From: Rafael Garcia-Suarez Date: Sun, 1 Mar 2009 22:37:53 +0000 (+0100) Subject: Add Compress-Raw-Bzip2, by Paul Marquess X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=commitdiff_plain;h=bdb7fd9ff27337264f70f75183d41d2995ccc57c;p=p5sagit%2Fp5-mst-13.2.git Add Compress-Raw-Bzip2, by Paul Marquess --- diff --git a/MANIFEST b/MANIFEST index 4aa99c4..e3a1df1 100644 --- a/MANIFEST +++ b/MANIFEST @@ -133,6 +133,36 @@ ext/B/t/showlex.t See if B::ShowLex works ext/B/t/terse.t See if B::Terse works ext/B/t/xref.t See if B::Xref works ext/B/typemap Compiler backend interface types +ext/Compress-Raw-Bzip2/README +ext/Compress-Raw-Bzip2/Makefile.PL +ext/Compress-Raw-Bzip2/Bzip2.xs +ext/Compress-Raw-Bzip2/typemap +ext/Compress-Raw-Bzip2/fallback/constants.h +ext/Compress-Raw-Bzip2/fallback/constants.xs +ext/Compress-Raw-Bzip2/MANIFEST +ext/Compress-Raw-Bzip2/private/MakeUtil.pm +ext/Compress-Raw-Bzip2/t/000prereq.t +ext/Compress-Raw-Bzip2/t/01bzip2.t +ext/Compress-Raw-Bzip2/t/99pod.t +ext/Compress-Raw-Bzip2/lib/Compress/Raw/Bzip2.pm +ext/Compress-Raw-Bzip2/bzip2-src/blocksort.c +ext/Compress-Raw-Bzip2/bzip2-src/huffman.c +ext/Compress-Raw-Bzip2/bzip2-src/crctable.c +ext/Compress-Raw-Bzip2/bzip2-src/randtable.c +ext/Compress-Raw-Bzip2/bzip2-src/compress.c +ext/Compress-Raw-Bzip2/bzip2-src/decompress.c +ext/Compress-Raw-Bzip2/bzip2-src/bzlib.c +ext/Compress-Raw-Bzip2/bzip2-src/bzip2.c +ext/Compress-Raw-Bzip2/bzip2-src/bzip2recover.c +ext/Compress-Raw-Bzip2/bzip2-src/bzlib.h +ext/Compress-Raw-Bzip2/bzip2-src/bzlib_private.h +ext/Compress-Raw-Bzip2/bzip2-src/dlltest.c +ext/Compress-Raw-Bzip2/bzip2-src/unzcrash.c +ext/Compress-Raw-Bzip2/bzip2-src/spewG.c +ext/Compress-Raw-Bzip2/bzip2-src/mk251.c +ext/Compress-Raw-Bzip2/bzip2-src/LICENSE +ext/Compress-Raw-Bzip2/Changes +ext/Compress-Raw-Bzip2/pod/FAQ.pod ext/Compress-Raw-Zlib/Changes Compress::Raw::Zlib ext/Compress-Raw-Zlib/config.in Compress::Raw::Zlib ext/Compress-Raw-Zlib/examples/filtdef Compress::Raw::Zlib diff --git a/Porting/Maintainers.pl b/Porting/Maintainers.pl index 736ab46..9bd4152 100644 --- a/Porting/Maintainers.pl +++ b/Porting/Maintainers.pl @@ -187,6 +187,14 @@ package Maintainers; 'UPSTREAM' => undef, }, + 'Compress::Raw::Bzip2' => + { + 'MAINTAINER' => 'pmqs', + 'FILES' => q[ext/Compress-Raw-Bzip2], + 'CPAN' => 1, + 'UPSTREAM' => undef, + }, + 'Compress::Raw::Zlib' => { 'MAINTAINER' => 'pmqs', diff --git a/ext/Compress-Raw-Bzip2/Bzip2.xs b/ext/Compress-Raw-Bzip2/Bzip2.xs new file mode 100644 index 0000000..1e6eeb9 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/Bzip2.xs @@ -0,0 +1,868 @@ +/* Filename: Bzip2.xs + * Author : Paul Marquess, + * Created : 5th October 2005 + * Version : 2.000 + * + * Copyright (c) 2005-2007 Paul Marquess. All rights reserved. + * This program is free software; you can redistribute it and/or + * modify it under the same terms as Perl itself. + * + */ + + +#include "EXTERN.h" +#include "perl.h" +#include "XSUB.h" + +#include "bzlib.h" + +#ifdef USE_PPPORT_H +# define NEED_sv_2pv_nolen +# include "ppport.h" +#endif + +#if PERL_REVISION == 5 && (PERL_VERSION < 8 || (PERL_VERSION == 8 && PERL_SUBVERSION < 4 )) + +# ifdef SvPVbyte_force +# undef SvPVbyte_force +# endif + +# define SvPVbyte_force(sv,lp) SvPV_force(sv,lp) + +#endif + +#ifndef SvPVbyte_nolen +# define SvPVbyte_nolen SvPV_nolen +#endif + + +#if PERL_REVISION == 5 && (PERL_VERSION >= 8 || (PERL_VERSION == 8 && PERL_SUBVERSION < 4 )) +# define UTF8_AVAILABLE +#endif + +typedef int DualType ; +typedef int int_undef ; + +typedef unsigned long uLong; +typedef unsigned int uInt; + +typedef struct di_stream { + int flags ; +#define FLAG_APPEND_OUTPUT 1 +#define FLAG_CONSUME_INPUT 8 + bz_stream stream; + uInt bufsize; + int last_error ; + uLong bytesInflated ; + uLong compressedBytes ; + uLong uncompressedBytes ; + +} di_stream; + +typedef di_stream * deflateStream ; +typedef di_stream * Compress__Raw__Bzip2 ; + +typedef di_stream * inflateStream ; +typedef di_stream * Compress__Raw__Bunzip2 ; + +#define COMPRESS_CLASS "Compress::Raw::Bzip2" +#define UNCOMPRESS_CLASS "Compress::Raw::Bunzip2" + +#define ZMALLOC(to, typ) ((to = (typ *)safemalloc(sizeof(typ))), \ + Zero(to,1,typ)) + + +/* static const char * const my_z_errmsg[] = { */ +static const char my_z_errmsg[][32] = { + "End of Stream", /* BZ_STREAM_END 4 */ + "Finish OK", /* BZ_FINISH_OK 3 */ + "Flush OK", /* BZ_FLUSH_OK 2 */ + "Run OK", /* BZ_RUN_OK 1 */ + "", /* BZ_OK 0 */ + "Sequence Error", /* BZ_SEQUENCE_ERROR (-1) */ + "Param Error", /* BZ_PARAM_ERROR (-2) */ + "Memory Error", /* BZ_MEM_ERROR (-3) */ + "Data Error", /* BZ_DATA_ERROR (-4) */ + "Magic Error", /* BZ_DATA_ERROR_MAGIC (-5) */ + "IO Error", /* BZ_IO_ERROR (-6) */ + "Unexpected EOF", /* BZ_UNEXPECTED_EOF (-7) */ + "Output Buffer Full", /* BZ_OUTBUFF_FULL (-8) */ + "Config Error", /* BZ_CONFIG_ERROR (-9) */ + ""}; + +#define setDUALstatus(var, err) \ + sv_setnv(var, (double)err) ; \ + sv_setpv(var, ((err) ? GetErrorString(err) : "")) ; \ + SvNOK_on(var); + + +#if defined(__SYMBIAN32__) +# define NO_WRITEABLE_DATA +#endif + +#define TRACE_DEFAULT 0 + +#ifdef NO_WRITEABLE_DATA +# define trace TRACE_DEFAULT +#else + static int trace = TRACE_DEFAULT ; +#endif + +/* Dodge PerlIO hiding of these functions. */ +#undef printf + +#if 1 +#define getInnerObject(x) (*av_fetch((AV*)SvRV(x), 0, FALSE)) +#else +#define getInnerObject(x) ((SV*)SvRV(sv)) +#endif + +#ifdef BZ_NO_STDIO +void bz_internal_error(int errorcode) +{ + croak("bz_internal_error %d\n", errorcode); +} +#endif + +static char * +#ifdef CAN_PROTOTYPE +GetErrorString(int error_no) +#else +GetErrorString(error_no) +int error_no ; +#endif +{ + dTHX; + char * errstr ; + +#if 0 + if (error_no == BZ_ERRNO) { + errstr = Strerror(errno) ; + } + else +#endif + errstr = (char*) my_z_errmsg[4 - error_no]; + + return errstr ; +} + +static void +#ifdef CAN_PROTOTYPE +DispHex(void * ptr, int length) +#else +DispHex(ptr, length) + void * ptr; + int length; +#endif +{ + char * p = (char*)ptr; + int i; + for (i = 0; i < length; ++i) { + printf(" %02x", 0xFF & *(p+i)); + } +} + + +static void +#ifdef CAN_PROTOTYPE +DispStream(di_stream * s, char * message) +#else +DispStream(s, message) + di_stream * s; + char * message; +#endif +{ + +#if 0 + if (! trace) + return ; +#endif + +#define EnDis(f) (s->flags & f ? "Enabled" : "Disabled") + + printf("DispStream 0x%p", s) ; + if (message) + printf("- %s \n", message) ; + printf("\n") ; + + if (!s) { + printf(" stream pointer is NULL\n"); + } + else { + printf(" stream 0x%p\n", &(s->stream)); + printf(" opaque 0x%p\n", s->stream.opaque); + printf(" next_in 0x%p", s->stream.next_in); + if (s->stream.next_in){ + printf(" =>"); + DispHex(s->stream.next_in, 4); + } + printf("\n"); + + printf(" next_out 0x%p", s->stream.next_out); + if (s->stream.next_out){ + printf(" =>"); + DispHex(s->stream.next_out, 4); + } + printf("\n"); + + printf(" avail_in %lu\n", (unsigned long)s->stream.avail_in); + printf(" avail_out %lu\n", (unsigned long)s->stream.avail_out); + printf(" bufsize %lu\n", (unsigned long)s->bufsize); + printf(" flags 0x%x\n", s->flags); + printf(" APPEND %s\n", EnDis(FLAG_APPEND_OUTPUT)); + printf(" CONSUME %s\n", EnDis(FLAG_CONSUME_INPUT)); + + printf("\n"); + + } +} + +static di_stream * +#ifdef CAN_PROTOTYPE +InitStream(void) +#else +InitStream() +#endif +{ + di_stream *s ; + + ZMALLOC(s, di_stream) ; + + return s ; + +} + +static void +#ifdef CAN_PROTOTYPE +PostInitStream(di_stream * s, int flags) +#else +PostInitStream(s, flags) + di_stream *s ; + int flags ; +#endif +{ + s->bufsize = 1024 * 16 ; + s->last_error = 0 ; + s->flags = flags ; +} + + +static SV* +#ifdef CAN_PROTOTYPE +deRef(SV * sv, const char * string) +#else +deRef(sv, string) +SV * sv ; +char * string; +#endif +{ + dTHX; + SvGETMAGIC(sv); + + if (SvROK(sv)) { + sv = SvRV(sv) ; + SvGETMAGIC(sv); + switch(SvTYPE(sv)) { + case SVt_PVAV: + case SVt_PVHV: + case SVt_PVCV: + croak("%s: buffer parameter is not a SCALAR reference", string); + } + if (SvROK(sv)) + croak("%s: buffer parameter is a reference to a reference", string) ; + } + + if (!SvOK(sv)) { + sv = newSVpv("", 0); + } + + return sv ; +} + +static SV* +#ifdef CAN_PROTOTYPE +deRef_l(SV * sv, const char * string) +#else +deRef_l(sv, string) +SV * sv ; +char * string ; +#endif +{ + dTHX; + bool wipe = 0 ; + + SvGETMAGIC(sv); + wipe = ! SvOK(sv) ; + + if (SvROK(sv)) { + sv = SvRV(sv) ; + SvGETMAGIC(sv); + wipe = ! SvOK(sv) ; + + switch(SvTYPE(sv)) { + case SVt_PVAV: + case SVt_PVHV: + case SVt_PVCV: + croak("%s: buffer parameter is not a SCALAR reference", string); + } + if (SvROK(sv)) + croak("%s: buffer parameter is a reference to a reference", string) ; + } + + if (SvREADONLY(sv) && PL_curcop != &PL_compiling) + croak("%s: buffer parameter is read-only", string); + + SvUPGRADE(sv, SVt_PV); + + if (wipe) + SvCUR_set(sv, 0); + + SvOOK_off(sv); + SvPOK_only(sv); + + return sv ; +} + + +#include "constants.h" + +MODULE = Compress::Raw::Bzip2 PACKAGE = Compress::Raw::Bzip2 PREFIX = Zip_ + +REQUIRE: 1.924 +PROTOTYPES: DISABLE + +INCLUDE: constants.xs + +BOOT: + /* Check this version of bzip2 is == 1 */ + if (BZ2_bzlibVersion()[0] != '1') + croak(COMPRESS_CLASS " needs bzip2 version 1.x, you have %s\n", BZ2_bzlibVersion()) ; + + +MODULE = Compress::Raw::Bzip2 PACKAGE = Compress::Raw::Bzip2 + +#define bzlibversion() BZ2_bzlibVersion() +const char * +bzlibversion() + +void +new(class, appendOut=1, blockSize100k=1, workfactor=0, verbosity=0) + const char * class + int appendOut + int blockSize100k + int workfactor + int verbosity + PPCODE: + { + int err ; + deflateStream s ; +#if 0 + /* if (trace) */ + warn("in Compress::Raw::Bzip2::_new(items=%d,appendOut=%d, blockSize100k=%d, workfactor=%d, verbosity=%d\n", + items, appendOut, blockSize100k, workfactor, verbosity); +#endif + if ((s = InitStream() )) { + + err = BZ2_bzCompressInit ( &(s->stream), + blockSize100k, + verbosity, + workfactor ); + + if (err != BZ_OK) { + Safefree(s) ; + s = NULL ; + } + else { + int flags = 0 ; + if (appendOut) + flags |= FLAG_APPEND_OUTPUT; + PostInitStream(s, appendOut ? FLAG_APPEND_OUTPUT :0) ; + } + } + else + err = BZ_MEM_ERROR ; + + { + SV* obj = sv_setref_pv(sv_newmortal(), class, (void*)s); + XPUSHs(obj); + } + if(0) + { + SV* obj = sv_2mortal(newSViv(PTR2IV(s))) ; + XPUSHs(obj); + } + if (GIMME == G_ARRAY) { + SV * sv = sv_2mortal(newSViv(err)) ; + setDUALstatus(sv, err); + XPUSHs(sv) ; + } + } + +MODULE = Compress::Raw::Bunzip2 PACKAGE = Compress::Raw::Bunzip2 + +void +new(class, appendOut=1 , consume=1, small=0, verbosity=0) + const char* class + int appendOut + int consume + int small + int verbosity + PPCODE: + { + int err = BZ_OK ; + inflateStream s ; +#if 0 + if (trace) + warn("in _inflateInit(windowBits=%d, bufsize=%lu, dictionary=%lu\n", + windowBits, bufsize, (unsigned long)SvCUR(dictionary)) ; +#endif + if ((s = InitStream() )) { + + err = BZ2_bzDecompressInit (&(s->stream), verbosity, small); + if (err != BZ_OK) { + Safefree(s) ; + s = NULL ; + } + if (s) { + int flags = 0; + if (appendOut) + flags |= FLAG_APPEND_OUTPUT; + if (consume) + flags |= FLAG_CONSUME_INPUT; + PostInitStream(s, flags) ; + } + } + else + err = BZ_MEM_ERROR ; + + { + SV* obj = sv_setref_pv(sv_newmortal(), class, (void*)s); + XPUSHs(obj); + } + if (0) + { + SV* obj = sv_2mortal(newSViv(PTR2IV(s))) ; + XPUSHs(obj); + } + if (GIMME == G_ARRAY) { + SV * sv = sv_2mortal(newSViv(err)) ; + setDUALstatus(sv, err); + XPUSHs(sv) ; + } + } + + + +MODULE = Compress::Raw::Bzip2 PACKAGE = Compress::Raw::Bzip2 + +void +DispStream(s, message=NULL) + Compress::Raw::Bzip2 s + char * message + +DualType +bzdeflate (s, buf, output) + Compress::Raw::Bzip2 s + SV * buf + SV * output + uInt cur_length = NO_INIT + uInt increment = NO_INIT + int RETVAL = 0; + uInt bufinc = NO_INIT + CODE: + bufinc = s->bufsize; + + /* If the input buffer is a reference, dereference it */ + buf = deRef(buf, "deflate") ; + + /* initialise the input buffer */ +#ifdef UTF8_AVAILABLE + if (DO_UTF8(buf) && !sv_utf8_downgrade(buf, 1)) + croak("Wide character in " COMPRESS_CLASS "::bzdeflate input parameter"); +#endif + s->stream.next_in = (char*)SvPVbyte_nolen(buf) ; + s->stream.avail_in = SvCUR(buf) ; + + /* and retrieve the output buffer */ + output = deRef_l(output, "deflate") ; +#ifdef UTF8_AVAILABLE + if (DO_UTF8(output) && !sv_utf8_downgrade(output, 1)) + croak("Wide character in " COMPRESS_CLASS "::bzdeflate output parameter"); +#endif + + if((s->flags & FLAG_APPEND_OUTPUT) != FLAG_APPEND_OUTPUT) { + SvCUR_set(output, 0); + /* sv_setpvn(output, "", 0); */ + } + cur_length = SvCUR(output) ; + s->stream.next_out = (char*) SvPVbyte_nolen(output) + cur_length; + increment = SvLEN(output) - cur_length; + s->stream.avail_out = increment; + while (s->stream.avail_in != 0) { + + if (s->stream.avail_out == 0) { + /* out of space in the output buffer so make it bigger */ + Sv_Grow(output, SvLEN(output) + bufinc) ; + cur_length += increment ; + s->stream.next_out = (char*) SvPVbyte_nolen(output) + cur_length ; + increment = bufinc ; + s->stream.avail_out = increment; + bufinc *= 2 ; + } + + RETVAL = BZ2_bzCompress(&(s->stream), BZ_RUN); + if (RETVAL != BZ_RUN_OK) + break; + } + + s->compressedBytes += cur_length + increment - s->stream.avail_out ; + s->uncompressedBytes += SvCUR(buf) - s->stream.avail_in ; + + s->last_error = RETVAL ; + if (RETVAL == BZ_RUN_OK) { + SvPOK_only(output); + SvCUR_set(output, cur_length + increment - s->stream.avail_out) ; + SvSETMAGIC(output); + } + OUTPUT: + RETVAL + + +void +DESTROY(s) + Compress::Raw::Bzip2 s + CODE: + BZ2_bzCompressEnd(&s->stream) ; + Safefree(s) ; + + +DualType +bzclose(s, output) + Compress::Raw::Bzip2 s + SV * output + uInt cur_length = NO_INIT + uInt increment = NO_INIT + uInt bufinc = NO_INIT + CODE: + bufinc = s->bufsize; + + s->stream.avail_in = 0; /* should be zero already anyway */ + + /* retrieve the output buffer */ + output = deRef_l(output, "close") ; +#ifdef UTF8_AVAILABLE + if (DO_UTF8(output) && !sv_utf8_downgrade(output, 1)) + croak("Wide character in " COMPRESS_CLASS "::bzclose input parameter"); +#endif + if(! s->flags & FLAG_APPEND_OUTPUT) { + SvCUR_set(output, 0); + /* sv_setpvn(output, "", 0); */ + } + cur_length = SvCUR(output) ; + s->stream.next_out = (char*) SvPVbyte_nolen(output) + cur_length; + increment = SvLEN(output) - cur_length; + s->stream.avail_out = increment; + + for (;;) { + if (s->stream.avail_out == 0) { + /* consumed all the available output, so extend it */ + Sv_Grow(output, SvLEN(output) + bufinc) ; + cur_length += increment ; + s->stream.next_out = (char*) SvPVbyte_nolen(output) + cur_length ; + increment = bufinc ; + s->stream.avail_out = increment; + bufinc *= 2 ; + } + RETVAL = BZ2_bzCompress(&(s->stream), BZ_FINISH); + + /* deflate has finished flushing only when it hasn't used up + * all the available space in the output buffer: + */ + /* if (s->stream.avail_out != 0 || RETVAL < 0 ) */ + if (RETVAL == BZ_STREAM_END || RETVAL < 0 ) + break; + } + + /* RETVAL = (RETVAL == BZ_STREAM_END ? BZ_OK : RETVAL) ; */ + s->last_error = RETVAL ; + + s->compressedBytes += cur_length + increment - s->stream.avail_out ; + + if (RETVAL == BZ_STREAM_END) { + SvPOK_only(output); + SvCUR_set(output, cur_length + increment - s->stream.avail_out) ; + SvSETMAGIC(output); + } + OUTPUT: + RETVAL + + +DualType +bzflush(s, output) + Compress::Raw::Bzip2 s + SV * output + uInt cur_length = NO_INIT + uInt increment = NO_INIT + uInt bufinc = NO_INIT + CODE: + bufinc = s->bufsize; + + s->stream.avail_in = 0; /* should be zero already anyway */ + + /* retrieve the output buffer */ + output = deRef_l(output, "close") ; +#ifdef UTF8_AVAILABLE + if (DO_UTF8(output) && !sv_utf8_downgrade(output, 1)) + croak("Wide character in " COMPRESS_CLASS "::bzflush input parameter"); +#endif + if(! s->flags & FLAG_APPEND_OUTPUT) { + SvCUR_set(output, 0); + /* sv_setpvn(output, "", 0); */ + } + cur_length = SvCUR(output) ; + s->stream.next_out = (char*) SvPVbyte_nolen(output) + cur_length; + increment = SvLEN(output) - cur_length; + s->stream.avail_out = increment; + + for (;;) { + if (s->stream.avail_out == 0) { + /* consumed all the available output, so extend it */ + Sv_Grow(output, SvLEN(output) + bufinc) ; + cur_length += increment ; + s->stream.next_out = (char*) SvPVbyte_nolen(output) + cur_length ; + increment = bufinc ; + s->stream.avail_out = increment; + bufinc *= 2 ; + } + RETVAL = BZ2_bzCompress(&(s->stream), BZ_FLUSH); + + if (RETVAL == BZ_RUN_OK || RETVAL < 0) + break; + + /* deflate has finished flushing only when it hasn't used up + * all the available space in the output buffer: + */ + /* RETVAL == if (s->stream.avail_out != 0 || RETVAL < 0 ) + break; */ + } + + /* RETVAL = (RETVAL == BZ_STREAM_END ? BZ_OK : RETVAL) ; */ + s->last_error = RETVAL ; + + s->compressedBytes += cur_length + increment - s->stream.avail_out ; + + if (RETVAL == BZ_RUN_OK) { + SvPOK_only(output); + SvCUR_set(output, cur_length + increment - s->stream.avail_out) ; + SvSETMAGIC(output); + } + OUTPUT: + RETVAL + +uLong +total_in_lo32(s) + Compress::Raw::Bzip2 s + CODE: + RETVAL = s->stream.total_in_lo32 ; + OUTPUT: + RETVAL + +uLong +total_out_lo32(s) + Compress::Raw::Bzip2 s + CODE: + RETVAL = s->stream.total_out_lo32 ; + OUTPUT: + RETVAL + +uLong +compressedBytes(s) + Compress::Raw::Bzip2 s + CODE: + RETVAL = s->compressedBytes; + OUTPUT: + RETVAL + +uLong +uncompressedBytes(s) + Compress::Raw::Bzip2 s + CODE: + RETVAL = s->uncompressedBytes; + OUTPUT: + RETVAL + + +MODULE = Compress::Raw::Bunzip2 PACKAGE = Compress::Raw::Bunzip2 + +void +DispStream(s, message=NULL) + Compress::Raw::Bunzip2 s + char * message + +DualType +bzinflate (s, buf, output) + Compress::Raw::Bunzip2 s + SV * buf + SV * output + uInt cur_length = 0; + uInt prefix_length = 0; + uInt increment = 0; + STRLEN stmp = NO_INIT + uInt bufinc = NO_INIT + PREINIT: +#ifdef UTF8_AVAILABLE + bool out_utf8 = FALSE; +#endif + CODE: + bufinc = s->bufsize; + /* If the buffer is a reference, dereference it */ + buf = deRef(buf, "inflate") ; + + if (s->flags & FLAG_CONSUME_INPUT && SvREADONLY(buf)) + croak(UNCOMPRESS_CLASS "::bzinflate input parameter cannot be read-only when ConsumeInput is specified"); +#ifdef UTF8_AVAILABLE + if (DO_UTF8(buf) && !sv_utf8_downgrade(buf, 1)) + croak("Wide character in " UNCOMPRESS_CLASS "::bzinflate input parameter"); +#endif + + /* initialise the input buffer */ + s->stream.next_in = (char*)SvPVbyte_force(buf, stmp) ; + s->stream.avail_in = SvCUR(buf); + + /* and retrieve the output buffer */ + output = deRef_l(output, "inflate") ; +#ifdef UTF8_AVAILABLE + if (DO_UTF8(output)) + out_utf8 = TRUE ; + if (DO_UTF8(output) && !sv_utf8_downgrade(output, 1)) + croak("Wide character in " UNCOMPRESS_CLASS "::bzinflate output parameter"); +#endif + if((s->flags & FLAG_APPEND_OUTPUT) != FLAG_APPEND_OUTPUT) { + SvCUR_set(output, 0); + } + if (SvLEN(output)) { + prefix_length = cur_length = SvCUR(output) ; + s->stream.next_out = (char*) SvPVbyte_nolen(output) + cur_length; + increment = SvLEN(output) - cur_length - 1; + s->stream.avail_out = increment; + } + else { + s->stream.avail_out = 0; + } + s->bytesInflated = 0; + + while (1) { + + if (s->stream.avail_out == 0) { + /* out of space in the output buffer so make it bigger */ + Sv_Grow(output, SvLEN(output) + bufinc) ; + cur_length += increment ; + s->stream.next_out = (char*) SvPVbyte_nolen(output) + cur_length ; + increment = bufinc ; + s->stream.avail_out = increment; + bufinc *= 2 ; + } + + RETVAL = BZ2_bzDecompress (&(s->stream)); + + if (RETVAL != BZ_OK) + break ; + + if (s->stream.avail_out == 0) + continue ; + + if (s->stream.avail_in == 0) { + RETVAL = BZ_OK ; + break ; + } + + } + + s->last_error = RETVAL ; + if (RETVAL == BZ_OK || RETVAL == BZ_STREAM_END) { + unsigned in ; + + s->bytesInflated = cur_length + increment - s->stream.avail_out - prefix_length; + s->uncompressedBytes += s->bytesInflated ; + s->compressedBytes += SvCUR(buf) - s->stream.avail_in ; + + SvPOK_only(output); + SvCUR_set(output, prefix_length + s->bytesInflated) ; + *SvEND(output) = '\0'; +#ifdef UTF8_AVAILABLE + if (out_utf8) + sv_utf8_upgrade(output); +#endif + SvSETMAGIC(output); + + /* fix the input buffer */ + if (s->flags & FLAG_CONSUME_INPUT) { + in = s->stream.avail_in ; + SvCUR_set(buf, in) ; + if (in) + Move(s->stream.next_in, SvPVbyte_nolen(buf), in, char) ; + *SvEND(buf) = '\0'; + SvSETMAGIC(buf); + } + } + OUTPUT: + RETVAL + +uLong +inflateCount(s) + Compress::Raw::Bunzip2 s + CODE: + RETVAL = s->bytesInflated; + OUTPUT: + RETVAL + + +void +DESTROY(s) + Compress::Raw::Bunzip2 s + CODE: + BZ2_bzDecompressEnd(&s->stream) ; + Safefree(s) ; + + +uLong +status(s) + Compress::Raw::Bunzip2 s + CODE: + RETVAL = s->last_error ; + OUTPUT: + RETVAL + +uLong +total_in_lo32(s) + Compress::Raw::Bunzip2 s + CODE: + RETVAL = s->stream.total_in_lo32 ; + OUTPUT: + RETVAL + +uLong +total_out_lo32(s) + Compress::Raw::Bunzip2 s + CODE: + RETVAL = s->stream.total_out_lo32 ; + OUTPUT: + RETVAL + +uLong +compressedBytes(s) + Compress::Raw::Bunzip2 s + CODE: + RETVAL = s->compressedBytes; + OUTPUT: + RETVAL + +uLong +uncompressedBytes(s) + Compress::Raw::Bunzip2 s + CODE: + RETVAL = s->uncompressedBytes; + OUTPUT: + RETVAL + +MODULE = Compress::Raw::Bzip2 PACKAGE = Compress::Raw::Bzip2 PREFIX = Zip_ diff --git a/ext/Compress-Raw-Bzip2/Changes b/ext/Compress-Raw-Bzip2/Changes new file mode 100644 index 0000000..f25e57a --- /dev/null +++ b/ext/Compress-Raw-Bzip2/Changes @@ -0,0 +1,82 @@ +CHANGES +------- + + 2.016 26 February 2009 + + * Minor changes to allow building in perl core. + + 2.015 3 September 2008 + + * Documented bzlibversion + + 2.015 3 September 2008 + + * Makefile.PL + Backout changes made in 2.014 + + 2.014 2 September 2008 + + * Makefile.PL + Updated to check for indirect dependencies. + + 2.012 15 July 2008 + + * No Changes + + 2.011 17 May 2008 + + * No Changes + + 2.010 5 May 2008 + + * No Changes + + 2.009 20 April 2008 + + * Updated to use bzip2 1.0.5 + + 2.008 2 November 2007 + + * Minor documentation changes in README + + 2.006 1 September 2007 + + * Added the bzip2 LICENSE file into bzip2-src. + [rt.cpan.org #28980] + + 2.005 18 June 2007 + + * Added patch from Fedora to allow the module to be built with an + existing bzip2 library. + [rt.cpan.org #25489] + + 2.004 3 March 2007 + + * Updated to use bzip2 1.0.4 + + * Remove redundant code from Bzip2.xs + + 2.003 2 January 2007 + + * Added explicit version checking + + 2.002 29 December 2006 + + * Documentation updates. + + 2.001 1 November 2006 + + * Remove beta status. + + 2.000_14 26 October 2006 + + * Fixed memory leak on realloc. + + 2.000_12 12 May 2006 + + * Documentation updates. + + 2.000_10 13 March 2006 + + * Created Compress::Raw::Bzip2 + diff --git a/ext/Compress-Raw-Bzip2/MANIFEST b/ext/Compress-Raw-Bzip2/MANIFEST new file mode 100644 index 0000000..fb4d781 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/MANIFEST @@ -0,0 +1,31 @@ +ext/Compress-Raw-Bzip2/README +ext/Compress-Raw-Bzip2/Makefile.PL +ext/Compress-Raw-Bzip2/Bzip2.xs +ext/Compress-Raw-Bzip2/typemap +ext/Compress-Raw-Bzip2/fallback/constants.h +ext/Compress-Raw-Bzip2/fallback/constants.xs +ext/Compress-Raw-Bzip2/MANIFEST +ext/Compress-Raw-Bzip2/private/MakeUtil.pm +ext/Compress-Raw-Bzip2/t/000prereq.t +ext/Compress-Raw-Bzip2/t/01bzip2.t +ext/Compress-Raw-Bzip2/t/99pod.t +ext/Compress-Raw-Bzip2/t/compress/CompTestUtils.pm +ext/Compress-Raw-Bzip2/lib/Compress/Raw/Bzip2.pm +ext/Compress-Raw-Bzip2/bzip2-src/blocksort.c +ext/Compress-Raw-Bzip2/bzip2-src/huffman.c +ext/Compress-Raw-Bzip2/bzip2-src/crctable.c +ext/Compress-Raw-Bzip2/bzip2-src/randtable.c +ext/Compress-Raw-Bzip2/bzip2-src/compress.c +ext/Compress-Raw-Bzip2/bzip2-src/decompress.c +ext/Compress-Raw-Bzip2/bzip2-src/bzlib.c +ext/Compress-Raw-Bzip2/bzip2-src/bzip2.c +ext/Compress-Raw-Bzip2/bzip2-src/bzip2recover.c +ext/Compress-Raw-Bzip2/bzip2-src/bzlib.h +ext/Compress-Raw-Bzip2/bzip2-src/bzlib_private.h +ext/Compress-Raw-Bzip2/bzip2-src/dlltest.c +ext/Compress-Raw-Bzip2/bzip2-src/unzcrash.c +ext/Compress-Raw-Bzip2/bzip2-src/spewG.c +ext/Compress-Raw-Bzip2/bzip2-src/mk251.c +ext/Compress-Raw-Bzip2/bzip2-src/LICENSE +ext/Compress-Raw-Bzip2/Changes +ext/Compress-Raw-Bzip2/pod/FAQ.pod diff --git a/ext/Compress-Raw-Bzip2/Makefile.PL b/ext/Compress-Raw-Bzip2/Makefile.PL new file mode 100644 index 0000000..192c28f --- /dev/null +++ b/ext/Compress-Raw-Bzip2/Makefile.PL @@ -0,0 +1,162 @@ +#! perl -w + +use strict ; +require 5.004 ; + +use private::MakeUtil; +use ExtUtils::MakeMaker 5.16 ; + +my $WALL= ''; +$WALL = ' -Wall -Wno-comment ' if $Config{'cc'} =~ /gcc/ ; +my $USE_PPPORT_H = ($ENV{PERL_CORE}) ? '' : '-DUSE_PPPORT_H'; + + +my $BUILD_BZIP2 = defined($ENV{BUILD_BZIP2}) ? $ENV{BUILD_BZIP2} : 1; +my $BZIP2_LIB = defined($ENV{BZIP2_LIB}) ? $ENV{BZIP2_LIB} : 'bzip2-src'; +my $BZIP2_INCLUDE = defined($ENV{BZIP2_INCLUDE}) ? $ENV{BZIP2_INCLUDE} : '.'; + +#ParseCONFIG() ; + +UpDowngrade(getPerlFiles('MANIFEST')) + unless $ENV{PERL_CORE}; + +WriteMakefile( + NAME => 'Compress::Raw::Bzip2', + VERSION_FROM => 'lib/Compress/Raw/Bzip2.pm', + INC => "-I$BZIP2_INCLUDE" , + DEFINE => "$WALL -DBZ_NO_STDIO $USE_PPPORT_H" , + XS => { 'Bzip2.xs' => 'Bzip2.c'}, + 'clean' => { FILES => '*.c bzip2.h bzlib.h bzlib_private.h constants.h constants.xs' }, + #'depend' => { 'Makefile' => 'config.in' }, + 'dist' => { COMPRESS => 'gzip', + TARFLAGS => '-chvf', + SUFFIX => 'gz', + DIST_DEFAULT => 'MyTrebleCheck tardist', + }, + + ( + $ENV{SKIP_FOR_CORE} + ? (MAN3PODS => {}) + : () + ), + + + ( + $BUILD_BZIP2 + ? bzip2_files($BZIP2_LIB) + : (LIBS => [ "-L$BZIP2_LIB -lbz2 " ]) + ), + + ( + $] >= 5.005 + ? (ABSTRACT_FROM => 'lib/Compress/Raw/Bzip2.pm', + AUTHOR => 'Paul Marquess ') + : () + ), + + ((ExtUtils::MakeMaker->VERSION() gt '6.30') ? + ('LICENSE' => 'perl') : ()), + +) ; + +my @names = qw( + BZ_RUN + BZ_FLUSH + BZ_FINISH + + BZ_OK + BZ_RUN_OK + BZ_FLUSH_OK + BZ_FINISH_OK + BZ_STREAM_END + BZ_SEQUENCE_ERROR + BZ_PARAM_ERROR + BZ_MEM_ERROR + BZ_DATA_ERROR + BZ_DATA_ERROR_MAGIC + BZ_IO_ERROR + BZ_UNEXPECTED_EOF + BZ_OUTBUFF_FULL + BZ_CONFIG_ERROR + ); + +if (eval {require ExtUtils::Constant; 1}) { + # Check the constants above all appear in @EXPORT in Bzip2.pm + my %names = map { $_, 1} @names ; #, 'BZ_VERSION'; + open F, ") + { + last if /^\s*\@EXPORT\s+=\s+qw\(/ ; + } + + while () + { + last if /^\s*\)/ ; + /(\S+)/ ; + delete $names{$1} if defined $1 ; + } + close F ; + + if ( keys %names ) + { + my $missing = join ("\n\t", sort keys %names) ; + die "The following names are missing from \@EXPORT in Bzip2.pm\n" . + "\t$missing\n" ; + } + + #push @names, {name => 'BZ_VERSION', type => 'PV' }; + + ExtUtils::Constant::WriteConstants( + NAME => 'Bzip2', + NAMES => \@names, + C_FILE => 'constants.h', + XS_FILE => 'constants.xs', + + ); +} +else { + foreach my $name (qw( constants.h constants.xs )) + { + my $from = catfile('fallback', $name); + copy ($from, $name) + or die "Can't copy $from to $name: $!"; + } +} + + +sub bzip2_files +{ + my $dir = shift ; + + my @c_files = qw( + blocksort.c + huffman.c + crctable.c + randtable.c + compress.c + decompress.c + bzlib.c + ); + + my @h_files = qw( bzlib.h bzlib_private.h ); + + foreach my $file (@c_files, @h_files) + { copy(catfile($dir, $file), '.') } + + + @h_files = map { catfile($dir, $_) } @h_files ; + my @o_files = map { "$_\$(OBJ_EXT)" } 'Bzip2', @c_files; + push @c_files, 'Bzip2.c' ; + + return ( + #'H' => [ @h_files ], + 'C' => [ @c_files ] , + #'OBJECT' => qq[ @o_files ], + 'OBJECT' => q[ $(O_FILES) ], + + + ) ; +} + + + diff --git a/ext/Compress-Raw-Bzip2/README b/ext/Compress-Raw-Bzip2/README new file mode 100644 index 0000000..158b897 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/README @@ -0,0 +1,181 @@ + + Compress-Raw-Bzip2 + + Version 2.015 + + 2nd September 2008 + + Copyright (c) 2005-2008 Paul Marquess. All rights reserved. + This program is free software; you can redistribute it + and/or modify it under the same terms as Perl itself. + + The directory bzip2-src contains a subset of the + source files copied directly from bzip2 version 1.0.5. + These files are Copyright(C) 1996-2007 Julian Seward. + See the file bzip2-src/LICENSE for licence details for these files. + Full source for the bzip2 library is available at + http://www.bzip.org/ + +DESCRIPTION +----------- + +Compress-Raw-Bzip2 provides the interface to the bzip2 library for the modules +IO::Compress::Bzip2 and IO::Compress::Bunzip2. + +PREREQUISITES +------------- + +Before you can build Compress-Raw-Bzip2 you need to have the following +installed on your system: + + * A C compiler + + * Perl 5.004 or better. + +BUILDING THE MODULE +------------------- + +Assuming you have met all the prerequisites, the module can now be built +using this sequence of commands: + + perl Makefile.PL + make + make test + +INSTALLATION +------------ + +To install Compress-Raw-Bzip2, run the command below: + + make install + +TROUBLESHOOTING +--------------- + +Solaris build fails with "language optional software package not installed" +--------------------------------------------------------------------------- + +If you are trying to build this module under Solaris and you get an +error message like this + + /usr/ucb/cc: language optional software package not installed + +it means that Perl cannot find the C compiler on your system. The cryptic +message is just Sun's way of telling you that you haven't bought their +C compiler. + +When you build a Perl module that needs a C compiler, the Perl build +system tries to use the same C compiler that was used to build perl +itself. In this case your Perl binary was built with a C compiler that +lived in /usr/ucb. + +To continue with building this module, you need to get a C compiler, +or tell Perl where your C compiler is, if you already have one. + +Assuming you have now got a C compiler, what you do next will be dependent +on what C compiler you have installed. If you have just installed Sun's +C compiler, you shouldn't have to do anything. Just try rebuilding +this module. + +If you have installed another C compiler, say gcc, you have to tell perl +how to use it instead of /usr/ucb/cc. + +This set of options seems to work if you want to use gcc. Your mileage +may vary. + + perl Makefile.PL CC=gcc CCCDLFLAGS=-fPIC OPTIMIZE=" " + make test + +If that doesn't work for you, it's time to make changes to the Makefile +by hand. Good luck! + +Solaris build fails with "gcc: unrecognized option `-KPIC'" +----------------------------------------------------------- + +You are running Solaris and you get an error like this when you try to +build this Perl module + + gcc: unrecognized option `-KPIC' + +This symptom usually means that you are using a Perl binary that has been +built with the Sun C compiler, but you are using gcc to build this module. + +When Perl builds modules that need a C compiler, it will attempt to use +the same C compiler and command line options that was used to build perl +itself. In this case "-KPIC" is a valid option for the Sun C compiler, +but not for gcc. The equivalent option for gcc is "-fPIC". + +The solution is either: + + 1. Build both Perl and this module with the same C compiler, either + by using the Sun C compiler for both or gcc for both. + + 2. Try generating the Makefile for this module like this perl + + perl Makefile.PL CC=gcc CCCDLFLAGS=-fPIC OPTIMIZE=" " LD=gcc + make test + + This second option seems to work when mixing a Perl binary built + with the Sun C compiler and this module built with gcc. Your + mileage may vary. + +HP-UX Notes +----------- + +I've had a report that when building Compress-Raw-Bzip2 under HP-UX that it +is necessary to have first built the bzip2 library with the -fpic +option. + +FEEDBACK +-------- + +How to report a problem with Compress-Raw-Bzip2. + +To help me help you, I need all of the following information: + + 1. The Versions of everything relevant. + This includes: + + a. The *complete* output from running this + + perl -V + + Do not edit the output in any way. + Note, I want you to run "perl -V" and NOT "perl -v". + + If your perl does not understand the "-V" option it is too + old. This module needs Perl version 5.004 or better. + + b. The version of Compress-Raw-Bzip2 you have. + If you have successfully installed Compress-Raw-Bzip2, this one-liner + will tell you: + + perl -MCompress::Raw::Bzip2 -e 'print qq[ver $Compress::Raw::Bzip2::VERSION\n]' + + If you are running windows use this + + perl -MCompress::Raw::Bzip2 -e "print qq[ver $Compress::Raw::Bzip2::VERSION\n]" + + If you haven't installed Compress-Raw-Bzip2 then search Compress::Raw::Bzip2.pm + for a line like this: + + $VERSION = "2.015" ; + + c. The version of bzip2 you have used. + If you have successfully installed Compress-Raw-Bzip2, this one-liner + will tell you: + + perl -MCompress::Raw::Bzip2 -e "print q[bzip2 ver ]. Compress::Raw::Bzip2::ZLIB_VERSION.qq[\n]" + + If not, look at the beginning of the file zlib.h. + + 2. If you are having problems building Compress-Raw-Bzip2, send me a + complete log of what happened. Start by unpacking the Compress-Raw-Bzip2 + module into a fresh directory and keep a log of all the steps + + [edit config.in, if necessary] + perl Makefile.PL + make + make test TEST_VERBOSE=1 + +Paul Marquess diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/.gitignore b/ext/Compress-Raw-Bzip2/bzip2-src/.gitignore new file mode 100644 index 0000000..970d78f --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/.gitignore @@ -0,0 +1 @@ +!