1 package IO::Uncompress::Inflate ;
8 use IO::Compress::Base::Common qw(:Status createSelfTiedObject);
9 use IO::Compress::Zlib::Constants;
11 use IO::Uncompress::RawInflate ;
14 our ($VERSION, @ISA, @EXPORT_OK, %EXPORT_TAGS, $InflateError);
16 $VERSION = '2.000_14';
19 @ISA = qw( Exporter IO::Uncompress::RawInflate );
20 @EXPORT_OK = qw( $InflateError inflate ) ;
21 %EXPORT_TAGS = %IO::Uncompress::RawInflate::DEFLATE_CONSTANTS ;
22 push @{ $EXPORT_TAGS{all} }, @EXPORT_OK ;
23 Exporter::export_ok_tags('all');
29 my $obj = createSelfTiedObject($class, \$InflateError);
31 $obj->_create(undef, 0, @_);
36 my $obj = createSelfTiedObject(undef, \$InflateError);
37 return $obj->_inf(@_);
50 # gunzip always needs adler32
51 $got->value('ADLER32' => 1);
61 $self->smartReadExact(\$magic, ZLIB_HEADER_SIZE);
63 *$self->{HeaderPending} = $magic ;
65 return $self->HeaderError("Header size is " .
66 ZLIB_HEADER_SIZE . " bytes")
67 if length $magic != ZLIB_HEADER_SIZE;
69 #return $self->HeaderError("CRC mismatch.")
71 if ! $self->isZlibMagic($magic) ;
73 *$self->{Type} = 'rfc1950';
82 return $self->_readDeflateHeader($magic) ;
90 my $ADLER32 = unpack("N", $trailer) ;
91 *$self->{Info}{ADLER32} = $ADLER32;
92 return $self->TrailerError("CRC mismatch")
93 if *$self->{Strict} && $ADLER32 != *$self->{Uncomp}->adler32() ;
106 if length $buffer < ZLIB_HEADER_SIZE ;
108 my $hdr = unpack("n", $buffer) ;
109 #return 0 if $hdr % 31 != 0 ;
110 return $self->HeaderError("CRC mismatch.")
113 my ($CMF, $FLG) = unpack "C C", $buffer;
114 my $cm = bits($CMF, ZLIB_CMF_CM_OFFSET, ZLIB_CMF_CM_BITS) ;
116 # Only Deflate supported
117 return $self->HeaderError("Not Deflate (CM is $cm)")
118 if $cm != ZLIB_CMF_CM_DEFLATED ;
120 # Max window value is 7 for Deflate.
121 my $cinfo = bits($CMF, ZLIB_CMF_CINFO_OFFSET, ZLIB_CMF_CINFO_BITS) ;
122 return $self->HeaderError("CINFO > " . ZLIB_CMF_CINFO_MAX .
123 " (CINFO is $cinfo)")
124 if $cinfo > ZLIB_CMF_CINFO_MAX ;
135 ($data >> $offset ) & $mask & 0xFF ;
139 sub _readDeflateHeader
141 my ($self, $buffer) = @_ ;
144 # $self->smartReadExact(\$buffer, ZLIB_HEADER_SIZE);
146 # *$self->{HeaderPending} = $buffer ;
148 # return $self->HeaderError("Header size is " .
149 # ZLIB_HEADER_SIZE . " bytes")
150 # if length $buffer != ZLIB_HEADER_SIZE;
152 # return $self->HeaderError("CRC mismatch.")
