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);
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 This module provides a Perl interface that allows the reading of
259 files/buffers that conform to RFC 1950.
261 For writing RFC 1950 files/buffers, see the companion module IO::Compress::Deflate.
267 =head1 Functional Interface
269 A top-level function, C<inflate>, is provided to carry out
270 "one-shot" uncompression between buffers and/or files. For finer
271 control over the uncompression process, see the L</"OO Interface">
274 use IO::Uncompress::Inflate qw(inflate $InflateError) ;
276 inflate $input => $output [,OPTS]
277 or die "inflate failed: $InflateError\n";
281 The functional interface needs Perl5.005 or better.
284 =head2 inflate $input => $output [, OPTS]
287 C<inflate> expects at least two parameters, C<$input> and C<$output>.
289 =head3 The C<$input> parameter
291 The parameter, C<$input>, is used to define the source of
294 It can take one of the following forms:
300 If the C<$input> parameter is a simple scalar, it is assumed to be a
301 filename. This file will be opened for reading and the input data
302 will be read from it.
306 If the C<$input> parameter is a filehandle, the input data will be
308 The string '-' can be used as an alias for standard input.
310 =item A scalar reference
312 If C<$input> is a scalar reference, the input data will be read
315 =item An array reference
317 If C<$input> is an array reference, each element in the array must be a
320 The input data will be read from each file in turn.
322 The complete array will be walked to ensure that it only
323 contains valid filenames before any data is uncompressed.
327 =item An Input FileGlob string
329 If C<$input> is a string that is delimited by the characters "<" and ">"
330 C<inflate> will assume that it is an I<input fileglob string>. The
331 input is the list of files that match the fileglob.
333 If the fileglob does not match any files ...
335 See L<File::GlobMapper|File::GlobMapper> for more details.
340 If the C<$input> parameter is any other type, C<undef> will be returned.
344 =head3 The C<$output> parameter
346 The parameter C<$output> is used to control the destination of the
347 uncompressed data. This parameter can take one of these forms.
353 If the C<$output> parameter is a simple scalar, it is assumed to be a
354 filename. This file will be opened for writing and the uncompressed
355 data will be written to it.
359 If the C<$output> parameter is a filehandle, the uncompressed data
360 will be written to it.
361 The string '-' can be used as an alias for standard output.
364 =item A scalar reference
366 If C<$output> is a scalar reference, the uncompressed data will be
367 stored in C<$$output>.
371 =item An Array Reference
373 If C<$output> is an array reference, the uncompressed data will be
374 pushed onto the array.
376 =item An Output FileGlob
378 If C<$output> is a string that is delimited by the characters "<" and ">"
379 C<inflate> will assume that it is an I<output fileglob string>. The
380 output is the list of files that match the fileglob.
382 When C<$output> is an fileglob string, C<$input> must also be a fileglob
383 string. Anything else is an error.
387 If the C<$output> parameter is any other type, C<undef> will be returned.
394 When C<$input> maps to multiple compressed files/buffers and C<$output> is
395 a single file/buffer, after uncompression C<$output> will contain a
396 concatenation of all the uncompressed data from each of the input
403 =head2 Optional Parameters
405 Unless specified below, the optional parameters for C<inflate>,
406 C<OPTS>, are the same as those used with the OO interface defined in the
407 L</"Constructor Options"> section below.
411 =item C<< AutoClose => 0|1 >>
413 This option applies to any input or output data streams to
414 C<inflate> that are filehandles.
416 If C<AutoClose> is specified, and the value is true, it will result in all
417 input and/or output filehandles being closed once C<inflate> has
420 This parameter defaults to 0.
423 =item C<< BinModeOut => 0|1 >>
425 When writing to a file or filehandle, set C<binmode> before writing to the
434 =item C<< Append => 0|1 >>
438 =item C<< MultiStream => 0|1 >>
441 If the input file/buffer contains multiple compressed data streams, this
442 option will uncompress the whole lot as a single data stream.
450 =item C<< TrailingData => $scalar >>
452 Returns the data, if any, that is present immediately after the compressed
453 data stream once uncompression is complete.
455 This option can be used when there is useful information immediately
456 following the compressed data stream, and you don't know the length of the
457 compressed data stream.