*.c diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/LICENSE b/ext/Compress-Raw-Bzip2/bzip2-src/LICENSE new file mode 100644 index 0000000..f420cff --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/LICENSE @@ -0,0 +1,42 @@ + +-------------------------------------------------------------------------- + +This program, "bzip2", the associated library "libbzip2", and all +documentation, are copyright (C) 1996-2007 Julian R Seward. All +rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions +are met: + +1. Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + +2. The origin of this software must not be misrepresented; you must + not claim that you wrote the original software. If you use this + software in a product, an acknowledgment in the product + documentation would be appreciated but is not required. + +3. Altered source versions must be plainly marked as such, and must + not be misrepresented as being the original software. + +4. The name of the author may not be used to endorse or promote + products derived from this software without specific prior written + permission. + +THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS +OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED +WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY +DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE +GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, +WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING +NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS +SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +Julian Seward, jseward@bzip.org +bzip2/libbzip2 version 1.0.5 of 10 December 2007 + +-------------------------------------------------------------------------- diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/blocksort.c b/ext/Compress-Raw-Bzip2/bzip2-src/blocksort.c new file mode 100644 index 0000000..bd2dec1 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/blocksort.c @@ -0,0 +1,1094 @@ + +/*-------------------------------------------------------------*/ +/*--- Block sorting machinery ---*/ +/*--- blocksort.c ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#include "bzlib_private.h" + +/*---------------------------------------------*/ +/*--- Fallback O(N log(N)^2) sorting ---*/ +/*--- algorithm, for repetitive blocks ---*/ +/*---------------------------------------------*/ + +/*---------------------------------------------*/ +static +__inline__ +void fallbackSimpleSort ( UInt32* fmap, + UInt32* eclass, + Int32 lo, + Int32 hi ) +{ + Int32 i, j, tmp; + UInt32 ec_tmp; + + if (lo == hi) return; + + if (hi - lo > 3) { + for ( i = hi-4; i >= lo; i-- ) { + tmp = fmap[i]; + ec_tmp = eclass[tmp]; + for ( j = i+4; j <= hi && ec_tmp > eclass[fmap[j]]; j += 4 ) + fmap[j-4] = fmap[j]; + fmap[j-4] = tmp; + } + } + + for ( i = hi-1; i >= lo; i-- ) { + tmp = fmap[i]; + ec_tmp = eclass[tmp]; + for ( j = i+1; j <= hi && ec_tmp > eclass[fmap[j]]; j++ ) + fmap[j-1] = fmap[j]; + fmap[j-1] = tmp; + } +} + + +/*---------------------------------------------*/ +#define fswap(zz1, zz2) \ + { Int32 zztmp = zz1; zz1 = zz2; zz2 = zztmp; } + +#define fvswap(zzp1, zzp2, zzn) \ +{ \ + Int32 yyp1 = (zzp1); \ + Int32 yyp2 = (zzp2); \ + Int32 yyn = (zzn); \ + while (yyn > 0) { \ + fswap(fmap[yyp1], fmap[yyp2]); \ + yyp1++; yyp2++; yyn--; \ + } \ +} + + +#define fmin(a,b) ((a) < (b)) ? (a) : (b) + +#define fpush(lz,hz) { stackLo[sp] = lz; \ + stackHi[sp] = hz; \ + sp++; } + +#define fpop(lz,hz) { sp--; \ + lz = stackLo[sp]; \ + hz = stackHi[sp]; } + +#define FALLBACK_QSORT_SMALL_THRESH 10 +#define FALLBACK_QSORT_STACK_SIZE 100 + + +static +void fallbackQSort3 ( UInt32* fmap, + UInt32* eclass, + Int32 loSt, + Int32 hiSt ) +{ + Int32 unLo, unHi, ltLo, gtHi, n, m; + Int32 sp, lo, hi; + UInt32 med, r, r3; + Int32 stackLo[FALLBACK_QSORT_STACK_SIZE]; + Int32 stackHi[FALLBACK_QSORT_STACK_SIZE]; + + r = 0; + + sp = 0; + fpush ( loSt, hiSt ); + + while (sp > 0) { + + AssertH ( sp < FALLBACK_QSORT_STACK_SIZE - 1, 1004 ); + + fpop ( lo, hi ); + if (hi - lo < FALLBACK_QSORT_SMALL_THRESH) { + fallbackSimpleSort ( fmap, eclass, lo, hi ); + continue; + } + + /* Random partitioning. Median of 3 sometimes fails to + avoid bad cases. Median of 9 seems to help but + looks rather expensive. This too seems to work but + is cheaper. Guidance for the magic constants + 7621 and 32768 is taken from Sedgewick's algorithms + book, chapter 35. + */ + r = ((r * 7621) + 1) % 32768; + r3 = r % 3; + if (r3 == 0) med = eclass[fmap[lo]]; else + if (r3 == 1) med = eclass[fmap[(lo+hi)>>1]]; else + med = eclass[fmap[hi]]; + + unLo = ltLo = lo; + unHi = gtHi = hi; + + while (1) { + while (1) { + if (unLo > unHi) break; + n = (Int32)eclass[fmap[unLo]] - (Int32)med; + if (n == 0) { + fswap(fmap[unLo], fmap[ltLo]); + ltLo++; unLo++; + continue; + }; + if (n > 0) break; + unLo++; + } + while (1) { + if (unLo > unHi) break; + n = (Int32)eclass[fmap[unHi]] - (Int32)med; + if (n == 0) { + fswap(fmap[unHi], fmap[gtHi]); + gtHi--; unHi--; + continue; + }; + if (n < 0) break; + unHi--; + } + if (unLo > unHi) break; + fswap(fmap[unLo], fmap[unHi]); unLo++; unHi--; + } + + AssertD ( unHi == unLo-1, "fallbackQSort3(2)" ); + + if (gtHi < ltLo) continue; + + n = fmin(ltLo-lo, unLo-ltLo); fvswap(lo, unLo-n, n); + m = fmin(hi-gtHi, gtHi-unHi); fvswap(unLo, hi-m+1, m); + + n = lo + unLo - ltLo - 1; + m = hi - (gtHi - unHi) + 1; + + if (n - lo > hi - m) { + fpush ( lo, n ); + fpush ( m, hi ); + } else { + fpush ( m, hi ); + fpush ( lo, n ); + } + } +} + +#undef fmin +#undef fpush +#undef fpop +#undef fswap +#undef fvswap +#undef FALLBACK_QSORT_SMALL_THRESH +#undef FALLBACK_QSORT_STACK_SIZE + + +/*---------------------------------------------*/ +/* Pre: + nblock > 0 + eclass exists for [0 .. nblock-1] + ((UChar*)eclass) [0 .. nblock-1] holds block + ptr exists for [0 .. nblock-1] + + Post: + ((UChar*)eclass) [0 .. nblock-1] holds block + All other areas of eclass destroyed + fmap [0 .. nblock-1] holds sorted order + bhtab [ 0 .. 2+(nblock/32) ] destroyed +*/ + +#define SET_BH(zz) bhtab[(zz) >> 5] |= (1 << ((zz) & 31)) +#define CLEAR_BH(zz) bhtab[(zz) >> 5] &= ~(1 << ((zz) & 31)) +#define ISSET_BH(zz) (bhtab[(zz) >> 5] & (1 << ((zz) & 31))) +#define WORD_BH(zz) bhtab[(zz) >> 5] +#define UNALIGNED_BH(zz) ((zz) & 0x01f) + +static +void fallbackSort ( UInt32* fmap, + UInt32* eclass, + UInt32* bhtab, + Int32 nblock, + Int32 verb ) +{ + Int32 ftab[257]; + Int32 ftabCopy[256]; + Int32 H, i, j, k, l, r, cc, cc1; + Int32 nNotDone; + Int32 nBhtab; + UChar* eclass8 = (UChar*)eclass; + + /*-- + Initial 1-char radix sort to generate + initial fmap and initial BH bits. + --*/ + if (verb >= 4) + VPrintf0 ( " bucket sorting ...\n" ); + for (i = 0; i < 257; i++) ftab[i] = 0; + for (i = 0; i < nblock; i++) ftab[eclass8[i]]++; + for (i = 0; i < 256; i++) ftabCopy[i] = ftab[i]; + for (i = 1; i < 257; i++) ftab[i] += ftab[i-1]; + + for (i = 0; i < nblock; i++) { + j = eclass8[i]; + k = ftab[j] - 1; + ftab[j] = k; + fmap[k] = i; + } + + nBhtab = 2 + (nblock / 32); + for (i = 0; i < nBhtab; i++) bhtab[i] = 0; + for (i = 0; i < 256; i++) SET_BH(ftab[i]); + + /*-- + Inductively refine the buckets. Kind-of an + "exponential radix sort" (!), inspired by the + Manber-Myers suffix array construction algorithm. + --*/ + + /*-- set sentinel bits for block-end detection --*/ + for (i = 0; i < 32; i++) { + SET_BH(nblock + 2*i); + CLEAR_BH(nblock + 2*i + 1); + } + + /*-- the log(N) loop --*/ + H = 1; + while (1) { + + if (verb >= 4) + VPrintf1 ( " depth %6d has ", H ); + + j = 0; + for (i = 0; i < nblock; i++) { + if (ISSET_BH(i)) j = i; + k = fmap[i] - H; if (k < 0) k += nblock; + eclass[k] = j; + } + + nNotDone = 0; + r = -1; + while (1) { + + /*-- find the next non-singleton bucket --*/ + k = r + 1; + while (ISSET_BH(k) && UNALIGNED_BH(k)) k++; + if (ISSET_BH(k)) { + while (WORD_BH(k) == 0xffffffff) k += 32; + while (ISSET_BH(k)) k++; + } + l = k - 1; + if (l >= nblock) break; + while (!ISSET_BH(k) && UNALIGNED_BH(k)) k++; + if (!ISSET_BH(k)) { + while (WORD_BH(k) == 0x00000000) k += 32; + while (!ISSET_BH(k)) k++; + } + r = k - 1; + if (r >= nblock) break; + + /*-- now [l, r] bracket current bucket --*/ + if (r > l) { + nNotDone += (r - l + 1); + fallbackQSort3 ( fmap, eclass, l, r ); + + /*-- scan bucket and generate header bits-- */ + cc = -1; + for (i = l; i <= r; i++) { + cc1 = eclass[fmap[i]]; + if (cc != cc1) { SET_BH(i); cc = cc1; }; + } + } + } + + if (verb >= 4) + VPrintf1 ( "%6d unresolved strings\n", nNotDone ); + + H *= 2; + if (H > nblock || nNotDone == 0) break; + } + + /*-- + Reconstruct the original block in + eclass8 [0 .. nblock-1], since the + previous phase destroyed it. + --*/ + if (verb >= 4) + VPrintf0 ( " reconstructing block ...\n" ); + j = 0; + for (i = 0; i < nblock; i++) { + while (ftabCopy[j] == 0) j++; + ftabCopy[j]--; + eclass8[fmap[i]] = (UChar)j; + } + AssertH ( j < 256, 1005 ); +} + +#undef SET_BH +#undef CLEAR_BH +#undef ISSET_BH +#undef WORD_BH +#undef UNALIGNED_BH + + +/*---------------------------------------------*/ +/*--- The main, O(N^2 log(N)) sorting ---*/ +/*--- algorithm. Faster for "normal" ---*/ +/*--- non-repetitive blocks. ---*/ +/*---------------------------------------------*/ + +/*---------------------------------------------*/ +static +__inline__ +Bool mainGtU ( UInt32 i1, + UInt32 i2, + UChar* block, + UInt16* quadrant, + UInt32 nblock, + Int32* budget ) +{ + Int32 k; + UChar c1, c2; + UInt16 s1, s2; + + AssertD ( i1 != i2, "mainGtU" ); + /* 1 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 2 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 3 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 4 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 5 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 6 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 7 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 8 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 9 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 10 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 11 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 12 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + + k = nblock + 8; + + do { + /* 1 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + /* 2 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + /* 3 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + /* 4 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + /* 5 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + /* 6 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + /* 7 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + /* 8 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + + if (i1 >= nblock) i1 -= nblock; + if (i2 >= nblock) i2 -= nblock; + + k -= 8; + (*budget)--; + } + while (k >= 0); + + return False; +} + + +/*---------------------------------------------*/ +/*-- + Knuth's increments seem to work better + than Incerpi-Sedgewick here. Possibly + because the number of elems to sort is + usually small, typically <= 20. +--*/ +static +Int32 incs[14] = { 1, 4, 13, 40, 121, 364, 1093, 3280, + 9841, 29524, 88573, 265720, + 797161, 2391484 }; + +static +void mainSimpleSort ( UInt32* ptr, + UChar* block, + UInt16* quadrant, + Int32 nblock, + Int32 lo, + Int32 hi, + Int32 d, + Int32* budget ) +{ + Int32 i, j, h, bigN, hp; + UInt32 v; + + bigN = hi - lo + 1; + if (bigN < 2) return; + + hp = 0; + while (incs[hp] < bigN) hp++; + hp--; + + for (; hp >= 0; hp--) { + h = incs[hp]; + + i = lo + h; + while (True) { + + /*-- copy 1 --*/ + if (i > hi) break; + v = ptr[i]; + j = i; + while ( mainGtU ( + ptr[j-h]+d, v+d, block, quadrant, nblock, budget + ) ) { + ptr[j] = ptr[j-h]; + j = j - h; + if (j <= (lo + h - 1)) break; + } + ptr[j] = v; + i++; + + /*-- copy 2 --*/ + if (i > hi) break; + v = ptr[i]; + j = i; + while ( mainGtU ( + ptr[j-h]+d, v+d, block, quadrant, nblock, budget + ) ) { + ptr[j] = ptr[j-h]; + j = j - h; + if (j <= (lo + h - 1)) break; + } + ptr[j] = v; + i++; + + /*-- copy 3 --*/ + if (i > hi) break; + v = ptr[i]; + j = i; + while ( mainGtU ( + ptr[j-h]+d, v+d, block, quadrant, nblock, budget + ) ) { + ptr[j] = ptr[j-h]; + j = j - h; + if (j <= (lo + h - 1)) break; + } + ptr[j] = v; + i++; + + if (*budget < 0) return; + } + } +} + + +/*---------------------------------------------*/ +/*-- + The following is an implementation of + an elegant 3-way quicksort for strings, + described in a paper "Fast Algorithms for + Sorting and Searching Strings", by Robert + Sedgewick and Jon L. Bentley. +--*/ + +#define mswap(zz1, zz2) \ + { Int32 zztmp = zz1; zz1 = zz2; zz2 = zztmp; } + +#define mvswap(zzp1, zzp2, zzn) \ +{ \ + Int32 yyp1 = (zzp1); \ + Int32 yyp2 = (zzp2); \ + Int32 yyn = (zzn); \ + while (yyn > 0) { \ + mswap(ptr[yyp1], ptr[yyp2]); \ + yyp1++; yyp2++; yyn--; \ + } \ +} + +static +__inline__ +UChar mmed3 ( UChar a, UChar b, UChar c ) +{ + UChar t; + if (a > b) { t = a; a = b; b = t; }; + if (b > c) { + b = c; + if (a > b) b = a; + } + return b; +} + +#define mmin(a,b) ((a) < (b)) ? (a) : (b) + +#define mpush(lz,hz,dz) { stackLo[sp] = lz; \ + stackHi[sp] = hz; \ + stackD [sp] = dz; \ + sp++; } + +#define mpop(lz,hz,dz) { sp--; \ + lz = stackLo[sp]; \ + hz = stackHi[sp]; \ + dz = stackD [sp]; } + + +#define mnextsize(az) (nextHi[az]-nextLo[az]) + +#define mnextswap(az,bz) \ + { Int32 tz; \ + tz = nextLo[az]; nextLo[az] = nextLo[bz]; nextLo[bz] = tz; \ + tz = nextHi[az]; nextHi[az] = nextHi[bz]; nextHi[bz] = tz; \ + tz = nextD [az]; nextD [az] = nextD [bz]; nextD [bz] = tz; } + + +#define MAIN_QSORT_SMALL_THRESH 20 +#define MAIN_QSORT_DEPTH_THRESH (BZ_N_RADIX + BZ_N_QSORT) +#define MAIN_QSORT_STACK_SIZE 100 + +static +void mainQSort3 ( UInt32* ptr, + UChar* block, + UInt16* quadrant, + Int32 nblock, + Int32 loSt, + Int32 hiSt, + Int32 dSt, + Int32* budget ) +{ + Int32 unLo, unHi, ltLo, gtHi, n, m, med; + Int32 sp, lo, hi, d; + + Int32 stackLo[MAIN_QSORT_STACK_SIZE]; + Int32 stackHi[MAIN_QSORT_STACK_SIZE]; + Int32 stackD [MAIN_QSORT_STACK_SIZE]; + + Int32 nextLo[3]; + Int32 nextHi[3]; + Int32 nextD [3]; + + sp = 0; + mpush ( loSt, hiSt, dSt ); + + while (sp > 0) { + + AssertH ( sp < MAIN_QSORT_STACK_SIZE - 2, 1001 ); + + mpop ( lo, hi, d ); + if (hi - lo < MAIN_QSORT_SMALL_THRESH || + d > MAIN_QSORT_DEPTH_THRESH) { + mainSimpleSort ( ptr, block, quadrant, nblock, lo, hi, d, budget ); + if (*budget < 0) return; + continue; + } + + med = (Int32) + mmed3 ( block[ptr[ lo ]+d], + block[ptr[ hi ]+d], + block[ptr[ (lo+hi)>>1 ]+d] ); + + unLo = ltLo = lo; + unHi = gtHi = hi; + + while (True) { + while (True) { + if (unLo > unHi) break; + n = ((Int32)block[ptr[unLo]+d]) - med; + if (n == 0) { + mswap(ptr[unLo], ptr[ltLo]); + ltLo++; unLo++; continue; + }; + if (n > 0) break; + unLo++; + } + while (True) { + if (unLo > unHi) break; + n = ((Int32)block[ptr[unHi]+d]) - med; + if (n == 0) { + mswap(ptr[unHi], ptr[gtHi]); + gtHi--; unHi--; continue; + }; + if (n < 0) break; + unHi--; + } + if (unLo > unHi) break; + mswap(ptr[unLo], ptr[unHi]); unLo++; unHi--; + } + + AssertD ( unHi == unLo-1, "mainQSort3(2)" ); + + if (gtHi < ltLo) { + mpush(lo, hi, d+1 ); + continue; + } + + n = mmin(ltLo-lo, unLo-ltLo); mvswap(lo, unLo-n, n); + m = mmin(hi-gtHi, gtHi-unHi); mvswap(unLo, hi-m+1, m); + + n = lo + unLo - ltLo - 1; + m = hi - (gtHi - unHi) + 1; + + nextLo[0] = lo; nextHi[0] = n; nextD[0] = d; + nextLo[1] = m; nextHi[1] = hi; nextD[1] = d; + nextLo[2] = n+1; nextHi[2] = m-1; nextD[2] = d+1; + + if (mnextsize(0) < mnextsize(1)) mnextswap(0,1); + if (mnextsize(1) < mnextsize(2)) mnextswap(1,2); + if (mnextsize(0) < mnextsize(1)) mnextswap(0,1); + + AssertD (mnextsize(0) >= mnextsize(1), "mainQSort3(8)" ); + AssertD (mnextsize(1) >= mnextsize(2), "mainQSort3(9)" ); + + mpush (nextLo[0], nextHi[0], nextD[0]); + mpush (nextLo[1], nextHi[1], nextD[1]); + mpush (nextLo[2], nextHi[2], nextD[2]); + } +} + +#undef mswap +#undef mvswap +#undef mpush +#undef mpop +#undef mmin +#undef mnextsize +#undef mnextswap +#undef MAIN_QSORT_SMALL_THRESH +#undef MAIN_QSORT_DEPTH_THRESH +#undef MAIN_QSORT_STACK_SIZE + + +/*---------------------------------------------*/ +/* Pre: + nblock > N_OVERSHOOT + block32 exists for [0 .. nblock-1 +N_OVERSHOOT] + ((UChar*)block32) [0 .. nblock-1] holds block + ptr exists for [0 .. nblock-1] + + Post: + ((UChar*)block32) [0 .. nblock-1] holds block + All other areas of block32 destroyed + ftab [0 .. 65536 ] destroyed + ptr [0 .. nblock-1] holds sorted order + if (*budget < 0), sorting was abandoned +*/ + +#define BIGFREQ(b) (ftab[((b)+1) << 8] - ftab[(b) << 8]) +#define SETMASK (1 << 21) +#define CLEARMASK (~(SETMASK)) + +static +void mainSort ( UInt32* ptr, + UChar* block, + UInt16* quadrant, + UInt32* ftab, + Int32 nblock, + Int32 verb, + Int32* budget ) +{ + Int32 i, j, k, ss, sb; + Int32 runningOrder[256]; + Bool bigDone[256]; + Int32 copyStart[256]; + Int32 copyEnd [256]; + UChar c1; + Int32 numQSorted; + UInt16 s; + if (verb >= 4) VPrintf0 ( " main sort initialise ...\n" ); + + /*-- set up the 2-byte frequency table --*/ + for (i = 65536; i >= 0; i--) ftab[i] = 0; + + j = block[0] << 8; + i = nblock-1; + for (; i >= 3; i -= 4) { + quadrant[i] = 0; + j = (j >> 8) | ( ((UInt16)block[i]) << 8); + ftab[j]++; + quadrant[i-1] = 0; + j = (j >> 8) | ( ((UInt16)block[i-1]) << 8); + ftab[j]++; + quadrant[i-2] = 0; + j = (j >> 8) | ( ((UInt16)block[i-2]) << 8); + ftab[j]++; + quadrant[i-3] = 0; + j = (j >> 8) | ( ((UInt16)block[i-3]) << 8); + ftab[j]++; + } + for (; i >= 0; i--) { + quadrant[i] = 0; + j = (j >> 8) | ( ((UInt16)block[i]) << 8); + ftab[j]++; + } + + /*-- (emphasises close relationship of block & quadrant) --*/ + for (i = 0; i < BZ_N_OVERSHOOT; i++) { + block [nblock+i] = block[i]; + quadrant[nblock+i] = 0; + } + + if (verb >= 4) VPrintf0 ( " bucket sorting ...\n" ); + + /*-- Complete the initial radix sort --*/ + for (i = 1; i <= 65536; i++) ftab[i] += ftab[i-1]; + + s = block[0] << 8; + i = nblock-1; + for (; i >= 3; i -= 4) { + s = (s >> 8) | (block[i] << 8); + j = ftab[s] -1; + ftab[s] = j; + ptr[j] = i; + s = (s >> 8) | (block[i-1] << 8); + j = ftab[s] -1; + ftab[s] = j; + ptr[j] = i-1; + s = (s >> 8) | (block[i-2] << 8); + j = ftab[s] -1; + ftab[s] = j; + ptr[j] = i-2; + s = (s >> 8) | (block[i-3] << 8); + j = ftab[s] -1; + ftab[s] = j; + ptr[j] = i-3; + } + for (; i >= 0; i--) { + s = (s >> 8) | (block[i] << 8); + j = ftab[s] -1; + ftab[s] = j; + ptr[j] = i; + } + + /*-- + Now ftab contains the first loc of every small bucket. + Calculate the running order, from smallest to largest + big bucket. + --*/ + for (i = 0; i <= 255; i++) { + bigDone [i] = False; + runningOrder[i] = i; + } + + { + Int32 vv; + Int32 h = 1; + do h = 3 * h + 1; while (h <= 256); + do { + h = h / 3; + for (i = h; i <= 255; i++) { + vv = runningOrder[i]; + j = i; + while ( BIGFREQ(runningOrder[j-h]) > BIGFREQ(vv) ) { + runningOrder[j] = runningOrder[j-h]; + j = j - h; + if (j <= (h - 1)) goto zero; + } + zero: + runningOrder[j] = vv; + } + } while (h != 1); + } + + /*-- + The main sorting loop. + --*/ + + numQSorted = 0; + + for (i = 0; i <= 255; i++) { + + /*-- + Process big buckets, starting with the least full. + Basically this is a 3-step process in which we call + mainQSort3 to sort the small buckets [ss, j], but + also make a big effort to avoid the calls if we can. + --*/ + ss = runningOrder[i]; + + /*-- + Step 1: + Complete the big bucket [ss] by quicksorting + any unsorted small buckets [ss, j], for j != ss. + Hopefully previous pointer-scanning phases have already + completed many of the small buckets [ss, j], so + we don't have to sort them at all. + --*/ + for (j = 0; j <= 255; j++) { + if (j != ss) { + sb = (ss << 8) + j; + if ( ! (ftab[sb] & SETMASK) ) { + Int32 lo = ftab[sb] & CLEARMASK; + Int32 hi = (ftab[sb+1] & CLEARMASK) - 1; + if (hi > lo) { + if (verb >= 4) + VPrintf4 ( " qsort [0x%x, 0x%x] " + "done %d this %d\n", + ss, j, numQSorted, hi - lo + 1 ); + mainQSort3 ( + ptr, block, quadrant, nblock, + lo, hi, BZ_N_RADIX, budget + ); + numQSorted += (hi - lo + 1); + if (*budget < 0) return; + } + } + ftab[sb] |= SETMASK; + } + } + + AssertH ( !bigDone[ss], 1006 ); + + /*-- + Step 2: + Now scan this big bucket [ss] so as to synthesise the + sorted order for small buckets [t, ss] for all t, + including, magically, the bucket [ss,ss] too. + This will avoid doing Real Work in subsequent Step 1's. + --*/ + { + for (j = 0; j <= 255; j++) { + copyStart[j] = ftab[(j << 8) + ss] & CLEARMASK; + copyEnd [j] = (ftab[(j << 8) + ss + 1] & CLEARMASK) - 1; + } + for (j = ftab[ss << 8] & CLEARMASK; j < copyStart[ss]; j++) { + k = ptr[j]-1; if (k < 0) k += nblock; + c1 = block[k]; + if (!bigDone[c1]) + ptr[ copyStart[c1]++ ] = k; + } + for (j = (ftab[(ss+1) << 8] & CLEARMASK) - 1; j > copyEnd[ss]; j--) { + k = ptr[j]-1; if (k < 0) k += nblock; + c1 = block[k]; + if (!bigDone[c1]) + ptr[ copyEnd[c1]-- ] = k; + } + } + + AssertH ( (copyStart[ss]-1 == copyEnd[ss]) + || + /* Extremely rare case missing in bzip2-1.0.0 and 1.0.1. + Necessity for this case is demonstrated by compressing + a sequence of approximately 48.5 million of character + 251; 1.0.0/1.0.1 will then die here. */ + (copyStart[ss] == 0 && copyEnd[ss] == nblock-1), + 1007 ) + + for (j = 0; j <= 255; j++) ftab[(j << 8) + ss] |= SETMASK; + + /*-- + Step 3: + The [ss] big bucket is now done. Record this fact, + and update the quadrant descriptors. Remember to + update quadrants in the overshoot area too, if + necessary. The "if (i < 255)" test merely skips + this updating for the last bucket processed, since + updating for the last bucket is pointless. + + The quadrant array provides a way to incrementally + cache sort orderings, as they appear, so as to + make subsequent comparisons in fullGtU() complete + faster. For repetitive blocks this makes a big + difference (but not big enough to be able to avoid + the fallback sorting mechanism, exponential radix sort). + + The precise meaning is: at all times: + + for 0 <= i < nblock and 0 <= j <= nblock + + if block[i] != block[j], + + then the relative values of quadrant[i] and + quadrant[j] are meaningless. + + else { + if quadrant[i] < quadrant[j] + then the string starting at i lexicographically + precedes the string starting at j + + else if quadrant[i] > quadrant[j] + then the string starting at j lexicographically + precedes the string starting at i + + else + the relative ordering of the strings starting + at i and j has not yet been determined. + } + --*/ + bigDone[ss] = True; + + if (i < 255) { + Int32 bbStart = ftab[ss << 8] & CLEARMASK; + Int32 bbSize = (ftab[(ss+1) << 8] & CLEARMASK) - bbStart; + Int32 shifts = 0; + + while ((bbSize >> shifts) > 65534) shifts++; + + for (j = bbSize-1; j >= 0; j--) { + Int32 a2update = ptr[bbStart + j]; + UInt16 qVal = (UInt16)(j >> shifts); + quadrant[a2update] = qVal; + if (a2update < BZ_N_OVERSHOOT) + quadrant[a2update + nblock] = qVal; + } + AssertH ( ((bbSize-1) >> shifts) <= 65535, 1002 ); + } + + } + + if (verb >= 4) + VPrintf3 ( " %d pointers, %d sorted, %d scanned\n", + nblock, numQSorted, nblock - numQSorted ); +} + +#undef BIGFREQ +#undef SETMASK +#undef CLEARMASK + + +/*---------------------------------------------*/ +/* Pre: + nblock > 0 + arr2 exists for [0 .. nblock-1 +N_OVERSHOOT] + ((UChar*)arr2) [0 .. nblock-1] holds block + arr1 exists for [0 .. nblock-1] + + Post: + ((UChar*)arr2) [0 .. nblock-1] holds block + All other areas of block destroyed + ftab [ 0 .. 65536 ] destroyed + arr1 [0 .. nblock-1] holds sorted order +*/ +void BZ2_blockSort ( EState* s ) +{ + UInt32* ptr = s->ptr; + UChar* block = s->block; + UInt32* ftab = s->ftab; + Int32 nblock = s->nblock; + Int32 verb = s->verbosity; + Int32 wfact = s->workFactor; + UInt16* quadrant; + Int32 budget; + Int32 budgetInit; + Int32 i; + + if (nblock < 10000) { + fallbackSort ( s->arr1, s->arr2, ftab, nblock, verb ); + } else { + /* Calculate the location for quadrant, remembering to get + the alignment right. Assumes that &(block[0]) is at least + 2-byte aligned -- this should be ok since block is really + the first section of arr2. + */ + i = nblock+BZ_N_OVERSHOOT; + if (i & 1) i++; + quadrant = (UInt16*)(&(block[i])); + + /* (wfact-1) / 3 puts the default-factor-30 + transition point at very roughly the same place as + with v0.1 and v0.9.0. + Not that it particularly matters any more, since the + resulting compressed stream is now the same regardless + of whether or not we use the main sort or fallback sort. + */ + if (wfact < 1 ) wfact = 1; + if (wfact > 100) wfact = 100; + budgetInit = nblock * ((wfact-1) / 3); + budget = budgetInit; + + mainSort ( ptr, block, quadrant, ftab, nblock, verb, &budget ); + if (verb >= 3) + VPrintf3 ( " %d work, %d block, ratio %5.2f\n", + budgetInit - budget, + nblock, + (float)(budgetInit - budget) / + (float)(nblock==0 ? 1 : nblock) ); + if (budget < 0) { + if (verb >= 2) + VPrintf0 ( " too repetitive; using fallback" + " sorting algorithm\n" ); + fallbackSort ( s->arr1, s->arr2, ftab, nblock, verb ); + } + } + + s->origPtr = -1; + for (i = 0; i < s->nblock; i++) + if (ptr[i] == 0) + { s->origPtr = i; break; }; + + AssertH( s->origPtr != -1, 1003 ); +} + + +/*-------------------------------------------------------------*/ +/*--- end blocksort.c ---*/ +/*-------------------------------------------------------------*/ diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/bzip2.c b/ext/Compress-Raw-Bzip2/bzip2-src/bzip2.c new file mode 100644 index 0000000..3904107 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/bzip2.c @@ -0,0 +1,2034 @@ + +/*-----------------------------------------------------------*/ +/*--- A block-sorting, lossless compressor bzip2.c ---*/ +/*-----------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +/* Place a 1 beside your platform, and 0 elsewhere. + Generic 32-bit Unix. + Also works on 64-bit Unix boxes. + This is the default. +*/ +#define BZ_UNIX 1 + +/*-- + Win32, as seen by Jacob Navia's excellent + port of (Chris Fraser & David Hanson)'s excellent + lcc compiler. Or with MS Visual C. + This is selected automatically if compiled by a compiler which + defines _WIN32, not including the Cygwin GCC. +--*/ +#define BZ_LCCWIN32 0 + +#if defined(_WIN32) && !defined(__CYGWIN__) +#undef BZ_LCCWIN32 +#define BZ_LCCWIN32 1 +#undef BZ_UNIX +#define BZ_UNIX 0 +#endif + + +/*---------------------------------------------*/ +/*-- + Some stuff for all platforms. +--*/ + +#include +#include +#include +#include +#include +#include +#include +#include "bzlib.h" + +#define ERROR_IF_EOF(i) { if ((i) == EOF) ioError(); } +#define ERROR_IF_NOT_ZERO(i) { if ((i) != 0) ioError(); } +#define ERROR_IF_MINUS_ONE(i) { if ((i) == (-1)) ioError(); } + + +/*---------------------------------------------*/ +/*-- + Platform-specific stuff. +--*/ + +#if BZ_UNIX +# include +# include +# include +# include +# include +# include + +# define PATH_SEP '/' +# define MY_LSTAT lstat +# define MY_STAT stat +# define MY_S_ISREG S_ISREG +# define MY_S_ISDIR S_ISDIR + +# define APPEND_FILESPEC(root, name) \ + root=snocString((root), (name)) + +# define APPEND_FLAG(root, name) \ + root=snocString((root), (name)) + +# define SET_BINARY_MODE(fd) /**/ + +# ifdef __GNUC__ +# define NORETURN __attribute__ ((noreturn)) +# else +# define NORETURN /**/ +# endif + +# ifdef __DJGPP__ +# include +# include +# undef MY_LSTAT +# undef MY_STAT +# define MY_LSTAT stat +# define MY_STAT stat +# undef SET_BINARY_MODE +# define SET_BINARY_MODE(fd) \ + do { \ + int retVal = setmode ( fileno ( fd ), \ + O_BINARY ); \ + ERROR_IF_MINUS_ONE ( retVal ); \ + } while ( 0 ) +# endif + +# ifdef __CYGWIN__ +# include +# include +# undef SET_BINARY_MODE +# define SET_BINARY_MODE(fd) \ + do { \ + int retVal = setmode ( fileno ( fd ), \ + O_BINARY ); \ + ERROR_IF_MINUS_ONE ( retVal ); \ + } while ( 0 ) +# endif +#endif /* BZ_UNIX */ + + + +#if BZ_LCCWIN32 +# include +# include +# include + +# define NORETURN /**/ +# define PATH_SEP '\\' +# define MY_LSTAT _stat +# define MY_STAT _stat +# define MY_S_ISREG(x) ((x) & _S_IFREG) +# define MY_S_ISDIR(x) ((x) & _S_IFDIR) + +# define APPEND_FLAG(root, name) \ + root=snocString((root), (name)) + +# define APPEND_FILESPEC(root, name) \ + root = snocString ((root), (name)) + +# define SET_BINARY_MODE(fd) \ + do { \ + int retVal = setmode ( fileno ( fd ), \ + O_BINARY ); \ + ERROR_IF_MINUS_ONE ( retVal ); \ + } while ( 0 ) + +#endif /* BZ_LCCWIN32 */ + + +/*---------------------------------------------*/ +/*-- + Some more stuff for all platforms :-) +--*/ + +typedef char Char; +typedef unsigned char Bool; +typedef unsigned char UChar; +typedef int Int32; +typedef unsigned int UInt32; +typedef short Int16; +typedef unsigned short UInt16; + +#define True ((Bool)1) +#define False ((Bool)0) + +/*-- + IntNative is your platform's `native' int size. + Only here to avoid probs with 64-bit platforms. +--*/ +typedef int IntNative; + + +/*---------------------------------------------------*/ +/*--- Misc (file handling) data decls ---*/ +/*---------------------------------------------------*/ + +Int32 verbosity; +Bool keepInputFiles, smallMode, deleteOutputOnInterrupt; +Bool forceOverwrite, testFailsExist, unzFailsExist, noisy; +Int32 numFileNames, numFilesProcessed, blockSize100k; +Int32 exitValue; + +/*-- source modes; F==file, I==stdin, O==stdout --*/ +#define SM_I2O 1 +#define SM_F2O 2 +#define SM_F2F 3 + +/*-- operation modes --*/ +#define OM_Z 1 +#define OM_UNZ 2 +#define OM_TEST 3 + +Int32 opMode; +Int32 srcMode; + +#define FILE_NAME_LEN 1034 + +Int32 longestFileName; +Char inName [FILE_NAME_LEN]; +Char outName[FILE_NAME_LEN]; +Char tmpName[FILE_NAME_LEN]; +Char *progName; +Char progNameReally[FILE_NAME_LEN]; +FILE *outputHandleJustInCase; +Int32 workFactor; + +static void panic ( const Char* ) NORETURN; +static void ioError ( void ) NORETURN; +static void outOfMemory ( void ) NORETURN; +static void configError ( void ) NORETURN; +static void crcError ( void ) NORETURN; +static void cleanUpAndFail ( Int32 ) NORETURN; +static void compressedStreamEOF ( void ) NORETURN; + +static void copyFileName ( Char*, Char* ); +static void* myMalloc ( Int32 ); +static void applySavedFileAttrToOutputFile ( IntNative fd ); + + + +/*---------------------------------------------------*/ +/*--- An implementation of 64-bit ints. Sigh. ---*/ +/*--- Roll on widespread deployment of ANSI C9X ! ---*/ +/*---------------------------------------------------*/ + +typedef + struct { UChar b[8]; } + UInt64; + + +static +void uInt64_from_UInt32s ( UInt64* n, UInt32 lo32, UInt32 hi32 ) +{ + n->b[7] = (UChar)((hi32 >> 24) & 0xFF); + n->b[6] = (UChar)((hi32 >> 16) & 0xFF); + n->b[5] = (UChar)((hi32 >> 8) & 0xFF); + n->b[4] = (UChar) (hi32 & 0xFF); + n->b[3] = (UChar)((lo32 >> 24) & 0xFF); + n->b[2] = (UChar)((lo32 >> 16) & 0xFF); + n->b[1] = (UChar)((lo32 >> 8) & 0xFF); + n->b[0] = (UChar) (lo32 & 0xFF); +} + + +static +double uInt64_to_double ( UInt64* n ) +{ + Int32 i; + double base = 1.0; + double sum = 0.0; + for (i = 0; i < 8; i++) { + sum += base * (double)(n->b[i]); + base *= 256.0; + } + return sum; +} + + +static +Bool uInt64_isZero ( UInt64* n ) +{ + Int32 i; + for (i = 0; i < 8; i++) + if (n->b[i] != 0) return 0; + return 1; +} + + +/* Divide *n by 10, and return the remainder. */ +static +Int32 uInt64_qrm10 ( UInt64* n ) +{ + UInt32 rem, tmp; + Int32 i; + rem = 0; + for (i = 7; i >= 0; i--) { + tmp = rem * 256 + n->b[i]; + n->b[i] = tmp / 10; + rem = tmp % 10; + } + return rem; +} + + +/* ... and the Whole Entire Point of all this UInt64 stuff is + so that we can supply the following function. +*/ +static +void uInt64_toAscii ( char* outbuf, UInt64* n ) +{ + Int32 i, q; + UChar buf[32]; + Int32 nBuf = 0; + UInt64 n_copy = *n; + do { + q = uInt64_qrm10 ( &n_copy ); + buf[nBuf] = q + '0'; + nBuf++; + } while (!uInt64_isZero(&n_copy)); + outbuf[nBuf] = 0; + for (i = 0; i < nBuf; i++) + outbuf[i] = buf[nBuf-i-1]; +} + + +/*---------------------------------------------------*/ +/*--- Processing of complete files and streams ---*/ +/*---------------------------------------------------*/ + +/*---------------------------------------------*/ +static +Bool myfeof ( FILE* f ) +{ + Int32 c = fgetc ( f ); + if (c == EOF) return True; + ungetc ( c, f ); + return False; +} + + +/*---------------------------------------------*/ +static +void compressStream ( FILE *stream, FILE *zStream ) +{ + BZFILE* bzf = NULL; + UChar ibuf[5000]; + Int32 nIbuf; + UInt32 nbytes_in_lo32, nbytes_in_hi32; + UInt32 nbytes_out_lo32, nbytes_out_hi32; + Int32 bzerr, bzerr_dummy, ret; + + SET_BINARY_MODE(stream); + SET_BINARY_MODE(zStream); + + if (ferror(stream)) goto errhandler_io; + if (ferror(zStream)) goto errhandler_io; + + bzf = BZ2_bzWriteOpen ( &bzerr, zStream, + blockSize100k, verbosity, workFactor ); + if (bzerr != BZ_OK) goto errhandler; + + if (verbosity >= 2) fprintf ( stderr, "\n" ); + + while (True) { + + if (myfeof(stream)) break; + nIbuf = fread ( ibuf, sizeof(UChar), 5000, stream ); + if (ferror(stream)) goto errhandler_io; + if (nIbuf > 0) BZ2_bzWrite ( &bzerr, bzf, (void*)ibuf, nIbuf ); + if (bzerr != BZ_OK) goto errhandler; + + } + + BZ2_bzWriteClose64 ( &bzerr, bzf, 0, + &nbytes_in_lo32, &nbytes_in_hi32, + &nbytes_out_lo32, &nbytes_out_hi32 ); + if (bzerr != BZ_OK) goto errhandler; + + if (ferror(zStream)) goto errhandler_io; + ret = fflush ( zStream ); + if (ret == EOF) goto errhandler_io; + if (zStream != stdout) { + Int32 fd = fileno ( zStream ); + if (fd < 0) goto errhandler_io; + applySavedFileAttrToOutputFile ( fd ); + ret = fclose ( zStream ); + outputHandleJustInCase = NULL; + if (ret == EOF) goto errhandler_io; + } + outputHandleJustInCase = NULL; + if (ferror(stream)) goto errhandler_io; + ret = fclose ( stream ); + if (ret == EOF) goto errhandler_io; + + if (verbosity >= 1) { + if (nbytes_in_lo32 == 0 && nbytes_in_hi32 == 0) { + fprintf ( stderr, " no data compressed.