153 # if ! isZlibMagic($buffer) ;
156 my ($CMF, $FLG) = unpack "C C", $buffer;
157 my $FDICT = bits($FLG, ZLIB_FLG_FDICT_OFFSET, ZLIB_FLG_FDICT_BITS ),
159 my $cm = bits($CMF, ZLIB_CMF_CM_OFFSET, ZLIB_CMF_CM_BITS) ;
160 $cm == ZLIB_CMF_CM_DEFLATED
161 or return $self->HeaderError("Not Deflate (CM is $cm)") ;
165 $self->smartReadExact(\$buffer, ZLIB_FDICT_SIZE)
166 or return $self->TruncatedHeader("FDICT");
168 $DICTID = unpack("N", $buffer) ;
171 *$self->{Type} = 'rfc1950';
175 'FingerprintLength' => ZLIB_HEADER_SIZE,
176 'HeaderLength' => ZLIB_HEADER_SIZE,
177 'TrailerLength' => ZLIB_TRAILER_SIZE,
181 CM => bits($CMF, ZLIB_CMF_CM_OFFSET, ZLIB_CMF_CM_BITS ),
182 CINFO => bits($CMF, ZLIB_CMF_CINFO_OFFSET, ZLIB_CMF_CINFO_BITS ),
184 FCHECK => bits($FLG, ZLIB_FLG_FCHECK_OFFSET, ZLIB_FLG_FCHECK_BITS),
185 FDICT => bits($FLG, ZLIB_FLG_FDICT_OFFSET, ZLIB_FLG_FDICT_BITS ),
186 FLEVEL => bits($FLG, ZLIB_FLG_LEVEL_OFFSET, ZLIB_FLG_LEVEL_BITS ),
204 IO::Uncompress::Inflate - Read RFC 1950 files/buffers
210 use IO::Uncompress::Inflate qw(inflate $InflateError) ;
212 my $status = inflate $input => $output [,OPTS]
213 or die "inflate failed: $InflateError\n";
215 my $z = new IO::Uncompress::Inflate $input [OPTS]
216 or die "inflate failed: $InflateError\n";
218 $status = $z->read($buffer)
219 $status = $z->read($buffer, $length)
220 $status = $z->read($buffer, $length, $offset)
221 $line = $z->getline()
226 $status = $z->inflateSync()
228 $data = $z->trailingData()
229 $status = $z->nextStream()
230 $data = $z->getHeaderInfo()
232 $z->seek($position, $whence)
244 read($z, $buffer, $length);
245 read($z, $buffer, $length, $offset);
247 seek($z, $position, $whence)
258 B<WARNING -- This is a Beta release>.
262 =item * DO NOT use in production code.
264 =item * The documentation is incomplete in places.
266 =item * Parts of the interface defined here are tentative.
268 =item * Please report any problems you find.
276 This module provides a Perl interface that allows the reading of
277 files/buffers that conform to RFC 1950.
279 For writing RFC 1950 files/buffers, see the companion module IO::Compress::Deflate.
285 =head1 Functional Interface
287 A top-level function, C<inflate>, is provided to carry out
288 "one-shot" uncompression between buffers and/or files. For finer
289 control over the uncompression process, see the L</"OO Interface">
292 use IO::Uncompress::Inflate qw(inflate $InflateError) ;
294 inflate $input => $output [,OPTS]
295 or die "inflate failed: $InflateError\n";
299 The functional interface needs Perl5.005 or better.
302 =head2 inflate $input => $output [, OPTS]
305 C<inflate> expects at least two parameters, C<$input> and C<$output>.
307 =head3 The C<$input> parameter
309 The parameter, C<$input>, is used to define the source of
312 It can take one of the following forms:
318 If the C<$input> parameter is a simple scalar, it is assumed to be a
319 filename. This file will be opened for reading and the input data
320 will be read from it.
324 If the C<$input> parameter is a filehandle, the input data will be
326 The string '-' can be used as an alias for standard input.
328 =item A scalar reference
330 If C<$input> is a scalar reference, the input data will be read
333 =item An array reference
335 If C<$input> is an array reference, each element in the array must be a
338 The input data will be read from each file in turn.
340 The complete array will be walked to ensure that it only
341 contains valid filenames before any data is uncompressed.
345 =item An Input FileGlob string
347 If C<$input> is a string that is delimited by the characters "<" and ">"
348 C<inflate> will assume that it is an I<input fileglob string>. The
349 input is the list of files that match the fileglob.
