1 package Scalar::Vec::Util;
10 Scalar::Vec::Util - Utility routines for vec strings.
23 XSLoader::load(__PACKAGE__, $VERSION);
26 *SVU_PP = sub () { 1 };
27 *SVU_SIZE = sub () { 1 };
36 use Scalar::Vec::Util qw/vfill vcopy veq/;
39 vfill $s, 0, 100, 1; # Fill with 100 bits 1 starting at 0.
41 vcopy $s, 20, $t, 10, 30; # Copy 30 bits from $s, starting at 20,
42 # to $t, starting at 10.
43 vcopy $t, 10, $t, 20, 30; # Overalapping areas DWIM.
44 if (veq $t, 10, $t, 20, 30) { ... } # Yes, they are equal now.
48 A set of utilities to manipulate bits in vec strings.
49 Highly optimized XS routines are used when available, but straightforward pure perl replacements are also provided for platforms without a C compiler.
51 This module doesn't reimplement bit vectors.
52 It can be used on the very same scalars that C<vec> builds, or actually on any Perl string (C<SVt_PV>).
58 True when pure perl fallbacks are used instead of XS functions.
62 Size in bits of the unit used for moves.
63 The higher this value is, the faster the XS functions are.
64 It's usually C<CHAR_BIT * $Config{alignbytes}>, except on non-little-endian architectures where it currently falls back to C<CHAR_BIT> (e.g. SPARC).
68 =head2 C<vfill $vec, $start, $length, $bit>
70 Starting at C<$start> in C<$vec>, fills C<$length> bits with C<$bit>.
71 Grows C<$vec> if necessary.
76 for (@_) { return 0 unless defined }
81 (undef, my $s, my $l, my $x) = @_;
82 croak "Invalid argument" unless _alldef @_;
86 my $t = int($s / $SIZE) + 1;
87 my $u = int(($s + $l) / $SIZE);
88 if ($SIZE * $t < $s + $l and $t <= $u) {
89 vec($_[0], $_, 1) = $x for $s .. $SIZE * $t - 1;
90 vec($_[0], $_, $SIZE) = $x for $t .. $u - 1;
91 vec($_[0], $_, 1) = $x for $SIZE * $u .. $s + $l - 1;
93 vec($_[0], $_, 1) = $x for $s .. $s + $l - 1;
97 =head2 C<< vcopy $from => $from_start, $to => $to_start, $length >>
99 Copies C<$length> bits starting at C<$from_start> in C<$from> to C<$to_start> in C<$to>.
100 If C<$from_start + $length> is too long for C<$from>, zeros are copied past C<$length>.
101 Grows C<$to> if necessary.
102 Doesn't need to allocate any extra memory.
107 my ($fs, $ts, $l) = @_[1, 3, 4];
108 croak "Invalid argument" unless _alldef @_;
110 my $step = $ts - $fs;
112 vec($_[2], $_ + $step, 1) = vec($_[0], $_, 1) for $fs .. $fs + $l - 1;
113 } else { # There's a risk of overwriting if $_[0] and $_[2] are the same SV.
114 vec($_[2], $_ + $step, 1) = vec($_[0], $_, 1) for reverse $fs .. $fs + $l - 1;
118 =head2 C<< vshift $v, $start, $length => $bits [, $insert ] >>
120 In the area starting at C<$start> and of length C<$length> in C<$v>, shift bits C<abs $bits> positions left if C<< $bits > 0 >> and right otherwise.
121 If C<$insert> is defined, also fills the resulting gap with ones if C<$insert> is true and zeros if it's false.
122 Bits outside of the specified area are left untouched.
123 Doesn't need to allocate any extra memory.
128 my ($start, $length, $bits, $insert) = @_[1 .. 4];
135 $bits = $length if $bits > $length;
138 vcopy($_[0], $start, $_[0], $start + $bits, $length);
139 vfill($_[0], $start, $bits, $insert) if defined $insert;
141 vcopy($_[0], $start + $bits, $_[0], $start, $length);
142 vfill($_[0], $start + $length, $bits, $insert) if defined $insert;
146 =head2 C<< veq $v1 => $v1_start, $v2 => $v2_start, $length >>
148 Returns true if the C<$length> bits starting at C<$v1_start> in C<$v1> and C<$v2_start> in C<$v2> are equal, and false otherwise.