\n"); + } else { + Char buf_nin[32], buf_nout[32]; + UInt64 nbytes_in, nbytes_out; + double nbytes_in_d, nbytes_out_d; + uInt64_from_UInt32s ( &nbytes_in, + nbytes_in_lo32, nbytes_in_hi32 ); + uInt64_from_UInt32s ( &nbytes_out, + nbytes_out_lo32, nbytes_out_hi32 ); + nbytes_in_d = uInt64_to_double ( &nbytes_in ); + nbytes_out_d = uInt64_to_double ( &nbytes_out ); + uInt64_toAscii ( buf_nin, &nbytes_in ); + uInt64_toAscii ( buf_nout, &nbytes_out ); + fprintf ( stderr, "%6.3f:1, %6.3f bits/byte, " + "%5.2f%% saved, %s in, %s out.\n", + nbytes_in_d / nbytes_out_d, + (8.0 * nbytes_out_d) / nbytes_in_d, + 100.0 * (1.0 - nbytes_out_d / nbytes_in_d), + buf_nin, + buf_nout + ); + } + } + + return; + + errhandler: + BZ2_bzWriteClose64 ( &bzerr_dummy, bzf, 1, + &nbytes_in_lo32, &nbytes_in_hi32, + &nbytes_out_lo32, &nbytes_out_hi32 ); + switch (bzerr) { + case BZ_CONFIG_ERROR: + configError(); break; + case BZ_MEM_ERROR: + outOfMemory (); break; + case BZ_IO_ERROR: + errhandler_io: + ioError(); break; + default: + panic ( "compress:unexpected error" ); + } + + panic ( "compress:end" ); + /*notreached*/ +} + + + +/*---------------------------------------------*/ +static +Bool uncompressStream ( FILE *zStream, FILE *stream ) +{ + BZFILE* bzf = NULL; + Int32 bzerr, bzerr_dummy, ret, nread, streamNo, i; + UChar obuf[5000]; + UChar unused[BZ_MAX_UNUSED]; + Int32 nUnused; + void* unusedTmpV; + UChar* unusedTmp; + + nUnused = 0; + streamNo = 0; + + SET_BINARY_MODE(stream); + SET_BINARY_MODE(zStream); + + if (ferror(stream)) goto errhandler_io; + if (ferror(zStream)) goto errhandler_io; + + while (True) { + + bzf = BZ2_bzReadOpen ( + &bzerr, zStream, verbosity, + (int)smallMode, unused, nUnused + ); + if (bzf == NULL || bzerr != BZ_OK) goto errhandler; + streamNo++; + + while (bzerr == BZ_OK) { + nread = BZ2_bzRead ( &bzerr, bzf, obuf, 5000 ); + if (bzerr == BZ_DATA_ERROR_MAGIC) goto trycat; + if ((bzerr == BZ_OK || bzerr == BZ_STREAM_END) && nread > 0) + fwrite ( obuf, sizeof(UChar), nread, stream ); + if (ferror(stream)) goto errhandler_io; + } + if (bzerr != BZ_STREAM_END) goto errhandler; + + BZ2_bzReadGetUnused ( &bzerr, bzf, &unusedTmpV, &nUnused ); + if (bzerr != BZ_OK) panic ( "decompress:bzReadGetUnused" ); + + unusedTmp = (UChar*)unusedTmpV; + for (i = 0; i < nUnused; i++) unused[i] = unusedTmp[i]; + + BZ2_bzReadClose ( &bzerr, bzf ); + if (bzerr != BZ_OK) panic ( "decompress:bzReadGetUnused" ); + + if (nUnused == 0 && myfeof(zStream)) break; + } + + closeok: + if (ferror(zStream)) goto errhandler_io; + if (stream != stdout) { + Int32 fd = fileno ( stream ); + if (fd < 0) goto errhandler_io; + applySavedFileAttrToOutputFile ( fd ); + } + ret = fclose ( zStream ); + if (ret == EOF) goto errhandler_io; + + if (ferror(stream)) goto errhandler_io; + ret = fflush ( stream ); + if (ret != 0) goto errhandler_io; + if (stream != stdout) { + ret = fclose ( stream ); + outputHandleJustInCase = NULL; + if (ret == EOF) goto errhandler_io; + } + outputHandleJustInCase = NULL; + if (verbosity >= 2) fprintf ( stderr, "\n " ); + return True; + + trycat: + if (forceOverwrite) { + rewind(zStream); + while (True) { + if (myfeof(zStream)) break; + nread = fread ( obuf, sizeof(UChar), 5000, zStream ); + if (ferror(zStream)) goto errhandler_io; + if (nread > 0) fwrite ( obuf, sizeof(UChar), nread, stream ); + if (ferror(stream)) goto errhandler_io; + } + goto closeok; + } + + errhandler: + BZ2_bzReadClose ( &bzerr_dummy, bzf ); + switch (bzerr) { + case BZ_CONFIG_ERROR: + configError(); break; + case BZ_IO_ERROR: + errhandler_io: + ioError(); break; + case BZ_DATA_ERROR: + crcError(); + case BZ_MEM_ERROR: + outOfMemory(); + case BZ_UNEXPECTED_EOF: + compressedStreamEOF(); + case BZ_DATA_ERROR_MAGIC: + if (zStream != stdin) fclose(zStream); + if (stream != stdout) fclose(stream); + if (streamNo == 1) { + return False; + } else { + if (noisy) + fprintf ( stderr, + "\n%s: %s: trailing garbage after EOF ignored\n", + progName, inName ); + return True; + } + default: + panic ( "decompress:unexpected error" ); + } + + panic ( "decompress:end" ); + return True; /*notreached*/ +} + + +/*---------------------------------------------*/ +static +Bool testStream ( FILE *zStream ) +{ + BZFILE* bzf = NULL; + Int32 bzerr, bzerr_dummy, ret, nread, streamNo, i; + UChar obuf[5000]; + UChar unused[BZ_MAX_UNUSED]; + Int32 nUnused; + void* unusedTmpV; + UChar* unusedTmp; + + nUnused = 0; + streamNo = 0; + + SET_BINARY_MODE(zStream); + if (ferror(zStream)) goto errhandler_io; + + while (True) { + + bzf = BZ2_bzReadOpen ( + &bzerr, zStream, verbosity, + (int)smallMode, unused, nUnused + ); + if (bzf == NULL || bzerr != BZ_OK) goto errhandler; + streamNo++; + + while (bzerr == BZ_OK) { + nread = BZ2_bzRead ( &bzerr, bzf, obuf, 5000 ); + if (bzerr == BZ_DATA_ERROR_MAGIC) goto errhandler; + } + if (bzerr != BZ_STREAM_END) goto errhandler; + + BZ2_bzReadGetUnused ( &bzerr, bzf, &unusedTmpV, &nUnused ); + if (bzerr != BZ_OK) panic ( "test:bzReadGetUnused" ); + + unusedTmp = (UChar*)unusedTmpV; + for (i = 0; i < nUnused; i++) unused[i] = unusedTmp[i]; + + BZ2_bzReadClose ( &bzerr, bzf ); + if (bzerr != BZ_OK) panic ( "test:bzReadGetUnused" ); + if (nUnused == 0 && myfeof(zStream)) break; + + } + + if (ferror(zStream)) goto errhandler_io; + ret = fclose ( zStream ); + if (ret == EOF) goto errhandler_io; + + if (verbosity >= 2) fprintf ( stderr, "\n " ); + return True; + + errhandler: + BZ2_bzReadClose ( &bzerr_dummy, bzf ); + if (verbosity == 0) + fprintf ( stderr, "%s: %s: ", progName, inName ); + switch (bzerr) { + case BZ_CONFIG_ERROR: + configError(); break; + case BZ_IO_ERROR: + errhandler_io: + ioError(); break; + case BZ_DATA_ERROR: + fprintf ( stderr, + "data integrity (CRC) error in data\n" ); + return False; + case BZ_MEM_ERROR: + outOfMemory(); + case BZ_UNEXPECTED_EOF: + fprintf ( stderr, + "file ends unexpectedly\n" ); + return False; + case BZ_DATA_ERROR_MAGIC: + if (zStream != stdin) fclose(zStream); + if (streamNo == 1) { + fprintf ( stderr, + "bad magic number (file not created by bzip2)\n" ); + return False; + } else { + if (noisy) + fprintf ( stderr, + "trailing garbage after EOF ignored\n" ); + return True; + } + default: + panic ( "test:unexpected error" ); + } + + panic ( "test:end" ); + return True; /*notreached*/ +} + + +/*---------------------------------------------------*/ +/*--- Error [non-] handling grunge ---*/ +/*---------------------------------------------------*/ + +/*---------------------------------------------*/ +static +void setExit ( Int32 v ) +{ + if (v > exitValue) exitValue = v; +} + + +/*---------------------------------------------*/ +static +void cadvise ( void ) +{ + if (noisy) + fprintf ( + stderr, + "\nIt is possible that the compressed file(s) have become corrupted.\n" + "You can use the -tvv option to test integrity of such files.\n\n" + "You can use the `bzip2recover' program to attempt to recover\n" + "data from undamaged sections of corrupted files.\n\n" + ); +} + + +/*---------------------------------------------*/ +static +void showFileNames ( void ) +{ + if (noisy) + fprintf ( + stderr, + "\tInput file = %s, output file = %s\n", + inName, outName + ); +} + + +/*---------------------------------------------*/ +static +void cleanUpAndFail ( Int32 ec ) +{ + IntNative retVal; + struct MY_STAT statBuf; + + if ( srcMode == SM_F2F + && opMode != OM_TEST + && deleteOutputOnInterrupt ) { + + /* Check whether input file still exists. Delete output file + only if input exists to avoid loss of data. Joerg Prante, 5 + January 2002. (JRS 06-Jan-2002: other changes in 1.0.2 mean + this is less likely to happen. But to be ultra-paranoid, we + do the check anyway.) */ + retVal = MY_STAT ( inName, &statBuf ); + if (retVal == 0) { + if (noisy) + fprintf ( stderr, + "%s: Deleting output file %s, if it exists.\n", + progName, outName ); + if (outputHandleJustInCase != NULL) + fclose ( outputHandleJustInCase ); + retVal = remove ( outName ); + if (retVal != 0) + fprintf ( stderr, + "%s: WARNING: deletion of output file " + "(apparently) failed.\n", + progName ); + } else { + fprintf ( stderr, + "%s: WARNING: deletion of output file suppressed\n", + progName ); + fprintf ( stderr, + "%s: since input file no longer exists. Output file\n", + progName ); + fprintf ( stderr, + "%s: `%s' may be incomplete.\n", + progName, outName ); + fprintf ( stderr, + "%s: I suggest doing an integrity test (bzip2 -tv)" + " of it.\n", + progName ); + } + } + + if (noisy && numFileNames > 0 && numFilesProcessed < numFileNames) { + fprintf ( stderr, + "%s: WARNING: some files have not been processed:\n" + "%s: %d specified on command line, %d not processed yet.\n\n", + progName, progName, + numFileNames, numFileNames - numFilesProcessed ); + } + setExit(ec); + exit(exitValue); +} + + +/*---------------------------------------------*/ +static +void panic ( const Char* s ) +{ + fprintf ( stderr, + "\n%s: PANIC -- internal consistency error:\n" + "\t%s\n" + "\tThis is a BUG. Please report it to me at:\n" + "\tjseward@bzip.org\n", + progName, s ); + showFileNames(); + cleanUpAndFail( 3 ); +} + + +/*---------------------------------------------*/ +static +void crcError ( void ) +{ + fprintf ( stderr, + "\n%s: Data integrity error when decompressing.\n", + progName ); + showFileNames(); + cadvise(); + cleanUpAndFail( 2 ); +} + + +/*---------------------------------------------*/ +static +void compressedStreamEOF ( void ) +{ + if (noisy) { + fprintf ( stderr, + "\n%s: Compressed file ends unexpectedly;\n\t" + "perhaps it is corrupted? *Possible* reason follows.\n", + progName ); + perror ( progName ); + showFileNames(); + cadvise(); + } + cleanUpAndFail( 2 ); +} + + +/*---------------------------------------------*/ +static +void ioError ( void ) +{ + fprintf ( stderr, + "\n%s: I/O or other error, bailing out. " + "Possible reason follows.\n", + progName ); + perror ( progName ); + showFileNames(); + cleanUpAndFail( 1 ); +} + + +/*---------------------------------------------*/ +static +void mySignalCatcher ( IntNative n ) +{ + fprintf ( stderr, + "\n%s: Control-C or similar caught, quitting.\n", + progName ); + cleanUpAndFail(1); +} + + +/*---------------------------------------------*/ +static +void mySIGSEGVorSIGBUScatcher ( IntNative n ) +{ + if (opMode == OM_Z) + fprintf ( + stderr, + "\n%s: Caught a SIGSEGV or SIGBUS whilst compressing.\n" + "\n" + " Possible causes are (most likely first):\n" + " (1) This computer has unreliable memory or cache hardware\n" + " (a surprisingly common problem; try a different machine.)\n" + " (2) A bug in the compiler used to create this executable\n" + " (unlikely, if you didn't compile bzip2 yourself.)\n" + " (3) A real bug in bzip2 -- I hope this should never be the case.\n" + " The user's manual, Section 4.3, has more info on (1) and (2).\n" + " \n" + " If you suspect this is a bug in bzip2, or are unsure about (1)\n" + " or (2), feel free to report it to me at: jseward@bzip.org.\n" + " Section 4.3 of the user's manual describes the info a useful\n" + " bug report should have. If the manual is available on your\n" + " system, please try and read it before mailing me. If you don't\n" + " have the manual or can't be bothered to read it, mail me anyway.\n" + "\n", + progName ); + else + fprintf ( + stderr, + "\n%s: Caught a SIGSEGV or SIGBUS whilst decompressing.\n" + "\n" + " Possible causes are (most likely first):\n" + " (1) The compressed data is corrupted, and bzip2's usual checks\n" + " failed to detect this. Try bzip2 -tvv my_file.bz2.\n" + " (2) This computer has unreliable memory or cache hardware\n" + " (a surprisingly common problem; try a different machine.)\n" + " (3) A bug in the compiler used to create this executable\n" + " (unlikely, if you didn't compile bzip2 yourself.)\n" + " (4) A real bug in bzip2 -- I hope this should never be the case.\n" + " The user's manual, Section 4.3, has more info on (2) and (3).\n" + " \n" + " If you suspect this is a bug in bzip2, or are unsure about (2)\n" + " or (3), feel free to report it to me at: jseward@bzip.org.\n" + " Section 4.3 of the user's manual describes the info a useful\n" + " bug report should have. If the manual is available on your\n" + " system, please try and read it before mailing me. If you don't\n" + " have the manual or can't be bothered to read it, mail me anyway.\n" + "\n", + progName ); + + showFileNames(); + if (opMode == OM_Z) + cleanUpAndFail( 3 ); else + { cadvise(); cleanUpAndFail( 2 ); } +} + + +/*---------------------------------------------*/ +static +void outOfMemory ( void ) +{ + fprintf ( stderr, + "\n%s: couldn't allocate enough memory\n", + progName ); + showFileNames(); + cleanUpAndFail(1); +} + + +/*---------------------------------------------*/ +static +void configError ( void ) +{ + fprintf ( stderr, + "bzip2: I'm not configured correctly for this platform!\n" + "\tI require Int32, Int16 and Char to have sizes\n" + "\tof 4, 2 and 1 bytes to run properly, and they don't.\n" + "\tProbably you can fix this by defining them correctly,\n" + "\tand recompiling. Bye!\n" ); + setExit(3); + exit(exitValue); +} + + +/*---------------------------------------------------*/ +/*--- The main driver machinery ---*/ +/*---------------------------------------------------*/ + +/* All rather crufty. The main problem is that input files + are stat()d multiple times before use. This should be + cleaned up. +*/ + +/*---------------------------------------------*/ +static +void pad ( Char *s ) +{ + Int32 i; + if ( (Int32)strlen(s) >= longestFileName ) return; + for (i = 1; i <= longestFileName - (Int32)strlen(s); i++) + fprintf ( stderr, " " ); +} + + +/*---------------------------------------------*/ +static +void copyFileName ( Char* to, Char* from ) +{ + if ( strlen(from) > FILE_NAME_LEN-10 ) { + fprintf ( + stderr, + "bzip2: file name\n`%s'\n" + "is suspiciously (more than %d chars) long.\n" + "Try using a reasonable file name instead. Sorry! :-)\n", + from, FILE_NAME_LEN-10 + ); + setExit(1); + exit(exitValue); + } + + strncpy(to,from,FILE_NAME_LEN-10); + to[FILE_NAME_LEN-10]='\0'; +} + + +/*---------------------------------------------*/ +static +Bool fileExists ( Char* name ) +{ + FILE *tmp = fopen ( name, "rb" ); + Bool exists = (tmp != NULL); + if (tmp != NULL) fclose ( tmp ); + return exists; +} + + +/*---------------------------------------------*/ +/* Open an output file safely with O_EXCL and good permissions. + This avoids a race condition in versions < 1.0.2, in which + the file was first opened and then had its interim permissions + set safely. We instead use open() to create the file with + the interim permissions required. (--- --- rw-). + + For non-Unix platforms, if we are not worrying about + security issues, simple this simply behaves like fopen. +*/ +static +FILE* fopen_output_safely ( Char* name, const char* mode ) +{ +# if BZ_UNIX + FILE* fp; + IntNative fh; + fh = open(name, O_WRONLY|O_CREAT|O_EXCL, S_IWUSR|S_IRUSR); + if (fh == -1) return NULL; + fp = fdopen(fh, mode); + if (fp == NULL) close(fh); + return fp; +# else + return fopen(name, mode); +# endif +} + + +/*---------------------------------------------*/ +/*-- + if in doubt, return True +--*/ +static +Bool notAStandardFile ( Char* name ) +{ + IntNative i; + struct MY_STAT statBuf; + + i = MY_LSTAT ( name, &statBuf ); + if (i != 0) return True; + if (MY_S_ISREG(statBuf.st_mode)) return False; + return True; +} + + +/*---------------------------------------------*/ +/*-- + rac 11/21/98 see if file has hard links to it +--*/ +static +Int32 countHardLinks ( Char* name ) +{ + IntNative i; + struct MY_STAT statBuf; + + i = MY_LSTAT ( name, &statBuf ); + if (i != 0) return 0; + return (statBuf.st_nlink - 1); +} + + +/*---------------------------------------------*/ +/* Copy modification date, access date, permissions and owner from the + source to destination file. We have to copy this meta-info off + into fileMetaInfo before starting to compress / decompress it, + because doing it afterwards means we get the wrong access time. + + To complicate matters, in compress() and decompress() below, the + sequence of tests preceding the call to saveInputFileMetaInfo() + involves calling fileExists(), which in turn establishes its result + by attempting to fopen() the file, and if successful, immediately + fclose()ing it again. So we have to assume that the fopen() call + does not cause the access time field to be updated. + + Reading of the man page for stat() (man 2 stat) on RedHat 7.2 seems + to imply that merely doing open() will not affect the access time. + Therefore we merely need to hope that the C library only does + open() as a result of fopen(), and not any kind of read()-ahead + cleverness. + + It sounds pretty fragile to me. Whether this carries across + robustly to arbitrary Unix-like platforms (or even works robustly + on this one, RedHat 7.2) is unknown to me. Nevertheless ... +*/ +#if BZ_UNIX +static +struct MY_STAT fileMetaInfo; +#endif + +static +void saveInputFileMetaInfo ( Char *srcName ) +{ +# if BZ_UNIX + IntNative retVal; + /* Note use of stat here, not lstat. */ + retVal = MY_STAT( srcName, &fileMetaInfo ); + ERROR_IF_NOT_ZERO ( retVal ); +# endif +} + + +static +void applySavedTimeInfoToOutputFile ( Char *dstName ) +{ +# if BZ_UNIX + IntNative retVal; + struct utimbuf uTimBuf; + + uTimBuf.actime = fileMetaInfo.st_atime; + uTimBuf.modtime = fileMetaInfo.st_mtime; + + retVal = utime ( dstName, &uTimBuf ); + ERROR_IF_NOT_ZERO ( retVal ); +# endif +} + +static +void applySavedFileAttrToOutputFile ( IntNative fd ) +{ +# if BZ_UNIX + IntNative retVal; + + retVal = fchmod ( fd, fileMetaInfo.st_mode ); + ERROR_IF_NOT_ZERO ( retVal ); + + (void) fchown ( fd, fileMetaInfo.st_uid, fileMetaInfo.st_gid ); + /* chown() will in many cases return with EPERM, which can + be safely ignored. + */ +# endif +} + + +/*---------------------------------------------*/ +static +Bool containsDubiousChars ( Char* name ) +{ +# if BZ_UNIX + /* On unix, files can contain any characters and the file expansion + * is performed by the shell. + */ + return False; +# else /* ! BZ_UNIX */ + /* On non-unix (Win* platforms), wildcard characters are not allowed in + * filenames. + */ + for (; *name != '\0'; name++) + if (*name == '?' || *name == '*') return True; + return False; +# endif /* BZ_UNIX */ +} + + +/*---------------------------------------------*/ +#define BZ_N_SUFFIX_PAIRS 4 + +const Char* zSuffix[BZ_N_SUFFIX_PAIRS] + = { ".bz2", ".bz", ".tbz2", ".tbz" }; +const Char* unzSuffix[BZ_N_SUFFIX_PAIRS] + = { "", "", ".tar", ".tar" }; + +static +Bool hasSuffix ( Char* s, const Char* suffix ) +{ + Int32 ns = strlen(s); + Int32 nx = strlen(suffix); + if (ns < nx) return False; + if (strcmp(s + ns - nx, suffix) == 0) return True; + return False; +} + +static +Bool mapSuffix ( Char* name, + const Char* oldSuffix, + const Char* newSuffix ) +{ + if (!hasSuffix(name,oldSuffix)) return False; + name[strlen(name)-strlen(oldSuffix)] = 0; + strcat ( name, newSuffix ); + return True; +} + + +/*---------------------------------------------*/ +static +void compress ( Char *name ) +{ + FILE *inStr; + FILE *outStr; + Int32 n, i; + struct MY_STAT statBuf; + + deleteOutputOnInterrupt = False; + + if (name == NULL && srcMode != SM_I2O) + panic ( "compress: bad modes\n" ); + + switch (srcMode) { + case SM_I2O: + copyFileName ( inName, (Char*)"(stdin)" ); + copyFileName ( outName, (Char*)"(stdout)" ); + break; + case SM_F2F: + copyFileName ( inName, name ); + copyFileName ( outName, name ); + strcat ( outName, ".bz2" ); + break; + case SM_F2O: + copyFileName ( inName, name ); + copyFileName ( outName, (Char*)"(stdout)" ); + break; + } + + if ( srcMode != SM_I2O && containsDubiousChars ( inName ) ) { + if (noisy) + fprintf ( stderr, "%s: There are no files matching `%s'.\n", + progName, inName ); + setExit(1); + return; + } + if ( srcMode != SM_I2O && !fileExists ( inName ) ) { + fprintf ( stderr, "%s: Can't open input file %s: %s.\n", + progName, inName, strerror(errno) ); + setExit(1); + return; + } + for (i = 0; i < BZ_N_SUFFIX_PAIRS; i++) { + if (hasSuffix(inName, zSuffix[i])) { + if (noisy) + fprintf ( stderr, + "%s: Input file %s already has %s suffix.\n", + progName, inName, zSuffix[i] ); + setExit(1); + return; + } + } + if ( srcMode == SM_F2F || srcMode == SM_F2O ) { + MY_STAT(inName, &statBuf); + if ( MY_S_ISDIR(statBuf.st_mode) ) { + fprintf( stderr, + "%s: Input file %s is a directory.\n", + progName,inName); + setExit(1); + return; + } + } + if ( srcMode == SM_F2F && !forceOverwrite && notAStandardFile ( inName )) { + if (noisy) + fprintf ( stderr, "%s: Input file %s is not a normal file.\n", + progName, inName ); + setExit(1); + return; + } + if ( srcMode == SM_F2F && fileExists ( outName ) ) { + if (forceOverwrite) { + remove(outName); + } else { + fprintf ( stderr, "%s: Output file %s already exists.\n", + progName, outName ); + setExit(1); + return; + } + } + if ( srcMode == SM_F2F && !forceOverwrite && + (n=countHardLinks ( inName )) > 0) { + fprintf ( stderr, "%s: Input file %s has %d other link%s.\n", + progName, inName, n, n > 1 ? "s" : "" ); + setExit(1); + return; + } + + if ( srcMode == SM_F2F ) { + /* Save the file's meta-info before we open it. Doing it later + means we mess up the access times. */ + saveInputFileMetaInfo ( inName ); + } + + switch ( srcMode ) { + + case SM_I2O: + inStr = stdin; + outStr = stdout; + if ( isatty ( fileno ( stdout ) ) ) { + fprintf ( stderr, + "%s: I won't write compressed data to a terminal.\n", + progName ); + fprintf ( stderr, "%s: For help, type: `%s --help'.\n", + progName, progName ); + setExit(1); + return; + }; + break; + + case SM_F2O: + inStr = fopen ( inName, "rb" ); + outStr = stdout; + if ( isatty ( fileno ( stdout ) ) ) { + fprintf ( stderr, + "%s: I won't write compressed data to a terminal.\n", + progName ); + fprintf ( stderr, "%s: For help, type: `%s --help'.\n", + progName, progName ); + if ( inStr != NULL ) fclose ( inStr ); + setExit(1); + return; + }; + if ( inStr == NULL ) { + fprintf ( stderr, "%s: Can't open input file %s: %s.\n", + progName, inName, strerror(errno) ); + setExit(1); + return; + }; + break; + + case SM_F2F: + inStr = fopen ( inName, "rb" ); + outStr = fopen_output_safely ( outName, "wb" ); + if ( outStr == NULL) { + fprintf ( stderr, "%s: Can't create output file %s: %s.\n", + progName, outName, strerror(errno) ); + if ( inStr != NULL ) fclose ( inStr ); + setExit(1); + return; + } + if ( inStr == NULL ) { + fprintf ( stderr, "%s: Can't open input file %s: %s.\n", + progName, inName, strerror(errno) ); + if ( outStr != NULL ) fclose ( outStr ); + setExit(1); + return; + }; + break; + + default: + panic ( "compress: bad srcMode" ); + break; + } + + if (verbosity >= 1) { + fprintf ( stderr, " %s: ", inName ); + pad ( inName ); + fflush ( stderr ); + } + + /*--- Now the input and output handles are sane. Do the Biz. ---*/ + outputHandleJustInCase = outStr; + deleteOutputOnInterrupt = True; + compressStream ( inStr, outStr ); + outputHandleJustInCase = NULL; + + /*--- If there was an I/O error, we won't get here. ---*/ + if ( srcMode == SM_F2F ) { + applySavedTimeInfoToOutputFile ( outName ); + deleteOutputOnInterrupt = False; + if ( !keepInputFiles ) { + IntNative retVal = remove ( inName ); + ERROR_IF_NOT_ZERO ( retVal ); + } + } + + deleteOutputOnInterrupt = False; +} + + +/*---------------------------------------------*/ +static +void uncompress ( Char *name ) +{ + FILE *inStr; + FILE *outStr; + Int32 n, i; + Bool magicNumberOK; + Bool cantGuess; + struct MY_STAT statBuf; + + deleteOutputOnInterrupt = False; + + if (name == NULL && srcMode != SM_I2O) + panic ( "uncompress: bad modes\n" ); + + cantGuess = False; + switch (srcMode) { + case SM_I2O: + copyFileName ( inName, (Char*)"(stdin)" ); + copyFileName ( outName, (Char*)"(stdout)" ); + break; + case SM_F2F: + copyFileName ( inName, name ); + copyFileName ( outName, name ); + for (i = 0; i < BZ_N_SUFFIX_PAIRS; i++) + if (mapSuffix(outName,zSuffix[i],unzSuffix[i])) + goto zzz; + cantGuess = True; + strcat ( outName, ".out" ); + break; + case SM_F2O: + copyFileName ( inName, name ); + copyFileName ( outName, (Char*)"(stdout)" ); + break; + } + + zzz: + if ( srcMode != SM_I2O && containsDubiousChars ( inName ) ) { + if (noisy) + fprintf ( stderr, "%s: There are no files matching `%s'.\n", + progName, inName ); + setExit(1); + return; + } + if ( srcMode != SM_I2O && !fileExists ( inName ) ) { + fprintf ( stderr, "%s: Can't open input file %s: %s.\n", + progName, inName, strerror(errno) ); + setExit(1); + return; + } + if ( srcMode == SM_F2F || srcMode == SM_F2O ) { + MY_STAT(inName, &statBuf); + if ( MY_S_ISDIR(statBuf.st_mode) ) { + fprintf( stderr, + "%s: Input file %s is a directory.\n", + progName,inName); + setExit(1); + return; + } + } + if ( srcMode == SM_F2F && !forceOverwrite && notAStandardFile ( inName )) { + if (noisy) + fprintf ( stderr, "%s: Input file %s is not a normal file.\n", + progName, inName ); + setExit(1); + return; + } + if ( /* srcMode == SM_F2F implied && */ cantGuess ) { + if (noisy) + fprintf ( stderr, + "%s: Can't guess original name for %s -- using %s\n", + progName, inName, outName ); + /* just a warning, no return */ + } + if ( srcMode == SM_F2F && fileExists ( outName ) ) { + if (forceOverwrite) { + remove(outName); + } else { + fprintf ( stderr, "%s: Output file %s already exists.\n", + progName, outName ); + setExit(1); + return; + } + } + if ( srcMode == SM_F2F && !forceOverwrite && + (n=countHardLinks ( inName ) ) > 0) { + fprintf ( stderr, "%s: Input file %s has %d other link%s.\n", + progName, inName, n, n > 1 ? "s" : "" ); + setExit(1); + return; + } + + if ( srcMode == SM_F2F ) { + /* Save the file's meta-info before we open it. Doing it later + means we mess up the access times. */ + saveInputFileMetaInfo ( inName ); + } + + switch ( srcMode ) { + + case SM_I2O: + inStr = stdin; + outStr = stdout; + if ( isatty ( fileno ( stdin ) ) ) { + fprintf ( stderr, + "%s: I won't read compressed data from a terminal.\n", + progName ); + fprintf ( stderr, "%s: For help, type: `%s --help'.\n", + progName, progName ); + setExit(1); + return; + }; + break; + + case SM_F2O: + inStr = fopen ( inName, "rb" ); + outStr = stdout; + if ( inStr == NULL ) { + fprintf ( stderr, "%s: Can't open input file %s:%s.\n", + progName, inName, strerror(errno) ); + if ( inStr != NULL ) fclose ( inStr ); + setExit(1); + return; + }; + break; + + case SM_F2F: + inStr = fopen ( inName, "rb" ); + outStr = fopen_output_safely ( outName, "wb" ); + if ( outStr == NULL) { + fprintf ( stderr, "%s: Can't create output file %s: %s.\n", + progName, outName, strerror(errno) ); + if ( inStr != NULL ) fclose ( inStr ); + setExit(1); + return; + } + if ( inStr == NULL ) { + fprintf ( stderr, "%s: Can't open input file %s: %s.\n", + progName, inName, strerror(errno) ); + if ( outStr != NULL ) fclose ( outStr ); + setExit(1); + return; + }; + break; + + default: + panic ( "uncompress: bad srcMode" ); + break; + } + + if (verbosity >= 1) { + fprintf ( stderr, " %s: ", inName ); + pad ( inName ); + fflush ( stderr ); + } + + /*--- Now the input and output handles are sane. Do the Biz. ---*/ + outputHandleJustInCase = outStr; + deleteOutputOnInterrupt = True; + magicNumberOK = uncompressStream ( inStr, outStr ); + outputHandleJustInCase = NULL; + + /*--- If there was an I/O error, we won't get here. ---*/ + if ( magicNumberOK ) { + if ( srcMode == SM_F2F ) { + applySavedTimeInfoToOutputFile ( outName ); + deleteOutputOnInterrupt = False; + if ( !keepInputFiles ) { + IntNative retVal = remove ( inName ); + ERROR_IF_NOT_ZERO ( retVal ); + } + } + } else { + unzFailsExist = True; + deleteOutputOnInterrupt = False; + if ( srcMode == SM_F2F ) { + IntNative retVal = remove ( outName ); + ERROR_IF_NOT_ZERO ( retVal ); + } + } + deleteOutputOnInterrupt = False; + + if ( magicNumberOK ) { + if (verbosity >= 1) + fprintf ( stderr, "done\n" ); + } else { + setExit(2); + if (verbosity >= 1) + fprintf ( stderr, "not a bzip2 file.\n" ); else + fprintf ( stderr, + "%s: %s is not a bzip2 file.\n", + progName, inName ); + } + +} + + +/*---------------------------------------------*/ +static +void testf ( Char *name ) +{ + FILE *inStr; + Bool allOK; + struct MY_STAT statBuf; + + deleteOutputOnInterrupt = False; + + if (name == NULL && srcMode != SM_I2O) + panic ( "testf: bad modes\n" ); + + copyFileName ( outName, (Char*)"(none)" ); + switch (srcMode) { + case SM_I2O: copyFileName ( inName, (Char*)"(stdin)" ); break; + case SM_F2F: copyFileName ( inName, name ); break; + case SM_F2O: copyFileName ( inName, name ); break; + } + + if ( srcMode != SM_I2O && containsDubiousChars ( inName ) ) { + if (noisy) + fprintf ( stderr, "%s: There are no files matching `%s'.\n", + progName, inName ); + setExit(1); + return; + } + if ( srcMode != SM_I2O && !fileExists ( inName ) ) { + fprintf ( stderr, "%s: Can't open input %s: %s.\n", + progName, inName, strerror(errno) ); + setExit(1); + return; + } + if ( srcMode != SM_I2O ) { + MY_STAT(inName, &statBuf); + if ( MY_S_ISDIR(statBuf.st_mode) ) { + fprintf( stderr, + "%s: Input file %s is a directory.\n", + progName,inName); + setExit(1); + return; + } + } + + switch ( srcMode ) { + + case SM_I2O: + if ( isatty ( fileno ( stdin ) ) ) { + fprintf ( stderr, + "%s: I won't read compressed data from a terminal.\n", + progName ); + fprintf ( stderr, "%s: For help, type: `%s --help'.\n", + progName, progName ); + setExit(1); + return; + }; + inStr = stdin; + break; + + case SM_F2O: case SM_F2F: + inStr = fopen ( inName, "rb" ); + if ( inStr == NULL ) { + fprintf ( stderr, "%s: Can't open input file %s:%s.