view release on metacpan or  search on metacpan

lib/Tie/Hash/MinPerfHashTwoLevel/OnDisk.pm  view on Meta::CPAN

    my ($self, $key)= @_;
    return fetch_by_key($self->{mount},$key);
}

sub FIRSTKEY {
    my ($self)= @_;
    $self->{iter_idx}= 0;
    return $self->NEXTKEY();
}

sub NEXTKEY {
    my ($self, $lastkey)= @_;
    fetch_by_index($self->{mount},$self->{iter_idx}++,my $key);
    return $key;
}

sub SCALAR {
    my ($self)= @_;
    my $buckets= $self->get_hdr_num_buckets();
    if ($scalar_has_slash) {
        $buckets .= "/" . $buckets;
    }
    return $buckets;
}

sub UNTIE {
    my ($self)= @_;
}

sub DESTROY {
    my ($self)= @_;
    unmount_file($self->{mount}) if $self->{mount};
}

sub STORE {
    my ($self, $key, $value)= @_;
    confess __PACKAGE__ . " is readonly, STORE operations are not supported";
}

sub DELETE {
    my ($self, $key)= @_;
    confess __PACKAGE__ . " is readonly, DELETE operations are not supported";
}

sub CLEAR {
    my ($self)= @_;
    confess __PACKAGE__ . " is readonly, CLEAR operations are not supported";
}

sub make_file {
    my ($class, %opts)= @_;

    my $ofile= $opts{file} 
        or die "file is a mandatory option to make_file";
    my $source_hash= $opts{source_hash}
        or die "source_hash is a mandatory option to make_file";
    $opts{comment}= "" unless defined $opts{comment};
    $opts{variant}= $DEFAULT_VARIANT unless defined $opts{variant};
    
    my $comment= $opts{comment}||"";
    my $debug= $opts{debug} || 0;
    my $variant= int($opts{variant});
    my $deterministic;
    $deterministic //= delete $opts{canonical};
    $deterministic //= delete $opts{deterministic};
    $deterministic //= 1;

                    #1234567812345678
    $opts{seed} = "MinPerfHash2Levl"
        if !defined($opts{seed}) and $deterministic;

    my $compute_flags= int($opts{compute_flags}||0);
    $compute_flags |= MPH_F_NO_DEDUPE if delete $opts{no_dedupe};
    $compute_flags |= MPH_F_DETERMINISTIC
        if $deterministic;
    $compute_flags |= MPH_F_FILTER_UNDEF
        if delete $opts{filter_undef};

    die "Unknown variant '$variant', max known is "
        . MAX_VARIANT . " default is " . $DEFAULT_VARIANT
        if $variant > MAX_VARIANT;
    die "Unknown variant '$variant', min known is "
        . MIN_VARIANT . " default is " . $DEFAULT_VARIANT
        if $variant < MIN_VARIANT;

    die "comment cannot contain null"
        if index($comment,"\0") >= 0;

    my $seed= $opts{seed};
    my $hasher= Algorithm::MinPerfHashTwoLevel->new(
        debug => $debug,
        seed => (ref $seed ? $$seed : $seed),
        variant => $variant,
        compute_flags => $compute_flags,
        max_tries => $opts{max_tries},
    );
    my $buckets= $hasher->compute($source_hash);
    my $buf_length= $hasher->{buf_length};
    my $state= $hasher->{state};
    my $buf= packed_xs($variant, $buf_length, $state, $comment, $compute_flags, @$buckets);
    $$seed= $hasher->get_seed if ref $seed;

    my $tmp_file= "$ofile.$$";
    open my $ofh, ">", $tmp_file
        or die "Failed to open $tmp_file for output";
    print $ofh $buf
        or die "failed to print to '$tmp_file': $!";
    close $ofh
        or die "failed to close '$tmp_file': $!";
    rename $tmp_file, $ofile
        or die "failed to rename '$tmp_file' to '$ofile': $!";
    return $ofile;
}

sub validate_file {
    my ($class, %opts)= @_;
    my $file= $opts{file}
        or die "file is a mandatory option to validate_file";
    my $verbose= $opts{verbose};
    my ($variant,$msg);

    my $error_sv;
    my $self= $class->new(file => $file, flags => MPH_F_VALIDATE, error_rsv => \$error_sv);
    if ($self) {
        $msg= sprintf "file '%s' is a valid '%s' file\n"
         . "  variant: %d\n"
         . "  keys: %d\n"
         . "  hash-state: %s\n"
         . "  table  checksum: %016x\n"
         . "  string checksum: %016x\n"
         . "  comment: %s"
         ,  $file,
            MAGIC_STR,
            $self->get_hdr_variant,
            $self->get_hdr_num_buckets,
            unpack("H*", $self->get_state),
            $self->get_hdr_table_checksum,
            $self->get_hdr_str_buf_checksum,
            $self->get_comment,
        ;
        $variant = $self->get_hdr_variant;
    } else {
        $msg= $error_sv;
    }
    if ($verbose) {
        if (defined $variant) {
            print $msg;
        } else {
            die $msg."\n";
        }
    }
    return ($variant, $msg);
}