149 If needed, C<$length> is decreased to fit inside C<$v1> and C<$v2> boundaries.
154 my ($s1, $s2, $l) = @_[1, 3, 4];
155 croak "Invalid argument" unless _alldef @_;
158 return 0 if vec($_[0], $s1 + $i, 1) != vec($_[2], $s2 + $i, 1);
166 The functions L</vfill>, L</vcopy>, L</vshift> and L</veq> are only exported on request.
167 All of them are exported by the tags C<':funcs'> and C<':all'>.
169 The constants L</SVU_PP> and L</SVU_SIZE> are also only exported on request.
170 They are all exported by the tags C<':consts'> and C<':all'>.
174 use base qw/Exporter/;
178 'funcs' => [ qw/vfill vcopy vshift veq/ ],
179 'consts' => [ qw/SVU_PP SVU_SIZE/ ]
181 our @EXPORT_OK = map { @$_ } values %EXPORT_TAGS;
182 $EXPORT_TAGS{'all'} = [ @EXPORT_OK ];
186 The following timings were obtained by running the C<samples/bench.pl> script.
187 The C<_pp> entries are the pure Perl versions, whereas C<_bv> are L<Bit::Vector> versions.
191 =item This is for perl 5.8.8 on a Core 2 Duo 2.66GHz machine (unit is 64 bits).
193 Filling bits at a given position :
194 Rate vfill_pp vfill_bv vfill
195 vfill_pp 80.3/s -- -100% -100%
196 vfill_bv 1053399/s 1312401% -- -11%
197 vfill 1180792/s 1471129% 12% --
199 Copying bits from a bit vector to a different one :
200 Rate vcopy_pp vcopy_bv vcopy
201 vcopy_pp 112/s -- -100% -100%
202 vcopy_bv 62599/s 55622% -- -89%
203 vcopy 558491/s 497036% 792% --
205 Moving bits in the same bit vector from a given position to a different one :
206 Rate vmove_pp vmove_bv vmove
207 vmove_pp 64.8/s -- -100% -100%
208 vmove_bv 64742/s 99751% -- -88%
209 vmove 547980/s 845043% 746% --
211 Testing bit equality from different positions of different bit vectors :
212 Rate veq_pp veq_bv veq
213 veq_pp 92.7/s -- -100% -100%
214 veq_bv 32777/s 35241% -- -94%
215 veq 505828/s 545300% 1443% --
217 =item This is for perl 5.10.0 on a Pentium 4 3.0GHz (unit is 32 bits).
219 Rate vfill_pp vfill_bv vfill
220 vfill_pp 185/s -- -100% -100%
221 vfill_bv 407979/s 220068% -- -16%
222 vfill 486022/s 262184% 19% --
224 Rate vcopy_pp vcopy_bv vcopy
225 vcopy_pp 61.5/s -- -100% -100%
226 vcopy_bv 32548/s 52853% -- -83%
227 vcopy 187360/s 304724% 476% --
229 Rate vmove_pp vmove_bv vmove
230 vmove_pp 63.1/s -- -100% -100%
231 vmove_bv 32829/s 51933% -- -83%
232 vmove 188572/s 298787% 474% --
234 Rate veq_pp veq_bv veq
235 veq_pp 34.2/s -- -100% -100%
236 veq_bv 17518/s 51190% -- -91%
237 veq 192181/s 562591% 997% --
239 =item This is for perl 5.10.0 on an UltraSPARC-IIi (unit is 8 bits).
241 Rate vfill_pp vfill vfill_bv
242 vfill_pp 4.23/s -- -100% -100%
243 vfill 30039/s 709283% -- -17%
244 vfill_bv 36022/s 850568% 20% --
246 Rate vcopy_pp vcopy_bv vcopy
247 vcopy_pp 2.74/s -- -100% -100%
248 vcopy_bv 8146/s 297694% -- -60%
249 vcopy 20266/s 740740% 149% --
251 Rate vmove_pp vmove_bv vmove
252 vmove_pp 2.66/s -- -100% -100%
253 vmove_bv 8274/s 311196% -- -59%
254 vmove 20287/s 763190% 145% --
256 Rate veq_pp veq_bv veq
257 veq_pp 7.33/s -- -100% -100%
258 veq_bv 2499/s 33978% -- -87%
259 veq 19675/s 268193% 687% --
265 Please report architectures where we can't use the alignment as the move unit.
266 I'll add exceptions for them.
270 L<Carp>, L<Exporter> (core modules since perl 5), L<XSLoader> (since perl 5.006).
274 L<Bit::Vector> gives a complete reimplementation of bit vectors.
278 Vincent Pit, C<< <perl at profvince.com> >>, L<http://www.profvince.com>.
280 You can contact me by mail or on C<irc.perl.org> (vincent).
284 Please report any bugs or feature requests to C<bug-scalar-vec-util at rt.cpan.org>, or through the web interface at L<http://rt.cpan.org/NoAuth/ReportBug.html?Queue=Scalar-Vec-Util>.
285 I will be notified, and then you'll automatically be notified of progress on your bug as I make changes.
289 You can find documentation for this module with the perldoc command.
291 perldoc Scalar::Vec::Util
293 Tests code coverage report is available at L<http://www.profvince.com/perl/cover/Scalar-Vec-Util>.
295 =head1 COPYRIGHT & LICENSE
297 Copyright 2008 Vincent Pit, all rights reserved.
299 This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.
303 1; # End of Scalar::Vec::Util