\n", + progName, inName, strerror(errno) ); + setExit(1); + return; + }; + break; + + default: + panic ( "testf: bad srcMode" ); + break; + } + + if (verbosity >= 1) { + fprintf ( stderr, " %s: ", inName ); + pad ( inName ); + fflush ( stderr ); + } + + /*--- Now the input handle is sane. Do the Biz. ---*/ + outputHandleJustInCase = NULL; + allOK = testStream ( inStr ); + + if (allOK && verbosity >= 1) fprintf ( stderr, "ok\n" ); + if (!allOK) testFailsExist = True; +} + + +/*---------------------------------------------*/ +static +void license ( void ) +{ + fprintf ( stderr, + + "bzip2, a block-sorting file compressor. " + "Version %s.\n" + " \n" + " Copyright (C) 1996-2007 by Julian Seward.\n" + " \n" + " This program is free software; you can redistribute it and/or modify\n" + " it under the terms set out in the LICENSE file, which is included\n" + " in the bzip2-1.0.5 source distribution.\n" + " \n" + " This program is distributed in the hope that it will be useful,\n" + " but WITHOUT ANY WARRANTY; without even the implied warranty of\n" + " MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the\n" + " LICENSE file for more details.\n" + " \n", + BZ2_bzlibVersion() + ); +} + + +/*---------------------------------------------*/ +static +void usage ( Char *fullProgName ) +{ + fprintf ( + stderr, + "bzip2, a block-sorting file compressor. " + "Version %s.\n" + "\n usage: %s [flags and input files in any order]\n" + "\n" + " -h --help print this message\n" + " -d --decompress force decompression\n" + " -z --compress force compression\n" + " -k --keep keep (don't delete) input files\n" + " -f --force overwrite existing output files\n" + " -t --test test compressed file integrity\n" + " -c --stdout output to standard out\n" + " -q --quiet suppress noncritical error messages\n" + " -v --verbose be verbose (a 2nd -v gives more)\n" + " -L --license display software version & license\n" + " -V --version display software version & license\n" + " -s --small use less memory (at most 2500k)\n" + " -1 .. -9 set block size to 100k .. 900k\n" + " --fast alias for -1\n" + " --best alias for -9\n" + "\n" + " If invoked as `bzip2', default action is to compress.\n" + " as `bunzip2', default action is to decompress.\n" + " as `bzcat', default action is to decompress to stdout.\n" + "\n" + " If no file names are given, bzip2 compresses or decompresses\n" + " from standard input to standard output. You can combine\n" + " short flags, so `-v -4' means the same as -v4 or -4v, &c.\n" +# if BZ_UNIX + "\n" +# endif + , + + BZ2_bzlibVersion(), + fullProgName + ); +} + + +/*---------------------------------------------*/ +static +void redundant ( Char* flag ) +{ + fprintf ( + stderr, + "%s: %s is redundant in versions 0.9.5 and above\n", + progName, flag ); +} + + +/*---------------------------------------------*/ +/*-- + All the garbage from here to main() is purely to + implement a linked list of command-line arguments, + into which main() copies argv[1 .. argc-1]. + + The purpose of this exercise is to facilitate + the expansion of wildcard characters * and ? in + filenames for OSs which don't know how to do it + themselves, like MSDOS, Windows 95 and NT. + + The actual Dirty Work is done by the platform- + specific macro APPEND_FILESPEC. +--*/ + +typedef + struct zzzz { + Char *name; + struct zzzz *link; + } + Cell; + + +/*---------------------------------------------*/ +static +void *myMalloc ( Int32 n ) +{ + void* p; + + p = malloc ( (size_t)n ); + if (p == NULL) outOfMemory (); + return p; +} + + +/*---------------------------------------------*/ +static +Cell *mkCell ( void ) +{ + Cell *c; + + c = (Cell*) myMalloc ( sizeof ( Cell ) ); + c->name = NULL; + c->link = NULL; + return c; +} + + +/*---------------------------------------------*/ +static +Cell *snocString ( Cell *root, Char *name ) +{ + if (root == NULL) { + Cell *tmp = mkCell(); + tmp->name = (Char*) myMalloc ( 5 + strlen(name) ); + strcpy ( tmp->name, name ); + return tmp; + } else { + Cell *tmp = root; + while (tmp->link != NULL) tmp = tmp->link; + tmp->link = snocString ( tmp->link, name ); + return root; + } +} + + +/*---------------------------------------------*/ +static +void addFlagsFromEnvVar ( Cell** argList, Char* varName ) +{ + Int32 i, j, k; + Char *envbase, *p; + + envbase = getenv(varName); + if (envbase != NULL) { + p = envbase; + i = 0; + while (True) { + if (p[i] == 0) break; + p += i; + i = 0; + while (isspace((Int32)(p[0]))) p++; + while (p[i] != 0 && !isspace((Int32)(p[i]))) i++; + if (i > 0) { + k = i; if (k > FILE_NAME_LEN-10) k = FILE_NAME_LEN-10; + for (j = 0; j < k; j++) tmpName[j] = p[j]; + tmpName[k] = 0; + APPEND_FLAG(*argList, tmpName); + } + } + } +} + + +/*---------------------------------------------*/ +#define ISFLAG(s) (strcmp(aa->name, (s))==0) + +IntNative main ( IntNative argc, Char *argv[] ) +{ + Int32 i, j; + Char *tmp; + Cell *argList; + Cell *aa; + Bool decode; + + /*-- Be really really really paranoid :-) --*/ + if (sizeof(Int32) != 4 || sizeof(UInt32) != 4 || + sizeof(Int16) != 2 || sizeof(UInt16) != 2 || + sizeof(Char) != 1 || sizeof(UChar) != 1) + configError(); + + /*-- Initialise --*/ + outputHandleJustInCase = NULL; + smallMode = False; + keepInputFiles = False; + forceOverwrite = False; + noisy = True; + verbosity = 0; + blockSize100k = 9; + testFailsExist = False; + unzFailsExist = False; + numFileNames = 0; + numFilesProcessed = 0; + workFactor = 30; + deleteOutputOnInterrupt = False; + exitValue = 0; + i = j = 0; /* avoid bogus warning from egcs-1.1.X */ + + /*-- Set up signal handlers for mem access errors --*/ + signal (SIGSEGV, mySIGSEGVorSIGBUScatcher); +# if BZ_UNIX +# ifndef __DJGPP__ + signal (SIGBUS, mySIGSEGVorSIGBUScatcher); +# endif +# endif + + copyFileName ( inName, (Char*)"(none)" ); + copyFileName ( outName, (Char*)"(none)" ); + + copyFileName ( progNameReally, argv[0] ); + progName = &progNameReally[0]; + for (tmp = &progNameReally[0]; *tmp != '\0'; tmp++) + if (*tmp == PATH_SEP) progName = tmp + 1; + + + /*-- Copy flags from env var BZIP2, and + expand filename wildcards in arg list. + --*/ + argList = NULL; + addFlagsFromEnvVar ( &argList, (Char*)"BZIP2" ); + addFlagsFromEnvVar ( &argList, (Char*)"BZIP" ); + for (i = 1; i <= argc-1; i++) + APPEND_FILESPEC(argList, argv[i]); + + + /*-- Find the length of the longest filename --*/ + longestFileName = 7; + numFileNames = 0; + decode = True; + for (aa = argList; aa != NULL; aa = aa->link) { + if (ISFLAG("--")) { decode = False; continue; } + if (aa->name[0] == '-' && decode) continue; + numFileNames++; + if (longestFileName < (Int32)strlen(aa->name) ) + longestFileName = (Int32)strlen(aa->name); + } + + + /*-- Determine source modes; flag handling may change this too. --*/ + if (numFileNames == 0) + srcMode = SM_I2O; else srcMode = SM_F2F; + + + /*-- Determine what to do (compress/uncompress/test/cat). --*/ + /*-- Note that subsequent flag handling may change this. --*/ + opMode = OM_Z; + + if ( (strstr ( progName, "unzip" ) != 0) || + (strstr ( progName, "UNZIP" ) != 0) ) + opMode = OM_UNZ; + + if ( (strstr ( progName, "z2cat" ) != 0) || + (strstr ( progName, "Z2CAT" ) != 0) || + (strstr ( progName, "zcat" ) != 0) || + (strstr ( progName, "ZCAT" ) != 0) ) { + opMode = OM_UNZ; + srcMode = (numFileNames == 0) ? SM_I2O : SM_F2O; + } + + + /*-- Look at the flags. --*/ + for (aa = argList; aa != NULL; aa = aa->link) { + if (ISFLAG("--")) break; + if (aa->name[0] == '-' && aa->name[1] != '-') { + for (j = 1; aa->name[j] != '\0'; j++) { + switch (aa->name[j]) { + case 'c': srcMode = SM_F2O; break; + case 'd': opMode = OM_UNZ; break; + case 'z': opMode = OM_Z; break; + case 'f': forceOverwrite = True; break; + case 't': opMode = OM_TEST; break; + case 'k': keepInputFiles = True; break; + case 's': smallMode = True; break; + case 'q': noisy = False; break; + case '1': blockSize100k = 1; break; + case '2': blockSize100k = 2; break; + case '3': blockSize100k = 3; break; + case '4': blockSize100k = 4; break; + case '5': blockSize100k = 5; break; + case '6': blockSize100k = 6; break; + case '7': blockSize100k = 7; break; + case '8': blockSize100k = 8; break; + case '9': blockSize100k = 9; break; + case 'V': + case 'L': license(); break; + case 'v': verbosity++; break; + case 'h': usage ( progName ); + exit ( 0 ); + break; + default: fprintf ( stderr, "%s: Bad flag `%s'\n", + progName, aa->name ); + usage ( progName ); + exit ( 1 ); + break; + } + } + } + } + + /*-- And again ... --*/ + for (aa = argList; aa != NULL; aa = aa->link) { + if (ISFLAG("--")) break; + if (ISFLAG("--stdout")) srcMode = SM_F2O; else + if (ISFLAG("--decompress")) opMode = OM_UNZ; else + if (ISFLAG("--compress")) opMode = OM_Z; else + if (ISFLAG("--force")) forceOverwrite = True; else + if (ISFLAG("--test")) opMode = OM_TEST; else + if (ISFLAG("--keep")) keepInputFiles = True; else + if (ISFLAG("--small")) smallMode = True; else + if (ISFLAG("--quiet")) noisy = False; else + if (ISFLAG("--version")) license(); else + if (ISFLAG("--license")) license(); else + if (ISFLAG("--exponential")) workFactor = 1; else + if (ISFLAG("--repetitive-best")) redundant(aa->name); else + if (ISFLAG("--repetitive-fast")) redundant(aa->name); else + if (ISFLAG("--fast")) blockSize100k = 1; else + if (ISFLAG("--best")) blockSize100k = 9; else + if (ISFLAG("--verbose")) verbosity++; else + if (ISFLAG("--help")) { usage ( progName ); exit ( 0 ); } + else + if (strncmp ( aa->name, "--", 2) == 0) { + fprintf ( stderr, "%s: Bad flag `%s'\n", progName, aa->name ); + usage ( progName ); + exit ( 1 ); + } + } + + if (verbosity > 4) verbosity = 4; + if (opMode == OM_Z && smallMode && blockSize100k > 2) + blockSize100k = 2; + + if (opMode == OM_TEST && srcMode == SM_F2O) { + fprintf ( stderr, "%s: -c and -t cannot be used together.\n", + progName ); + exit ( 1 ); + } + + if (srcMode == SM_F2O && numFileNames == 0) + srcMode = SM_I2O; + + if (opMode != OM_Z) blockSize100k = 0; + + if (srcMode == SM_F2F) { + signal (SIGINT, mySignalCatcher); + signal (SIGTERM, mySignalCatcher); +# if BZ_UNIX + signal (SIGHUP, mySignalCatcher); +# endif + } + + if (opMode == OM_Z) { + if (srcMode == SM_I2O) { + compress ( NULL ); + } else { + decode = True; + for (aa = argList; aa != NULL; aa = aa->link) { + if (ISFLAG("--")) { decode = False; continue; } + if (aa->name[0] == '-' && decode) continue; + numFilesProcessed++; + compress ( aa->name ); + } + } + } + else + + if (opMode == OM_UNZ) { + unzFailsExist = False; + if (srcMode == SM_I2O) { + uncompress ( NULL ); + } else { + decode = True; + for (aa = argList; aa != NULL; aa = aa->link) { + if (ISFLAG("--")) { decode = False; continue; } + if (aa->name[0] == '-' && decode) continue; + numFilesProcessed++; + uncompress ( aa->name ); + } + } + if (unzFailsExist) { + setExit(2); + exit(exitValue); + } + } + + else { + testFailsExist = False; + if (srcMode == SM_I2O) { + testf ( NULL ); + } else { + decode = True; + for (aa = argList; aa != NULL; aa = aa->link) { + if (ISFLAG("--")) { decode = False; continue; } + if (aa->name[0] == '-' && decode) continue; + numFilesProcessed++; + testf ( aa->name ); + } + } + if (testFailsExist && noisy) { + fprintf ( stderr, + "\n" + "You can use the `bzip2recover' program to attempt to recover\n" + "data from undamaged sections of corrupted files.\n\n" + ); + setExit(2); + exit(exitValue); + } + } + + /* Free the argument list memory to mollify leak detectors + (eg) Purify, Checker. Serves no other useful purpose. + */ + aa = argList; + while (aa != NULL) { + Cell* aa2 = aa->link; + if (aa->name != NULL) free(aa->name); + free(aa); + aa = aa2; + } + + return exitValue; +} + + +/*-----------------------------------------------------------*/ +/*--- end bzip2.c ---*/ +/*-----------------------------------------------------------*/ diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/bzip2recover.c b/ext/Compress-Raw-Bzip2/bzip2-src/bzip2recover.c new file mode 100644 index 0000000..5f6d621 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/bzip2recover.c @@ -0,0 +1,514 @@ +/*-----------------------------------------------------------*/ +/*--- Block recoverer program for bzip2 ---*/ +/*--- bzip2recover.c ---*/ +/*-----------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + +/* This program is a complete hack and should be rewritten properly. + It isn't very complicated. */ + +#include +#include +#include +#include + + +/* This program records bit locations in the file to be recovered. + That means that if 64-bit ints are not supported, we will not + be able to recover .bz2 files over 512MB (2^32 bits) long. + On GNU supported platforms, we take advantage of the 64-bit + int support to circumvent this problem. Ditto MSVC. + + This change occurred in version 1.0.2; all prior versions have + the 512MB limitation. +*/ +#ifdef __GNUC__ + typedef unsigned long long int MaybeUInt64; +# define MaybeUInt64_FMT "%Lu" +#else +#ifdef _MSC_VER + typedef unsigned __int64 MaybeUInt64; +# define MaybeUInt64_FMT "%I64u" +#else + typedef unsigned int MaybeUInt64; +# define MaybeUInt64_FMT "%u" +#endif +#endif + +typedef unsigned int UInt32; +typedef int Int32; +typedef unsigned char UChar; +typedef char Char; +typedef unsigned char Bool; +#define True ((Bool)1) +#define False ((Bool)0) + + +#define BZ_MAX_FILENAME 2000 + +Char inFileName[BZ_MAX_FILENAME]; +Char outFileName[BZ_MAX_FILENAME]; +Char progName[BZ_MAX_FILENAME]; + +MaybeUInt64 bytesOut = 0; +MaybeUInt64 bytesIn = 0; + + +/*---------------------------------------------------*/ +/*--- Header bytes ---*/ +/*---------------------------------------------------*/ + +#define BZ_HDR_B 0x42 /* 'B' */ +#define BZ_HDR_Z 0x5a /* 'Z' */ +#define BZ_HDR_h 0x68 /* 'h' */ +#define BZ_HDR_0 0x30 /* '0' */ + + +/*---------------------------------------------------*/ +/*--- I/O errors ---*/ +/*---------------------------------------------------*/ + +/*---------------------------------------------*/ +static void readError ( void ) +{ + fprintf ( stderr, + "%s: I/O error reading `%s', possible reason follows.\n", + progName, inFileName ); + perror ( progName ); + fprintf ( stderr, "%s: warning: output file(s) may be incomplete.\n", + progName ); + exit ( 1 ); +} + + +/*---------------------------------------------*/ +static void writeError ( void ) +{ + fprintf ( stderr, + "%s: I/O error reading `%s', possible reason follows.\n", + progName, inFileName ); + perror ( progName ); + fprintf ( stderr, "%s: warning: output file(s) may be incomplete.\n", + progName ); + exit ( 1 ); +} + + +/*---------------------------------------------*/ +static void mallocFail ( Int32 n ) +{ + fprintf ( stderr, + "%s: malloc failed on request for %d bytes.\n", + progName, n ); + fprintf ( stderr, "%s: warning: output file(s) may be incomplete.\n", + progName ); + exit ( 1 ); +} + + +/*---------------------------------------------*/ +static void tooManyBlocks ( Int32 max_handled_blocks ) +{ + fprintf ( stderr, + "%s: `%s' appears to contain more than %d blocks\n", + progName, inFileName, max_handled_blocks ); + fprintf ( stderr, + "%s: and cannot be handled. To fix, increase\n", + progName ); + fprintf ( stderr, + "%s: BZ_MAX_HANDLED_BLOCKS in bzip2recover.c, and recompile.\n", + progName ); + exit ( 1 ); +} + + + +/*---------------------------------------------------*/ +/*--- Bit stream I/O ---*/ +/*---------------------------------------------------*/ + +typedef + struct { + FILE* handle; + Int32 buffer; + Int32 buffLive; + Char mode; + } + BitStream; + + +/*---------------------------------------------*/ +static BitStream* bsOpenReadStream ( FILE* stream ) +{ + BitStream *bs = malloc ( sizeof(BitStream) ); + if (bs == NULL) mallocFail ( sizeof(BitStream) ); + bs->handle = stream; + bs->buffer = 0; + bs->buffLive = 0; + bs->mode = 'r'; + return bs; +} + + +/*---------------------------------------------*/ +static BitStream* bsOpenWriteStream ( FILE* stream ) +{ + BitStream *bs = malloc ( sizeof(BitStream) ); + if (bs == NULL) mallocFail ( sizeof(BitStream) ); + bs->handle = stream; + bs->buffer = 0; + bs->buffLive = 0; + bs->mode = 'w'; + return bs; +} + + +/*---------------------------------------------*/ +static void bsPutBit ( BitStream* bs, Int32 bit ) +{ + if (bs->buffLive == 8) { + Int32 retVal = putc ( (UChar) bs->buffer, bs->handle ); + if (retVal == EOF) writeError(); + bytesOut++; + bs->buffLive = 1; + bs->buffer = bit & 0x1; + } else { + bs->buffer = ( (bs->buffer << 1) | (bit & 0x1) ); + bs->buffLive++; + }; +} + + +/*---------------------------------------------*/ +/*-- + Returns 0 or 1, or 2 to indicate EOF. +--*/ +static Int32 bsGetBit ( BitStream* bs ) +{ + if (bs->buffLive > 0) { + bs->buffLive --; + return ( ((bs->buffer) >> (bs->buffLive)) & 0x1 ); + } else { + Int32 retVal = getc ( bs->handle ); + if ( retVal == EOF ) { + if (errno != 0) readError(); + return 2; + } + bs->buffLive = 7; + bs->buffer = retVal; + return ( ((bs->buffer) >> 7) & 0x1 ); + } +} + + +/*---------------------------------------------*/ +static void bsClose ( BitStream* bs ) +{ + Int32 retVal; + + if ( bs->mode == 'w' ) { + while ( bs->buffLive < 8 ) { + bs->buffLive++; + bs->buffer <<= 1; + }; + retVal = putc ( (UChar) (bs->buffer), bs->handle ); + if (retVal == EOF) writeError(); + bytesOut++; + retVal = fflush ( bs->handle ); + if (retVal == EOF) writeError(); + } + retVal = fclose ( bs->handle ); + if (retVal == EOF) { + if (bs->mode == 'w') writeError(); else readError(); + } + free ( bs ); +} + + +/*---------------------------------------------*/ +static void bsPutUChar ( BitStream* bs, UChar c ) +{ + Int32 i; + for (i = 7; i >= 0; i--) + bsPutBit ( bs, (((UInt32) c) >> i) & 0x1 ); +} + + +/*---------------------------------------------*/ +static void bsPutUInt32 ( BitStream* bs, UInt32 c ) +{ + Int32 i; + + for (i = 31; i >= 0; i--) + bsPutBit ( bs, (c >> i) & 0x1 ); +} + + +/*---------------------------------------------*/ +static Bool endsInBz2 ( Char* name ) +{ + Int32 n = strlen ( name ); + if (n <= 4) return False; + return + (name[n-4] == '.' && + name[n-3] == 'b' && + name[n-2] == 'z' && + name[n-1] == '2'); +} + + +/*---------------------------------------------------*/ +/*--- ---*/ +/*---------------------------------------------------*/ + +/* This logic isn't really right when it comes to Cygwin. */ +#ifdef _WIN32 +# define BZ_SPLIT_SYM '\\' /* path splitter on Windows platform */ +#else +# define BZ_SPLIT_SYM '/' /* path splitter on Unix platform */ +#endif + +#define BLOCK_HEADER_HI 0x00003141UL +#define BLOCK_HEADER_LO 0x59265359UL + +#define BLOCK_ENDMARK_HI 0x00001772UL +#define BLOCK_ENDMARK_LO 0x45385090UL + +/* Increase if necessary. However, a .bz2 file with > 50000 blocks + would have an uncompressed size of at least 40GB, so the chances + are low you'll need to up this. +*/ +#define BZ_MAX_HANDLED_BLOCKS 50000 + +MaybeUInt64 bStart [BZ_MAX_HANDLED_BLOCKS]; +MaybeUInt64 bEnd [BZ_MAX_HANDLED_BLOCKS]; +MaybeUInt64 rbStart[BZ_MAX_HANDLED_BLOCKS]; +MaybeUInt64 rbEnd [BZ_MAX_HANDLED_BLOCKS]; + +Int32 main ( Int32 argc, Char** argv ) +{ + FILE* inFile; + FILE* outFile; + BitStream* bsIn, *bsWr; + Int32 b, wrBlock, currBlock, rbCtr; + MaybeUInt64 bitsRead; + + UInt32 buffHi, buffLo, blockCRC; + Char* p; + + strcpy ( progName, argv[0] ); + inFileName[0] = outFileName[0] = 0; + + fprintf ( stderr, + "bzip2recover 1.0.5: extracts blocks from damaged .bz2 files.\n" ); + + if (argc != 2) { + fprintf ( stderr, "%s: usage is `%s damaged_file_name'.\n", + progName, progName ); + switch (sizeof(MaybeUInt64)) { + case 8: + fprintf(stderr, + "\trestrictions on size of recovered file: None\n"); + break; + case 4: + fprintf(stderr, + "\trestrictions on size of recovered file: 512 MB\n"); + fprintf(stderr, + "\tto circumvent, recompile with MaybeUInt64 as an\n" + "\tunsigned 64-bit int.\n"); + break; + default: + fprintf(stderr, + "\tsizeof(MaybeUInt64) is not 4 or 8 -- " + "configuration error.\n"); + break; + } + exit(1); + } + + if (strlen(argv[1]) >= BZ_MAX_FILENAME-20) { + fprintf ( stderr, + "%s: supplied filename is suspiciously (>= %d chars) long. Bye!\n", + progName, (int)strlen(argv[1]) ); + exit(1); + } + + strcpy ( inFileName, argv[1] ); + + inFile = fopen ( inFileName, "rb" ); + if (inFile == NULL) { + fprintf ( stderr, "%s: can't read `%s'\n", progName, inFileName ); + exit(1); + } + + bsIn = bsOpenReadStream ( inFile ); + fprintf ( stderr, "%s: searching for block boundaries ...\n", progName ); + + bitsRead = 0; + buffHi = buffLo = 0; + currBlock = 0; + bStart[currBlock] = 0; + + rbCtr = 0; + + while (True) { + b = bsGetBit ( bsIn ); + bitsRead++; + if (b == 2) { + if (bitsRead >= bStart[currBlock] && + (bitsRead - bStart[currBlock]) >= 40) { + bEnd[currBlock] = bitsRead-1; + if (currBlock > 0) + fprintf ( stderr, " block %d runs from " MaybeUInt64_FMT + " to " MaybeUInt64_FMT " (incomplete)\n", + currBlock, bStart[currBlock], bEnd[currBlock] ); + } else + currBlock--; + break; + } + buffHi = (buffHi << 1) | (buffLo >> 31); + buffLo = (buffLo << 1) | (b & 1); + if ( ( (buffHi & 0x0000ffff) == BLOCK_HEADER_HI + && buffLo == BLOCK_HEADER_LO) + || + ( (buffHi & 0x0000ffff) == BLOCK_ENDMARK_HI + && buffLo == BLOCK_ENDMARK_LO) + ) { + if (bitsRead > 49) { + bEnd[currBlock] = bitsRead-49; + } else { + bEnd[currBlock] = 0; + } + if (currBlock > 0 && + (bEnd[currBlock] - bStart[currBlock]) >= 130) { + fprintf ( stderr, " block %d runs from " MaybeUInt64_FMT + " to " MaybeUInt64_FMT "\n", + rbCtr+1, bStart[currBlock], bEnd[currBlock] ); + rbStart[rbCtr] = bStart[currBlock]; + rbEnd[rbCtr] = bEnd[currBlock]; + rbCtr++; + } + if (currBlock >= BZ_MAX_HANDLED_BLOCKS) + tooManyBlocks(BZ_MAX_HANDLED_BLOCKS); + currBlock++; + + bStart[currBlock] = bitsRead; + } + } + + bsClose ( bsIn ); + + /*-- identified blocks run from 1 to rbCtr inclusive. --*/ + + if (rbCtr < 1) { + fprintf ( stderr, + "%s: sorry, I couldn't find any block boundaries.\n", + progName ); + exit(1); + }; + + fprintf ( stderr, "%s: splitting into blocks\n", progName ); + + inFile = fopen ( inFileName, "rb" ); + if (inFile == NULL) { + fprintf ( stderr, "%s: can't open `%s'\n", progName, inFileName ); + exit(1); + } + bsIn = bsOpenReadStream ( inFile ); + + /*-- placate gcc's dataflow analyser --*/ + blockCRC = 0; bsWr = 0; + + bitsRead = 0; + outFile = NULL; + wrBlock = 0; + while (True) { + b = bsGetBit(bsIn); + if (b == 2) break; + buffHi = (buffHi << 1) | (buffLo >> 31); + buffLo = (buffLo << 1) | (b & 1); + if (bitsRead == 47+rbStart[wrBlock]) + blockCRC = (buffHi << 16) | (buffLo >> 16); + + if (outFile != NULL && bitsRead >= rbStart[wrBlock] + && bitsRead <= rbEnd[wrBlock]) { + bsPutBit ( bsWr, b ); + } + + bitsRead++; + + if (bitsRead == rbEnd[wrBlock]+1) { + if (outFile != NULL) { + bsPutUChar ( bsWr, 0x17 ); bsPutUChar ( bsWr, 0x72 ); + bsPutUChar ( bsWr, 0x45 ); bsPutUChar ( bsWr, 0x38 ); + bsPutUChar ( bsWr, 0x50 ); bsPutUChar ( bsWr, 0x90 ); + bsPutUInt32 ( bsWr, blockCRC ); + bsClose ( bsWr ); + } + if (wrBlock >= rbCtr) break; + wrBlock++; + } else + if (bitsRead == rbStart[wrBlock]) { + /* Create the output file name, correctly handling leading paths. + (31.10.2001 by Sergey E. Kusikov) */ + Char* split; + Int32 ofs, k; + for (k = 0; k < BZ_MAX_FILENAME; k++) + outFileName[k] = 0; + strcpy (outFileName, inFileName); + split = strrchr (outFileName, BZ_SPLIT_SYM); + if (split == NULL) { + split = outFileName; + } else { + ++split; + } + /* Now split points to the start of the basename. */ + ofs = split - outFileName; + sprintf (split, "rec%5d", wrBlock+1); + for (p = split; *p != 0; p++) if (*p == ' ') *p = '0'; + strcat (outFileName, inFileName + ofs); + + if ( !endsInBz2(outFileName)) strcat ( outFileName, ".bz2" ); + + fprintf ( stderr, " writing block %d to `%s' ...\n", + wrBlock+1, outFileName ); + + outFile = fopen ( outFileName, "wb" ); + if (outFile == NULL) { + fprintf ( stderr, "%s: can't write `%s'\n", + progName, outFileName ); + exit(1); + } + bsWr = bsOpenWriteStream ( outFile ); + bsPutUChar ( bsWr, BZ_HDR_B ); + bsPutUChar ( bsWr, BZ_HDR_Z ); + bsPutUChar ( bsWr, BZ_HDR_h ); + bsPutUChar ( bsWr, BZ_HDR_0 + 9 ); + bsPutUChar ( bsWr, 0x31 ); bsPutUChar ( bsWr, 0x41 ); + bsPutUChar ( bsWr, 0x59 ); bsPutUChar ( bsWr, 0x26 ); + bsPutUChar ( bsWr, 0x53 ); bsPutUChar ( bsWr, 0x59 ); + } + } + + fprintf ( stderr, "%s: finished\n", progName ); + return 0; +} + + + +/*-----------------------------------------------------------*/ +/*--- end bzip2recover.c ---*/ +/*-----------------------------------------------------------*/ diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/bzlib.c b/ext/Compress-Raw-Bzip2/bzip2-src/bzlib.c new file mode 100644 index 0000000..ef86c91 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/bzlib.c @@ -0,0 +1,1572 @@ + +/*-------------------------------------------------------------*/ +/*--- Library top-level functions. ---*/ +/*--- bzlib.c ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + +/* CHANGES + 0.9.0 -- original version. + 0.9.0a/b -- no changes in this file. + 0.9.0c -- made zero-length BZ_FLUSH work correctly in bzCompress(). + fixed bzWrite/bzRead to ignore zero-length requests. + fixed bzread to correctly handle read requests after EOF. + wrong parameter order in call to bzDecompressInit in + bzBuffToBuffDecompress. Fixed. +*/ + +#include "bzlib_private.h" + + +/*---------------------------------------------------*/ +/*--- Compression stuff ---*/ +/*---------------------------------------------------*/ + + +/*---------------------------------------------------*/ +#ifndef BZ_NO_STDIO +void BZ2_bz__AssertH__fail ( int errcode ) +{ + fprintf(stderr, + "\n\nbzip2/libbzip2: internal error number %d.\n" + "This is a bug in bzip2/libbzip2, %s.\n" + "Please report it to me at: jseward@bzip.org. If this happened\n" + "when you were using some program which uses libbzip2 as a\n" + "component, you should also report this bug to the author(s)\n" + "of that program. Please make an effort to report this bug;\n" + "timely and accurate bug reports eventually lead to higher\n" + "quality software. Thanks. Julian Seward, 10 December 2007.\n\n", + errcode, + BZ2_bzlibVersion() + ); + + if (errcode == 1007) { + fprintf(stderr, + "\n*** A special note about internal error number 1007 ***\n" + "\n" + "Experience suggests that a common cause of i.e. 1007\n" + "is unreliable memory or other hardware. The 1007 assertion\n" + "just happens to cross-check the results of huge numbers of\n" + "memory reads/writes, and so acts (unintendedly) as a stress\n" + "test of your memory system.\n" + "\n" + "I suggest the following: try compressing the file again,\n" + "possibly monitoring progress in detail with the -vv flag.\n" + "\n" + "* If the error cannot be reproduced, and/or happens at different\n" + " points in compression, you may have a flaky memory system.\n" + " Try a memory-test program. I have used Memtest86\n" + " (www.memtest86.com). At the time of writing it is free (GPLd).\n" + " Memtest86 tests memory much more thorougly than your BIOSs\n" + " power-on test, and may find failures that the BIOS doesn't.\n" + "\n" + "* If the error can be repeatably reproduced, this is a bug in\n" + " bzip2, and I would very much like to hear about it. Please\n" + " let me know, and, ideally, save a copy of the file causing the\n" + " problem -- without which I will be unable to investigate it.\n" + "\n" + ); + } + + exit(3); +} +#endif + + +/*---------------------------------------------------*/ +static +int bz_config_ok ( void ) +{ + if (sizeof(int) != 4) return 0; + if (sizeof(short) != 2) return 0; + if (sizeof(char) != 1) return 0; + return 1; +} + + +/*---------------------------------------------------*/ +static +void* default_bzalloc ( void* opaque, Int32 items, Int32 size ) +{ + void* v = malloc ( items * size ); + return v; +} + +static +void default_bzfree ( void* opaque, void* addr ) +{ + if (addr != NULL) free ( addr ); +} + + +/*---------------------------------------------------*/ +static +void prepare_new_block ( EState* s ) +{ + Int32 i; + s->nblock = 0; + s->numZ = 0; + s->state_out_pos = 0; + BZ_INITIALISE_CRC ( s->blockCRC ); + for (i = 0; i < 256; i++) s->inUse[i] = False; + s->blockNo++; +} + + +/*---------------------------------------------------*/ +static +void init_RL ( EState* s ) +{ + s->state_in_ch = 256; + s->state_in_len = 0; +} + + +static +Bool isempty_RL ( EState* s ) +{ + if (s->state_in_ch < 256 && s->state_in_len > 0) + return False; else + return True; +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzCompressInit) + ( bz_stream* strm, + int blockSize100k, + int verbosity, + int workFactor ) +{ + Int32 n; + EState* s; + + if (!bz_config_ok()) return BZ_CONFIG_ERROR; + + if (strm == NULL || + blockSize100k < 1 || blockSize100k > 9 || + workFactor < 0 || workFactor > 250) + return BZ_PARAM_ERROR; + + if (workFactor == 0) workFactor = 30; + if (strm->bzalloc == NULL) strm->bzalloc = default_bzalloc; + if (strm->bzfree == NULL) strm->bzfree = default_bzfree; + + s = BZALLOC( sizeof(EState) ); + if (s == NULL) return BZ_MEM_ERROR; + s->strm = strm; + + s->arr1 = NULL; + s->arr2 = NULL; + s->ftab = NULL; + + n = 100000 * blockSize100k; + s->arr1 = BZALLOC( n * sizeof(UInt32) ); + s->arr2 = BZALLOC( (n+BZ_N_OVERSHOOT) * sizeof(UInt32) ); + s->ftab = BZALLOC( 65537 * sizeof(UInt32) ); + + if (s->arr1 == NULL || s->arr2 == NULL || s->ftab == NULL) { + if (s->arr1 != NULL) BZFREE(s->arr1); + if (s->arr2 != NULL) BZFREE(s->arr2); + if (s->ftab != NULL) BZFREE(s->ftab); + if (s != NULL) BZFREE(s); + return BZ_MEM_ERROR; + } + + s->blockNo = 0; + s->state = BZ_S_INPUT; + s->mode = BZ_M_RUNNING; + s->combinedCRC = 0; + s->blockSize100k = blockSize100k; + s->nblockMAX = 100000 * blockSize100k - 19; + s->verbosity = verbosity; + s->workFactor = workFactor; + + s->block = (UChar*)s->arr2; + s->mtfv = (UInt16*)s->arr1; + s->zbits = NULL; + s->ptr = (UInt32*)s->arr1; + + strm->state = s; + strm->total_in_lo32 = 0; + strm->total_in_hi32 = 0; + strm->total_out_lo32 = 0; + strm->total_out_hi32 = 0; + init_RL ( s ); + prepare_new_block ( s ); + return BZ_OK; +} + + +/*---------------------------------------------------*/ +static +void add_pair_to_block ( EState* s ) +{ + Int32 i; + UChar ch = (UChar)(s->state_in_ch); + for (i = 0; i < s->state_in_len; i++) { + BZ_UPDATE_CRC( s->blockCRC, ch ); + } + s->inUse[s->state_in_ch] = True; + switch (s->state_in_len) { + case 1: + s->block[s->nblock] = (UChar)ch; s->nblock++; + break; + case 2: + s->block[s->nblock] = (UChar)ch; s->nblock++; + s->block[s->nblock] = (UChar)ch; s->nblock++; + break; + case 3: + s->block[s->nblock] = (UChar)ch; s->nblock++; + s->block[s->nblock] = (UChar)ch; s->nblock++; + s->block[s->nblock] = (UChar)ch; s->nblock++; + break; + default: + s->inUse[s->state_in_len-4] = True; + s->block[s->nblock] = (UChar)ch; s->nblock++; + s->block[s->nblock] = (UChar)ch; s->nblock++; + s->block[s->nblock] = (UChar)ch; s->nblock++; + s->block[s->nblock] = (UChar)ch; s->nblock++; + s->block[s->nblock] = ((UChar)(s->state_in_len-4)); + s->nblock++; + break; + } +} + + +/*---------------------------------------------------*/ +static +void flush_RL ( EState* s ) +{ + if (s->state_in_ch < 256) add_pair_to_block ( s ); + init_RL ( s ); +} + + +/*---------------------------------------------------*/ +#define ADD_CHAR_TO_BLOCK(zs,zchh0) \ +{ \ + UInt32 zchh = (UInt32)(zchh0); \ + /*-- fast track the common case --*/ \ + if (zchh != zs->state_in_ch && \ + zs->state_in_len == 1) { \ + UChar ch = (UChar)(zs->state_in_ch); \ + BZ_UPDATE_CRC( zs->blockCRC, ch ); \ + zs->inUse[zs->state_in_ch] = True; \ + zs->block[zs->nblock] = (UChar)ch; \ + zs->nblock++; \ + zs->state_in_ch = zchh; \ + } \ + else \ + /*-- general, uncommon cases --*/ \ + if (zchh != zs->state_in_ch || \ + zs->state_in_len == 255) { \ + if (zs->state_in_ch < 256) \ + add_pair_to_block ( zs ); \ + zs->state_in_ch = zchh; \ + zs->state_in_len = 1; \ + } else { \ + zs->state_in_len++; \ + } \ +} + + +/*---------------------------------------------------*/ +static +Bool copy_input_until_stop ( EState* s ) +{ + Bool progress_in = False; + + if (s->mode == BZ_M_RUNNING) { + + /*-- fast track the common case --*/ + while (True) { + /*-- block full? --*/ + if (s->nblock >= s->nblockMAX) break; + /*-- no input? --*/ + if (s->strm->avail_in == 0) break; + progress_in = True; + ADD_CHAR_TO_BLOCK ( s, (UInt32)(*((UChar*)(s->strm->next_in))) ); + s->strm->next_in++; + s->strm->avail_in--; + s->strm->total_in_lo32++; + if (s->strm->total_in_lo32 == 0) s->strm->total_in_hi32++; + } + + } else { + + /*-- general, uncommon case --*/ + while (True) { + /*-- block full? --*/ + if (s->nblock >= s->nblockMAX) break; + /*-- no input? --*/ + if (s->strm->avail_in == 0) break; + /*-- flush/finish end? --*/ + if (s->avail_in_expect == 0) break; + progress_in = True; + ADD_CHAR_TO_BLOCK ( s, (UInt32)(*((UChar*)(s->strm->next_in))) ); + s->strm->next_in++; + s->strm->avail_in--; + s->strm->total_in_lo32++; + if (s->strm->total_in_lo32 == 0) s->strm->total_in_hi32++; + s->avail_in_expect--; + } + } + return progress_in; +} + + +/*---------------------------------------------------*/ +static +Bool copy_output_until_stop ( EState* s ) +{ + Bool progress_out = False; + + while (True) { + + /*-- no output space? --*/ + if (s->strm->avail_out == 0) break; + + /*-- block done? --*/ + if (s->state_out_pos >= s->numZ) break; + + progress_out = True; + *(s->strm->next_out) = s->zbits[s->state_out_pos]; + s->state_out_pos++; + s->strm->avail_out--; + s->strm->next_out++; + s->strm->total_out_lo32++; + if (s->strm->total_out_lo32 == 0) s->strm->total_out_hi32++; + } + + return progress_out; +} + + +/*---------------------------------------------------*/ +static +Bool handle_compress ( bz_stream* strm ) +{ + Bool progress_in = False; + Bool progress_out = False; + EState* s = strm->state; + + while (True) { + + if (s->state == BZ_S_OUTPUT) { + progress_out |= copy_output_until_stop ( s ); + if (s->state_out_pos < s->numZ) break; + if (s->mode == BZ_M_FINISHING && + s->avail_in_expect == 0 && + isempty_RL(s)) break; + prepare_new_block ( s ); + s->state = BZ_S_INPUT; + if (s->mode == BZ_M_FLUSHING && + s->avail_in_expect == 0 && + isempty_RL(s)) break; + } + + if (s->state == BZ_S_INPUT) { + progress_in |= copy_input_until_stop ( s ); + if (s->mode != BZ_M_RUNNING && s->avail_in_expect == 0) { + flush_RL ( s ); + BZ2_compressBlock ( s, (Bool)(s->mode == BZ_M_FINISHING) ); + s->state = BZ_S_OUTPUT; + } + else + if (s->nblock >= s->nblockMAX) { + BZ2_compressBlock ( s, False ); + s->state = BZ_S_OUTPUT; + } + else + if (s->strm->avail_in == 0) { + break; + } + } + + } + + return progress_in || progress_out; +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzCompress) ( bz_stream *strm, int action ) +{ + Bool progress; + EState* s; + if (strm == NULL) return BZ_PARAM_ERROR; + s = strm->state; + if (s == NULL) return BZ_PARAM_ERROR; + if (s->strm != strm) return BZ_PARAM_ERROR; + + preswitch: + switch (s->mode) { + + case BZ_M_IDLE: + return BZ_SEQUENCE_ERROR; + + case BZ_M_RUNNING: + if (action == BZ_RUN) { + progress = handle_compress ( strm ); + return progress ? BZ_RUN_OK : BZ_PARAM_ERROR; + } + else + if (action == BZ_FLUSH) { + s->avail_in_expect = strm->avail_in; + s->mode = BZ_M_FLUSHING; + goto preswitch; + } + else + if (action == BZ_FINISH) { + s->avail_in_expect = strm->avail_in; + s->mode = BZ_M_FINISHING; + goto preswitch; + } + else + return BZ_PARAM_ERROR; + + case BZ_M_FLUSHING: + if (action != BZ_FLUSH) return BZ_SEQUENCE_ERROR; + if (s->avail_in_expect != s->strm->avail_in) + return BZ_SEQUENCE_ERROR; + progress = handle_compress ( strm ); + if (s->avail_in_expect > 0 || !isempty_RL(s) || + s->state_out_pos < s->numZ) return BZ_FLUSH_OK; + s->mode = BZ_M_RUNNING; + return BZ_RUN_OK; + + case BZ_M_FINISHING: + if (action != BZ_FINISH) return BZ_SEQUENCE_ERROR; + if (s->avail_in_expect != s->strm->avail_in) + return BZ_SEQUENCE_ERROR; + progress = handle_compress ( strm ); + if (!progress) return BZ_SEQUENCE_ERROR; + if (s->avail_in_expect > 0 || !isempty_RL(s) || + s->state_out_pos < s->numZ) return BZ_FINISH_OK; + s->mode = BZ_M_IDLE; + return BZ_STREAM_END; + } + return BZ_OK; /*--not reached--*/ +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzCompressEnd) ( bz_stream *strm ) +{ + EState* s; + if (strm == NULL) return BZ_PARAM_ERROR; + s = strm->state; + if (s == NULL) return BZ_PARAM_ERROR; + if (s->strm != strm) return BZ_PARAM_ERROR; + + if (s->arr1 != NULL) BZFREE(s->arr1); + if (s->arr2 != NULL) BZFREE(s->arr2); + if (s->ftab != NULL) BZFREE(s->ftab); + BZFREE(strm->state); + + strm->state = NULL; + + return BZ_OK; +} + + +/*---------------------------------------------------*/ +/*--- Decompression stuff ---*/ +/*---------------------------------------------------*/ + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzDecompressInit) + ( bz_stream* strm, + int verbosity, + int small ) +{ + DState* s; + + if (!bz_config_ok()) return BZ_CONFIG_ERROR; + + if (strm == NULL) return BZ_PARAM_ERROR; + if (small != 0 && small != 1) return BZ_PARAM_ERROR; + if (verbosity < 0 || verbosity > 4) return BZ_PARAM_ERROR; + + if (strm->bzalloc == NULL) strm->bzalloc = default_bzalloc; + if (strm->bzfree == NULL) strm->bzfree = default_bzfree; + + s = BZALLOC( sizeof(DState) ); + if (s == NULL) return BZ_MEM_ERROR; + s->strm = strm; + strm->state = s; + s->state = BZ_X_MAGIC_1; + s->bsLive = 0; + s->bsBuff = 0; + s->calculatedCombinedCRC = 0; + strm->total_in_lo32 = 0; + strm->total_in_hi32 = 0; + strm->total_out_lo32 = 0; + strm->total_out_hi32 = 0; + s->smallDecompress = (Bool)small; + s->ll4 = NULL; + s->ll16 = NULL; + s->tt = NULL; + s->currBlockNo = 0; + s->verbosity = verbosity; + + return BZ_OK; +} + + +/*---------------------------------------------------*/ +/* Return True iff data corruption is discovered. + Returns False if there is no problem. +*/ +static +Bool unRLE_obuf_to_output_FAST ( DState* s ) +{ + UChar k1; + + if (s->blockRandomised) { + + while (True) { + /* try to finish existing run */ + while (True) { + if (s->strm->avail_out == 0) return False; + if (s->state_out_len == 0) break; + *( (UChar*)(s->strm->next_out) ) = s->state_out_ch; + BZ_UPDATE_CRC ( s->calculatedBlockCRC, s->state_out_ch ); + s->state_out_len--; + s->strm->next_out++; + s->strm->avail_out--; + s->strm->total_out_lo32++; + if (s->strm->total_out_lo32 == 0) s->strm->total_out_hi32++; + } + + /* can a new run be started? */ + if (s->nblock_used == s->save_nblock+1) return False; + + /* Only caused by corrupt data stream? */ + if (s->nblock_used > s->save_nblock+1) + return True; + + s->state_out_len = 1; + s->state_out_ch = s->k0; + BZ_GET_FAST(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + s->state_out_len = 2; + BZ_GET_FAST(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + s->state_out_len = 3; + BZ_GET_FAST(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + BZ_GET_FAST(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + s->state_out_len = ((Int32)k1) + 4; + BZ_GET_FAST(s->k0); BZ_RAND_UPD_MASK; + s->k0 ^= BZ_RAND_MASK; s->nblock_used++; + } + + } else { + + /* restore */ + UInt32 c_calculatedBlockCRC = s->calculatedBlockCRC; + UChar c_state_out_ch = s->state_out_ch; + Int32 c_state_out_len = s->state_out_len; + Int32 c_nblock_used = s->nblock_used; + Int32 c_k0 = s->k0; + UInt32* c_tt = s->tt; + UInt32 c_tPos = s->tPos; + char* cs_next_out = s->strm->next_out; + unsigned int cs_avail_out = s->strm->avail_out; + Int32 ro_blockSize100k = s->blockSize100k; + /* end restore */ + + UInt32 avail_out_INIT = cs_avail_out; + Int32 s_save_nblockPP = s->save_nblock+1; + unsigned int total_out_lo32_old; + + while (True) { + + /* try to finish existing run */ + if (c_state_out_len > 0) { + while (True) { + if (cs_avail_out == 0) goto return_notr; + if (c_state_out_len == 1) break; + *( (UChar*)(cs_next_out) ) = c_state_out_ch; + BZ_UPDATE_CRC ( c_calculatedBlockCRC, c_state_out_ch ); + c_state_out_len--; + cs_next_out++; + cs_avail_out--; + } + s_state_out_len_eq_one: + { + if (cs_avail_out == 0) { + c_state_out_len = 1; goto return_notr; + }; + *( (UChar*)(cs_next_out) ) = c_state_out_ch; + BZ_UPDATE_CRC ( c_calculatedBlockCRC, c_state_out_ch ); + cs_next_out++; + cs_avail_out--; + } + } + /* Only caused by corrupt data stream? */ + if (c_nblock_used > s_save_nblockPP) + return True; + + /* can a new run be started? */ + if (c_nblock_used == s_save_nblockPP) { + c_state_out_len = 0; goto return_notr; + }; + c_state_out_ch = c_k0; + BZ_GET_FAST_C(k1); c_nblock_used++; + if (k1 != c_k0) { + c_k0 = k1; goto s_state_out_len_eq_one; + }; + if (c_nblock_used == s_save_nblockPP) + goto s_state_out_len_eq_one; + + c_state_out_len = 2; + BZ_GET_FAST_C(k1); c_nblock_used++; + if (c_nblock_used == s_save_nblockPP) continue; + if (k1 != c_k0) { c_k0 = k1; continue; }; + + c_state_out_len = 3; + BZ_GET_FAST_C(k1); c_nblock_used++; + if (c_nblock_used == s_save_nblockPP) continue; + if (k1 != c_k0) { c_k0 = k1; continue; }; + + BZ_GET_FAST_C(k1); c_nblock_used++; + c_state_out_len = ((Int32)k1) + 4; + BZ_GET_FAST_C(c_k0); c_nblock_used++; + } + + return_notr: + total_out_lo32_old = s->strm->total_out_lo32; + s->strm->total_out_lo32 += (avail_out_INIT - cs_avail_out); + if (s->strm->total_out_lo32 < total_out_lo32_old) + s->strm->total_out_hi32++; + + /* save */ + s->calculatedBlockCRC = c_calculatedBlockCRC; + s->state_out_ch = c_state_out_ch; + s->state_out_len = c_state_out_len; + s->nblock_used = c_nblock_used; + s->k0 = c_k0; + s->tt = c_tt; + s->tPos = c_tPos; + s->strm->next_out = cs_next_out; + s->strm->avail_out = cs_avail_out; + /* end save */ + } + return False; +} + + + +/*---------------------------------------------------*/ +__inline__ Int32 BZ2_indexIntoF ( Int32 indx, Int32 *cftab ) +{ + Int32 nb, na, mid; + nb = 0; + na = 256; + do { + mid = (nb + na) >> 1; + if (indx >= cftab[mid]) nb = mid; else na = mid; + } + while (na - nb != 1); + return nb; +} + + +/*---------------------------------------------------*/ +/* Return True iff data corruption is discovered. + Returns False if there is no problem. +*/ +static +Bool unRLE_obuf_to_output_SMALL ( DState* s ) +{ + UChar k1; + + if (s->blockRandomised) { + + while (True) { + /* try to finish existing run */ + while (True) { + if (s->strm->avail_out == 0) return False; + if (s->state_out_len == 0) break; + *( (UChar*)(s->strm->next_out) ) = s->state_out_ch; + BZ_UPDATE_CRC ( s->calculatedBlockCRC, s->state_out_ch ); + s->state_out_len--; + s->strm->next_out++; + s->strm->avail_out--; + s->strm->total_out_lo32++; + if (s->strm->total_out_lo32 == 0) s->strm->total_out_hi32++; + } + + /* can a new run be started? */ + if (s->nblock_used == s->save_nblock+1) return False; + + /* Only caused by corrupt data stream? */ + if (s->nblock_used > s->save_nblock+1) + return True; + + s->state_out_len = 1; + s->state_out_ch = s->k0; + BZ_GET_SMALL(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + s->state_out_len = 2; + BZ_GET_SMALL(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + s->state_out_len = 3; + BZ_GET_SMALL(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + BZ_GET_SMALL(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + s->state_out_len = ((Int32)k1) + 4; + BZ_GET_SMALL(s->k0); BZ_RAND_UPD_MASK; + s->k0 ^= BZ_RAND_MASK; s->nblock_used++; + } + + } else { + + while (True) { + /* try to finish existing run */ + while (True) { + if (s->strm->avail_out == 0) return False; + if (s->state_out_len == 0) break; + *( (UChar*)(s->strm->next_out) ) = s->state_out_ch; + BZ_UPDATE_CRC ( s->calculatedBlockCRC, s->state_out_ch ); + s->state_out_len--; + s->strm->next_out++; + s->strm->avail_out--; + s->strm->total_out_lo32++; + if (s->strm->total_out_lo32 == 0) s->strm->total_out_hi32++; + } + + /* can a new run be started? */ + if (s->nblock_used == s->save_nblock+1) return False; + + /* Only caused by corrupt data stream? */ + if (s->nblock_used > s->save_nblock+1) + return True; + + s->state_out_len = 1; + s->state_out_ch = s->k0; + BZ_GET_SMALL(k1); s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + s->state_out_len = 2; + BZ_GET_SMALL(k1); s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + s->state_out_len = 3; + BZ_GET_SMALL(k1); s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + BZ_GET_SMALL(k1); s->nblock_used++; + s->state_out_len = ((Int32)k1) + 4; + BZ_GET_SMALL(s->k0); s->nblock_used++; + } + + } +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzDecompress) ( bz_stream *strm ) +{ + Bool corrupt; + DState* s; + if (strm == NULL) return BZ_PARAM_ERROR; + s = strm->state; + if (s == NULL) return BZ_PARAM_ERROR; + if (s->strm != strm) return BZ_PARAM_ERROR; + + while (True) { + if (s->state == BZ_X_IDLE) return BZ_SEQUENCE_ERROR; + if (s->state == BZ_X_OUTPUT) { + if (s->smallDecompress) + corrupt = unRLE_obuf_to_output_SMALL ( s ); else + corrupt = unRLE_obuf_to_output_FAST ( s ); + if (corrupt) return BZ_DATA_ERROR; + if (s->nblock_used == s->save_nblock+1 && s->state_out_len == 0) { + BZ_FINALISE_CRC ( s->calculatedBlockCRC ); + if (s->verbosity >= 3) + VPrintf2 ( " {0x%08x, 0x%08x}", s->storedBlockCRC, + s->calculatedBlockCRC ); + if (s->verbosity >= 2) VPrintf0 ( "]" ); + if (s->calculatedBlockCRC != s->storedBlockCRC) + return BZ_DATA_ERROR; + s->calculatedCombinedCRC + = (s->calculatedCombinedCRC << 1) | + (s->calculatedCombinedCRC >> 31); + s->calculatedCombinedCRC ^= s->calculatedBlockCRC; + s->state = BZ_X_BLKHDR_1; + } else { + return BZ_OK; + } + } + if (s->state >= BZ_X_MAGIC_1) { + Int32 r = BZ2_decompress ( s ); + if (r == BZ_STREAM_END) { + if (s->verbosity >= 3) + VPrintf2 ( "\n combined CRCs: stored = 0x%08x, computed = 0x%08x", + s->storedCombinedCRC, s->calculatedCombinedCRC ); + if (s->calculatedCombinedCRC != s->storedCombinedCRC) + return BZ_DATA_ERROR; + return r; + } + if (s->state != BZ_X_OUTPUT) return r; + } + } + + AssertH ( 0, 6001 ); + + return 0; /*NOTREACHED*/ +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzDecompressEnd) ( bz_stream *strm ) +{ + DState* s; + if (strm == NULL) return BZ_PARAM_ERROR; + s = strm->state; + if (s == NULL) return BZ_PARAM_ERROR; + if (s->strm != strm) return BZ_PARAM_ERROR; + + if (s->tt != NULL) BZFREE(s->tt); + if (s->ll16 != NULL) BZFREE(s->ll16); + if (s->ll4 != NULL) BZFREE(s->ll4); + + BZFREE(strm->state); + strm->state = NULL; + + return BZ_OK; +} + + +#ifndef BZ_NO_STDIO +/*---------------------------------------------------*/ +/*--- File I/O stuff ---*/ +/*---------------------------------------------------*/ + +#define BZ_SETERR(eee) \ +{ \ + if (bzerror != NULL) *bzerror = eee; \ + if (bzf != NULL) bzf->lastErr = eee; \ +} + +typedef + struct { + FILE* handle; + Char buf[BZ_MAX_UNUSED]; + Int32 bufN; + Bool writing; + bz_stream strm; + Int32 lastErr; + Bool initialisedOk; + } + bzFile; + + +/*---------------------------------------------*/ +static Bool myfeof ( FILE* f ) +{ + Int32 c = fgetc ( f ); + if (c == EOF) return True; + ungetc ( c, f ); + return False; +} + + +/*---------------------------------------------------*/ +BZFILE* BZ_API(BZ2_bzWriteOpen) + ( int* bzerror, + FILE* f, + int blockSize100k, + int verbosity, + int workFactor ) +{ + Int32 ret; + bzFile* bzf = NULL; + + BZ_SETERR(BZ_OK); + + if (f == NULL || + (blockSize100k < 1 || blockSize100k > 9) || + (workFactor < 0 || workFactor > 250) || + (verbosity < 0 || verbosity > 4)) + { BZ_SETERR(BZ_PARAM_ERROR); return NULL; }; + + if (ferror(f)) + { BZ_SETERR(BZ_IO_ERROR); return NULL; }; + + bzf = malloc ( sizeof(bzFile) ); + if (bzf == NULL) + { BZ_SETERR(BZ_MEM_ERROR); return NULL; }; + + BZ_SETERR(BZ_OK); + bzf->initialisedOk = False; + bzf->bufN = 0; + bzf->handle = f; + bzf->writing = True; + bzf->strm.bzalloc = NULL; + bzf->strm.bzfree = NULL; + bzf->strm.opaque = NULL; + + if (workFactor == 0) workFactor = 30; + ret = BZ2_bzCompressInit ( &(bzf->strm), blockSize100k, + verbosity, workFactor ); + if (ret != BZ_OK) + { BZ_SETERR(ret); free(bzf); return NULL; }; + + bzf->strm.avail_in = 0; + bzf->initialisedOk = True; + return bzf; +} + + + +/*---------------------------------------------------*/ +void BZ_API(BZ2_bzWrite) + ( int* bzerror, + BZFILE* b, + void* buf, + int len ) +{ + Int32 n, n2, ret; + bzFile* bzf = (bzFile*)b; + + BZ_SETERR(BZ_OK); + if (bzf == NULL || buf == NULL || len < 0) + { BZ_SETERR(BZ_PARAM_ERROR); return; }; + if (!(bzf->writing)) + { BZ_SETERR(BZ_SEQUENCE_ERROR); return; }; + if (ferror(bzf->handle)) + { BZ_SETERR(BZ_IO_ERROR); return; }; + + if (len == 0) + { BZ_SETERR(BZ_OK); return; }; + + bzf->strm.avail_in = len; + bzf->strm.next_in = buf; + + while (True) { + bzf->strm.avail_out = BZ_MAX_UNUSED; + bzf->strm.next_out = bzf->buf; + ret = BZ2_bzCompress ( &(bzf->strm), BZ_RUN ); + if (ret != BZ_RUN_OK) + { BZ_SETERR(ret); return; }; + + if (bzf->strm.avail_out < BZ_MAX_UNUSED) { + n = BZ_MAX_UNUSED - bzf->strm.avail_out; + n2 = fwrite ( (void*)(bzf->buf), sizeof(UChar), + n, bzf->handle ); + if (n != n2 || ferror(bzf->handle)) + { BZ_SETERR(BZ_IO_ERROR); return; }; + } + + if (bzf->strm.avail_in == 0) + { BZ_SETERR(BZ_OK); return; }; + } +} + + +/*---------------------------------------------------*/ +void BZ_API(BZ2_bzWriteClose) + ( int* bzerror, + BZFILE* b, + int abandon, + unsigned int* nbytes_in, + unsigned int* nbytes_out ) +{ + BZ2_bzWriteClose64 ( bzerror, b, abandon, + nbytes_in, NULL, nbytes_out, NULL ); +} + + +void BZ_API(BZ2_bzWriteClose64) + ( int* bzerror, + BZFILE* b, + int abandon, + unsigned int* nbytes_in_lo32, + unsigned int* nbytes_in_hi32, + unsigned int* nbytes_out_lo32, + unsigned int* nbytes_out_hi32 ) +{ + Int32 n, n2, ret; + bzFile* bzf = (bzFile*)b; + + if (bzf == NULL) + { BZ_SETERR(BZ_OK); return; }; + if (!(bzf->writing)) + { BZ_SETERR(BZ_SEQUENCE_ERROR); return; }; + if (ferror(bzf->handle)) + { BZ_SETERR(BZ_IO_ERROR); return; }; + + if (nbytes_in_lo32 != NULL) *nbytes_in_lo32 = 0; + if (nbytes_in_hi32 != NULL) *nbytes_in_hi32 = 0; + if (nbytes_out_lo32 != NULL) *nbytes_out_lo32 = 0; + if (nbytes_out_hi32 != NULL) *nbytes_out_hi32 = 0; + + if ((!abandon) && bzf->lastErr == BZ_OK) { + while (True) { + bzf->strm.avail_out = BZ_MAX_UNUSED; + bzf->strm.next_out = bzf->buf; + ret = BZ2_bzCompress ( &(bzf->strm), BZ_FINISH ); + if (ret != BZ_FINISH_OK && ret != BZ_STREAM_END) + { BZ_SETERR(ret); return; }; + + if (bzf->strm.avail_out < BZ_MAX_UNUSED) { + n = BZ_MAX_UNUSED - bzf->strm.avail_out; + n2 = fwrite ( (void*)(bzf->buf), sizeof(UChar), + n, bzf->handle ); + if (n != n2 || ferror(bzf->handle)) + { BZ_SETERR(BZ_IO_ERROR); return; }; + } + + if (ret == BZ_STREAM_END) break; + } + } + + if ( !abandon && !ferror ( bzf->handle ) ) { + fflush ( bzf->handle ); + if (ferror(bzf->handle)) + { BZ_SETERR(BZ_IO_ERROR); return; }; + } + + if (nbytes_in_lo32 != NULL) + *nbytes_in_lo32 = bzf->strm.total_in_lo32; + if (nbytes_in_hi32 != NULL) + *nbytes_in_hi32 = bzf->strm.total_in_hi32; + if (nbytes_out_lo32 != NULL) + *nbytes_out_lo32 = bzf->strm.total_out_lo32; + if (nbytes_out_hi32 != NULL) + *nbytes_out_hi32 = bzf->strm.total_out_hi32; + + BZ_SETERR(BZ_OK); + BZ2_bzCompressEnd ( &(bzf->strm) ); + free ( bzf ); +} + + +/*---------------------------------------------------*/ +BZFILE* BZ_API(BZ2_bzReadOpen) + ( int* bzerror, + FILE* f, + int verbosity, + int small, + void* unused, + int nUnused ) +{ + bzFile* bzf = NULL; + int ret; + + BZ_SETERR(BZ_OK); + + if (f == NULL || + (small != 0 && small != 1) || + (verbosity < 0 || verbosity > 4) || + (unused == NULL && nUnused != 0) || + (unused != NULL && (nUnused < 0 || nUnused > BZ_MAX_UNUSED))) + { BZ_SETERR(BZ_PARAM_ERROR); return NULL; }; + + if (ferror(f)) + { BZ_SETERR(BZ_IO_ERROR); return NULL; }; + + bzf = malloc ( sizeof(bzFile) ); + if (bzf == NULL) + { BZ_SETERR(BZ_MEM_ERROR); return NULL; }; + + BZ_SETERR(BZ_OK); + + bzf->initialisedOk = False; + bzf->handle = f; + bzf->bufN = 0; + bzf->writing = False; + bzf->strm.bzalloc = NULL; + bzf->strm.bzfree = NULL; + bzf->strm.opaque = NULL; + + while (nUnused > 0) { + bzf->buf[bzf->bufN] = *((UChar*)(unused)); bzf->bufN++; + unused = ((void*)( 1 + ((UChar*)(unused)) )); + nUnused--; + } + + ret = BZ2_bzDecompressInit ( &(bzf->strm), verbosity, small ); + if (ret != BZ_OK) + { BZ_SETERR(ret); free(bzf); return NULL; }; + + bzf->strm.avail_in = bzf->bufN; + bzf->strm.next_in = bzf->buf; + + bzf->initialisedOk = True; + return bzf; +} + + +/*---------------------------------------------------*/ +void BZ_API(BZ2_bzReadClose) ( int *bzerror, BZFILE *b ) +{ + bzFile* bzf = (bzFile*)b; + + BZ_SETERR(BZ_OK); + if (bzf == NULL) + { BZ_SETERR(BZ_OK); return; }; + + if (bzf->writing) + { BZ_SETERR(BZ_SEQUENCE_ERROR); return; }; + + if (bzf->initialisedOk) + (void)BZ2_bzDecompressEnd ( &(bzf->strm) ); + free ( bzf ); +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzRead) + ( int* bzerror, + BZFILE* b, + void* buf, + int len ) +{ + Int32 n, ret; + bzFile* bzf = (bzFile*)b; + + BZ_SETERR(BZ_OK); + + if (bzf == NULL || buf == NULL || len < 0) + { BZ_SETERR(BZ_PARAM_ERROR); return 0; }; + + if (bzf->writing) + { BZ_SETERR(BZ_SEQUENCE_ERROR); return 0; }; + + if (len == 0) + { BZ_SETERR(BZ_OK); return 0; }; + + bzf->strm.avail_out = len; + bzf->strm.next_out = buf; + + while (True) { + + if (ferror(bzf->handle)) + { BZ_SETERR(BZ_IO_ERROR); return 0; }; + + if (bzf->strm.avail_in == 0 && !myfeof(bzf->handle)) { + n = fread ( bzf->buf, sizeof(UChar), + BZ_MAX_UNUSED, bzf->handle ); + if (ferror(bzf->handle)) + { BZ_SETERR(BZ_IO_ERROR); return 0; }; + bzf->bufN = n; + bzf->strm.avail_in = bzf->bufN; + bzf->strm.next_in = bzf->buf; + } + + ret = BZ2_bzDecompress ( &(bzf->strm) ); + + if (ret != BZ_OK && ret != BZ_STREAM_END) + { BZ_SETERR(ret); return 0; }; + + if (ret == BZ_OK && myfeof(bzf->handle) && + bzf->strm.avail_in == 0 && bzf->strm.avail_out > 0) + { BZ_SETERR(BZ_UNEXPECTED_EOF); return 0; }; + + if (ret == BZ_STREAM_END) + { BZ_SETERR(BZ_STREAM_END); + return len - bzf->strm.avail_out; }; + if (bzf->strm.avail_out == 0) + { BZ_SETERR(BZ_OK); return len; }; + + } + + return 0; /*not reached*/ +} + + +/*---------------------------------------------------*/ +void BZ_API(BZ2_bzReadGetUnused) + ( int* bzerror, + BZFILE* b, + void** unused, + int* nUnused ) +{ + bzFile* bzf = (bzFile*)b; + if (bzf == NULL) + { BZ_SETERR(BZ_PARAM_ERROR); return; }; + if (bzf->lastErr != BZ_STREAM_END) + { BZ_SETERR(BZ_SEQUENCE_ERROR); return; }; + if (unused == NULL || nUnused == NULL) + { BZ_SETERR(BZ_PARAM_ERROR); return; }; + + BZ_SETERR(BZ_OK); + *nUnused = bzf->strm.avail_in; + *unused = bzf->strm.next_in; +} +#endif + + +/*---------------------------------------------------*/ +/*--- Misc convenience stuff ---*/ +/*---------------------------------------------------*/ + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzBuffToBuffCompress) + ( char* dest, + unsigned int* destLen, + char* source, + unsigned int sourceLen, + int blockSize100k, + int verbosity, + int workFactor ) +{ + bz_stream strm; + int ret; + + if (dest == NULL || destLen == NULL || + source == NULL || + blockSize100k < 1 || blockSize100k > 9 || + verbosity < 0 || verbosity > 4 || + workFactor < 0 || workFactor > 250) + return BZ_PARAM_ERROR; + + if (workFactor == 0) workFactor = 30; + strm.bzalloc = NULL; + strm.bzfree = NULL; + strm.opaque = NULL; + ret = BZ2_bzCompressInit ( &strm, blockSize100k, + verbosity, workFactor ); + if (ret != BZ_OK) return ret; + + strm.next_in = source; + strm.next_out = dest; + strm.avail_in = sourceLen; + strm.avail_out = *destLen; + + ret = BZ2_bzCompress ( &strm, BZ_FINISH ); + if (ret == BZ_FINISH_OK) goto output_overflow; + if (ret != BZ_STREAM_END) goto errhandler; + + /* normal termination */ + *destLen -= strm.avail_out; + BZ2_bzCompressEnd ( &strm ); + return BZ_OK; + + output_overflow: + BZ2_bzCompressEnd ( &strm ); + return BZ_OUTBUFF_FULL; + + errhandler: + BZ2_bzCompressEnd ( &strm ); + return ret; +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzBuffToBuffDecompress) + ( char* dest, + unsigned int* destLen, + char* source, + unsigned int sourceLen, + int small, + int verbosity ) +{ + bz_stream strm; + int ret; + + if (dest == NULL || destLen == NULL || + source == NULL || + (small != 0 && small != 1) || + verbosity < 0 || verbosity > 4) + return BZ_PARAM_ERROR; + + strm.bzalloc = NULL; + strm.bzfree = NULL; + strm.opaque = NULL; + ret = BZ2_bzDecompressInit ( &strm, verbosity, small ); + if (ret != BZ_OK) return ret; + + strm.next_in = source; + strm.next_out = dest; + strm.avail_in = sourceLen; + strm.avail_out = *destLen; + + ret = BZ2_bzDecompress ( &strm ); + if (ret == BZ_OK) goto output_overflow_or_eof; + if (ret != BZ_STREAM_END) goto errhandler; + + /* normal termination */ + *destLen -= strm.avail_out; + BZ2_bzDecompressEnd ( &strm ); + return BZ_OK; + + output_overflow_or_eof: + if (strm.avail_out > 0) { + BZ2_bzDecompressEnd ( &strm ); + return BZ_UNEXPECTED_EOF; + } else { + BZ2_bzDecompressEnd ( &strm ); + return BZ_OUTBUFF_FULL; + }; + + errhandler: + BZ2_bzDecompressEnd ( &strm ); + return ret; +} + + +/*---------------------------------------------------*/ +/*-- + Code contributed by Yoshioka Tsuneo (tsuneo@rr.iij4u.or.jp) + to support better zlib compatibility. + This code is not _officially_ part of libbzip2 (yet); + I haven't tested it, documented it, or considered the + threading-safeness of it. + If this code breaks, please contact both Yoshioka and me. +--*/ +/*---------------------------------------------------*/ + +/*---------------------------------------------------*/ +/*-- + return version like "0.9.5d, 4-Sept-1999". +--*/ +const char * BZ_API(BZ2_bzlibVersion)(void) +{ + return BZ_VERSION; +} + + +#ifndef BZ_NO_STDIO +/*---------------------------------------------------*/ + +#if defined(_WIN32) || defined(OS2) || defined(MSDOS) +# include +# include +# define SET_BINARY_MODE(file) setmode(fileno(file),O_BINARY) +#else +# define SET_BINARY_MODE(file) +#endif +static +BZFILE * bzopen_or_bzdopen + ( const char *path, /* no use when bzdopen */ + int fd, /* no use when bzdopen */ + const char *mode, + int open_mode) /* bzopen: 0, bzdopen:1 */ +{ + int bzerr; + char unused[BZ_MAX_UNUSED]; + int blockSize100k = 9; + int writing = 0; + char mode2[10] = ""; + FILE *fp = NULL; + BZFILE *bzfp = NULL; + int verbosity = 0; + int workFactor = 30; + int smallMode = 0; + int nUnused = 0; + + if (mode == NULL) return NULL; + while (*mode) { + switch (*mode) { + case 'r': + writing = 0; break; + case 'w': + writing = 1; break; + case 's': + smallMode = 1; break; + default: + if (isdigit((int)(*mode))) { + blockSize100k = *mode-BZ_HDR_0; + } + } + mode++; + } + strcat(mode2, writing ? "w" : "r" ); + strcat(mode2,"b"); /* binary mode */ + + if (open_mode==0) { + if (path==NULL || strcmp(path,"")==0) { + fp = (writing ? stdout : stdin); + SET_BINARY_MODE(fp); + } else { + fp = fopen(path,mode2); + } + } else { +#ifdef BZ_STRICT_ANSI + fp = NULL; +#else + fp = fdopen(fd,mode2); +#endif + } + if (fp == NULL) return NULL; + + if (writing) { + /* Guard against total chaos and anarchy -- JRS */ + if (blockSize100k < 1) blockSize100k = 1; + if (blockSize100k > 9) blockSize100k = 9; + bzfp = BZ2_bzWriteOpen(&bzerr,fp,blockSize100k, + verbosity,workFactor); + } else { + bzfp = BZ2_bzReadOpen(&bzerr,fp,verbosity,smallMode, + unused,nUnused); + } + if (bzfp == NULL) { + if (fp != stdin && fp != stdout) fclose(fp); + return NULL; + } + return bzfp; +} + + +/*---------------------------------------------------*/ +/*-- + open file for read or write. + ex) bzopen("file","w9") + case path="" or NULL => use stdin or stdout. +--*/ +BZFILE * BZ_API(BZ2_bzopen) + ( const char *path, + const char *mode ) +{ + return bzopen_or_bzdopen(path,-1,mode,/*bzopen*/0); +} + + +/*---------------------------------------------------*/ +BZFILE * BZ_API(BZ2_bzdopen) + ( int fd, + const char *mode ) +{ + return bzopen_or_bzdopen(NULL,fd,mode,/*bzdopen*/1); +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzread) (BZFILE* b, void* buf, int len ) +{ + int bzerr, nread; + if (((bzFile*)b)->lastErr == BZ_STREAM_END) return 0; + nread = BZ2_bzRead(&bzerr,b,buf,len); + if (bzerr == BZ_OK || bzerr == BZ_STREAM_END) { + return nread; + } else { + return -1; + } +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzwrite) (BZFILE* b, void* buf, int len ) +{ + int bzerr; + + BZ2_bzWrite(&bzerr,b,buf,len); + if(bzerr == BZ_OK){ + return len; + }else{ + return -1; + } +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzflush) (BZFILE *b) +{ + /* do nothing now... */ + return 0; +} + + +/*---------------------------------------------------*/ +void BZ_API(BZ2_bzclose) (BZFILE* b) +{ + int bzerr; + FILE *fp; + + if (b==NULL) {return;} + fp = ((bzFile *)b)->handle; + if(((bzFile*)b)->writing){ + BZ2_bzWriteClose(&bzerr,b,0,NULL,NULL); + if(bzerr != BZ_OK){ + BZ2_bzWriteClose(NULL,b,1,NULL,NULL); + } + }else{ + BZ2_bzReadClose(&bzerr,b); + } + if(fp!=stdin && fp!=stdout){ + fclose(fp); + } +} + + +/*---------------------------------------------------*/ +/*-- + return last error code +--*/ +static const char *bzerrorstrings[] = { + "OK" + ,"SEQUENCE_ERROR" + ,"PARAM_ERROR" + ,"MEM_ERROR" + ,"DATA_ERROR" + ,"DATA_ERROR_MAGIC" + ,"IO_ERROR" + ,"UNEXPECTED_EOF" + ,"OUTBUFF_FULL" + ,"CONFIG_ERROR" + ,"???" /* for future */ + ,"???" /* for future */ + ,"???" /* for future */ + ,"???" /* for future */ + ,"???" /* for future */ + ,"???" /* for future */ +}; + + +const char * BZ_API(BZ2_bzerror) (BZFILE *b, int *errnum) +{ + int err = ((bzFile *)b)->lastErr; + + if(err>0) err = 0; + *errnum = err; + return bzerrorstrings[err*-1]; +} +#endif + + +/*-------------------------------------------------------------*/ +/*--- end bzlib.c ---*/ +/*-------------------------------------------------------------*/ diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/bzlib.h b/ext/Compress-Raw-Bzip2/bzip2-src/bzlib.h new file mode 100644 index 0000000..c5b75d6 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/bzlib.h @@ -0,0 +1,282 @@ + +/*-------------------------------------------------------------*/ +/*--- Public header file for the library. ---*/ +/*--- bzlib.h ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#ifndef _BZLIB_H +#define _BZLIB_H + +#ifdef __cplusplus +extern "C" { +#endif + +#define BZ_RUN 0 +#define BZ_FLUSH 1 +#define BZ_FINISH 2 + +#define BZ_OK 0 +#define BZ_RUN_OK 1 +#define BZ_FLUSH_OK 2 +#define BZ_FINISH_OK 3 +#define BZ_STREAM_END 4 +#define BZ_SEQUENCE_ERROR (-1) +#define BZ_PARAM_ERROR (-2) +#define BZ_MEM_ERROR (-3) +#define BZ_DATA_ERROR (-4) +#define BZ_DATA_ERROR_MAGIC (-5) +#define BZ_IO_ERROR (-6) +#define BZ_UNEXPECTED_EOF (-7) +#define BZ_OUTBUFF_FULL (-8) +#define BZ_CONFIG_ERROR (-9) + +typedef + struct { + char *next_in; + unsigned int avail_in; + unsigned int total_in_lo32; + unsigned int total_in_hi32; + + char *next_out; + unsigned int avail_out; + unsigned int total_out_lo32; + unsigned int total_out_hi32; + + void *state; + + void *(*bzalloc)(void *,int,int); + void (*bzfree)(void *,void *); + void *opaque; + } + bz_stream; + + +#ifndef BZ_IMPORT +#define BZ_EXPORT +#endif + +#ifndef BZ_NO_STDIO +/* Need a definitition for FILE */ +#include +#endif + +#ifdef _WIN32 +# include +# ifdef small + /* windows.h define small to char */ +# undef small +# endif +# ifdef BZ_EXPORT +# define BZ_API(func) WINAPI func +# define BZ_EXTERN extern +# else + /* import windows dll dynamically */ +# define BZ_API(func) (WINAPI * func) +# define BZ_EXTERN +# endif +#else +# define BZ_API(func) func +# define BZ_EXTERN extern +#endif + + +/*-- Core (low-level) library functions --*/ + +BZ_EXTERN int BZ_API(BZ2_bzCompressInit) ( + bz_stream* strm, + int blockSize100k, + int verbosity, + int workFactor + ); + +BZ_EXTERN int BZ_API(BZ2_bzCompress) ( + bz_stream* strm, + int action + ); + +BZ_EXTERN int BZ_API(BZ2_bzCompressEnd) ( + bz_stream* strm + ); + +BZ_EXTERN int BZ_API(BZ2_bzDecompressInit) ( + bz_stream *strm, + int verbosity, + int small + ); + +BZ_EXTERN int BZ_API(BZ2_bzDecompress) ( + bz_stream* strm + ); + +BZ_EXTERN int BZ_API(BZ2_bzDecompressEnd) ( + bz_stream *strm + ); + + + +/*-- High(er) level library functions --*/ + +#ifndef BZ_NO_STDIO +#define BZ_MAX_UNUSED 5000 + +typedef void BZFILE; + +BZ_EXTERN BZFILE* BZ_API(BZ2_bzReadOpen) ( + int* bzerror, + FILE* f, + int verbosity, + int small, + void* unused, + int nUnused + ); + +BZ_EXTERN void BZ_API(BZ2_bzReadClose) ( + int* bzerror, + BZFILE* b + ); + +BZ_EXTERN void BZ_API(BZ2_bzReadGetUnused) ( + int* bzerror, + BZFILE* b, + void** unused, + int* nUnused + ); + +BZ_EXTERN int BZ_API(BZ2_bzRead) ( + int* bzerror, + BZFILE* b, + void* buf, + int len + ); + +BZ_EXTERN BZFILE* BZ_API(BZ2_bzWriteOpen) ( + int* bzerror, + FILE* f, + int blockSize100k, + int verbosity, + int workFactor + ); + +BZ_EXTERN void BZ_API(BZ2_bzWrite) ( + int* bzerror, + BZFILE* b, + void* buf, + int len + ); + +BZ_EXTERN void BZ_API(BZ2_bzWriteClose) ( + int* bzerror, + BZFILE* b, + int abandon, + unsigned int* nbytes_in, + unsigned int* nbytes_out + ); + +BZ_EXTERN void BZ_API(BZ2_bzWriteClose64) ( + int* bzerror, + BZFILE* b, + int abandon, + unsigned int* nbytes_in_lo32, + unsigned int* nbytes_in_hi32, + unsigned int* nbytes_out_lo32, + unsigned int* nbytes_out_hi32 + ); +#endif + + +/*-- Utility functions --*/ + +BZ_EXTERN int BZ_API(BZ2_bzBuffToBuffCompress) ( + char* dest, + unsigned int* destLen, + char* source, + unsigned int sourceLen, + int blockSize100k, + int verbosity, + int workFactor + ); + +BZ_EXTERN int BZ_API(BZ2_bzBuffToBuffDecompress) ( + char* dest, + unsigned int* destLen, + char* source, + unsigned int sourceLen, + int small, + int verbosity + ); + + +/*-- + Code contributed by Yoshioka Tsuneo (tsuneo@rr.iij4u.or.jp) + to support better zlib compatibility. + This code is not _officially_ part of libbzip2 (yet); + I haven't tested it, documented it, or considered the + threading-safeness of it. + If this code breaks, please contact both Yoshioka and me. +--*/ + +BZ_EXTERN const char * BZ_API(BZ2_bzlibVersion) ( + void + ); + +#ifndef BZ_NO_STDIO +BZ_EXTERN BZFILE * BZ_API(BZ2_bzopen) ( + const char *path, + const char *mode + ); + +BZ_EXTERN BZFILE * BZ_API(BZ2_bzdopen) ( + int fd, + const char *mode + ); + +BZ_EXTERN int BZ_API(BZ2_bzread) ( + BZFILE* b, + void* buf, + int len + ); + +BZ_EXTERN int BZ_API(BZ2_bzwrite) ( + BZFILE* b, + void* buf, + int len + ); + +BZ_EXTERN int BZ_API(BZ2_bzflush) ( + BZFILE* b + ); + +BZ_EXTERN void BZ_API(BZ2_bzclose) ( + BZFILE* b + ); + +BZ_EXTERN const char * BZ_API(BZ2_bzerror) ( + BZFILE *b, + int *errnum + ); +#endif + +#ifdef __cplusplus +} +#endif + +#endif + +/*-------------------------------------------------------------*/ +/*--- end bzlib.h ---*/ +/*-------------------------------------------------------------*/ diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/bzlib_private.h b/ext/Compress-Raw-Bzip2/bzip2-src/bzlib_private.h new file mode 100644 index 0000000..2342787 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/bzlib_private.h @@ -0,0 +1,509 @@ + +/*-------------------------------------------------------------*/ +/*--- Private header file for the library. ---*/ +/*--- bzlib_private.h ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#ifndef _BZLIB_PRIVATE_H +#define _BZLIB_PRIVATE_H + +#include + +#ifndef BZ_NO_STDIO +#include +#include +#include +#endif + +#include "bzlib.h" + + + +/*-- General stuff. --*/ + +#define BZ_VERSION "1.0.5, 10-Dec-2007" + +typedef char Char; +typedef unsigned char Bool; +typedef unsigned char UChar; +typedef int Int32; +typedef unsigned int UInt32; +typedef short Int16; +typedef unsigned short UInt16; + +#define True ((Bool)1) +#define False ((Bool)0) + +#ifndef __GNUC__ +#define __inline__ /* */ +#endif + +#ifndef BZ_NO_STDIO + +extern void BZ2_bz__AssertH__fail ( int errcode ); +#define AssertH(cond,errcode) \ + { if (!