459 If the input is a buffer, C<trailingData> will return everything from the
460 end of the compressed data stream to the end of the buffer.
462 If the input is a filehandle, C<trailingData> will return the data that is
463 left in the filehandle input buffer once the end of the compressed data
464 stream has been reached. You can then use the filehandle to read the rest
467 Don't bother using C<trailingData> if the input is a filename.
471 If you know the length of the compressed data stream before you start
472 uncompressing, you can avoid having to use C<trailingData> by setting the
473 C<InputLength> option.
484 To read the contents of the file C<file1.txt.1950> and write the
485 compressed data to the file C<file1.txt>.
489 use IO::Uncompress::Inflate qw(inflate $InflateError) ;
491 my $input = "file1.txt.1950";
492 my $output = "file1.txt";
493 inflate $input => $output
494 or die "inflate failed: $InflateError\n";
497 To read from an existing Perl filehandle, C<$input>, and write the
498 uncompressed data to a buffer, C<$buffer>.
502 use IO::Uncompress::Inflate qw(inflate $InflateError) ;
505 my $input = new IO::File "<file1.txt.1950"
506 or die "Cannot open 'file1.txt.1950': $!\n" ;
508 inflate $input => \$buffer
509 or die "inflate failed: $InflateError\n";
511 To uncompress all files in the directory "/my/home" that match "*.txt.1950" and store the compressed data in the same directory
515 use IO::Uncompress::Inflate qw(inflate $InflateError) ;
517 inflate '</my/home/*.txt.1950>' => '</my/home/#1.txt>'
518 or die "inflate failed: $InflateError\n";
520 and if you want to compress each file one at a time, this will do the trick
524 use IO::Uncompress::Inflate qw(inflate $InflateError) ;
526 for my $input ( glob "/my/home/*.txt.1950" )
529 $output =~ s/.1950// ;
530 inflate $input => $output
531 or die "Error compressing '$input': $InflateError\n";
538 The format of the constructor for IO::Uncompress::Inflate is shown below
541 my $z = new IO::Uncompress::Inflate $input [OPTS]
542 or die "IO::Uncompress::Inflate failed: $InflateError\n";
544 Returns an C<IO::Uncompress::Inflate> object on success and undef on failure.
545 The variable C<$InflateError> will contain an error message on failure.
547 If you are running Perl 5.005 or better the object, C<$z>, returned from
548 IO::Uncompress::Inflate can be used exactly like an L<IO::File|IO::File> filehandle.
549 This means that all normal input file operations can be carried out with
550 C<$z>. For example, to read a line from a compressed file/buffer you can
551 use either of these forms
553 $line = $z->getline();
556 The mandatory parameter C<$input> is used to determine the source of the
557 compressed data. This parameter can take one of three forms.
563 If the C<$input> parameter is a scalar, it is assumed to be a filename. This
564 file will be opened for reading and the compressed data will be read from it.
568 If the C<$input> parameter is a filehandle, the compressed data will be
570 The string '-' can be used as an alias for standard input.
573 =item A scalar reference
575 If C<$input> is a scalar reference, the compressed data will be read from
580 =head2 Constructor Options
583 The option names defined below are case insensitive and can be optionally
584 prefixed by a '-'. So all of the following are valid
591 OPTS is a combination of the following options:
595 =item C<< AutoClose => 0|1 >>
597 This option is only valid when the C<$input> parameter is a filehandle. If
598 specified, and the value is true, it will result in the file being closed once
599 either the C<close> method is called or the IO::Uncompress::Inflate object is
602 This parameter defaults to 0.
604 =item C<< MultiStream => 0|1 >>
608 Allows multiple concatenated compressed streams to be treated as a single
609 compressed stream. Decompression will stop once either the end of the
610 file/buffer is reached, an error is encountered (premature eof, corrupt
611 compressed data) or the end of a stream is not immediately followed by the
612 start of another stream.
614 This parameter defaults to 0.
617 =item C<< Prime => $string >>
619 This option will uncompress the contents of C<$string> before processing the
622 This option can be useful when the compressed data is embedded in another
623 file/data structure and it is not possible to work out where the compressed
624 data begins without having to read the first few bytes. If this is the
625 case, the uncompression can be I<primed> with these bytes using this
628 =item C<< Transparent => 0|1 >>
630 If this option is set and the input file or buffer is not compressed data,
631 the module will allow reading of it anyway.