351 If the fileglob does not match any files ...
353 See L<File::GlobMapper|File::GlobMapper> for more details.
358 If the C<$input> parameter is any other type, C<undef> will be returned.
362 =head3 The C<$output> parameter
364 The parameter C<$output> is used to control the destination of the
365 uncompressed data. This parameter can take one of these forms.
371 If the C<$output> parameter is a simple scalar, it is assumed to be a
372 filename. This file will be opened for writing and the uncompressed
373 data will be written to it.
377 If the C<$output> parameter is a filehandle, the uncompressed data
378 will be written to it.
379 The string '-' can be used as an alias for standard output.
382 =item A scalar reference
384 If C<$output> is a scalar reference, the uncompressed data will be
385 stored in C<$$output>.
389 =item An Array Reference
391 If C<$output> is an array reference, the uncompressed data will be
392 pushed onto the array.
394 =item An Output FileGlob
396 If C<$output> is a string that is delimited by the characters "<" and ">"
397 C<inflate> will assume that it is an I<output fileglob string>. The
398 output is the list of files that match the fileglob.
400 When C<$output> is an fileglob string, C<$input> must also be a fileglob
401 string. Anything else is an error.
405 If the C<$output> parameter is any other type, C<undef> will be returned.
412 When C<$input> maps to multiple compressed files/buffers and C<$output> is
413 a single file/buffer, after uncompression C<$output> will contain a
414 concatenation of all the uncompressed data from each of the input
421 =head2 Optional Parameters
423 Unless specified below, the optional parameters for C<inflate>,
424 C<OPTS>, are the same as those used with the OO interface defined in the
425 L</"Constructor Options"> section below.
429 =item C<< AutoClose => 0|1 >>
431 This option applies to any input or output data streams to
432 C<inflate> that are filehandles.
434 If C<AutoClose> is specified, and the value is true, it will result in all
435 input and/or output filehandles being closed once C<inflate> has
438 This parameter defaults to 0.
441 =item C<< BinModeOut => 0|1 >>
443 When writing to a file or filehandle, set C<binmode> before writing to the
452 =item C<< Append => 0|1 >>
456 =item C<< MultiStream => 0|1 >>
459 If the input file/buffer contains multiple compressed data streams, this
460 option will uncompress the whole lot as a single data stream.
468 =item C<< TrailingData => $scalar >>
470 Returns the data, if any, that is present immediately after the compressed
471 data stream once uncompression is complete.
473 This option can be used when there is useful information immediately
474 following the compressed data stream, and you don't know the length of the
475 compressed data stream.
477 If the input is a buffer, C<trailingData> will return everything from the
478 end of the compressed data stream to the end of the buffer.
480 If the input is a filehandle, C<trailingData> will return the data that is
481 left in the filehandle input buffer once the end of the compressed data
482 stream has been reached. You can then use the filehandle to read the rest
485 Don't bother using C<trailingData> if the input is a filename.
489 If you know the length of the compressed data stream before you start
490 uncompressing, you can avoid having to use C<trailingData> by setting the
491 C<InputLength> option.
502 To read the contents of the file C<file1.txt.1950> and write the
503 compressed data to the file C<file1.txt>.
507 use IO::Uncompress::Inflate qw(inflate $InflateError) ;
509 my $input = "file1.txt.1950";
510 my $output = "file1.txt";
511 inflate $input => $output
512 or die "inflate failed: $InflateError\n";
515 To read from an existing Perl filehandle, C<$input>, and write the
516 uncompressed data to a buffer, C<$buffer>.