(cond)) BZ2_bz__AssertH__fail ( errcode ); } + +#if BZ_DEBUG +#define AssertD(cond,msg) \ + { if (!(cond)) { \ + fprintf ( stderr, \ + "\n\nlibbzip2(debug build): internal error\n\t%s\n", msg );\ + exit(1); \ + }} +#else +#define AssertD(cond,msg) /* */ +#endif + +#define VPrintf0(zf) \ + fprintf(stderr,zf) +#define VPrintf1(zf,za1) \ + fprintf(stderr,zf,za1) +#define VPrintf2(zf,za1,za2) \ + fprintf(stderr,zf,za1,za2) +#define VPrintf3(zf,za1,za2,za3) \ + fprintf(stderr,zf,za1,za2,za3) +#define VPrintf4(zf,za1,za2,za3,za4) \ + fprintf(stderr,zf,za1,za2,za3,za4) +#define VPrintf5(zf,za1,za2,za3,za4,za5) \ + fprintf(stderr,zf,za1,za2,za3,za4,za5) + +#else + +extern void bz_internal_error ( int errcode ); +#define AssertH(cond,errcode) \ + { if (!(cond)) bz_internal_error ( errcode ); } +#define AssertD(cond,msg) do { } while (0) +#define VPrintf0(zf) do { } while (0) +#define VPrintf1(zf,za1) do { } while (0) +#define VPrintf2(zf,za1,za2) do { } while (0) +#define VPrintf3(zf,za1,za2,za3) do { } while (0) +#define VPrintf4(zf,za1,za2,za3,za4) do { } while (0) +#define VPrintf5(zf,za1,za2,za3,za4,za5) do { } while (0) + +#endif + + +#define BZALLOC(nnn) (strm->bzalloc)(strm->opaque,(nnn),1) +#define BZFREE(ppp) (strm->bzfree)(strm->opaque,(ppp)) + + +/*-- Header bytes. --*/ + +#define BZ_HDR_B 0x42 /* 'B' */ +#define BZ_HDR_Z 0x5a /* 'Z' */ +#define BZ_HDR_h 0x68 /* 'h' */ +#define BZ_HDR_0 0x30 /* '0' */ + +/*-- Constants for the back end. --*/ + +#define BZ_MAX_ALPHA_SIZE 258 +#define BZ_MAX_CODE_LEN 23 + +#define BZ_RUNA 0 +#define BZ_RUNB 1 + +#define BZ_N_GROUPS 6 +#define BZ_G_SIZE 50 +#define BZ_N_ITERS 4 + +#define BZ_MAX_SELECTORS (2 + (900000 / BZ_G_SIZE)) + + + +/*-- Stuff for randomising repetitive blocks. --*/ + +extern Int32 BZ2_rNums[512]; + +#define BZ_RAND_DECLS \ + Int32 rNToGo; \ + Int32 rTPos \ + +#define BZ_RAND_INIT_MASK \ + s->rNToGo = 0; \ + s->rTPos = 0 \ + +#define BZ_RAND_MASK ((s->rNToGo == 1) ? 1 : 0) + +#define BZ_RAND_UPD_MASK \ + if (s->rNToGo == 0) { \ + s->rNToGo = BZ2_rNums[s->rTPos]; \ + s->rTPos++; \ + if (s->rTPos == 512) s->rTPos = 0; \ + } \ + s->rNToGo--; + + + +/*-- Stuff for doing CRCs. --*/ + +extern UInt32 BZ2_crc32Table[256]; + +#define BZ_INITIALISE_CRC(crcVar) \ +{ \ + crcVar = 0xffffffffL; \ +} + +#define BZ_FINALISE_CRC(crcVar) \ +{ \ + crcVar = ~(crcVar); \ +} + +#define BZ_UPDATE_CRC(crcVar,cha) \ +{ \ + crcVar = (crcVar << 8) ^ \ + BZ2_crc32Table[(crcVar >> 24) ^ \ + ((UChar)cha)]; \ +} + + + +/*-- States and modes for compression. --*/ + +#define BZ_M_IDLE 1 +#define BZ_M_RUNNING 2 +#define BZ_M_FLUSHING 3 +#define BZ_M_FINISHING 4 + +#define BZ_S_OUTPUT 1 +#define BZ_S_INPUT 2 + +#define BZ_N_RADIX 2 +#define BZ_N_QSORT 12 +#define BZ_N_SHELL 18 +#define BZ_N_OVERSHOOT (BZ_N_RADIX + BZ_N_QSORT + BZ_N_SHELL + 2) + + + + +/*-- Structure holding all the compression-side stuff. --*/ + +typedef + struct { + /* pointer back to the struct bz_stream */ + bz_stream* strm; + + /* mode this stream is in, and whether inputting */ + /* or outputting data */ + Int32 mode; + Int32 state; + + /* remembers avail_in when flush/finish requested */ + UInt32 avail_in_expect; + + /* for doing the block sorting */ + UInt32* arr1; + UInt32* arr2; + UInt32* ftab; + Int32 origPtr; + + /* aliases for arr1 and arr2 */ + UInt32* ptr; + UChar* block; + UInt16* mtfv; + UChar* zbits; + + /* for deciding when to use the fallback sorting algorithm */ + Int32 workFactor; + + /* run-length-encoding of the input */ + UInt32 state_in_ch; + Int32 state_in_len; + BZ_RAND_DECLS; + + /* input and output limits and current posns */ + Int32 nblock; + Int32 nblockMAX; + Int32 numZ; + Int32 state_out_pos; + + /* map of bytes used in block */ + Int32 nInUse; + Bool inUse[256]; + UChar unseqToSeq[256]; + + /* the buffer for bit stream creation */ + UInt32 bsBuff; + Int32 bsLive; + + /* block and combined CRCs */ + UInt32 blockCRC; + UInt32 combinedCRC; + + /* misc administratium */ + Int32 verbosity; + Int32 blockNo; + Int32 blockSize100k; + + /* stuff for coding the MTF values */ + Int32 nMTF; + Int32 mtfFreq [BZ_MAX_ALPHA_SIZE]; + UChar selector [BZ_MAX_SELECTORS]; + UChar selectorMtf[BZ_MAX_SELECTORS]; + + UChar len [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + Int32 code [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + Int32 rfreq [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + /* second dimension: only 3 needed; 4 makes index calculations faster */ + UInt32 len_pack[BZ_MAX_ALPHA_SIZE][4]; + + } + EState; + + + +/*-- externs for compression. --*/ + +extern void +BZ2_blockSort ( EState* ); + +extern void +BZ2_compressBlock ( EState*, Bool ); + +extern void +BZ2_bsInitWrite ( EState* ); + +extern void +BZ2_hbAssignCodes ( Int32*, UChar*, Int32, Int32, Int32 ); + +extern void +BZ2_hbMakeCodeLengths ( UChar*, Int32*, Int32, Int32 ); + + + +/*-- states for decompression. --*/ + +#define BZ_X_IDLE 1 +#define BZ_X_OUTPUT 2 + +#define BZ_X_MAGIC_1 10 +#define BZ_X_MAGIC_2 11 +#define BZ_X_MAGIC_3 12 +#define BZ_X_MAGIC_4 13 +#define BZ_X_BLKHDR_1 14 +#define BZ_X_BLKHDR_2 15 +#define BZ_X_BLKHDR_3 16 +#define BZ_X_BLKHDR_4 17 +#define BZ_X_BLKHDR_5 18 +#define BZ_X_BLKHDR_6 19 +#define BZ_X_BCRC_1 20 +#define BZ_X_BCRC_2 21 +#define BZ_X_BCRC_3 22 +#define BZ_X_BCRC_4 23 +#define BZ_X_RANDBIT 24 +#define BZ_X_ORIGPTR_1 25 +#define BZ_X_ORIGPTR_2 26 +#define BZ_X_ORIGPTR_3 27 +#define BZ_X_MAPPING_1 28 +#define BZ_X_MAPPING_2 29 +#define BZ_X_SELECTOR_1 30 +#define BZ_X_SELECTOR_2 31 +#define BZ_X_SELECTOR_3 32 +#define BZ_X_CODING_1 33 +#define BZ_X_CODING_2 34 +#define BZ_X_CODING_3 35 +#define BZ_X_MTF_1 36 +#define BZ_X_MTF_2 37 +#define BZ_X_MTF_3 38 +#define BZ_X_MTF_4 39 +#define BZ_X_MTF_5 40 +#define BZ_X_MTF_6 41 +#define BZ_X_ENDHDR_2 42 +#define BZ_X_ENDHDR_3 43 +#define BZ_X_ENDHDR_4 44 +#define BZ_X_ENDHDR_5 45 +#define BZ_X_ENDHDR_6 46 +#define BZ_X_CCRC_1 47 +#define BZ_X_CCRC_2 48 +#define BZ_X_CCRC_3 49 +#define BZ_X_CCRC_4 50 + + + +/*-- Constants for the fast MTF decoder. --*/ + +#define MTFA_SIZE 4096 +#define MTFL_SIZE 16 + + + +/*-- Structure holding all the decompression-side stuff. --*/ + +typedef + struct { + /* pointer back to the struct bz_stream */ + bz_stream* strm; + + /* state indicator for this stream */ + Int32 state; + + /* for doing the final run-length decoding */ + UChar state_out_ch; + Int32 state_out_len; + Bool blockRandomised; + BZ_RAND_DECLS; + + /* the buffer for bit stream reading */ + UInt32 bsBuff; + Int32 bsLive; + + /* misc administratium */ + Int32 blockSize100k; + Bool smallDecompress; + Int32 currBlockNo; + Int32 verbosity; + + /* for undoing the Burrows-Wheeler transform */ + Int32 origPtr; + UInt32 tPos; + Int32 k0; + Int32 unzftab[256]; + Int32 nblock_used; + Int32 cftab[257]; + Int32 cftabCopy[257]; + + /* for undoing the Burrows-Wheeler transform (FAST) */ + UInt32 *tt; + + /* for undoing the Burrows-Wheeler transform (SMALL) */ + UInt16 *ll16; + UChar *ll4; + + /* stored and calculated CRCs */ + UInt32 storedBlockCRC; + UInt32 storedCombinedCRC; + UInt32 calculatedBlockCRC; + UInt32 calculatedCombinedCRC; + + /* map of bytes used in block */ + Int32 nInUse; + Bool inUse[256]; + Bool inUse16[16]; + UChar seqToUnseq[256]; + + /* for decoding the MTF values */ + UChar mtfa [MTFA_SIZE]; + Int32 mtfbase[256 / MTFL_SIZE]; + UChar selector [BZ_MAX_SELECTORS]; + UChar selectorMtf[BZ_MAX_SELECTORS]; + UChar len [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + + Int32 limit [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + Int32 base [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + Int32 perm [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + Int32 minLens[BZ_N_GROUPS]; + + /* save area for scalars in the main decompress code */ + Int32 save_i; + Int32 save_j; + Int32 save_t; + Int32 save_alphaSize; + Int32 save_nGroups; + Int32 save_nSelectors; + Int32 save_EOB; + Int32 save_groupNo; + Int32 save_groupPos; + Int32 save_nextSym; + Int32 save_nblockMAX; + Int32 save_nblock; + Int32 save_es; + Int32 save_N; + Int32 save_curr; + Int32 save_zt; + Int32 save_zn; + Int32 save_zvec; + Int32 save_zj; + Int32 save_gSel; + Int32 save_gMinlen; + Int32* save_gLimit; + Int32* save_gBase; + Int32* save_gPerm; + + } + DState; + + + +/*-- Macros for decompression. --*/ + +#define BZ_GET_FAST(cccc) \ + /* c_tPos is unsigned, hence test < 0 is pointless. */ \ + if (s->tPos >= (UInt32)100000 * (UInt32)s->blockSize100k) return True; \ + s->tPos = s->tt[s->tPos]; \ + cccc = (UChar)(s->tPos & 0xff); \ + s->tPos >>= 8; + +#define BZ_GET_FAST_C(cccc) \ + /* c_tPos is unsigned, hence test < 0 is pointless. */ \ + if (c_tPos >= (UInt32)100000 * (UInt32)ro_blockSize100k) return True; \ + c_tPos = c_tt[c_tPos]; \ + cccc = (UChar)(c_tPos & 0xff); \ + c_tPos >>= 8; + +#define SET_LL4(i,n) \ + { if (((i) & 0x1) == 0) \ + s->ll4[(i) >> 1] = (s->ll4[(i) >> 1] & 0xf0) | (n); else \ + s->ll4[(i) >> 1] = (s->ll4[(i) >> 1] & 0x0f) | ((n) << 4); \ + } + +#define GET_LL4(i) \ + ((((UInt32)(s->ll4[(i) >> 1])) >> (((i) << 2) & 0x4)) & 0xF) + +#define SET_LL(i,n) \ + { s->ll16[i] = (UInt16)(n & 0x0000ffff); \ + SET_LL4(i, n >> 16); \ + } + +#define GET_LL(i) \ + (((UInt32)s->ll16[i]) | (GET_LL4(i) << 16)) + +#define BZ_GET_SMALL(cccc) \ + /* c_tPos is unsigned, hence test < 0 is pointless. */ \ + if (s->tPos >= (UInt32)100000 * (UInt32)s->blockSize100k) return True; \ + cccc = BZ2_indexIntoF ( s->tPos, s->cftab ); \ + s->tPos = GET_LL(s->tPos); + + +/*-- externs for decompression. --*/ + +extern Int32 +BZ2_indexIntoF ( Int32, Int32* ); + +extern Int32 +BZ2_decompress ( DState* ); + +extern void +BZ2_hbCreateDecodeTables ( Int32*, Int32*, Int32*, UChar*, + Int32, Int32, Int32 ); + + +#endif + + +/*-- BZ_NO_STDIO seems to make NULL disappear on some platforms. --*/ + +#ifdef BZ_NO_STDIO +#ifndef NULL +#define NULL 0 +#endif +#endif + + +/*-------------------------------------------------------------*/ +/*--- end bzlib_private.h ---*/ +/*-------------------------------------------------------------*/ diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/compress.c b/ext/Compress-Raw-Bzip2/bzip2-src/compress.c new file mode 100644 index 0000000..8c80a07 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/compress.c @@ -0,0 +1,672 @@ + +/*-------------------------------------------------------------*/ +/*--- Compression machinery (not incl block sorting) ---*/ +/*--- compress.c ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +/* CHANGES + 0.9.0 -- original version. + 0.9.0a/b -- no changes in this file. + 0.9.0c -- changed setting of nGroups in sendMTFValues() + so as to do a bit better on small files +*/ + +#include "bzlib_private.h" + + +/*---------------------------------------------------*/ +/*--- Bit stream I/O ---*/ +/*---------------------------------------------------*/ + +/*---------------------------------------------------*/ +void BZ2_bsInitWrite ( EState* s ) +{ + s->bsLive = 0; + s->bsBuff = 0; +} + + +/*---------------------------------------------------*/ +static +void bsFinishWrite ( EState* s ) +{ + while (s->bsLive > 0) { + s->zbits[s->numZ] = (UChar)(s->bsBuff >> 24); + s->numZ++; + s->bsBuff <<= 8; + s->bsLive -= 8; + } +} + + +/*---------------------------------------------------*/ +#define bsNEEDW(nz) \ +{ \ + while (s->bsLive >= 8) { \ + s->zbits[s->numZ] \ + = (UChar)(s->bsBuff >> 24); \ + s->numZ++; \ + s->bsBuff <<= 8; \ + s->bsLive -= 8; \ + } \ +} + + +/*---------------------------------------------------*/ +static +__inline__ +void bsW ( EState* s, Int32 n, UInt32 v ) +{ + bsNEEDW ( n ); + s->bsBuff |= (v << (32 - s->bsLive - n)); + s->bsLive += n; +} + + +/*---------------------------------------------------*/ +static +void bsPutUInt32 ( EState* s, UInt32 u ) +{ + bsW ( s, 8, (u >> 24) & 0xffL ); + bsW ( s, 8, (u >> 16) & 0xffL ); + bsW ( s, 8, (u >> 8) & 0xffL ); + bsW ( s, 8, u & 0xffL ); +} + + +/*---------------------------------------------------*/ +static +void bsPutUChar ( EState* s, UChar c ) +{ + bsW( s, 8, (UInt32)c ); +} + + +/*---------------------------------------------------*/ +/*--- The back end proper ---*/ +/*---------------------------------------------------*/ + +/*---------------------------------------------------*/ +static +void makeMaps_e ( EState* s ) +{ + Int32 i; + s->nInUse = 0; + for (i = 0; i < 256; i++) + if (s->inUse[i]) { + s->unseqToSeq[i] = s->nInUse; + s->nInUse++; + } +} + + +/*---------------------------------------------------*/ +static +void generateMTFValues ( EState* s ) +{ + UChar yy[256]; + Int32 i, j; + Int32 zPend; + Int32 wr; + Int32 EOB; + + /* + After sorting (eg, here), + s->arr1 [ 0 .. s->nblock-1 ] holds sorted order, + and + ((UChar*)s->arr2) [ 0 .. s->nblock-1 ] + holds the original block data. + + The first thing to do is generate the MTF values, + and put them in + ((UInt16*)s->arr1) [ 0 .. s->nblock-1 ]. + Because there are strictly fewer or equal MTF values + than block values, ptr values in this area are overwritten + with MTF values only when they are no longer needed. + + The final compressed bitstream is generated into the + area starting at + (UChar*) (&((UChar*)s->arr2)[s->nblock]) + + These storage aliases are set up in bzCompressInit(), + except for the last one, which is arranged in + compressBlock(). + */ + UInt32* ptr = s->ptr; + UChar* block = s->block; + UInt16* mtfv = s->mtfv; + + makeMaps_e ( s ); + EOB = s->nInUse+1; + + for (i = 0; i <= EOB; i++) s->mtfFreq[i] = 0; + + wr = 0; + zPend = 0; + for (i = 0; i < s->nInUse; i++) yy[i] = (UChar) i; + + for (i = 0; i < s->nblock; i++) { + UChar ll_i; + AssertD ( wr <= i, "generateMTFValues(1)" ); + j = ptr[i]-1; if (j < 0) j += s->nblock; + ll_i = s->unseqToSeq[block[j]]; + AssertD ( ll_i < s->nInUse, "generateMTFValues(2a)" ); + + if (yy[0] == ll_i) { + zPend++; + } else { + + if (zPend > 0) { + zPend--; + while (True) { + if (zPend & 1) { + mtfv[wr] = BZ_RUNB; wr++; + s->mtfFreq[BZ_RUNB]++; + } else { + mtfv[wr] = BZ_RUNA; wr++; + s->mtfFreq[BZ_RUNA]++; + } + if (zPend < 2) break; + zPend = (zPend - 2) / 2; + }; + zPend = 0; + } + { + register UChar rtmp; + register UChar* ryy_j; + register UChar rll_i; + rtmp = yy[1]; + yy[1] = yy[0]; + ryy_j = &(yy[1]); + rll_i = ll_i; + while ( rll_i != rtmp ) { + register UChar rtmp2; + ryy_j++; + rtmp2 = rtmp; + rtmp = *ryy_j; + *ryy_j = rtmp2; + }; + yy[0] = rtmp; + j = ryy_j - &(yy[0]); + mtfv[wr] = j+1; wr++; s->mtfFreq[j+1]++; + } + + } + } + + if (zPend > 0) { + zPend--; + while (True) { + if (zPend & 1) { + mtfv[wr] = BZ_RUNB; wr++; + s->mtfFreq[BZ_RUNB]++; + } else { + mtfv[wr] = BZ_RUNA; wr++; + s->mtfFreq[BZ_RUNA]++; + } + if (zPend < 2) break; + zPend = (zPend - 2) / 2; + }; + zPend = 0; + } + + mtfv[wr] = EOB; wr++; s->mtfFreq[EOB]++; + + s->nMTF = wr; +} + + +/*---------------------------------------------------*/ +#define BZ_LESSER_ICOST 0 +#define BZ_GREATER_ICOST 15 + +static +void sendMTFValues ( EState* s ) +{ + Int32 v, t, i, j, gs, ge, totc, bt, bc, iter; + Int32 nSelectors, alphaSize, minLen, maxLen, selCtr; + Int32 nGroups, nBytes; + + /*-- + UChar len [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + is a global since the decoder also needs it. + + Int32 code[BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + Int32 rfreq[BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + are also globals only used in this proc. + Made global to keep stack frame size small. + --*/ + + + UInt16 cost[BZ_N_GROUPS]; + Int32 fave[BZ_N_GROUPS]; + + UInt16* mtfv = s->mtfv; + + if (s->verbosity >= 3) + VPrintf3( " %d in block, %d after MTF & 1-2 coding, " + "%d+2 syms in use\n", + s->nblock, s->nMTF, s->nInUse ); + + alphaSize = s->nInUse+2; + for (t = 0; t < BZ_N_GROUPS; t++) + for (v = 0; v < alphaSize; v++) + s->len[t][v] = BZ_GREATER_ICOST; + + /*--- Decide how many coding tables to use ---*/ + AssertH ( s->nMTF > 0, 3001 ); + if (s->nMTF < 200) nGroups = 2; else + if (s->nMTF < 600) nGroups = 3; else + if (s->nMTF < 1200) nGroups = 4; else + if (s->nMTF < 2400) nGroups = 5; else + nGroups = 6; + + /*--- Generate an initial set of coding tables ---*/ + { + Int32 nPart, remF, tFreq, aFreq; + + nPart = nGroups; + remF = s->nMTF; + gs = 0; + while (nPart > 0) { + tFreq = remF / nPart; + ge = gs-1; + aFreq = 0; + while (aFreq < tFreq && ge < alphaSize-1) { + ge++; + aFreq += s->mtfFreq[ge]; + } + + if (ge > gs + && nPart != nGroups && nPart != 1 + && ((nGroups-nPart) % 2 == 1)) { + aFreq -= s->mtfFreq[ge]; + ge--; + } + + if (s->verbosity >= 3) + VPrintf5( " initial group %d, [%d .. %d], " + "has %d syms (%4.1f%%)\n", + nPart, gs, ge, aFreq, + (100.0 * (float)aFreq) / (float)(s->nMTF) ); + + for (v = 0; v < alphaSize; v++) + if (v >= gs && v <= ge) + s->len[nPart-1][v] = BZ_LESSER_ICOST; else + s->len[nPart-1][v] = BZ_GREATER_ICOST; + + nPart--; + gs = ge+1; + remF -= aFreq; + } + } + + /*--- + Iterate up to BZ_N_ITERS times to improve the tables. + ---*/ + for (iter = 0; iter < BZ_N_ITERS; iter++) { + + for (t = 0; t < nGroups; t++) fave[t] = 0; + + for (t = 0; t < nGroups; t++) + for (v = 0; v < alphaSize; v++) + s->rfreq[t][v] = 0; + + /*--- + Set up an auxiliary length table which is used to fast-track + the common case (nGroups == 6). + ---*/ + if (nGroups == 6) { + for (v = 0; v < alphaSize; v++) { + s->len_pack[v][0] = (s->len[1][v] << 16) | s->len[0][v]; + s->len_pack[v][1] = (s->len[3][v] << 16) | s->len[2][v]; + s->len_pack[v][2] = (s->len[5][v] << 16) | s->len[4][v]; + } + } + + nSelectors = 0; + totc = 0; + gs = 0; + while (True) { + + /*--- Set group start & end marks. --*/ + if (gs >= s->nMTF) break; + ge = gs + BZ_G_SIZE - 1; + if (ge >= s->nMTF) ge = s->nMTF-1; + + /*-- + Calculate the cost of this group as coded + by each of the coding tables. + --*/ + for (t = 0; t < nGroups; t++) cost[t] = 0; + + if (nGroups == 6 && 50 == ge-gs+1) { + /*--- fast track the common case ---*/ + register UInt32 cost01, cost23, cost45; + register UInt16 icv; + cost01 = cost23 = cost45 = 0; + +# define BZ_ITER(nn) \ + icv = mtfv[gs+(nn)]; \ + cost01 += s->len_pack[icv][0]; \ + cost23 += s->len_pack[icv][1]; \ + cost45 += s->len_pack[icv][2]; \ + + BZ_ITER(0); BZ_ITER(1); BZ_ITER(2); BZ_ITER(3); BZ_ITER(4); + BZ_ITER(5); BZ_ITER(6); BZ_ITER(7); BZ_ITER(8); BZ_ITER(9); + BZ_ITER(10); BZ_ITER(11); BZ_ITER(12); BZ_ITER(13); BZ_ITER(14); + BZ_ITER(15); BZ_ITER(16); BZ_ITER(17); BZ_ITER(18); BZ_ITER(19); + BZ_ITER(20); BZ_ITER(21); BZ_ITER(22); BZ_ITER(23); BZ_ITER(24); + BZ_ITER(25); BZ_ITER(26); BZ_ITER(27); BZ_ITER(28); BZ_ITER(29); + BZ_ITER(30); BZ_ITER(31); BZ_ITER(32); BZ_ITER(33); BZ_ITER(34); + BZ_ITER(35); BZ_ITER(36); BZ_ITER(37); BZ_ITER(38); BZ_ITER(39); + BZ_ITER(40); BZ_ITER(41); BZ_ITER(42); BZ_ITER(43); BZ_ITER(44); + BZ_ITER(45); BZ_ITER(46); BZ_ITER(47); BZ_ITER(48); BZ_ITER(49); + +# undef BZ_ITER + + cost[0] = cost01 & 0xffff; cost[1] = cost01 >> 16; + cost[2] = cost23 & 0xffff; cost[3] = cost23 >> 16; + cost[4] = cost45 & 0xffff; cost[5] = cost45 >> 16; + + } else { + /*--- slow version which correctly handles all situations ---*/ + for (i = gs; i <= ge; i++) { + UInt16 icv = mtfv[i]; + for (t = 0; t < nGroups; t++) cost[t] += s->len[t][icv]; + } + } + + /*-- + Find the coding table which is best for this group, + and record its identity in the selector table. + --*/ + bc = 999999999; bt = -1; + for (t = 0; t < nGroups; t++) + if (cost[t] < bc) { bc = cost[t]; bt = t; }; + totc += bc; + fave[bt]++; + s->selector[nSelectors] = bt; + nSelectors++; + + /*-- + Increment the symbol frequencies for the selected table. + --*/ + if (nGroups == 6 && 50 == ge-gs+1) { + /*--- fast track the common case ---*/ + +# define BZ_ITUR(nn) s->rfreq[bt][ mtfv[gs+(nn)] ]++ + + BZ_ITUR(0); BZ_ITUR(1); BZ_ITUR(2); BZ_ITUR(3); BZ_ITUR(4); + BZ_ITUR(5); BZ_ITUR(6); BZ_ITUR(7); BZ_ITUR(8); BZ_ITUR(9); + BZ_ITUR(10); BZ_ITUR(11); BZ_ITUR(12); BZ_ITUR(13); BZ_ITUR(14); + BZ_ITUR(15); BZ_ITUR(16); BZ_ITUR(17); BZ_ITUR(18); BZ_ITUR(19); + BZ_ITUR(20); BZ_ITUR(21); BZ_ITUR(22); BZ_ITUR(23); BZ_ITUR(24); + BZ_ITUR(25); BZ_ITUR(26); BZ_ITUR(27); BZ_ITUR(28); BZ_ITUR(29); + BZ_ITUR(30); BZ_ITUR(31); BZ_ITUR(32); BZ_ITUR(33); BZ_ITUR(34); + BZ_ITUR(35); BZ_ITUR(36); BZ_ITUR(37); BZ_ITUR(38); BZ_ITUR(39); + BZ_ITUR(40); BZ_ITUR(41); BZ_ITUR(42); BZ_ITUR(43); BZ_ITUR(44); + BZ_ITUR(45); BZ_ITUR(46); BZ_ITUR(47); BZ_ITUR(48); BZ_ITUR(49); + +# undef BZ_ITUR + + } else { + /*--- slow version which correctly handles all situations ---*/ + for (i = gs; i <= ge; i++) + s->rfreq[bt][ mtfv[i] ]++; + } + + gs = ge+1; + } + if (s->verbosity >= 3) { + VPrintf2 ( " pass %d: size is %d, grp uses are ", + iter+1, totc/8 ); + for (t = 0; t < nGroups; t++) + VPrintf1 ( "%d ", fave[t] ); + VPrintf0 ( "\n" ); + } + + /*-- + Recompute the tables based on the accumulated frequencies. + --*/ + /* maxLen was changed from 20 to 17 in bzip2-1.0.3. See + comment in huffman.c for details. */ + for (t = 0; t < nGroups; t++) + BZ2_hbMakeCodeLengths ( &(s->len[t][0]), &(s->rfreq[t][0]), + alphaSize, 17 /*20*/ ); + } + + + AssertH( nGroups < 8, 3002 ); + AssertH( nSelectors < 32768 && + nSelectors <= (2 + (900000 / BZ_G_SIZE)), + 3003 ); + + + /*--- Compute MTF values for the selectors. ---*/ + { + UChar pos[BZ_N_GROUPS], ll_i, tmp2, tmp; + for (i = 0; i < nGroups; i++) pos[i] = i; + for (i = 0; i < nSelectors; i++) { + ll_i = s->selector[i]; + j = 0; + tmp = pos[j]; + while ( ll_i != tmp ) { + j++; + tmp2 = tmp; + tmp = pos[j]; + pos[j] = tmp2; + }; + pos[0] = tmp; + s->selectorMtf[i] = j; + } + }; + + /*--- Assign actual codes for the tables. --*/ + for (t = 0; t < nGroups; t++) { + minLen = 32; + maxLen = 0; + for (i = 0; i < alphaSize; i++) { + if (s->len[t][i] > maxLen) maxLen = s->len[t][i]; + if (s->len[t][i] < minLen) minLen = s->len[t][i]; + } + AssertH ( !(maxLen > 17 /*20*/ ), 3004 ); + AssertH ( !(minLen < 1), 3005 ); + BZ2_hbAssignCodes ( &(s->code[t][0]), &(s->len[t][0]), + minLen, maxLen, alphaSize ); + } + + /*--- Transmit the mapping table. ---*/ + { + Bool inUse16[16]; + for (i = 0; i < 16; i++) { + inUse16[i] = False; + for (j = 0; j < 16; j++) + if (s->inUse[i * 16 + j]) inUse16[i] = True; + } + + nBytes = s->numZ; + for (i = 0; i < 16; i++) + if (inUse16[i]) bsW(s,1,1); else bsW(s,1,0); + + for (i = 0; i < 16; i++) + if (inUse16[i]) + for (j = 0; j < 16; j++) { + if (s->inUse[i * 16 + j]) bsW(s,1,1); else bsW(s,1,0); + } + + if (s->verbosity >= 3) + VPrintf1( " bytes: mapping %d, ", s->numZ-nBytes ); + } + + /*--- Now the selectors. ---*/ + nBytes = s->numZ; + bsW ( s, 3, nGroups ); + bsW ( s, 15, nSelectors ); + for (i = 0; i < nSelectors; i++) { + for (j = 0; j < s->selectorMtf[i]; j++) bsW(s,1,1); + bsW(s,1,0); + } + if (s->verbosity >= 3) + VPrintf1( "selectors %d, ", s->numZ-nBytes ); + + /*--- Now the coding tables. ---*/ + nBytes = s->numZ; + + for (t = 0; t < nGroups; t++) { + Int32 curr = s->len[t][0]; + bsW ( s, 5, curr ); + for (i = 0; i < alphaSize; i++) { + while (curr < s->len[t][i]) { bsW(s,2,2); curr++; /* 10 */ }; + while (curr > s->len[t][i]) { bsW(s,2,3); curr--; /* 11 */ }; + bsW ( s, 1, 0 ); + } + } + + if (s->verbosity >= 3) + VPrintf1 ( "code lengths %d, ", s->numZ-nBytes ); + + /*--- And finally, the block data proper ---*/ + nBytes = s->numZ; + selCtr = 0; + gs = 0; + while (True) { + if (gs >= s->nMTF) break; + ge = gs + BZ_G_SIZE - 1; + if (ge >= s->nMTF) ge = s->nMTF-1; + AssertH ( s->selector[selCtr] < nGroups, 3006 ); + + if (nGroups == 6 && 50 == ge-gs+1) { + /*--- fast track the common case ---*/ + UInt16 mtfv_i; + UChar* s_len_sel_selCtr + = &(s->len[s->selector[selCtr]][0]); + Int32* s_code_sel_selCtr + = &(s->code[s->selector[selCtr]][0]); + +# define BZ_ITAH(nn) \ + mtfv_i = mtfv[gs+(nn)]; \ + bsW ( s, \ + s_len_sel_selCtr[mtfv_i], \ + s_code_sel_selCtr[mtfv_i] ) + + BZ_ITAH(0); BZ_ITAH(1); BZ_ITAH(2); BZ_ITAH(3); BZ_ITAH(4); + BZ_ITAH(5); BZ_ITAH(6); BZ_ITAH(7); BZ_ITAH(8); BZ_ITAH(9); + BZ_ITAH(10); BZ_ITAH(11); BZ_ITAH(12); BZ_ITAH(13); BZ_ITAH(14); + BZ_ITAH(15); BZ_ITAH(16); BZ_ITAH(17); BZ_ITAH(18); BZ_ITAH(19); + BZ_ITAH(20); BZ_ITAH(21); BZ_ITAH(22); BZ_ITAH(23); BZ_ITAH(24); + BZ_ITAH(25); BZ_ITAH(26); BZ_ITAH(27); BZ_ITAH(28); BZ_ITAH(29); + BZ_ITAH(30); BZ_ITAH(31); BZ_ITAH(32); BZ_ITAH(33); BZ_ITAH(34); + BZ_ITAH(35); BZ_ITAH(36); BZ_ITAH(37); BZ_ITAH(38); BZ_ITAH(39); + BZ_ITAH(40); BZ_ITAH(41); BZ_ITAH(42); BZ_ITAH(43); BZ_ITAH(44); + BZ_ITAH(45); BZ_ITAH(46); BZ_ITAH(47); BZ_ITAH(48); BZ_ITAH(49); + +# undef BZ_ITAH + + } else { + /*--- slow version which correctly handles all situations ---*/ + for (i = gs; i <= ge; i++) { + bsW ( s, + s->len [s->selector[selCtr]] [mtfv[i]], + s->code [s->selector[selCtr]] [mtfv[i]] ); + } + } + + + gs = ge+1; + selCtr++; + } + AssertH( selCtr == nSelectors, 3007 ); + + if (s->verbosity >= 3) + VPrintf1( "codes %d\n", s->numZ-nBytes ); +} + + +/*---------------------------------------------------*/ +void BZ2_compressBlock ( EState* s, Bool is_last_block ) +{ + if (s->nblock > 0) { + + BZ_FINALISE_CRC ( s->blockCRC ); + s->combinedCRC = (s->combinedCRC << 1) | (s->combinedCRC >> 31); + s->combinedCRC ^= s->blockCRC; + if (s->blockNo > 1) s->numZ = 0; + + if (s->verbosity >= 2) + VPrintf4( " block %d: crc = 0x%08x, " + "combined CRC = 0x%08x, size = %d\n", + s->blockNo, s->blockCRC, s->combinedCRC, s->nblock ); + + BZ2_blockSort ( s ); + } + + s->zbits = (UChar*) (&((UChar*)s->arr2)[s->nblock]); + + /*-- If this is the first block, create the stream header. --*/ + if (s->blockNo == 1) { + BZ2_bsInitWrite ( s ); + bsPutUChar ( s, BZ_HDR_B ); + bsPutUChar ( s, BZ_HDR_Z ); + bsPutUChar ( s, BZ_HDR_h ); + bsPutUChar ( s, (UChar)(BZ_HDR_0 + s->blockSize100k) ); + } + + if (s->nblock > 0) { + + bsPutUChar ( s, 0x31 ); bsPutUChar ( s, 0x41 ); + bsPutUChar ( s, 0x59 ); bsPutUChar ( s, 0x26 ); + bsPutUChar ( s, 0x53 ); bsPutUChar ( s, 0x59 ); + + /*-- Now the block's CRC, so it is in a known place. --*/ + bsPutUInt32 ( s, s->blockCRC ); + + /*-- + Now a single bit indicating (non-)randomisation. + As of version 0.9.5, we use a better sorting algorithm + which makes randomisation unnecessary. So always set + the randomised bit to 'no'. Of course, the decoder + still needs to be able to handle randomised blocks + so as to maintain backwards compatibility with + older versions of bzip2. + --*/ + bsW(s,1,0); + + bsW ( s, 24, s->origPtr ); + generateMTFValues ( s ); + sendMTFValues ( s ); + } + + + /*-- If this is the last block, add the stream trailer. --*/ + if (is_last_block) { + + bsPutUChar ( s, 0x17 ); bsPutUChar ( s, 0x72 ); + bsPutUChar ( s, 0x45 ); bsPutUChar ( s, 0x38 ); + bsPutUChar ( s, 0x50 ); bsPutUChar ( s, 0x90 ); + bsPutUInt32 ( s, s->combinedCRC ); + if (s->verbosity >= 2) + VPrintf1( " final combined CRC = 0x%08x\n ", s->combinedCRC ); + bsFinishWrite ( s ); + } +} + + +/*-------------------------------------------------------------*/ +/*--- end compress.c ---*/ +/*-------------------------------------------------------------*/ diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/crctable.c b/ext/Compress-Raw-Bzip2/bzip2-src/crctable.c new file mode 100644 index 0000000..215687b --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/crctable.c @@ -0,0 +1,104 @@ + +/*-------------------------------------------------------------*/ +/*--- Table for doing CRCs ---*/ +/*--- crctable.c ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#include "bzlib_private.h" + +/*-- + I think this is an implementation of the AUTODIN-II, + Ethernet & FDDI 32-bit CRC standard. Vaguely derived + from code by Rob Warnock, in Section 51 of the + comp.compression FAQ. +--*/ + +UInt32 BZ2_crc32Table[256] = { + + /*-- Ugly, innit? --*/ + + 0x00000000L, 0x04c11db7L, 0x09823b6eL, 0x0d4326d9L, + 0x130476dcL, 0x17c56b6bL, 0x1a864db2L, 0x1e475005L, + 0x2608edb8L, 0x22c9f00fL, 0x2f8ad6d6L, 0x2b4bcb61L, + 0x350c9b64L, 0x31cd86d3L, 0x3c8ea00aL, 0x384fbdbdL, + 0x4c11db70L, 0x48d0c6c7L, 0x4593e01eL, 0x4152fda9L, + 0x5f15adacL, 0x5bd4b01bL, 0x569796c2L, 0x52568b75L, + 0x6a1936c8L, 0x6ed82b7fL, 0x639b0da6L, 0x675a1011L, + 0x791d4014L, 0x7ddc5da3L, 0x709f7b7aL, 0x745e66cdL, + 0x9823b6e0L, 0x9ce2ab57L, 0x91a18d8eL, 0x95609039L, + 0x8b27c03cL, 0x8fe6dd8bL, 0x82a5fb52L, 0x8664e6e5L, + 0xbe2b5b58L, 0xbaea46efL, 0xb7a96036L, 0xb3687d81L, + 0xad2f2d84L, 0xa9ee3033L, 0xa4ad16eaL, 0xa06c0b5dL, + 0xd4326d90L, 0xd0f37027L, 0xddb056feL, 0xd9714b49L, + 0xc7361b4cL, 0xc3f706fbL, 0xceb42022L, 0xca753d95L, + 0xf23a8028L, 0xf6fb9d9fL, 0xfbb8bb46L, 0xff79a6f1L, + 0xe13ef6f4L, 0xe5ffeb43L, 0xe8bccd9aL, 0xec7dd02dL, + 0x34867077L, 0x30476dc0L, 0x3d044b19L, 0x39c556aeL, + 0x278206abL, 0x23431b1cL, 0x2e003dc5L, 0x2ac12072L, + 0x128e9dcfL, 0x164f8078L, 0x1b0ca6a1L, 0x1fcdbb16L, + 0x018aeb13L, 0x054bf6a4L, 0x0808d07dL, 0x0cc9cdcaL, + 0x7897ab07L, 0x7c56b6b0L, 0x71159069L, 0x75d48ddeL, + 0x6b93dddbL, 0x6f52c06cL, 0x6211e6b5L, 0x66d0fb02L, + 0x5e9f46bfL, 0x5a5e5b08L, 0x571d7dd1L, 0x53dc6066L, + 0x4d9b3063L, 0x495a2dd4L, 0x44190b0dL, 0x40d816baL, + 0xaca5c697L, 0xa864db20L, 0xa527fdf9L, 0xa1e6e04eL, + 0xbfa1b04bL, 0xbb60adfcL, 0xb6238b25L, 0xb2e29692L, + 0x8aad2b2fL, 0x8e6c3698L, 0x832f1041L, 0x87ee0df6L, + 0x99a95df3L, 0x9d684044L, 0x902b669dL, 0x94ea7b2aL, + 0xe0b41de7L, 0xe4750050L, 0xe9362689L, 0xedf73b3eL, + 0xf3b06b3bL, 0xf771768cL, 0xfa325055L, 0xfef34de2L, + 0xc6bcf05fL, 0xc27dede8L, 0xcf3ecb31L, 0xcbffd686L, + 0xd5b88683L, 0xd1799b34L, 0xdc3abdedL, 0xd8fba05aL, + 0x690ce0eeL, 0x6dcdfd59L, 0x608edb80L, 0x644fc637L, + 0x7a089632L, 0x7ec98b85L, 0x738aad5cL, 0x774bb0ebL, + 0x4f040d56L, 0x4bc510e1L, 0x46863638L, 0x42472b8fL, + 0x5c007b8aL, 0x58c1663dL, 0x558240e4L, 0x51435d53L, + 0x251d3b9eL, 0x21dc2629L, 0x2c9f00f0L, 0x285e1d47L, + 0x36194d42L, 