Provided by: clc-intercal_1.0~4pre1.-94.-2-3_all 
NAME
Charset::Hollerith - allows to use Hollerith string constants in ASCII programs (and v.v.)
SYNOPSIS
use Charset::Hollerith qw(hollerith2ascii);
my $a = hollerith2ascii "(Hollerith text)";
DESCRIPTION
Charset::Hollerith defines functions to convert between a subset of ASCII and a subset of
nonstandard Hollerith (since there isn't such a thing as a standard Hollerith we defined
our own variant which is guaranteed to be incompatible with all versions of Hollerith used
by IBM hardware - however, for each character code we have used the code used by some (but
not all) IBM card reader, if the code exists in Hollerith at all, or we have made one up
in some logical way (such as overpunching) if no IBM hardware had that particular
character.
The two functions hollerith2ascii and ascii2hollerith are exportable but not exported by
default. They do the obvious thing to their argument.
HOLLERITH CHARACTER TABLE
A Hollerith string is a sequence of 12-bit characters; they are encoded as two ASCII
characters, containing 6 bits each: the first character contains punches 12, 0, 2, 4, 6, 8
and the second character contains punches 11, 1, 3, 5, 7, 9; interleaving the two
characters gives the original 12 bits. To make the characters printable on ASCII
terminals, bit 7 is always set to 0, and bit 6 is set to the complement of bit 5. These
two bits are ignored when reading Hollerith cards.
Some Hollerith characters (produced by overpunching) can be converted to sequences of
ASCII characters; ascii2hollerith will correctly recognise the sequences.
The following punched cards document the encoding of characters (the last three symbols at
the end nongraphic symbols in ASCII; the previous two symbols correspond to multicharacter
sequences):
' !"#$%&()*+,-./:;<=>?@[\]^_`{|}~XX0123456789
12 * * * * * * * * * * 12
11 * * * * ** ** * * * 11
0 * * * * **** * * *** 0
1 * * * 1
2 * * * * * 2
3 ** * * * * 3
4 *** * * ** * * * * 4
5 * * * * * * 5
6 * * ** * * 6
7 * *** * * 7
8 * ******** * * ******* * * * * * 8
9 * * * 9
ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrs
12 ********* ********* 12
11 ********* ********* 11
0 ********* * * 0
1 * * ** ** * 1
2 * * * ** ** * 2
3 * * * ** ** 3
4 * * * ** ** 4
5 * * * ** ** 5
6 * * * ** ** 6
7 * * * ** ** 7
8 * * * ** ** 8
9 * * * * * 9
tuvwxyz [] ". NL CR HT
12 * * * 12
11 * 11
0 ******* * * 0
1 * * * 1
2 * * * * 2
3 ** * * * * 3
4 ** * * * * 4
5 ** * * * 5
6 ** * * * 6
7 ** * * * * * 7
8 ** * * * * 8
9 * * * * 9
PLEASE NOTE that versions of CLC-INTERCAL before 1.-94.-2 had a bug which caused a rabbit
to be represented as 12-3-2-8 instead of 12-3-7-8. Cards punched with such older versions,
and containing rabbits, will need to be copied with one of the rabbit holes moved from row
2 to row 7.
COPYRIGHT
This module is part of CLC-INTERCAL.
Copyright (C) 2000, 2002, 2006, 2007 Claudio Calvelli, all rights reserved
See the files README and COPYING in the distribution for information.
SEE ALSO
A qualified psychiatrist.
POD ERRORS
Hey! The above document had some coding errors, which are explained below:
Around line 253:
Non-ASCII character seen before =encoding in
'!"#$%&()*+,-./:;<=>?@[\]^_`{|}~XX0123456789'. Assuming ISO8859-1