1;
__END__

=head1 NAME

Tie::Hash::MinPerfHashTwoLevel::OnDisk - construct or tie a "two level" minimal perfect hash based on disk

=head1 SYNOPSIS

  use Tie::Hash::MinPerfHashTwoLevel::OnDisk;

  Tie::Hash::MinPerfHashTwoLevel::OnDisk->make_file(
    file => $some_file,
    source_hash => $some_hash,
    comment => "this is a comment",
    debug => 0,
  );

  my %hash;
  tie %hash, "Tie::Hash::MinPerfHashTwoLevel::OnDisk", $some_file;

=head1 DESCRIPTION

This module allows one to either construct, or use a precomputed minimal
perfect hash on disk via tied interface. The disk image of the hash is
loaded by using mmap, which means that multiple processes may use the
same disk image at the same time without memory duplication. The hash
is readonly, and may only contain string values.

=head2 METHODS

=over 4

=item make_file

Construct a new file from a given 'source_hash' argument. Takes the following arguments:

=over 4

=item file

The file name to produce, mandatory.

=item comment

An arbitrary piece of text of your choosing. Can be extracted from
the file later if needed. Only practical restriction on the value is
that it cannot contain a null.

=item seed

A 16 byte string (or a reference to one) to use as the seed for
the hashing and generation process. If this is omitted a standard default
is chosen.

If it should prove impossible to construct a solution using the seed chosen
then a new one will be constructed deterministically from the old until a
solution is found (see L<max_tries>) (prior to version v0.10 this used rand()).
Should you wish to access the seed actually used for the final solution
then you can pass in a reference to a scalar containing your chosen seed.
The reference scalar will be updated after successful construction.

Thus both of the following are valid:

    Tie::Hash::MinPerfHashTwoLevel::OnDisk->make_file(seed => "1234567812345678", ...);
    Tie::Hash::MinPerfHashTwoLevel::OnDisk->make_file(seed => \my $seed= "1234567812345678", ...);

=item compute_flags

This is an integer which contains various flags which control construction.
They are as follows:

       MPH_F_FILTER_UNDEF   =>  1 - filter keys with undef values
       MPH_F_DETERMINISTIC  =>  2 - repeatable results (sort keys during processing)
       MPH_F_NO_DEDUPE      =>  4 - do not dedupe strings in final buffer

These constants can be imported via the ":flags" tag, but there are also options that
have the equivalent result, see C<no_dedupe>, C<deterministic> and C<filter_undef>.

=item no_dedupe

Speed up construction at the cost of a larger string buffer by disabling
deduplication of values and keys.  Same as setting the MPH_F_NO_DEDUPE bit
in compute_flags.

=item deterministic

=item canonical

Produce a canonical result from the source data set, albeit somewhat less quickly
than normal. Note this is independent of supplying a seed, the same seed may produce
a different result for the same set of keys without this option.  Same
as setting the MPH_F_DETERMINISTIC bit in compute_flags.

=item filter_undef

Ignore keys with undef values during construction. This means that exists() checks
may differ between source and the constructed hash table, but avoids the need to
store such keys in the resulting file, saving space. Same as setting the
MPH_F_FILTER_UNDEF bit in compute_flags.

=item max_tries

The maximum number of attempts to make to find a solution for this keyset.
Defaults to 3. 

=item debug

Enable debug during generation.

=item variant

Select which variant of construction algorithm and file format to produce.
When omitted the variant is determined by the global var

   $Tie::Hash::MinPerfHashTwoLevel::DEFAULT_VARIANT

which itself defaults to the latest version. This is mostly for testing,
Older variants will be deprecated and removed eventually.

The list of supported variants is as follows:

    5 - Xor, siphash, with inthash, 8 byte alignment, one checksum.

In version 0.15 we switched hash functions to use SipHash(1-3), which
unfortunately made supporting variants prior to 5 impossible.

=back

=item validate_file

Validate the file specified by the 'file' argument. Returns a list of
two values, 'variant' and 'message'. If the file fails validation the 'variant'
will be undef and the 'message' will contain an error message. If the file
passes validation the 'variant' will specify the variant of the file
(currently only 0 is valid), and 'message' will contain some basic information
about the file, such as how many keys it contains, the comment it was
created with, etc.

=back

=head2 SUBS

=over 4

=item mph2l_tied_hashref

Simple wrapper to replace the cumbersome

    tie my %hash, "Tie::Hash::MinPerfHashTwoLevel::OnDisk", $file;

with a simple sub that can be imported

    my $hashref= mph2l_tied_hashref($file,$validate);

The validate flag causes MPH_F_VALIDATE validations to occur on load.

=item mph2l_make_file

Sub form of L<make_file>. Eg:

  use Tie::Hash::MinPerfHashTwoLevel::OnDisk;
  Tie::Hash::MinPerfHashTwoLevel::OnDisk->make_file(@args);

is identical to

  use Tie::Hash::MinPerfHashTwoLevel::OnDisk qw(mph2l_make_file);
  mph2l_make_file(@args);