0x32d850f5L, 0x3f9b762cL, 0x3b5a6b9bL, + 0x0315d626L, 0x07d4cb91L, 0x0a97ed48L, 0x0e56f0ffL, + 0x1011a0faL, 0x14d0bd4dL, 0x19939b94L, 0x1d528623L, + 0xf12f560eL, 0xf5ee4bb9L, 0xf8ad6d60L, 0xfc6c70d7L, + 0xe22b20d2L, 0xe6ea3d65L, 0xeba91bbcL, 0xef68060bL, + 0xd727bbb6L, 0xd3e6a601L, 0xdea580d8L, 0xda649d6fL, + 0xc423cd6aL, 0xc0e2d0ddL, 0xcda1f604L, 0xc960ebb3L, + 0xbd3e8d7eL, 0xb9ff90c9L, 0xb4bcb610L, 0xb07daba7L, + 0xae3afba2L, 0xaafbe615L, 0xa7b8c0ccL, 0xa379dd7bL, + 0x9b3660c6L, 0x9ff77d71L, 0x92b45ba8L, 0x9675461fL, + 0x8832161aL, 0x8cf30badL, 0x81b02d74L, 0x857130c3L, + 0x5d8a9099L, 0x594b8d2eL, 0x5408abf7L, 0x50c9b640L, + 0x4e8ee645L, 0x4a4ffbf2L, 0x470cdd2bL, 0x43cdc09cL, + 0x7b827d21L, 0x7f436096L, 0x7200464fL, 0x76c15bf8L, + 0x68860bfdL, 0x6c47164aL, 0x61043093L, 0x65c52d24L, + 0x119b4be9L, 0x155a565eL, 0x18197087L, 0x1cd86d30L, + 0x029f3d35L, 0x065e2082L, 0x0b1d065bL, 0x0fdc1becL, + 0x3793a651L, 0x3352bbe6L, 0x3e119d3fL, 0x3ad08088L, + 0x2497d08dL, 0x2056cd3aL, 0x2d15ebe3L, 0x29d4f654L, + 0xc5a92679L, 0xc1683bceL, 0xcc2b1d17L, 0xc8ea00a0L, + 0xd6ad50a5L, 0xd26c4d12L, 0xdf2f6bcbL, 0xdbee767cL, + 0xe3a1cbc1L, 0xe760d676L, 0xea23f0afL, 0xeee2ed18L, + 0xf0a5bd1dL, 0xf464a0aaL, 0xf9278673L, 0xfde69bc4L, + 0x89b8fd09L, 0x8d79e0beL, 0x803ac667L, 0x84fbdbd0L, + 0x9abc8bd5L, 0x9e7d9662L, 0x933eb0bbL, 0x97ffad0cL, + 0xafb010b1L, 0xab710d06L, 0xa6322bdfL, 0xa2f33668L, + 0xbcb4666dL, 0xb8757bdaL, 0xb5365d03L, 0xb1f740b4L +}; + + +/*-------------------------------------------------------------*/ +/*--- end crctable.c ---*/ +/*-------------------------------------------------------------*/ diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/decompress.c b/ext/Compress-Raw-Bzip2/bzip2-src/decompress.c new file mode 100644 index 0000000..bba5e0f --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/decompress.c @@ -0,0 +1,626 @@ + +/*-------------------------------------------------------------*/ +/*--- Decompression machinery ---*/ +/*--- decompress.c ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#include "bzlib_private.h" + + +/*---------------------------------------------------*/ +static +void makeMaps_d ( DState* s ) +{ + Int32 i; + s->nInUse = 0; + for (i = 0; i < 256; i++) + if (s->inUse[i]) { + s->seqToUnseq[s->nInUse] = i; + s->nInUse++; + } +} + + +/*---------------------------------------------------*/ +#define RETURN(rrr) \ + { retVal = rrr; goto save_state_and_return; }; + +#define GET_BITS(lll,vvv,nnn) \ + case lll: s->state = lll; \ + while (True) { \ + if (s->bsLive >= nnn) { \ + UInt32 v; \ + v = (s->bsBuff >> \ + (s->bsLive-nnn)) & ((1 << nnn)-1); \ + s->bsLive -= nnn; \ + vvv = v; \ + break; \ + } \ + if (s->strm->avail_in == 0) RETURN(BZ_OK); \ + s->bsBuff \ + = (s->bsBuff << 8) | \ + ((UInt32) \ + (*((UChar*)(s->strm->next_in)))); \ + s->bsLive += 8; \ + s->strm->next_in++; \ + s->strm->avail_in--; \ + s->strm->total_in_lo32++; \ + if (s->strm->total_in_lo32 == 0) \ + s->strm->total_in_hi32++; \ + } + +#define GET_UCHAR(lll,uuu) \ + GET_BITS(lll,uuu,8) + +#define GET_BIT(lll,uuu) \ + GET_BITS(lll,uuu,1) + +/*---------------------------------------------------*/ +#define GET_MTF_VAL(label1,label2,lval) \ +{ \ + if (groupPos == 0) { \ + groupNo++; \ + if (groupNo >= nSelectors) \ + RETURN(BZ_DATA_ERROR); \ + groupPos = BZ_G_SIZE; \ + gSel = s->selector[groupNo]; \ + gMinlen = s->minLens[gSel]; \ + gLimit = &(s->limit[gSel][0]); \ + gPerm = &(s->perm[gSel][0]); \ + gBase = &(s->base[gSel][0]); \ + } \ + groupPos--; \ + zn = gMinlen; \ + GET_BITS(label1, zvec, zn); \ + while (1) { \ + if (zn > 20 /* the longest code */) \ + RETURN(BZ_DATA_ERROR); \ + if (zvec <= gLimit[zn]) break; \ + zn++; \ + GET_BIT(label2, zj); \ + zvec = (zvec << 1) | zj; \ + }; \ + if (zvec - gBase[zn] < 0 \ + || zvec - gBase[zn] >= BZ_MAX_ALPHA_SIZE) \ + RETURN(BZ_DATA_ERROR); \ + lval = gPerm[zvec - gBase[zn]]; \ +} + + +/*---------------------------------------------------*/ +Int32 BZ2_decompress ( DState* s ) +{ + UChar uc; + Int32 retVal; + Int32 minLen, maxLen; + bz_stream* strm = s->strm; + + /* stuff that needs to be saved/restored */ + Int32 i; + Int32 j; + Int32 t; + Int32 alphaSize; + Int32 nGroups; + Int32 nSelectors; + Int32 EOB; + Int32 groupNo; + Int32 groupPos; + Int32 nextSym; + Int32 nblockMAX; + Int32 nblock; + Int32 es; + Int32 N; + Int32 curr; + Int32 zt; + Int32 zn; + Int32 zvec; + Int32 zj; + Int32 gSel; + Int32 gMinlen; + Int32* gLimit; + Int32* gBase; + Int32* gPerm; + + if (s->state == BZ_X_MAGIC_1) { + /*initialise the save area*/ + s->save_i = 0; + s->save_j = 0; + s->save_t = 0; + s->save_alphaSize = 0; + s->save_nGroups = 0; + s->save_nSelectors = 0; + s->save_EOB = 0; + s->save_groupNo = 0; + s->save_groupPos = 0; + s->save_nextSym = 0; + s->save_nblockMAX = 0; + s->save_nblock = 0; + s->save_es = 0; + s->save_N = 0; + s->save_curr = 0; + s->save_zt = 0; + s->save_zn = 0; + s->save_zvec = 0; + s->save_zj = 0; + s->save_gSel = 0; + s->save_gMinlen = 0; + s->save_gLimit = NULL; + s->save_gBase = NULL; + s->save_gPerm = NULL; + } + + /*restore from the save area*/ + i = s->save_i; + j = s->save_j; + t = s->save_t; + alphaSize = s->save_alphaSize; + nGroups = s->save_nGroups; + nSelectors = s->save_nSelectors; + EOB = s->save_EOB; + groupNo = s->save_groupNo; + groupPos = s->save_groupPos; + nextSym = s->save_nextSym; + nblockMAX = s->save_nblockMAX; + nblock = s->save_nblock; + es = s->save_es; + N = s->save_N; + curr = s->save_curr; + zt = s->save_zt; + zn = s->save_zn; + zvec = s->save_zvec; + zj = s->save_zj; + gSel = s->save_gSel; + gMinlen = s->save_gMinlen; + gLimit = s->save_gLimit; + gBase = s->save_gBase; + gPerm = s->save_gPerm; + + retVal = BZ_OK; + + switch (s->state) { + + GET_UCHAR(BZ_X_MAGIC_1, uc); + if (uc != BZ_HDR_B) RETURN(BZ_DATA_ERROR_MAGIC); + + GET_UCHAR(BZ_X_MAGIC_2, uc); + if (uc != BZ_HDR_Z) RETURN(BZ_DATA_ERROR_MAGIC); + + GET_UCHAR(BZ_X_MAGIC_3, uc) + if (uc != BZ_HDR_h) RETURN(BZ_DATA_ERROR_MAGIC); + + GET_BITS(BZ_X_MAGIC_4, s->blockSize100k, 8) + if (s->blockSize100k < (BZ_HDR_0 + 1) || + s->blockSize100k > (BZ_HDR_0 + 9)) RETURN(BZ_DATA_ERROR_MAGIC); + s->blockSize100k -= BZ_HDR_0; + + if (s->smallDecompress) { + s->ll16 = BZALLOC( s->blockSize100k * 100000 * sizeof(UInt16) ); + s->ll4 = BZALLOC( + ((1 + s->blockSize100k * 100000) >> 1) * sizeof(UChar) + ); + if (s->ll16 == NULL || s->ll4 == NULL) RETURN(BZ_MEM_ERROR); + } else { + s->tt = BZALLOC( s->blockSize100k * 100000 * sizeof(Int32) ); + if (s->tt == NULL) RETURN(BZ_MEM_ERROR); + } + + GET_UCHAR(BZ_X_BLKHDR_1, uc); + + if (uc == 0x17) goto endhdr_2; + if (uc != 0x31) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_BLKHDR_2, uc); + if (uc != 0x41) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_BLKHDR_3, uc); + if (uc != 0x59) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_BLKHDR_4, uc); + if (uc != 0x26) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_BLKHDR_5, uc); + if (uc != 0x53) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_BLKHDR_6, uc); + if (uc != 0x59) RETURN(BZ_DATA_ERROR); + + s->currBlockNo++; + if (s->verbosity >= 2) + VPrintf1 ( "\n [%d: huff+mtf ", s->currBlockNo ); + + s->storedBlockCRC = 0; + GET_UCHAR(BZ_X_BCRC_1, uc); + s->storedBlockCRC = (s->storedBlockCRC << 8) | ((UInt32)uc); + GET_UCHAR(BZ_X_BCRC_2, uc); + s->storedBlockCRC = (s->storedBlockCRC << 8) | ((UInt32)uc); + GET_UCHAR(BZ_X_BCRC_3, uc); + s->storedBlockCRC = (s->storedBlockCRC << 8) | ((UInt32)uc); + GET_UCHAR(BZ_X_BCRC_4, uc); + s->storedBlockCRC = (s->storedBlockCRC << 8) | ((UInt32)uc); + + GET_BITS(BZ_X_RANDBIT, s->blockRandomised, 1); + + s->origPtr = 0; + GET_UCHAR(BZ_X_ORIGPTR_1, uc); + s->origPtr = (s->origPtr << 8) | ((Int32)uc); + GET_UCHAR(BZ_X_ORIGPTR_2, uc); + s->origPtr = (s->origPtr << 8) | ((Int32)uc); + GET_UCHAR(BZ_X_ORIGPTR_3, uc); + s->origPtr = (s->origPtr << 8) | ((Int32)uc); + + if (s->origPtr < 0) + RETURN(BZ_DATA_ERROR); + if (s->origPtr > 10 + 100000*s->blockSize100k) + RETURN(BZ_DATA_ERROR); + + /*--- Receive the mapping table ---*/ + for (i = 0; i < 16; i++) { + GET_BIT(BZ_X_MAPPING_1, uc); + if (uc == 1) + s->inUse16[i] = True; else + s->inUse16[i] = False; + } + + for (i = 0; i < 256; i++) s->inUse[i] = False; + + for (i = 0; i < 16; i++) + if (s->inUse16[i]) + for (j = 0; j < 16; j++) { + GET_BIT(BZ_X_MAPPING_2, uc); + if (uc == 1) s->inUse[i * 16 + j] = True; + } + makeMaps_d ( s ); + if (s->nInUse == 0) RETURN(BZ_DATA_ERROR); + alphaSize = s->nInUse+2; + + /*--- Now the selectors ---*/ + GET_BITS(BZ_X_SELECTOR_1, nGroups, 3); + if (nGroups < 2 || nGroups > 6) RETURN(BZ_DATA_ERROR); + GET_BITS(BZ_X_SELECTOR_2, nSelectors, 15); + if (nSelectors < 1) RETURN(BZ_DATA_ERROR); + for (i = 0; i < nSelectors; i++) { + j = 0; + while (True) { + GET_BIT(BZ_X_SELECTOR_3, uc); + if (uc == 0) break; + j++; + if (j >= nGroups) RETURN(BZ_DATA_ERROR); + } + s->selectorMtf[i] = j; + } + + /*--- Undo the MTF values for the selectors. ---*/ + { + UChar pos[BZ_N_GROUPS], tmp, v; + for (v = 0; v < nGroups; v++) pos[v] = v; + + for (i = 0; i < nSelectors; i++) { + v = s->selectorMtf[i]; + tmp = pos[v]; + while (v > 0) { pos[v] = pos[v-1]; v--; } + pos[0] = tmp; + s->selector[i] = tmp; + } + } + + /*--- Now the coding tables ---*/ + for (t = 0; t < nGroups; t++) { + GET_BITS(BZ_X_CODING_1, curr, 5); + for (i = 0; i < alphaSize; i++) { + while (True) { + if (curr < 1 || curr > 20) RETURN(BZ_DATA_ERROR); + GET_BIT(BZ_X_CODING_2, uc); + if (uc == 0) break; + GET_BIT(BZ_X_CODING_3, uc); + if (uc == 0) curr++; else curr--; + } + s->len[t][i] = curr; + } + } + + /*--- Create the Huffman decoding tables ---*/ + for (t = 0; t < nGroups; t++) { + minLen = 32; + maxLen = 0; + for (i = 0; i < alphaSize; i++) { + if (s->len[t][i] > maxLen) maxLen = s->len[t][i]; + if (s->len[t][i] < minLen) minLen = s->len[t][i]; + } + BZ2_hbCreateDecodeTables ( + &(s->limit[t][0]), + &(s->base[t][0]), + &(s->perm[t][0]), + &(s->len[t][0]), + minLen, maxLen, alphaSize + ); + s->minLens[t] = minLen; + } + + /*--- Now the MTF values ---*/ + + EOB = s->nInUse+1; + nblockMAX = 100000 * s->blockSize100k; + groupNo = -1; + groupPos = 0; + + for (i = 0; i <= 255; i++) s->unzftab[i] = 0; + + /*-- MTF init --*/ + { + Int32 ii, jj, kk; + kk = MTFA_SIZE-1; + for (ii = 256 / MTFL_SIZE - 1; ii >= 0; ii--) { + for (jj = MTFL_SIZE-1; jj >= 0; jj--) { + s->mtfa[kk] = (UChar)(ii * MTFL_SIZE + jj); + kk--; + } + s->mtfbase[ii] = kk + 1; + } + } + /*-- end MTF init --*/ + + nblock = 0; + GET_MTF_VAL(BZ_X_MTF_1, BZ_X_MTF_2, nextSym); + + while (True) { + + if (nextSym == EOB) break; + + if (nextSym == BZ_RUNA || nextSym == BZ_RUNB) { + + es = -1; + N = 1; + do { + if (nextSym == BZ_RUNA) es = es + (0+1) * N; else + if (nextSym == BZ_RUNB) es = es + (1+1) * N; + N = N * 2; + GET_MTF_VAL(BZ_X_MTF_3, BZ_X_MTF_4, nextSym); + } + while (nextSym == BZ_RUNA || nextSym == BZ_RUNB); + + es++; + uc = s->seqToUnseq[ s->mtfa[s->mtfbase[0]] ]; + s->unzftab[uc] += es; + + if (s->smallDecompress) + while (es > 0) { + if (nblock >= nblockMAX) RETURN(BZ_DATA_ERROR); + s->ll16[nblock] = (UInt16)uc; + nblock++; + es--; + } + else + while (es > 0) { + if (nblock >= nblockMAX) RETURN(BZ_DATA_ERROR); + s->tt[nblock] = (UInt32)uc; + nblock++; + es--; + }; + + continue; + + } else { + + if (nblock >= nblockMAX) RETURN(BZ_DATA_ERROR); + + /*-- uc = MTF ( nextSym-1 ) --*/ + { + Int32 ii, jj, kk, pp, lno, off; + UInt32 nn; + nn = (UInt32)(nextSym - 1); + + if (nn < MTFL_SIZE) { + /* avoid general-case expense */ + pp = s->mtfbase[0]; + uc = s->mtfa[pp+nn]; + while (nn > 3) { + Int32 z = pp+nn; + s->mtfa[(z) ] = s->mtfa[(z)-1]; + s->mtfa[(z)-1] = s->mtfa[(z)-2]; + s->mtfa[(z)-2] = s->mtfa[(z)-3]; + s->mtfa[(z)-3] = s->mtfa[(z)-4]; + nn -= 4; + } + while (nn > 0) { + s->mtfa[(pp+nn)] = s->mtfa[(pp+nn)-1]; nn--; + }; + s->mtfa[pp] = uc; + } else { + /* general case */ + lno = nn / MTFL_SIZE; + off = nn % MTFL_SIZE; + pp = s->mtfbase[lno] + off; + uc = s->mtfa[pp]; + while (pp > s->mtfbase[lno]) { + s->mtfa[pp] = s->mtfa[pp-1]; pp--; + }; + s->mtfbase[lno]++; + while (lno > 0) { + s->mtfbase[lno]--; + s->mtfa[s->mtfbase[lno]] + = s->mtfa[s->mtfbase[lno-1] + MTFL_SIZE - 1]; + lno--; + } + s->mtfbase[0]--; + s->mtfa[s->mtfbase[0]] = uc; + if (s->mtfbase[0] == 0) { + kk = MTFA_SIZE-1; + for (ii = 256 / MTFL_SIZE-1; ii >= 0; ii--) { + for (jj = MTFL_SIZE-1; jj >= 0; jj--) { + s->mtfa[kk] = s->mtfa[s->mtfbase[ii] + jj]; + kk--; + } + s->mtfbase[ii] = kk + 1; + } + } + } + } + /*-- end uc = MTF ( nextSym-1 ) --*/ + + s->unzftab[s->seqToUnseq[uc]]++; + if (s->smallDecompress) + s->ll16[nblock] = (UInt16)(s->seqToUnseq[uc]); else + s->tt[nblock] = (UInt32)(s->seqToUnseq[uc]); + nblock++; + + GET_MTF_VAL(BZ_X_MTF_5, BZ_X_MTF_6, nextSym); + continue; + } + } + + /* Now we know what nblock is, we can do a better sanity + check on s->origPtr. + */ + if (s->origPtr < 0 || s->origPtr >= nblock) + RETURN(BZ_DATA_ERROR); + + /*-- Set up cftab to facilitate generation of T^(-1) --*/ + s->cftab[0] = 0; + for (i = 1; i <= 256; i++) s->cftab[i] = s->unzftab[i-1]; + for (i = 1; i <= 256; i++) s->cftab[i] += s->cftab[i-1]; + for (i = 0; i <= 256; i++) { + if (s->cftab[i] < 0 || s->cftab[i] > nblock) { + /* s->cftab[i] can legitimately be == nblock */ + RETURN(BZ_DATA_ERROR); + } + } + + s->state_out_len = 0; + s->state_out_ch = 0; + BZ_INITIALISE_CRC ( s->calculatedBlockCRC ); + s->state = BZ_X_OUTPUT; + if (s->verbosity >= 2) VPrintf0 ( "rt+rld" ); + + if (s->smallDecompress) { + + /*-- Make a copy of cftab, used in generation of T --*/ + for (i = 0; i <= 256; i++) s->cftabCopy[i] = s->cftab[i]; + + /*-- compute the T vector --*/ + for (i = 0; i < nblock; i++) { + uc = (UChar)(s->ll16[i]); + SET_LL(i, s->cftabCopy[uc]); + s->cftabCopy[uc]++; + } + + /*-- Compute T^(-1) by pointer reversal on T --*/ + i = s->origPtr; + j = GET_LL(i); + do { + Int32 tmp = GET_LL(j); + SET_LL(j, i); + i = j; + j = tmp; + } + while (i != s->origPtr); + + s->tPos = s->origPtr; + s->nblock_used = 0; + if (s->blockRandomised) { + BZ_RAND_INIT_MASK; + BZ_GET_SMALL(s->k0); s->nblock_used++; + BZ_RAND_UPD_MASK; s->k0 ^= BZ_RAND_MASK; + } else { + BZ_GET_SMALL(s->k0); s->nblock_used++; + } + + } else { + + /*-- compute the T^(-1) vector --*/ + for (i = 0; i < nblock; i++) { + uc = (UChar)(s->tt[i] & 0xff); + s->tt[s->cftab[uc]] |= (i << 8); + s->cftab[uc]++; + } + + s->tPos = s->tt[s->origPtr] >> 8; + s->nblock_used = 0; + if (s->blockRandomised) { + BZ_RAND_INIT_MASK; + BZ_GET_FAST(s->k0); s->nblock_used++; + BZ_RAND_UPD_MASK; s->k0 ^= BZ_RAND_MASK; + } else { + BZ_GET_FAST(s->k0); s->nblock_used++; + } + + } + + RETURN(BZ_OK); + + + + endhdr_2: + + GET_UCHAR(BZ_X_ENDHDR_2, uc); + if (uc != 0x72) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_ENDHDR_3, uc); + if (uc != 0x45) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_ENDHDR_4, uc); + if (uc != 0x38) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_ENDHDR_5, uc); + if (uc != 0x50) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_ENDHDR_6, uc); + if (uc != 0x90) RETURN(BZ_DATA_ERROR); + + s->storedCombinedCRC = 0; + GET_UCHAR(BZ_X_CCRC_1, uc); + s->storedCombinedCRC = (s->storedCombinedCRC << 8) | ((UInt32)uc); + GET_UCHAR(BZ_X_CCRC_2, uc); + s->storedCombinedCRC = (s->storedCombinedCRC << 8) | ((UInt32)uc); + GET_UCHAR(BZ_X_CCRC_3, uc); + s->storedCombinedCRC = (s->storedCombinedCRC << 8) | ((UInt32)uc); + GET_UCHAR(BZ_X_CCRC_4, uc); + s->storedCombinedCRC = (s->storedCombinedCRC << 8) | ((UInt32)uc); + + s->state = BZ_X_IDLE; + RETURN(BZ_STREAM_END); + + default: AssertH ( False, 4001 ); + } + + AssertH ( False, 4002 ); + + save_state_and_return: + + s->save_i = i; + s->save_j = j; + s->save_t = t; + s->save_alphaSize = alphaSize; + s->save_nGroups = nGroups; + s->save_nSelectors = nSelectors; + s->save_EOB = EOB; + s->save_groupNo = groupNo; + s->save_groupPos = groupPos; + s->save_nextSym = nextSym; + s->save_nblockMAX = nblockMAX; + s->save_nblock = nblock; + s->save_es = es; + s->save_N = N; + s->save_curr = curr; + s->save_zt = zt; + s->save_zn = zn; + s->save_zvec = zvec; + s->save_zj = zj; + s->save_gSel = gSel; + s->save_gMinlen = gMinlen; + s->save_gLimit = gLimit; + s->save_gBase = gBase; + s->save_gPerm = gPerm; + + return retVal; +} + + +/*-------------------------------------------------------------*/ +/*--- end decompress.c ---*/ +/*-------------------------------------------------------------*/ diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/dlltest.c b/ext/Compress-Raw-Bzip2/bzip2-src/dlltest.c new file mode 100644 index 0000000..03fa146 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/dlltest.c @@ -0,0 +1,175 @@ +/* + minibz2 + libbz2.dll test program. + by Yoshioka Tsuneo (tsuneo@rr.iij4u.or.jp) + This file is Public Domain. Welcome any email to me. + + usage: minibz2 [-d] [-{1,2,..9}] [[srcfilename] destfilename] +*/ + +#define BZ_IMPORT +#include +#include +#include "bzlib.h" +#ifdef _WIN32 +#include +#endif + + +#ifdef _WIN32 + +#define BZ2_LIBNAME "libbz2-1.0.2.DLL" + +#include +static int BZ2DLLLoaded = 0; +static HINSTANCE BZ2DLLhLib; +int BZ2DLLLoadLibrary(void) +{ + HINSTANCE hLib; + + if(BZ2DLLLoaded==1){return 0;} + hLib=LoadLibrary(BZ2_LIBNAME); + if(hLib == NULL){ + fprintf(stderr,"Can't load %s\n",BZ2_LIBNAME); + return -1; + } + BZ2_bzlibVersion=GetProcAddress(hLib,"BZ2_bzlibVersion"); + BZ2_bzopen=GetProcAddress(hLib,"BZ2_bzopen"); + BZ2_bzdopen=GetProcAddress(hLib,"BZ2_bzdopen"); + BZ2_bzread=GetProcAddress(hLib,"BZ2_bzread"); + BZ2_bzwrite=GetProcAddress(hLib,"BZ2_bzwrite"); + BZ2_bzflush=GetProcAddress(hLib,"BZ2_bzflush"); + BZ2_bzclose=GetProcAddress(hLib,"BZ2_bzclose"); + BZ2_bzerror=GetProcAddress(hLib,"BZ2_bzerror"); + + if (!BZ2_bzlibVersion || !BZ2_bzopen || !BZ2_bzdopen + || !BZ2_bzread || !BZ2_bzwrite || !BZ2_bzflush + || !BZ2_bzclose || !BZ2_bzerror) { + fprintf(stderr,"GetProcAddress failed.\n"); + return -1; + } + BZ2DLLLoaded=1; + BZ2DLLhLib=hLib; + return 0; + +} +int BZ2DLLFreeLibrary(void) +{ + if(BZ2DLLLoaded==0){return 0;} + FreeLibrary(BZ2DLLhLib); + BZ2DLLLoaded=0; +} +#endif /* WIN32 */ + +void usage(void) +{ + puts("usage: minibz2 [-d] [-{1,2,..9}] [[srcfilename] destfilename]"); +} + +int main(int argc,char *argv[]) +{ + int decompress = 0; + int level = 9; + char *fn_r = NULL; + char *fn_w = NULL; + +#ifdef _WIN32 + if(BZ2DLLLoadLibrary()<0){ + fprintf(stderr,"Loading of %s failed. Giving up.\n", BZ2_LIBNAME); + exit(1); + } + printf("Loading of %s succeeded. Library version is %s.\n", + BZ2_LIBNAME, BZ2_bzlibVersion() ); +#endif + while(++argv,--argc){ + if(**argv =='-' || **argv=='/'){ + char *p; + + for(p=*argv+1;*p;p++){ + if(*p=='d'){ + decompress = 1; + }else if('1'<=*p && *p<='9'){ + level = *p - '0'; + }else{ + usage(); + exit(1); + } + } + }else{ + break; + } + } + if(argc>=1){ + fn_r = *argv; + argc--;argv++; + }else{ + fn_r = NULL; + } + if(argc>=1){ + fn_w = *argv; + argc--;argv++; + }else{ + fn_w = NULL; + } + { + int len; + char buff[0x1000]; + char mode[10]; + + if(decompress){ + BZFILE *BZ2fp_r = NULL; + FILE *fp_w = NULL; + + if(fn_w){ + if((fp_w = fopen(fn_w,"wb"))==NULL){ + printf("can't open [%s]\n",fn_w); + perror("reason:"); + exit(1); + } + }else{ + fp_w = stdout; + } + if((fn_r == NULL && (BZ2fp_r = BZ2_bzdopen(fileno(stdin),"rb"))==NULL) + || (fn_r != NULL && (BZ2fp_r = BZ2_bzopen(fn_r,"rb"))==NULL)){ + printf("can't bz2openstream\n"); + exit(1); + } + while((len=BZ2_bzread(BZ2fp_r,buff,0x1000))>0){ + fwrite(buff,1,len,fp_w); + } + BZ2_bzclose(BZ2fp_r); + if(fp_w != stdout) fclose(fp_w); + }else{ + BZFILE *BZ2fp_w = NULL; + FILE *fp_r = NULL; + + if(fn_r){ + if((fp_r = fopen(fn_r,"rb"))==NULL){ + printf("can't open [%s]\n",fn_r); + perror("reason:"); + exit(1); + } + }else{ + fp_r = stdin; + } + mode[0]='w'; + mode[1] = '0' + level; + mode[2] = '\0'; + + if((fn_w == NULL && (BZ2fp_w = BZ2_bzdopen(fileno(stdout),mode))==NULL) + || (fn_w !=NULL && (BZ2fp_w = BZ2_bzopen(fn_w,mode))==NULL)){ + printf("can't bz2openstream\n"); + exit(1); + } + while((len=fread(buff,1,0x1000,fp_r))>0){ + BZ2_bzwrite(BZ2fp_w,buff,len); + } + BZ2_bzclose(BZ2fp_w); + if(fp_r!=stdin)fclose(fp_r); + } + } +#ifdef _WIN32 + BZ2DLLFreeLibrary(); +#endif + return 0; +} diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/huffman.c b/ext/Compress-Raw-Bzip2/bzip2-src/huffman.c new file mode 100644 index 0000000..87e79e3 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/huffman.c @@ -0,0 +1,205 @@ + +/*-------------------------------------------------------------*/ +/*--- Huffman coding low-level stuff ---*/ +/*--- huffman.c ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#include "bzlib_private.h" + +/*---------------------------------------------------*/ +#define WEIGHTOF(zz0) ((zz0) & 0xffffff00) +#define DEPTHOF(zz1) ((zz1) & 0x000000ff) +#define MYMAX(zz2,zz3) ((zz2) > (zz3) ? (zz2) : (zz3)) + +#define ADDWEIGHTS(zw1,zw2) \ + (WEIGHTOF(zw1)+WEIGHTOF(zw2)) | \ + (1 + MYMAX(DEPTHOF(zw1),DEPTHOF(zw2))) + +#define UPHEAP(z) \ +{ \ + Int32 zz, tmp; \ + zz = z; tmp = heap[zz]; \ + while (weight[tmp] < weight[heap[zz >> 1]]) { \ + heap[zz] = heap[zz >> 1]; \ + zz >>= 1; \ + } \ + heap[zz] = tmp; \ +} + +#define DOWNHEAP(z) \ +{ \ + Int32 zz, yy, tmp; \ + zz = z; tmp = heap[zz]; \ + while (True) { \ + yy = zz << 1; \ + if (yy > nHeap) break; \ + if (yy < nHeap && \ + weight[heap[yy+1]] < weight[heap[yy]]) \ + yy++; \ + if (weight[tmp] < weight[heap[yy]]) break; \ + heap[zz] = heap[yy]; \ + zz = yy; \ + } \ + heap[zz] = tmp; \ +} + + +/*---------------------------------------------------*/ +void BZ2_hbMakeCodeLengths ( UChar *len, + Int32 *freq, + Int32 alphaSize, + Int32 maxLen ) +{ + /*-- + Nodes and heap entries run from 1. Entry 0 + for both the heap and nodes is a sentinel. + --*/ + Int32 nNodes, nHeap, n1, n2, i, j, k; + Bool tooLong; + + Int32 heap [ BZ_MAX_ALPHA_SIZE + 2 ]; + Int32 weight [ BZ_MAX_ALPHA_SIZE * 2 ]; + Int32 parent [ BZ_MAX_ALPHA_SIZE * 2 ]; + + for (i = 0; i < alphaSize; i++) + weight[i+1] = (freq[i] == 0 ? 1 : freq[i]) << 8; + + while (True) { + + nNodes = alphaSize; + nHeap = 0; + + heap[0] = 0; + weight[0] = 0; + parent[0] = -2; + + for (i = 1; i <= alphaSize; i++) { + parent[i] = -1; + nHeap++; + heap[nHeap] = i; + UPHEAP(nHeap); + } + + AssertH( nHeap < (BZ_MAX_ALPHA_SIZE+2), 2001 ); + + while (nHeap > 1) { + n1 = heap[1]; heap[1] = heap[nHeap]; nHeap--; DOWNHEAP(1); + n2 = heap[1]; heap[1] = heap[nHeap]; nHeap--; DOWNHEAP(1); + nNodes++; + parent[n1] = parent[n2] = nNodes; + weight[nNodes] = ADDWEIGHTS(weight[n1], weight[n2]); + parent[nNodes] = -1; + nHeap++; + heap[nHeap] = nNodes; + UPHEAP(nHeap); + } + + AssertH( nNodes < (BZ_MAX_ALPHA_SIZE * 2), 2002 ); + + tooLong = False; + for (i = 1; i <= alphaSize; i++) { + j = 0; + k = i; + while (parent[k] >= 0) { k = parent[k]; j++; } + len[i-1] = j; + if (j > maxLen) tooLong = True; + } + + if (! tooLong) break; + + /* 17 Oct 04: keep-going condition for the following loop used + to be 'i < alphaSize', which missed the last element, + theoretically leading to the possibility of the compressor + looping. However, this count-scaling step is only needed if + one of the generated Huffman code words is longer than + maxLen, which up to and including version 1.0.2 was 20 bits, + which is extremely unlikely. In version 1.0.3 maxLen was + changed to 17 bits, which has minimal effect on compression + ratio, but does mean this scaling step is used from time to + time, enough to verify that it works. + + This means that bzip2-1.0.3 and later will only produce + Huffman codes with a maximum length of 17 bits. However, in + order to preserve backwards compatibility with bitstreams + produced by versions pre-1.0.3, the decompressor must still + handle lengths of up to 20. */ + + for (i = 1; i <= alphaSize; i++) { + j = weight[i] >> 8; + j = 1 + (j / 2); + weight[i] = j << 8; + } + } +} + + +/*---------------------------------------------------*/ +void BZ2_hbAssignCodes ( Int32 *code, + UChar *length, + Int32 minLen, + Int32 maxLen, + Int32 alphaSize ) +{ + Int32 n, vec, i; + + vec = 0; + for (n = minLen; n <= maxLen; n++) { + for (i = 0; i < alphaSize; i++) + if (length[i] == n) { code[i] = vec; vec++; }; + vec <<= 1; + } +} + + +/*---------------------------------------------------*/ +void BZ2_hbCreateDecodeTables ( Int32 *limit, + Int32 *base, + Int32 *perm, + UChar *length, + Int32 minLen, + Int32 maxLen, + Int32 alphaSize ) +{ + Int32 pp, i, j, vec; + + pp = 0; + for (i = minLen; i <= maxLen; i++) + for (j = 0; j < alphaSize; j++) + if (length[j] == i) { perm[pp] = j; pp++; }; + + for (i = 0; i < BZ_MAX_CODE_LEN; i++) base[i] = 0; + for (i = 0; i < alphaSize; i++) base[length[i]+1]++; + + for (i = 1; i < BZ_MAX_CODE_LEN; i++) base[i] += base[i-1]; + + for (i = 0; i < BZ_MAX_CODE_LEN; i++) limit[i] = 0; + vec = 0; + + for (i = minLen; i <= maxLen; i++) { + vec += (base[i+1] - base[i]); + limit[i] = vec-1; + vec <<= 1; + } + for (i = minLen + 1; i <= maxLen; i++) + base[i] = ((limit[i-1] + 1) << 1) - base[i]; +} + + +/*-------------------------------------------------------------*/ +/*--- end huffman.c ---*/ +/*-------------------------------------------------------------*/ diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/mk251.c b/ext/Compress-Raw-Bzip2/bzip2-src/mk251.c new file mode 100644 index 0000000..39e94c0 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/mk251.c @@ -0,0 +1,31 @@ + +/* Spew out a long sequence of the byte 251. When fed to bzip2 + versions 1.0.0 or 1.0.1, causes it to die with internal error + 1007 in blocksort.c. This assertion misses an extremely rare + case, which is fixed in this version (1.0.2) and above. +*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#include + +int main () +{ + int i; + for (i = 0; i < 48500000 ; i++) + putchar(251); + return 0; +} diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/randtable.c b/ext/Compress-Raw-Bzip2/bzip2-src/randtable.c new file mode 100644 index 0000000..068b763 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/randtable.c @@ -0,0 +1,84 @@ + +/*-------------------------------------------------------------*/ +/*--- Table for randomising repetitive blocks ---*/ +/*--- randtable.c ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#include "bzlib_private.h" + + +/*---------------------------------------------*/ +Int32 BZ2_rNums[512] = { + 619, 720, 127, 481, 931, 816, 813, 233, 566, 247, + 985, 724, 205, 454, 863, 491, 741, 242, 949, 214, + 733, 859, 335, 708, 621, 574, 73, 654, 730, 472, + 419, 436, 278, 496, 867, 210, 399, 680, 480, 51, + 878, 465, 811, 169, 869, 675, 611, 697, 867, 561, + 862, 687, 507, 283, 482, 129, 807, 591, 733, 623, + 150, 238, 59, 379, 684, 877, 625, 169, 643, 105, + 170, 607, 520, 932, 727, 476, 693, 425, 174, 647, + 73, 122, 335, 530, 442, 853, 695, 249, 445, 515, + 909, 545, 703, 919, 874, 474, 882, 500, 594, 612, + 641, 801, 220, 162, 819, 984, 589, 513, 495, 799, + 161, 604, 958, 533, 221, 400, 386, 867, 600, 782, + 382, 596, 414, 171, 516, 375, 682, 485, 911, 276, + 98, 553, 163, 354, 666, 933, 424, 341, 533, 870, + 227, 730, 475, 186, 263, 647, 537, 686, 600, 224, + 469, 68, 770, 919, 190, 373, 294, 822, 808, 206, + 184, 943, 795, 384, 383, 461, 404, 758, 839, 887, + 715, 67, 618, 276, 204, 918, 873, 777, 604, 560, + 951, 160, 578, 722, 79, 804, 96, 409, 713, 940, + 652, 934, 970, 447, 318, 353, 859, 672, 112, 785, + 645, 863, 803, 350, 139, 93, 354, 99, 820, 908, + 609, 772, 154, 274, 580, 184, 79, 626, 630, 742, + 653, 282, 762, 623, 680, 81, 927, 626, 789, 125, + 411, 521, 938, 300, 821, 78, 343, 175, 128, 250, + 170, 774, 972, 275, 999, 639, 495, 78, 352, 126, + 857, 956, 358, 619, 580, 124, 737, 594, 701, 612, + 669, 112, 134, 694, 363, 992, 809, 743, 168, 974, + 944, 375, 748, 52, 600, 747, 642, 182, 862, 81, + 344, 805, 988, 739, 511, 655, 814, 334, 249, 515, + 897, 955, 664, 981, 649, 113, 974, 459, 893, 228, + 433, 837, 553, 268, 926, 240, 102, 654, 459, 51, + 686, 754, 806, 760, 493, 403, 415, 394, 687, 700, + 946, 670, 656, 610, 738, 392, 760, 799, 887, 653, + 978, 321, 576, 617, 626, 502, 894, 679, 243, 440, + 680, 879, 194, 572, 640, 724, 926, 56, 204, 700, + 707, 151, 457, 449, 797, 195, 791, 558, 945, 679, + 297, 59, 87, 824, 713, 663, 412, 693, 342, 606, + 134, 108, 571, 364, 631, 212, 174, 643, 304, 329, + 343, 97, 430, 751, 497, 314, 983, 374, 822, 928, + 140, 206, 73, 263, 980, 736, 876, 478, 430, 305, + 170, 514, 364, 692, 829, 82, 855, 953, 676, 246, + 369, 970, 294, 750, 807, 827, 150, 790, 288, 923, + 804, 378, 215, 828, 592, 281, 565, 555, 710, 82, + 896, 831, 547, 261, 524, 462, 293, 465, 502, 56, + 661, 821, 976, 991, 658, 869, 905, 758, 745, 193, + 768, 550, 608, 933, 378, 286, 215, 979, 792, 961, + 61, 688, 793, 644, 986, 403, 106, 366, 905, 644, + 372, 567, 466, 434, 645, 210, 389, 550, 919, 135, + 780, 773, 635, 389, 707, 100, 626, 958, 165, 504, + 920, 176, 193, 713, 857, 265, 203, 50, 668, 108, + 645, 990, 626, 197, 510, 357, 358, 850, 858, 364, + 936, 638 +}; + + +/*-------------------------------------------------------------*/ +/*--- end randtable.c ---*/ +/*-------------------------------------------------------------*/ diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/spewG.c b/ext/Compress-Raw-Bzip2/bzip2-src/spewG.c new file mode 100644 index 0000000..5892b92 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/spewG.c @@ -0,0 +1,54 @@ + +/* spew out a thoroughly gigantic file designed so that bzip2 + can compress it reasonably rapidly. This is to help test + support for large files (> 2GB) in a reasonable amount of time. + I suggest you use the undocumented --exponential option to + bzip2 when compressing the resulting file; this saves a bit of + time. Note: *don't* bother with --exponential when compressing + Real Files; it'll just waste a lot of CPU time :-) + (but is otherwise harmless). +*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#define _FILE_OFFSET_BITS 64 + +#include +#include + +/* The number of megabytes of junk to spew out (roughly) */ +#define MEGABYTES 5000 + +#define N_BUF 1000000 +char buf[N_BUF]; + +int main ( int argc, char** argv ) +{ + int ii, kk, p; + srandom(1); + setbuffer ( stdout, buf, N_BUF ); + for (kk = 0; kk < MEGABYTES * 515; kk+=3) { + p = 25+random()%50; + for (ii = 0; ii < p; ii++) + printf ( "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa" ); + for (ii = 0; ii < p-1; ii++) + printf ( "bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb" ); + for (ii = 0; ii < p+1; ii++) + printf ( "ccccccccccccccccccccccccccccccccccccc" ); + } + fflush(stdout); + return 0; +} diff --git a/ext/Compress-Raw-Bzip2/bzip2-src/unzcrash.c b/ext/Compress-Raw-Bzip2/bzip2-src/unzcrash.c new file mode 100644 index 0000000..a1b7546 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/bzip2-src/unzcrash.c @@ -0,0 +1,141 @@ + +/* A test program written to test robustness to decompression of + corrupted data. Usage is + unzcrash filename + and the program will read the specified file, compress it (in memory), + and then repeatedly decompress it, each time with a different bit of + the compressed data inverted, so as to test all possible one-bit errors. + This should not cause any invalid memory accesses. If it does, + I want to know about it! + + PS. As you can see from the above description, the process is + incredibly slow. A file of size eg 5KB will cause it to run for + many hours. +*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#include +#include +#include "bzlib.h" + +#define M_BLOCK 1000000 + +typedef unsigned char uchar; + +#define M_BLOCK_OUT (M_BLOCK + 1000000) +uchar inbuf[M_BLOCK]; +uchar outbuf[M_BLOCK_OUT]; +uchar zbuf[M_BLOCK + 600 + (M_BLOCK / 100)]; + +int nIn, nOut, nZ; + +static char *bzerrorstrings[] = { + "OK" + ,"SEQUENCE_ERROR" + ,"PARAM_ERROR" + ,"MEM_ERROR" + ,"DATA_ERROR" + ,"DATA_ERROR_MAGIC" + ,"IO_ERROR" + ,"UNEXPECTED_EOF" + ,"OUTBUFF_FULL" + ,"???" /* for future */ + ,"???" /* for future */ + ,"???" /* for future */ + ,"???" /* for future */ + ,"???" /* for future */ + ,"???" /* for future */ +}; + +void flip_bit ( int bit ) +{ + int byteno = bit / 8; + int bitno = bit % 8; + uchar mask = 1 << bitno; + //fprintf ( stderr, "(byte %d bit %d mask %d)", + // byteno, bitno, (int)mask ); + zbuf[byteno] ^= mask; +} + +int main ( int argc, char** argv ) +{ + FILE* f; + int r; + int bit; + int i; + + if (argc != 2) { + fprintf ( stderr, "usage: unzcrash filename\n" ); + return 1; + } + + f = fopen ( argv[1], "r" ); + if (!f) { + fprintf ( stderr, "unzcrash: can't open %s\n", argv[1] ); + return 1; + } + + nIn = fread ( inbuf, 1, M_BLOCK, f ); + fprintf ( stderr, "%d bytes read\n", nIn ); + + nZ = M_BLOCK; + r = BZ2_bzBuffToBuffCompress ( + zbuf, &nZ, inbuf, nIn, 9, 0, 30 ); + + assert (r == BZ_OK); + fprintf ( stderr, "%d after compression\n", nZ ); + + for (bit = 0; bit < nZ*8; bit++) { + fprintf ( stderr, "bit %d ", bit ); + flip_bit ( bit ); + nOut = M_BLOCK_OUT; + r = BZ2_bzBuffToBuffDecompress ( + outbuf, &nOut, zbuf, nZ, 0, 0 ); + fprintf ( stderr, " %d %s ", r, bzerrorstrings[-r] ); + + if (r != BZ_OK) { + fprintf ( stderr, "\n" ); + } else { + if (nOut != nIn) { + fprintf(stderr, "nIn/nOut mismatch %d %d\n", nIn, nOut ); + return 1; + } else { + for (i = 0; i < nOut; i++) + if (inbuf[i] != outbuf[i]) { + fprintf(stderr, "mismatch at %d\n", i ); + return 1; + } + if (i == nOut) fprintf(stderr, "really ok!