From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00 autolearn=ham
	autolearn_force=no version=3.4.4
X-Google-Thread: 103376,223570f8a5a85f0a
X-Google-Attributes: gid103376,public
X-Google-Language: ENGLISH,ASCII-7-bit
Path: 
 g2news1.google.com!news1.google.com!news.glorb.com!border1.nntp.dca.giganews.com!border2.nntp.dca.giganews.com!nntp.giganews.com!wn14feed!worldnet.att.net!attbi_s02.POSTED!53ab2750!not-for-mail
From: tmoran@acm.org
Newsgroups: comp.lang.ada
Subject: Re: unbound array?
References: <41046a70@dnews.tpgi.com.au>
X-Newsreader: Tom's custom newsreader
Message-ID: <LR0Nc.163873$IQ4.146435@attbi_s02>
NNTP-Posting-Host: 24.6.132.82
X-Complaints-To: abuse@comcast.net
X-Trace: attbi_s02 1090820587 24.6.132.82 (Mon, 26 Jul 2004 05:43:07 GMT)
NNTP-Posting-Date: Mon, 26 Jul 2004 05:43:07 GMT
Organization: Comcast Online
Date: Mon, 26 Jul 2004 05:43:07 GMT
Xref: g2news1.google.com comp.lang.ada:2388
Date: 2004-07-26T05:43:07+00:00
List-Id: <comp.lang.ada>

>The problem I am facing is that I don't know how many words I will encounter
>in the text file. I do not want to constrain size to 100 as I have above. I

> But it would be a little ridiculous assigning "size" = "number of
> characters in text file".
   If "words" must be separated by blanks or other punctuation, then
size := size of text file/2 will always do.
   If it's real text, with a known statistical word length distribution,
you can use
size = a large but reasonable size + file_size/(avg_word_and_punctuation_size)
and, by choosing the first parameter you can make it arbitrarily
improbable that the data will overflow your array.

A q&d approach for small files could use recursion:

 type word_lengths is range 1 .. whatever is reasonable

 type length_type is array (integer range <>) of word_lengths;

 function add_a_count(current_list : length_type) return length_type is
 begin
   if end of file then
     return current_list;
   else
     return add_a_count(current_list & (read_next_length));
   end if;
 end add_a_count;

 null_length : length_type(1 .. 0);
 word_length : length_type := add_a_count(null_length);

Perhaps more practical would be to do it in chunks of some suitable
size, recursing only if the current chunk became full.