PerlIO::via::Base64 - PerlIO layer for base64 (MIME) encoded strings
This documentation describes version 0.08.
use PerlIO::via::Base64; PerlIO::via::Base64->eol( "\n" ); # default, write lines 76 bytes long PerlIO::via::Base64->eol( '' ); # no line endings, write one long string use PerlIO::via::Base64 eol => "\n"; open( my $in,'<:via(Base64)','file.mime' ) or die "Can't open file.mime for reading: $!\n"; open( my $out,'>:via(Base64)','file.mime' ) or die "Can't open file.mime for writing: $!\n";
This module implements a PerlIO layer that works on files encoded in the Base64 format (as described in RFC 2045). It will decode from base64 format while reading from a handle, and it will encode to base64 while writing to a handle.
There is one class method. It can also be specified as a key value pair in the
use PerlIO::via::Base64 eol => ''; PerlIO::via::Base64->eol( '' ); # no line endings, one long string open( my $out,'>:via(Base64)','file.mime' ); # no line endings $eol= PerlIO::via::Base64->eol; # obtain current setting
MIME (Base64) encoded files can be written with line endings, causing all lines (except the last) to be exactly 76 bytes long. By default a linefeed ("\n") will be assumed.
Calling this class method with a new value will cause all subsequently opened files to assume that new setting. The eol value however is remembered within the layer, so that it becomes part of the information that is associated with that file.
If it were possible to pass parameters such as this to the layer while opening the file, that would have been the approach taken. Since that is not possible yet, this way of doing it seems to be the next best thing.
The current implementation slurps the whole contents of a handle into memory before doing any encoding or decoding. This may change in the future when I finally figured out how READ and WRITE are supposed to work on incompletely processed buffers.
Copyright (c) 2002, 2003, 2004, 2009, 2012 Elizabeth Mattijsen <firstname.lastname@example.org>. All rights reserved. This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.