520 use IO::Uncompress::Inflate qw(inflate $InflateError) ;
523 my $input = new IO::File "<file1.txt.1950"
524 or die "Cannot open 'file1.txt.1950': $!\n" ;
526 inflate $input => \$buffer
527 or die "inflate failed: $InflateError\n";
529 To uncompress all files in the directory "/my/home" that match "*.txt.1950" and store the compressed data in the same directory
533 use IO::Uncompress::Inflate qw(inflate $InflateError) ;
535 inflate '</my/home/*.txt.1950>' => '</my/home/#1.txt>'
536 or die "inflate failed: $InflateError\n";
538 and if you want to compress each file one at a time, this will do the trick
542 use IO::Uncompress::Inflate qw(inflate $InflateError) ;
544 for my $input ( glob "/my/home/*.txt.1950" )
547 $output =~ s/.1950// ;
548 inflate $input => $output
549 or die "Error compressing '$input': $InflateError\n";
556 The format of the constructor for IO::Uncompress::Inflate is shown below
559 my $z = new IO::Uncompress::Inflate $input [OPTS]
560 or die "IO::Uncompress::Inflate failed: $InflateError\n";
562 Returns an C<IO::Uncompress::Inflate> object on success and undef on failure.
563 The variable C<$InflateError> will contain an error message on failure.
565 If you are running Perl 5.005 or better the object, C<$z>, returned from
566 IO::Uncompress::Inflate can be used exactly like an L<IO::File|IO::File> filehandle.
567 This means that all normal input file operations can be carried out with
568 C<$z>. For example, to read a line from a compressed file/buffer you can
569 use either of these forms
571 $line = $z->getline();
574 The mandatory parameter C<$input> is used to determine the source of the
575 compressed data. This parameter can take one of three forms.
581 If the C<$input> parameter is a scalar, it is assumed to be a filename. This
582 file will be opened for reading and the compressed data will be read from it.
586 If the C<$input> parameter is a filehandle, the compressed data will be
588 The string '-' can be used as an alias for standard input.
591 =item A scalar reference
593 If C<$input> is a scalar reference, the compressed data will be read from
598 =head2 Constructor Options
601 The option names defined below are case insensitive and can be optionally
602 prefixed by a '-'. So all of the following are valid
609 OPTS is a combination of the following options:
613 =item C<< AutoClose => 0|1 >>
615 This option is only valid when the C<$input> parameter is a filehandle. If
616 specified, and the value is true, it will result in the file being closed once
617 either the C<close> method is called or the IO::Uncompress::Inflate object is
620 This parameter defaults to 0.
622 =item C<< MultiStream => 0|1 >>
626 Allows multiple concatenated compressed streams to be treated as a single
627 compressed stream. Decompression will stop once either the end of the
628 file/buffer is reached, an error is encountered (premature eof, corrupt
629 compressed data) or the end of a stream is not immediately followed by the
630 start of another stream.
632 This parameter defaults to 0.
635 =item C<< Prime => $string >>
637 This option will uncompress the contents of C<$string> before processing the
640 This option can be useful when the compressed data is embedded in another
641 file/data structure and it is not possible to work out where the compressed
642 data begins without having to read the first few bytes. If this is the
643 case, the uncompression can be I<primed> with these bytes using this
646 =item C<< Transparent => 0|1 >>
648 If this option is set and the input file or buffer is not compressed data,
649 the module will allow reading of it anyway.
651 This option defaults to 1.
653 =item C<< BlockSize => $num >>
655 When reading the compressed input data, IO::Uncompress::Inflate will read it in
656 blocks of C<$num> bytes.
658 This option defaults to 4096.
660 =item C<< InputLength => $size >>
662 When present this option will limit the number of compressed bytes read
663 from the input file/buffer to C<$size>. This option can be used in the
664 situation where there is useful data directly after the compressed data
665 stream and you know beforehand the exact length of the compressed data
668 This option is mostly used when reading from a filehandle, in which case
669 the file pointer will be left pointing to the first byte directly after the
670 compressed data stream.
674 This option defaults to off.
676 =item C<< Append => 0|1 >>
678 This option controls what the C<read> method does with uncompressed data.
680 If set to 1, all uncompressed data will be appended to the output parameter
681 of the C<read> method.