633 This option defaults to 1.
635 =item C<< BlockSize => $num >>
637 When reading the compressed input data, IO::Uncompress::Inflate will read it in
638 blocks of C<$num> bytes.
640 This option defaults to 4096.
642 =item C<< InputLength => $size >>
644 When present this option will limit the number of compressed bytes read
645 from the input file/buffer to C<$size>. This option can be used in the
646 situation where there is useful data directly after the compressed data
647 stream and you know beforehand the exact length of the compressed data
650 This option is mostly used when reading from a filehandle, in which case
651 the file pointer will be left pointing to the first byte directly after the
652 compressed data stream.
656 This option defaults to off.
658 =item C<< Append => 0|1 >>
660 This option controls what the C<read> method does with uncompressed data.
662 If set to 1, all uncompressed data will be appended to the output parameter
663 of the C<read> method.
665 If set to 0, the contents of the output parameter of the C<read> method
666 will be overwritten by the uncompressed data.
670 =item C<< Strict => 0|1 >>
674 This option controls whether the extra checks defined below are used when
675 carrying out the decompression. When Strict is on, the extra tests are
676 carried out, when Strict is off they are not.
678 The default for this option is off.
688 The ADLER32 checksum field must be present.
692 The value of the ADLER32 field read must match the adler32 value of the
693 uncompressed data actually contained in the file.
720 $status = $z->read($buffer)
722 Reads a block of compressed data (the size the the compressed block is
723 determined by the C<Buffer> option in the constructor), uncompresses it and
724 writes any uncompressed data into C<$buffer>. If the C<Append> parameter is
725 set in the constructor, the uncompressed data will be appended to the
726 C<$buffer> parameter. Otherwise C<$buffer> will be overwritten.
728 Returns the number of uncompressed bytes written to C<$buffer>, zero if eof
729 or a negative number on error.
735 $status = $z->read($buffer, $length)
736 $status = $z->read($buffer, $length, $offset)
738 $status = read($z, $buffer, $length)
739 $status = read($z, $buffer, $length, $offset)
741 Attempt to read C<$length> bytes of uncompressed data into C<$buffer>.
743 The main difference between this form of the C<read> method and the
744 previous one, is that this one will attempt to return I<exactly> C<$length>
745 bytes. The only circumstances that this function will not is if end-of-file
746 or an IO error is encountered.
748 Returns the number of uncompressed bytes written to C<$buffer>, zero if eof
749 or a negative number on error.
756 $line = $z->getline()
761 This method fully supports the use of of the variable C<$/> (or
762 C<$INPUT_RECORD_SEPARATOR> or C<$RS> when C<English> is in use) to
763 determine what constitutes an end of line. Paragraph mode, record mode and
764 file slurp mode are all supported.
773 Read a single character.
779 $char = $z->ungetc($string)
787 $status = $z->inflateSync()
796 $hdr = $z->getHeaderInfo();
797 @hdrs = $z->getHeaderInfo();
799 This method returns either a hash reference (in scalar context) or a list
800 or hash references (in array context) that contains information about each
801 of the header fields in the compressed data stream(s).
813 Returns the uncompressed file offset.
824 Returns true if the end of the compressed input stream has been reached.
830 $z->seek($position, $whence);
831 seek($z, $position, $whence);
836 Provides a sub-set of the C<seek> functionality, with the restriction
837 that it is only legal to seek forward in the input file/buffer.
838 It is a fatal error to attempt to seek backward.
842 The C<$whence> parameter takes one the usual values, namely SEEK_SET,
843 SEEK_CUR or SEEK_END.
845 Returns 1 on success, 0 on failure.
854 This is a noop provided for completeness.
860 Returns true if the object currently refers to a opened file/buffer.
864 my $prev = $z->autoflush()
865 my $prev = $z->autoflush(EXPR)
867 If the C<$z> object is associated with a file or a filehandle, this method
868 returns the current autoflush setting for the underlying filehandle. If
869 C<EXPR> is present, and is non-zero, it will enable flushing after every
870 write/print operation.
