticals.

However, the details depend on the actual data.

You must provide some concrete instances of the actual catenated web pages,
rather than "uA|vB|wC|xD|".

Also show the exact output that you want from each input sample (i.e. each set
of catenated web pages).

Since these things are probably large, you should put them on some file hosting
site (perhaps as a compressed archive) and give a URL.

Without these reference input/output pairs, it is impossible to write, debug,
test, and refine a piece of software.

If you keep the exact input and output pairs to yourself (like you did
throughout the entire large thread in the other newsgroups where you posted
this) it will just be another big waste of time.

> So then whith the garbage represented as 8, it looks like:
> u8A|v8B|wC8|x8D|
>
> The algorithm that I see is:
>
> the human starts reading/editing accumulatorFile,
>   and notices the/some repeated/redundant/garbage,
>   which be pastes out to FileH;

That, right off the bat, is not an algorithm. An algorithm must describe how
garbage is recognized and delimited.  The above part is essentially
programming. The human prepares a specification of what is "redundant garbage".

> and then the program does:

And now the algorithm begins:

>  scan accumulatorFile, and delete all copies of the
>  text-block:H, except the frst copy.

It's probably much more productive to look for what to *keep*, rather
than what to *delete*.

That is to say, take the catenated web pages and scrape them for interesting
content, skipping uninteresting content.

> That should'nt be difficult, but the following refinement
>     is also required:
>    for matching purposes, ignore  all "["d{d}"]" and 
>    spaces and tabs.

What is this notation "[" d{d} "]"  ? Is it BNF?

You can't just drop random notations in the middle of a sentence without
explaining what they are, and expect to be perfectly understood.

> So the following 2 lines should match:
> the    [4] cat sat  on the mat
>   the [27] cat sat on the mat

Ah, of course, d means digit? Oh, stupid me!  Would it kill you to write
something like:

  ignore all "[" d { d } "]" where this is EBNF notation, and the
  grammar symbol d stands for a digit?