683 If set to 0, the contents of the output parameter of the C<read> method
684 will be overwritten by the uncompressed data.
688 =item C<< Strict => 0|1 >>
692 This option controls whether the extra checks defined below are used when
693 carrying out the decompression. When Strict is on, the extra tests are
694 carried out, when Strict is off they are not.
696 The default for this option is off.
706 The ADLER32 checksum field must be present.
710 The value of the ADLER32 field read must match the adler32 value of the
711 uncompressed data actually contained in the file.
738 $status = $z->read($buffer)
740 Reads a block of compressed data (the size the the compressed block is
741 determined by the C<Buffer> option in the constructor), uncompresses it and
742 writes any uncompressed data into C<$buffer>. If the C<Append> parameter is
743 set in the constructor, the uncompressed data will be appended to the
744 C<$buffer> parameter. Otherwise C<$buffer> will be overwritten.
746 Returns the number of uncompressed bytes written to C<$buffer>, zero if eof
747 or a negative number on error.
753 $status = $z->read($buffer, $length)
754 $status = $z->read($buffer, $length, $offset)
756 $status = read($z, $buffer, $length)
757 $status = read($z, $buffer, $length, $offset)
759 Attempt to read C<$length> bytes of uncompressed data into C<$buffer>.
761 The main difference between this form of the C<read> method and the
762 previous one, is that this one will attempt to return I<exactly> C<$length>
763 bytes. The only circumstances that this function will not is if end-of-file
764 or an IO error is encountered.
766 Returns the number of uncompressed bytes written to C<$buffer>, zero if eof
767 or a negative number on error.
774 $line = $z->getline()
779 This method fully supports the use of of the variable C<$/> (or
780 C<$INPUT_RECORD_SEPARATOR> or C<$RS> when C<English> is in use) to
781 determine what constitutes an end of line. Paragraph mode, record mode and
782 file slurp mode are all supported.
791 Read a single character.
797 $char = $z->ungetc($string)
805 $status = $z->inflateSync()
814 $hdr = $z->getHeaderInfo();
815 @hdrs = $z->getHeaderInfo();
817 This method returns either a hash reference (in scalar context) or a list
818 or hash references (in array context) that contains information about each
819 of the header fields in the compressed data stream(s).
831 Returns the uncompressed file offset.
842 Returns true if the end of the compressed input stream has been reached.
848 $z->seek($position, $whence);
849 seek($z, $position, $whence);
854 Provides a sub-set of the C<seek> functionality, with the restriction
855 that it is only legal to seek forward in the input file/buffer.
856 It is a fatal error to attempt to seek backward.
860 The C<$whence> parameter takes one the usual values, namely SEEK_SET,
861 SEEK_CUR or SEEK_END.
863 Returns 1 on success, 0 on failure.
872 This is a noop provided for completeness.
878 Returns true if the object currently refers to a opened file/buffer.
882 my $prev = $z->autoflush()
883 my $prev = $z->autoflush(EXPR)
885 If the C<$z> object is associated with a file or a filehandle, this method
886 returns the current autoflush setting for the underlying filehandle. If
887 C<EXPR> is present, and is non-zero, it will enable flushing after every
888 write/print operation.
890 If C<$z> is associated with a buffer, this method has no effect and always
893 B<Note> that the special variable C<$|> B<cannot> be used to set or
894 retrieve the autoflush setting.
896 =head2 input_line_number
898 $z->input_line_number()
899 $z->input_line_number(EXPR)
903 Returns the current uncompressed line number. If C<EXPR> is present it has
904 the effect of setting the line number. Note that setting the line number
905 does not change the current position within the file/buffer being read.
907 The contents of C<$/> are used to to determine what constitutes a line
917 If the C<$z> object is associated with a file or a filehandle, this method
918 will return the underlying file descriptor.