872 If C<$z> is associated with a buffer, this method has no effect and always
875 B<Note> that the special variable C<$|> B<cannot> be used to set or
876 retrieve the autoflush setting.
878 =head2 input_line_number
880 $z->input_line_number()
881 $z->input_line_number(EXPR)
885 Returns the current uncompressed line number. If C<EXPR> is present it has
886 the effect of setting the line number. Note that setting the line number
887 does not change the current position within the file/buffer being read.
889 The contents of C<$/> are used to to determine what constitutes a line
899 If the C<$z> object is associated with a file or a filehandle, this method
900 will return the underlying file descriptor.
902 If the C<$z> object is is associated with a buffer, this method will
912 Closes the output file/buffer.
916 For most versions of Perl this method will be automatically invoked if
917 the IO::Uncompress::Inflate object is destroyed (either explicitly or by the
918 variable with the reference to the object going out of scope). The
919 exceptions are Perl versions 5.005 through 5.00504 and 5.8.0. In
920 these cases, the C<close> method will be called automatically, but
921 not until global destruction of all live objects when the program is
924 Therefore, if you want your scripts to be able to run on all versions
925 of Perl, you should call C<close> explicitly and not rely on automatic
928 Returns true on success, otherwise 0.
930 If the C<AutoClose> option has been enabled when the IO::Uncompress::Inflate
931 object was created, and the object is associated with a file, the
932 underlying file will also be closed.
941 my $status = $z->nextStream();
943 Skips to the next compressed data stream in the input file/buffer. If a new
944 compressed data stream is found, the eof marker will be cleared and C<$.>
947 Returns 1 if a new stream was found, 0 if none was found, and -1 if an
948 error was encountered.
954 my $data = $z->trailingData();
956 Returns the data, if any, that is present immediately after the compressed
957 data stream once uncompression is complete. It only makes sense to call
958 this method once the end of the compressed data stream has been
961 This option can be used when there is useful information immediately
962 following the compressed data stream, and you don't know the length of the
963 compressed data stream.
965 If the input is a buffer, C<trailingData> will return everything from the
966 end of the compressed data stream to the end of the buffer.
968 If the input is a filehandle, C<trailingData> will return the data that is
969 left in the filehandle input buffer once the end of the compressed data
970 stream has been reached. You can then use the filehandle to read the rest
973 Don't bother using C<trailingData> if the input is a filename.
977 If you know the length of the compressed data stream before you start
978 uncompressing, you can avoid having to use C<trailingData> by setting the
979 C<InputLength> option in the constructor.
983 No symbolic constants are required by this IO::Uncompress::Inflate at present.
989 Imports C<inflate> and C<$InflateError>.
992 use IO::Uncompress::Inflate qw(inflate $InflateError) ;
1003 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>
1005 L<Compress::Zlib::FAQ|Compress::Zlib::FAQ>
1007 L<File::GlobMapper|File::GlobMapper>, L<Archive::Zip|Archive::Zip>,
1008 L<Archive::Tar|Archive::Tar>,
1009 L<IO::Zlib|IO::Zlib>
1012 For RFC 1950, 1951 and 1952 see
1013 F<http://www.faqs.org/rfcs/rfc1950.html>,
1014 F<http://www.faqs.org/rfcs/rfc1951.html> and
1015 F<http://www.faqs.org/rfcs/rfc1952.html>
1017 The I<zlib> compression library was written by Jean-loup Gailly
1018 F<gzip@prep.ai.mit.edu> and Mark Adler F<madler@alumni.caltech.edu>.
1020 The primary site for the I<zlib> compression library is
1021 F<http://www.zlib.org>.
1023 The primary site for gzip is F<http://www.gzip.org>.
1030 This module was written by Paul Marquess, F<pmqs@cpan.org>.
1034 =head1 MODIFICATION HISTORY
1036 See the Changes file.
1038 =head1 COPYRIGHT AND LICENSE
1040 Copyright (c) 2005-2006 Paul Marquess. All rights reserved.
1042 This program is free software; you can redistribute it and/or
1043 modify it under the same terms as Perl itself.