920 If the C<$z> object is is associated with a buffer, this method will
930 Closes the output file/buffer.
934 For most versions of Perl this method will be automatically invoked if
935 the IO::Uncompress::Inflate object is destroyed (either explicitly or by the
936 variable with the reference to the object going out of scope). The
937 exceptions are Perl versions 5.005 through 5.00504 and 5.8.0. In
938 these cases, the C<close> method will be called automatically, but
939 not until global destruction of all live objects when the program is
942 Therefore, if you want your scripts to be able to run on all versions
943 of Perl, you should call C<close> explicitly and not rely on automatic
946 Returns true on success, otherwise 0.
948 If the C<AutoClose> option has been enabled when the IO::Uncompress::Inflate
949 object was created, and the object is associated with a file, the
950 underlying file will also be closed.
959 my $status = $z->nextStream();
961 Skips to the next compressed data stream in the input file/buffer. If a new
962 compressed data stream is found, the eof marker will be cleared and C<$.>
965 Returns 1 if a new stream was found, 0 if none was found, and -1 if an
966 error was encountered.
972 my $data = $z->trailingData();
974 Returns the data, if any, that is present immediately after the compressed
975 data stream once uncompression is complete. It only makes sense to call
976 this method once the end of the compressed data stream has been
979 This option can be used when there is useful information immediately
980 following the compressed data stream, and you don't know the length of the
981 compressed data stream.
983 If the input is a buffer, C<trailingData> will return everything from the
984 end of the compressed data stream to the end of the buffer.
986 If the input is a filehandle, C<trailingData> will return the data that is
987 left in the filehandle input buffer once the end of the compressed data
988 stream has been reached. You can then use the filehandle to read the rest
991 Don't bother using C<trailingData> if the input is a filename.
995 If you know the length of the compressed data stream before you start
996 uncompressing, you can avoid having to use C<trailingData> by setting the
997 C<InputLength> option in the constructor.
1001 No symbolic constants are required by this IO::Uncompress::Inflate at present.
1007 Imports C<inflate> and C<$InflateError>.
1010 use IO::Uncompress::Inflate qw(inflate $InflateError) ;
1021 L<Compress::Zlib>, L<IO::Compress::Gzip>, L<IO::Uncompress::Gunzip>, L<IO::Compress::Deflate>, L<IO::Compress::RawDeflate>, L<IO::Uncompress::RawInflate>, L<IO::Compress::Bzip2>, L<IO::Uncompress::Bunzip2>, L<IO::Compress::Lzop>, L<IO::Uncompress::UnLzop>, L<IO::Compress::Lzf>, L<IO::Uncompress::UnLzf>, L<IO::Uncompress::AnyInflate>, L<IO::Uncompress::AnyUncompress>
1023 L<Compress::Zlib::FAQ|Compress::Zlib::FAQ>
1025 L<File::GlobMapper|File::GlobMapper>, L<Archive::Zip|Archive::Zip>,
1026 L<Archive::Tar|Archive::Tar>,
1027 L<IO::Zlib|IO::Zlib>
1030 For RFC 1950, 1951 and 1952 see
1031 F<http://www.faqs.org/rfcs/rfc1950.html>,
1032 F<http://www.faqs.org/rfcs/rfc1951.html> and
1033 F<http://www.faqs.org/rfcs/rfc1952.html>
1035 The I<zlib> compression library was written by Jean-loup Gailly
1036 F<gzip@prep.ai.mit.edu> and Mark Adler F<madler@alumni.caltech.edu>.
1038 The primary site for the I<zlib> compression library is
1039 F<http://www.zlib.org>.
1041 The primary site for gzip is F<http://www.gzip.org>.
1048 This module was written by Paul Marquess, F<pmqs@cpan.org>.
1052 =head1 MODIFICATION HISTORY
1054 See the Changes file.
1056 =head1 COPYRIGHT AND LICENSE
1058 Copyright (c) 2005-2006 Paul Marquess. All rights reserved.
1060 This program is free software; you can redistribute it and/or
1061 modify it under the same terms as Perl itself.