\n" ); + } + } + + flip_bit ( bit ); + } + +#if 0 + assert (nOut == nIn); + for (i = 0; i < nOut; i++) { + if (inbuf[i] != outbuf[i]) { + fprintf ( stderr, "difference at %d !\n", i ); + return 1; + } + } +#endif + + fprintf ( stderr, "all ok\n" ); + return 0; +} diff --git a/ext/Compress-Raw-Bzip2/fallback/constants.h b/ext/Compress-Raw-Bzip2/fallback/constants.h new file mode 100644 index 0000000..f7f8d6c --- /dev/null +++ b/ext/Compress-Raw-Bzip2/fallback/constants.h @@ -0,0 +1,282 @@ +#define PERL_constant_NOTFOUND 1 +#define PERL_constant_NOTDEF 2 +#define PERL_constant_ISIV 3 +#define PERL_constant_ISNO 4 +#define PERL_constant_ISNV 5 +#define PERL_constant_ISPV 6 +#define PERL_constant_ISPVN 7 +#define PERL_constant_ISSV 8 +#define PERL_constant_ISUNDEF 9 +#define PERL_constant_ISUV 10 +#define PERL_constant_ISYES 11 + +#ifndef NVTYPE +typedef double NV; /* 5.6 and later define NVTYPE, and typedef NV to it. */ +#endif +#ifndef aTHX_ +#define aTHX_ /* 5.6 or later define this for threading support. */ +#endif +#ifndef pTHX_ +#define pTHX_ /* 5.6 or later define this for threading support. */ +#endif + +static int +constant (pTHX_ const char *name, STRLEN len, IV *iv_return) { + /* Initially switch on the length of the name. */ + /* When generated this function returned values for the list of names given + in this section of perl code. Rather than manually editing these functions + to add or remove constants, which would result in this comment and section + of code becoming inaccurate, we recommend that you edit this section of + code, and use it to regenerate a new set of constant functions which you + then use to replace the originals. + + Regenerate these constant functions by feeding this entire source file to + perl -x + +#!/spare/local/perls/5.8.6/bin/perl5.8.6 -w +use ExtUtils::Constant qw (constant_types C_constant XS_constant); + +my $types = {map {($_, 1)} qw(IV)}; +my @names = (qw(BZ_CONFIG_ERROR BZ_DATA_ERROR BZ_DATA_ERROR_MAGIC BZ_FINISH + BZ_FINISH_OK BZ_FLUSH BZ_FLUSH_OK BZ_IO_ERROR BZ_MEM_ERROR BZ_OK + BZ_OUTBUFF_FULL BZ_PARAM_ERROR BZ_RUN BZ_RUN_OK + BZ_SEQUENCE_ERROR BZ_STREAM_END BZ_UNEXPECTED_EOF)); + +print constant_types(); # macro defs +foreach (C_constant ("Bzip2", 'constant', 'IV', $types, undef, 3, @names) ) { + print $_, "\n"; # C constant subs +} +print "#### XS Section:\n"; +print XS_constant ("Bzip2", $types); +__END__ + */ + + switch (len) { + case 5: + if (memEQ(name, "BZ_OK", 5)) { +#ifdef BZ_OK + *iv_return = BZ_OK; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + case 6: + if (memEQ(name, "BZ_RUN", 6)) { +#ifdef BZ_RUN + *iv_return = BZ_RUN; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + case 8: + if (memEQ(name, "BZ_FLUSH", 8)) { +#ifdef BZ_FLUSH + *iv_return = BZ_FLUSH; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + case 9: + /* Names all of length 9. */ + /* BZ_FINISH BZ_RUN_OK */ + /* Offset 8 gives the best switch position. */ + switch (name[8]) { + case 'H': + if (memEQ(name, "BZ_FINIS", 8)) { + /* H */ +#ifdef BZ_FINISH + *iv_return = BZ_FINISH; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + case 'K': + if (memEQ(name, "BZ_RUN_O", 8)) { + /* K */ +#ifdef BZ_RUN_OK + *iv_return = BZ_RUN_OK; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + } + break; + case 11: + /* Names all of length 11. */ + /* BZ_FLUSH_OK BZ_IO_ERROR */ + /* Offset 3 gives the best switch position. */ + switch (name[3]) { + case 'F': + if (memEQ(name, "BZ_FLUSH_OK", 11)) { + /* ^ */ +#ifdef BZ_FLUSH_OK + *iv_return = BZ_FLUSH_OK; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + case 'I': + if (memEQ(name, "BZ_IO_ERROR", 11)) { + /* ^ */ +#ifdef BZ_IO_ERROR + *iv_return = BZ_IO_ERROR; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + } + break; + case 12: + /* Names all of length 12. */ + /* BZ_FINISH_OK BZ_MEM_ERROR */ + /* Offset 5 gives the best switch position. */ + switch (name[5]) { + case 'M': + if (memEQ(name, "BZ_MEM_ERROR", 12)) { + /* ^ */ +#ifdef BZ_MEM_ERROR + *iv_return = BZ_MEM_ERROR; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + case 'N': + if (memEQ(name, "BZ_FINISH_OK", 12)) { + /* ^ */ +#ifdef BZ_FINISH_OK + *iv_return = BZ_FINISH_OK; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + } + break; + case 13: + /* Names all of length 13. */ + /* BZ_DATA_ERROR BZ_STREAM_END */ + /* Offset 11 gives the best switch position. */ + switch (name[11]) { + case 'N': + if (memEQ(name, "BZ_STREAM_END", 13)) { + /* ^ */ +#ifdef BZ_STREAM_END + *iv_return = BZ_STREAM_END; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + case 'O': + if (memEQ(name, "BZ_DATA_ERROR", 13)) { + /* ^ */ +#ifdef BZ_DATA_ERROR + *iv_return = BZ_DATA_ERROR; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + } + break; + case 14: + if (memEQ(name, "BZ_PARAM_ERROR", 14)) { +#ifdef BZ_PARAM_ERROR + *iv_return = BZ_PARAM_ERROR; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + case 15: + /* Names all of length 15. */ + /* BZ_CONFIG_ERROR BZ_OUTBUFF_FULL */ + /* Offset 8 gives the best switch position. */ + switch (name[8]) { + case 'F': + if (memEQ(name, "BZ_OUTBUFF_FULL", 15)) { + /* ^ */ +#ifdef BZ_OUTBUFF_FULL + *iv_return = BZ_OUTBUFF_FULL; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + case 'G': + if (memEQ(name, "BZ_CONFIG_ERROR", 15)) { + /* ^ */ +#ifdef BZ_CONFIG_ERROR + *iv_return = BZ_CONFIG_ERROR; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + } + break; + case 17: + /* Names all of length 17. */ + /* BZ_SEQUENCE_ERROR BZ_UNEXPECTED_EOF */ + /* Offset 12 gives the best switch position. */ + switch (name[12]) { + case 'D': + if (memEQ(name, "BZ_UNEXPECTED_EOF", 17)) { + /* ^ */ +#ifdef BZ_UNEXPECTED_EOF + *iv_return = BZ_UNEXPECTED_EOF; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + case 'E': + if (memEQ(name, "BZ_SEQUENCE_ERROR", 17)) { + /* ^ */ +#ifdef BZ_SEQUENCE_ERROR + *iv_return = BZ_SEQUENCE_ERROR; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + } + break; + case 19: + if (memEQ(name, "BZ_DATA_ERROR_MAGIC", 19)) { +#ifdef BZ_DATA_ERROR_MAGIC + *iv_return = BZ_DATA_ERROR_MAGIC; + return PERL_constant_ISIV; +#else + return PERL_constant_NOTDEF; +#endif + } + break; + } + return PERL_constant_NOTFOUND; +} + diff --git a/ext/Compress-Raw-Bzip2/fallback/constants.xs b/ext/Compress-Raw-Bzip2/fallback/constants.xs new file mode 100644 index 0000000..9a4d7c2 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/fallback/constants.xs @@ -0,0 +1,88 @@ +void +constant(sv) + PREINIT: +#ifdef dXSTARG + dXSTARG; /* Faster if we have it. */ +#else + dTARGET; +#endif + STRLEN len; + int type; + IV iv; + /* NV nv; Uncomment this if you need to return NVs */ + /* const char *pv; Uncomment this if you need to return PVs */ + INPUT: + SV * sv; + const char * s = SvPV(sv, len); + PPCODE: + /* Change this to constant(aTHX_ s, len, &iv, &nv); + if you need to return both NVs and IVs */ + type = constant(aTHX_ s, len, &iv); + /* Return 1 or 2 items. First is error message, or undef if no error. + Second, if present, is found value */ + switch (type) { + case PERL_constant_NOTFOUND: + sv = sv_2mortal(newSVpvf("%s is not a valid Bzip2 macro", s)); + PUSHs(sv); + break; + case PERL_constant_NOTDEF: + sv = sv_2mortal(newSVpvf( + "Your vendor has not defined Bzip2 macro %s, used", s)); + PUSHs(sv); + break; + case PERL_constant_ISIV: + EXTEND(SP, 1); + PUSHs(&PL_sv_undef); + PUSHi(iv); + break; + /* Uncomment this if you need to return NOs + case PERL_constant_ISNO: + EXTEND(SP, 1); + PUSHs(&PL_sv_undef); + PUSHs(&PL_sv_no); + break; */ + /* Uncomment this if you need to return NVs + case PERL_constant_ISNV: + EXTEND(SP, 1); + PUSHs(&PL_sv_undef); + PUSHn(nv); + break; */ + /* Uncomment this if you need to return PVs + case PERL_constant_ISPV: + EXTEND(SP, 1); + PUSHs(&PL_sv_undef); + PUSHp(pv, strlen(pv)); + break; */ + /* Uncomment this if you need to return PVNs + case PERL_constant_ISPVN: + EXTEND(SP, 1); + PUSHs(&PL_sv_undef); + PUSHp(pv, iv); + break; */ + /* Uncomment this if you need to return SVs + case PERL_constant_ISSV: + EXTEND(SP, 1); + PUSHs(&PL_sv_undef); + PUSHs(sv); + break; */ + /* Uncomment this if you need to return UNDEFs + case PERL_constant_ISUNDEF: + break; */ + /* Uncomment this if you need to return UVs + case PERL_constant_ISUV: + EXTEND(SP, 1); + PUSHs(&PL_sv_undef); + PUSHu((UV)iv); + break; */ + /* Uncomment this if you need to return YESs + case PERL_constant_ISYES: + EXTEND(SP, 1); + PUSHs(&PL_sv_undef); + PUSHs(&PL_sv_yes); + break; */ + default: + sv = sv_2mortal(newSVpvf( + "Unexpected return type %d while processing Bzip2 macro %s, used", + type, s)); + PUSHs(sv); + } diff --git a/ext/Compress-Raw-Bzip2/lib/Compress/Raw/Bzip2.pm b/ext/Compress-Raw-Bzip2/lib/Compress/Raw/Bzip2.pm new file mode 100644 index 0000000..ebc60a0 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/lib/Compress/Raw/Bzip2.pm @@ -0,0 +1,329 @@ + +package Compress::Raw::Bzip2; + +use strict ; +use warnings ; + +require 5.004 ; +require Exporter; +use AutoLoader; +use Carp ; + +use bytes ; +our ($VERSION, $XS_VERSION, @ISA, @EXPORT, $AUTOLOAD); + +$VERSION = '2.015'; +$XS_VERSION = $VERSION; +$VERSION = eval $VERSION; + +@ISA = qw(Exporter); +# Items to export into callers namespace by default. Note: do not export +# names by default without a very good reason. Use EXPORT_OK instead. +# Do not simply export all your public functions/methods/constants. +@EXPORT = qw( + BZ_RUN + BZ_FLUSH + BZ_FINISH + + BZ_OK + BZ_RUN_OK + BZ_FLUSH_OK + BZ_FINISH_OK + BZ_STREAM_END + BZ_SEQUENCE_ERROR + BZ_PARAM_ERROR + BZ_MEM_ERROR + BZ_DATA_ERROR + BZ_DATA_ERROR_MAGIC + BZ_IO_ERROR + BZ_UNEXPECTED_EOF + BZ_OUTBUFF_FULL + BZ_CONFIG_ERROR + + ); + +sub AUTOLOAD { + my($constname); + ($constname = $AUTOLOAD) =~ s/.*:://; + my ($error, $val) = constant($constname); + Carp::croak $error if $error; + no strict 'refs'; + *{$AUTOLOAD} = sub { $val }; + goto &{$AUTOLOAD}; + +} + +use constant FLAG_APPEND => 1 ; +use constant FLAG_CRC => 2 ; +use constant FLAG_ADLER => 4 ; +use constant FLAG_CONSUME_INPUT => 8 ; + +eval { + require XSLoader; + XSLoader::load('Compress::Raw::Bzip2', $XS_VERSION); + 1; +} +or do { + require DynaLoader; + local @ISA = qw(DynaLoader); + bootstrap Compress::Raw::Bzip2 $XS_VERSION ; +}; + +#sub Compress::Raw::Bzip2::new +#{ +# my $class = shift ; +# my ($ptr, $status) = _new(@_); +# return wantarray ? (undef, $status) : undef +# unless $ptr ; +# my $obj = bless [$ptr], $class ; +# return wantarray ? ($obj, $status) : $obj; +#} +# +#package Compress::Raw::Bunzip2 ; +# +#sub Compress::Raw::Bunzip2::new +#{ +# my $class = shift ; +# my ($ptr, $status) = _new(@_); +# return wantarray ? (undef, $status) : undef +# unless $ptr ; +# my $obj = bless [$ptr], $class ; +# return wantarray ? ($obj, $status) : $obj; +#} + +package Compress::Raw::Bzip2; + +1; + +__END__ + + +=head1 NAME + +Compress::Raw::Bzip2 - Low-Level Interface to bzip2 compression library + +=head1 SYNOPSIS + + use Compress::Raw::Bzip2 ; + + my ($bz, $status) = new Compress::Raw::Bzip2 [OPTS] + or die "Cannot create bzip2 object: $bzerno\n"; + + $status = $bz->bzdeflate($input, $output); + $status = $bz->bzflush($output); + $status = $bz->bzclose($output); + + my ($bz, $status) = new Compress::Raw::Bunzip2 [OPTS] + or die "Cannot create bunzip2 object: $bzerno\n"; + + $status = $bz->bzinflate($input, $output); + + my $version = Compress::Raw::Bzip2::bzlibversion(); + +=head1 DESCRIPTION + +C provides an interface to the in-memory +compression/uncompression functions from the bzip2 compression library. + +Although the primary purpose for the existence of C +is for use by the C and C +modules, it can be used on its own for simple compression/uncompression +tasks. + +=head1 Compression + +=head2 ($z, $status) = new Compress::Raw::Bzip2 $appendOutput, $blockSize100k, $workfactor; + +Creates a new compression object. + +If successful, it will return the initialised compression object, C<$z> +and a C<$status> of C in a list context. In scalar context it +returns the deflation object, C<$z>, only. + +If not successful, the returned compression object, C<$z>, will be +I and C<$status> will hold the a I error code. + +Below is a list of the valid options: + +=over 5 + +=item B<$appendOutput> + +Controls whether the compressed data is appended to the output buffer in +the C, C and C methods. + +Defaults to 1. + +=item B<$blockSize100k> + +To quote the bzip2 documentation + + blockSize100k specifies the block size to be used for compression. It + should be a value between 1 and 9 inclusive, and the actual block size + used is 100000 x this figure. 9 gives the best compression but takes + most memory. + +Defaults to 1. + +=item B<$workfactor> + +To quote the bzip2 documentation + + This parameter controls how the compression phase behaves when + presented with worst case, highly repetitive, input data. If + compression runs into difficulties caused by repetitive data, the + library switches from the standard sorting algorithm to a fallback + algorithm. The fallback is slower than the standard algorithm by + perhaps a factor of three, but always behaves reasonably, no matter how + bad the input. + + Lower values of workFactor reduce the amount of effort the standard + algorithm will expend before resorting to the fallback. You should set + this parameter carefully; too low, and many inputs will be handled by + the fallback algorithm and so compress rather slowly, too high, and + your average-to-worst case compression times can become very large. The + default value of 30 gives reasonable behaviour over a wide range of + circumstances. + + Allowable values range from 0 to 250 inclusive. 0 is a special case, + equivalent to using the default value of 30. + +Defaults to 0. + +=back + +=head2 $status = $bz->bzdeflate($input, $output); + +Reads the contents of C<$input>, compresses it and writes the compressed +data to C<$output>. + +Returns C on success and a C error code on failure. + +If C is enabled in the constructor for the bzip2 object, the +compressed data will be appended to C<$output>. If not enabled, C<$output> +will be truncated before the compressed data is written to it. + +=head2 $status = $bz->bzflush($output); + +Flushes any pending compressed data to C<$output>. + +Returns C on success and a C error code on failure. + +=head2 $status = $bz->bzclose($output); + +Terminates the compressed data stream and flushes any pending compressed +data to C<$output>. + +Returns C on success and a C error code on failure. + +=head2 Example + +=head1 Uncompression + +=head2 ($z, $status) = new Compress::Raw::Bunzip2 $appendOutput, $consumeInput, $small; + +If successful, it will return the initialised uncompression object, C<$z> +and a C<$status> of C in a list context. In scalar context it +returns the deflation object, C<$z>, only. + +If not successful, the returned uncompression object, C<$z>, will be +I and C<$status> will hold the a I error code. + +Below is a list of the valid options: + +=over 5 + +=item B<$appendOutput> + +Controls whether the compressed data is appended to the output buffer in the +C, C and C methods. + +Defaults to 1. + +=item B<$consumeInput> + +=item B<$small> + +To quote the bzip2 documentation + + If small is nonzero, the library will use an alternative decompression + algorithm which uses less memory but at the cost of decompressing more + slowly (roughly speaking, half the speed, but the maximum memory + requirement drops to around 2300k). + +Defaults to 0. + +=back + +=head2 $status = $z->bzinflate($input, $output); + +Uncompresses C<$input> and writes the uncompressed data to C<$output>. + +Returns C if the uncompression was successful, but the end of the +compressed data stream has not been reached. Returns C on +successful uncompression and the end of the compression stream has been +reached. + +If C is enabled in the constructor for the bunzip2 object, +C<$input> will have all compressed data removed from it after +uncompression. On C return this will mean that C<$input> will be an +empty string; when C C<$input> will either be an empty +string or will contain whatever data immediately followed the compressed +data stream. + +If C is enabled in the constructor for the bunzip2 object, +the uncompressed data will be appended to C<$output>. If not enabled, +C<$output> will be truncated before the uncompressed data is written to it. + +=head1 Constants + +The following bzip2 constants are exported by this module + + BZ_RUN + BZ_FLUSH + BZ_FINISH + + BZ_OK + BZ_RUN_OK + BZ_FLUSH_OK + BZ_FINISH_OK + BZ_STREAM_END + BZ_SEQUENCE_ERROR + BZ_PARAM_ERROR + BZ_MEM_ERROR + BZ_DATA_ERROR + BZ_DATA_ERROR_MAGIC + BZ_IO_ERROR + BZ_UNEXPECTED_EOF + BZ_OUTBUFF_FULL + BZ_CONFIG_ERROR + +=head1 SEE ALSO + +L, L, L, L, L, L, L, L, L, L, L, L, L, L, L + +L + +L, L, +L, +L + +The primary site for the bzip2 program is F. + +See the module L + +=head1 AUTHOR + +This module was written by Paul Marquess, F. + +=head1 MODIFICATION HISTORY + +See the Changes file. + +=head1 COPYRIGHT AND LICENSE + +Copyright (c) 2005-2008 Paul Marquess. All rights reserved. + +This program is free software; you can redistribute it and/or +modify it under the same terms as Perl itself. + diff --git a/ext/Compress-Raw-Bzip2/pod/FAQ.pod b/ext/Compress-Raw-Bzip2/pod/FAQ.pod new file mode 100644 index 0000000..f429807 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/pod/FAQ.pod @@ -0,0 +1,119 @@ + +=head1 NAME + +Compress::Raw::Bzip2::FAQ -- Frequently Asked Questions about Compress::Raw::Bzip2 + +=head1 DESCRIPTION + +Common questions answered. + +=head2 Compatibility with Unix compress/uncompress. + +This module is not compatible with Unix C. + +If you have the C program available, you can use this to read +compressed files + + open F, "uncompress -c $filename |"; + while () + { + ... + +Alternatively, if you have the C program available, you can use +this to read compressed files + + open F, "gunzip -c $filename |"; + while () + { + ... + +and this to write compress files, if you have the C program +available + + open F, "| compress -c $filename "; + print F "data"; + ... + close F ; + +=head2 Accessing .tar.Z files + +See previous FAQ item. + +If the C module is installed and either the C or +C programs are available, you can use one of these workarounds to +read C<.tar.Z> files. + +Firstly with C + + use strict; + use warnings; + use Archive::Tar; + + open F, "uncompress -c $filename |"; + my $tar = Archive::Tar->new(*F); + ... + +and this with C + + use strict; + use warnings; + use Archive::Tar; + + open F, "gunzip -c $filename |"; + my $tar = Archive::Tar->new(*F); + ... + +Similarly, if the C program is available, you can use this to +write a C<.tar.Z> file + + use strict; + use warnings; + use Archive::Tar; + use IO::File; + + my $fh = new IO::File "| compress -c >$filename"; + my $tar = Archive::Tar->new(); + ... + $tar->write($fh); + $fh->close ; + +=head2 Accessing Zip Files + +This module does not support reading/writing zip files. + +Support for reading/writing zip files is included with the +C and C modules. + +The primary focus of the C and C +modules is to provide an C compatible streaming read/write +interface to zip files/buffers. They are not fully flegged archivers. If +you are looking for an archiver check out the C module. You +can find it on CPAN at + + http://www.cpan.org/modules/by-module/Archive/Archive-Zip-*.tar.gz + +=head1 SEE ALSO + +L, L, L, L, L, L, L, L, L, L, L, L, L, L, L + +L + +L, L, +L, +L + +=head1 AUTHOR + +This module was written by Paul Marquess, F. + +=head1 MODIFICATION HISTORY + +See the Changes file. + +=head1 COPYRIGHT AND LICENSE + +Copyright (c) 2005-2008 Paul Marquess. All rights reserved. + +This program is free software; you can redistribute it and/or +modify it under the same terms as Perl itself. + diff --git a/ext/Compress-Raw-Bzip2/private/MakeUtil.pm b/ext/Compress-Raw-Bzip2/private/MakeUtil.pm new file mode 100644 index 0000000..e595910 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/private/MakeUtil.pm @@ -0,0 +1,381 @@ +package MakeUtil ; +package main ; + +use strict ; + +use Config qw(%Config); +use File::Copy; + +my $VERSION = '1.0'; + + +BEGIN +{ + eval { require File::Spec::Functions ; File::Spec::Functions->import() } ; + if ($@) + { + *catfile = sub { return "$_[0]/$_[1]" } + } +} + +require VMS::Filespec if $^O eq 'VMS'; + + +unless($ENV{PERL_CORE}) { + $ENV{PERL_CORE} = 1 if grep { $_ eq 'PERL_CORE=1' } @ARGV; +} + +$ENV{SKIP_FOR_CORE} = 1 if $ENV{PERL_CORE} || $ENV{MY_PERL_CORE} ; + + + +sub MY::libscan +{ + my $self = shift; + my $path = shift; + + return undef + if $path =~ /(~|\.bak|_bak)$/ || + $path =~ /\..*\.sw(o|p)$/ || + $path =~ /\B\.svn\b/; + + return $path; +} + +sub MY::postamble +{ + return '' + if $ENV{PERL_CORE} ; + + my @files = getPerlFiles('MANIFEST'); + + # Note: Once you remove all the layers of shell/makefile escaping + # the regular expression below reads + # + # /^\s*local\s*\(\s*\$^W\s*\)/ + # + my $postamble = ' + +MyTrebleCheck: + @echo Checking for $$^W in files: '. "@files" . ' + @perl -ne \' \ + exit 1 if /^\s*local\s*\(\s*\$$\^W\s*\)/; \ + \' ' . " @files || " . ' \ + (echo found unexpected $$^W ; exit 1) + @echo All is ok. + +'; + + return $postamble; +} + +sub getPerlFiles +{ + my @manifests = @_ ; + + my @files = (); + + for my $manifest (@manifests) + { + my $prefix = './'; + + $prefix = $1 + if $manifest =~ m#^(.*/)#; + + open M, "<$manifest" + or die "Cannot open '$manifest': $!\n"; + while () + { + chomp ; + next if /^\s*#/ || /^\s*$/ ; + + s/^\s+//; + s/\s+$//; + + /^(\S+)\s*(.*)$/; + + my ($file, $rest) = ($1, $2); + + if ($file =~ /\.(pm|pl|t)$/ and $file !~ /MakeUtil.pm/) + { + push @files, "$prefix$file"; + } + elsif ($rest =~ /perl/i) + { + push @files, "$prefix$file"; + } + + } + close M; + } + + return @files; +} + +sub UpDowngrade +{ + return if defined $ENV{TipTop}; + + my @files = @_ ; + + # our and use bytes/utf8 is stable from 5.6.0 onward + # warnings is stable from 5.6.1 onward + + # Note: this code assumes that each statement it modifies is not + # split across multiple lines. + + + my $warn_sub = ''; + my $our_sub = '' ; + + my $upgrade ; + my $downgrade ; + my $do_downgrade ; + + my $caller = (caller(1))[3] || ''; + + if ($caller =~ /downgrade/) + { + $downgrade = 1; + } + elsif ($caller =~ /upgrade/) + { + $upgrade = 1; + } + else + { + $do_downgrade = 1 + if $] < 5.006001 ; + } + +# else +# { +# my $opt = shift @ARGV || '' ; +# $upgrade = ($opt =~ /^-upgrade/i); +# $downgrade = ($opt =~ /^-downgrade/i); +# push @ARGV, $opt unless $downgrade || $upgrade; +# } + + + if ($downgrade || $do_downgrade) { + # From: use|no warnings "blah" + # To: local ($^W) = 1; # use|no warnings "blah" + $warn_sub = sub { + s/^(\s*)(no\s+warnings)/${1}local (\$^W) = 0; #$2/ ; + s/^(\s*)(use\s+warnings)/${1}local (\$^W) = 1; #$2/ ; + }; + } + #elsif ($] >= 5.006001 || $upgrade) { + elsif ($upgrade) { + # From: local ($^W) = 1; # use|no warnings "blah" + # To: use|no warnings "blah" + $warn_sub = sub { + s/^(\s*)local\s*\(\$\^W\)\s*=\s*\d+\s*;\s*#\s*((no|use)\s+warnings.*)/$1$2/ ; + }; + } + + if ($downgrade || $do_downgrade) { + $our_sub = sub { + if ( /^(\s*)our\s+\(\s*([^)]+\s*)\)/ ) { + my $indent = $1; + my $vars = join ' ', split /\s*,\s*/, $2; + $_ = "${indent}use vars qw($vars);\n"; + } + elsif ( /^(\s*)((use|no)\s+(bytes|utf8)\s*;.*)$/) + { + $_ = "$1# $2\n"; + } + }; + } + #elsif ($] >= 5.006000 || $upgrade) { + elsif ($upgrade) { + $our_sub = sub { + if ( /^(\s*)use\s+vars\s+qw\((.*?)\)/ ) { + my $indent = $1; + my $vars = join ', ', split ' ', $2; + $_ = "${indent}our ($vars);\n"; + } + elsif ( /^(\s*)#\s*((use|no)\s+(bytes|utf8)\s*;.*)$/) + { + $_ = "$1$2\n"; + } + }; + } + + if (! $our_sub && ! $warn_sub) { + warn "Up/Downgrade not needed.\n"; + if ($upgrade || $downgrade) + { exit 0 } + else + { return } + } + + foreach (@files) { + #if (-l $_ ) + { doUpDown($our_sub, $warn_sub, $_) } + #else + #{ doUpDownViaCopy($our_sub, $warn_sub, $_) } + } + + warn "Up/Downgrade complete.\n" ; + exit 0 if $upgrade || $downgrade; + +} + + +sub doUpDown +{ + my $our_sub = shift; + my $warn_sub = shift; + + return if -d $_[0]; + + local ($^I) = ($^O eq 'VMS') ? "_bak" : ".bak"; + local (@ARGV) = shift; + + while (<>) + { + print, last if /^__(END|DATA)__/ ; + + &{ $our_sub }() if $our_sub ; + &{ $warn_sub }() if $warn_sub ; + print ; + } + + return if eof ; + + while (<>) + { print } +} + +sub doUpDownViaCopy +{ + my $our_sub = shift; + my $warn_sub = shift; + my $file = shift ; + + use File::Copy ; + + return if -d $file ; + + my $backup = $file . ($^O eq 'VMS') ? "_bak" : ".bak"; + + copy($file, $backup) + or die "Cannot copy $file to $backup: $!"; + + my @keep = (); + + { + open F, "<$file" + or die "Cannot open $file: $!\n" ; + while () + { + if (/^__(END|DATA)__/) + { + push @keep, $_; + last ; + } + + &{ $our_sub }() if $our_sub ; + &{ $warn_sub }() if $warn_sub ; + push @keep, $_; + } + + if (! eof F) + { + while () + { push @keep, $_ } + } + close F; + } + + { + open F, ">$file" + or die "Cannot open $file: $!\n"; + print F @keep ; + close F; + } +} + + +sub FindBrokenDependencies +{ + my $version = shift ; + my %thisModule = map { $_ => 1} @_; + + my @modules = qw( + IO::Compress::Base + IO::Compress::Base::Common + IO::Uncompress::Base + + Compress::Raw::Zlib + Compress::Raw::Bzip2 + + IO::Compress::RawDeflate + IO::Uncompress::RawInflate + IO::Compress::Deflate + IO::Uncompress::Inflate + IO::Compress::Gzip + IO::Compress::Gzip::Constants + IO::Uncompress::Gunzip + IO::Compress::Zip + IO::Uncompress::Unzip + + IO::Compress::Bzip2 + IO::Uncompress::Bunzip2 + + IO::Compress::Lzf + IO::Uncompress::UnLzf + + IO::Compress::Lzop + IO::Uncompress::UnLzop + + Compress::Zlib + ); + + my @broken = (); + + foreach my $module ( grep { ! $thisModule{$_} } @modules) + { + my $hasVersion = getInstalledVersion($module); + + # No need to upgrade if the module isn't installed at all + next + if ! defined $hasVersion; + + # If already have C::Z version 1, then an upgrade to any of the + # IO::Compress modules will not break it. + next + if $module eq 'Compress::Zlib' && $hasVersion < 2; + + if ($hasVersion < $version) + { + push @broken, $module + } + } + + return @broken; +} + +sub getInstalledVersion +{ + my $module = shift; + my $version; + + eval " require $module; "; + + if ($@ eq '') + { + no strict 'refs'; + $version = ${ $module . "::VERSION" }; + $version = 0 + } + + return $version; +} + +package MakeUtil ; + +1; + + diff --git a/ext/Compress-Raw-Bzip2/t/000prereq.t b/ext/Compress-Raw-Bzip2/t/000prereq.t new file mode 100644 index 0000000..99d249b --- /dev/null +++ b/ext/Compress-Raw-Bzip2/t/000prereq.t @@ -0,0 +1,57 @@ +BEGIN { + if ($ENV{PERL_CORE}) { + chdir 't' if -d 't'; + @INC = ("../lib", "lib/compress"); + } +} + +use lib qw(t t/compress); +use strict ; +use warnings ; + +use Test::More ; + +BEGIN +{ + # use Test::NoWarnings, if available + my $extra = 0 ; + $extra = 1 + if eval { require Test::NoWarnings ; import Test::NoWarnings; 1 }; + + + my $VERSION = '2.015'; + my @NAMES = qw( + + ); + + my @OPT = qw( + + ); + + plan tests => 1 + @NAMES + @OPT + $extra ; + + ok 1; + + foreach my $name (@NAMES) + { + use_ok($name, $VERSION); + } + + + foreach my $name (@OPT) + { + eval " require $name " ; + if ($@) + { + ok 1, "$name not available" + } + else + { + my $ver = eval("\$${name}::VERSION"); + is $ver, $VERSION, "$name version should be $VERSION" + or diag "$name version is $ver, need $VERSION" ; + } + } + +} + diff --git a/ext/Compress-Raw-Bzip2/t/01bzip2.t b/ext/Compress-Raw-Bzip2/t/01bzip2.t new file mode 100644 index 0000000..407d16a --- /dev/null +++ b/ext/Compress-Raw-Bzip2/t/01bzip2.t @@ -0,0 +1,567 @@ +BEGIN { + if ($ENV{PERL_CORE}) { + chdir 't' if -d 't'; + #@INC = ("../lib", "lib/compress"); + @INC = ("../lib"); + } +} + +use lib 't'; +use strict; +use warnings; +use bytes; + +use Test::More ; +#use CompTestUtils; + + +BEGIN +{ + # use Test::NoWarnings, if available + my $extra = 0 ; + $extra = 1 + if eval { require Test::NoWarnings ; import Test::NoWarnings; 1 }; + + + my $count = 0 ; + if ($] < 5.005) { + $count = 103 ; + } + elsif ($] >= 5.006) { + $count = 157 ; + } + else { + $count = 115 ; + } + + plan tests => $count + $extra; + + use_ok('Compress::Raw::Bzip2') ; +} + +sub title +{ + #diag "" ; + ok 1, $_[0] ; + #diag "" ; +} + +sub mkErr +{ + my $string = shift ; + my ($dummy, $file, $line) = caller ; + -- $line ; + + $string = quotemeta $string; + $file = quotemeta($file); + + #return "/$string\\s+at $file line $line/" if $] >= 5.006 ; + return "/$string\\s+at /" ; +} + +sub mkEvalErr +{ + my $string = shift ; + + return "/$string\\s+at \\(eval /" if $] > 5.006 ; + return "/$string\\s+at /" ; +} + + + +my $hello = <uncompressedBytes(), 0, "uncompressedBytes() == 0" ; + is $x->compressedBytes(), 0, "compressedBytes() == 0" ; + + $X = "" ; + my $Answer = ''; + foreach (@hello) + { + $status = $x->bzdeflate($_, $X) ; + last unless $status == BZ_RUN_OK ; + + $Answer .= $X ; + } + + cmp_ok $status, '==', BZ_RUN_OK, "bzdeflate returned BZ_RUN_OK" ; + + cmp_ok $x->bzflush($X), '==', BZ_RUN_OK, "bzflush returned BZ_RUN_OK" ; + $Answer .= $X ; + + is $x->uncompressedBytes(), length $hello, "uncompressedBytes ok" ; + is $x->compressedBytes(), length $Answer, "compressedBytes ok" ; + + cmp_ok $x->bzclose($X), '==', BZ_STREAM_END, "bzclose returned BZ_STREAM_END"; + $Answer .= $X ; + + #open F, ">/tmp/xx1"; print F $Answer ; close F; + my @Answer = split('', $Answer) ; + + my $k; + ok(($k, $err) = new Compress::Raw::Bunzip2(0, 0)); + ok $k, "Compress::Raw::Bunzip2 ok" ; + cmp_ok $err, '==', BZ_OK, "status is BZ_OK" ; + + is $k->compressedBytes(), 0, "compressedBytes() == 0" ; + is $k->uncompressedBytes(), 0, "uncompressedBytes() == 0" ; + my $GOT = ''; + my $Z; + $Z = 1 ;#x 2000 ; + foreach (@Answer) + { + $status = $k->bzinflate($_, $Z) ; + $GOT .= $Z ; + last if $status == BZ_STREAM_END or $status != BZ_OK ; + + } + + cmp_ok $status, '==', BZ_STREAM_END, "Got BZ_STREAM_END" ; + is $GOT, $hello, "uncompressed data matches ok" ; + is $k->compressedBytes(), length $Answer, "compressedBytes ok" ; + is $k->uncompressedBytes(), length $hello , "uncompressedBytes ok"; + +} + + +{ + # bzdeflate/bzinflate - small buffer with a number + # ============================== + + my $hello = 6529 ; + + ok my ($x, $err) = new Compress::Raw::Bzip2 (1) ; + ok $x ; + cmp_ok $err, '==', BZ_OK ; + + my $status; + my $Answer = ''; + + cmp_ok $x->bzdeflate($hello, $Answer), '==', BZ_RUN_OK ; + + cmp_ok $x->bzclose($Answer), '==', BZ_STREAM_END, "bzclose returned BZ_STREAM_END"; + + my @Answer = split('', $Answer) ; + + my $k; + ok(($k, $err) = new Compress::Raw::Bunzip2(1, 0) ); + ok $k ; + cmp_ok $err, '==', BZ_OK ; + + #my $GOT = ''; + my $GOT ; + foreach (@Answer) + { + $status = $k->bzinflate($_, $GOT) ; + last if $status == BZ_STREAM_END or $status != BZ_OK ; + + } + + cmp_ok $status, '==', BZ_STREAM_END ; + is $GOT, $hello ; + +} + +{ + +# bzdeflate/bzinflate options - AppendOutput +# ================================ + + # AppendOutput + # CRC + + my $hello = "I am a HAL 9000 computer" ; + my @hello = split('', $hello) ; + + ok my ($x, $err) = new Compress::Raw::Bzip2 (1) ; + ok $x ; + cmp_ok $err, '==', BZ_OK ; + + my $status; + my $X; + foreach (@hello) + { + $status = $x->bzdeflate($_, $X) ; + last unless $status == BZ_RUN_OK ; + } + + cmp_ok $status, '==', BZ_RUN_OK ; + + cmp_ok $x->bzclose($X), '==', BZ_STREAM_END ; + + + my @Answer = split('', $X) ; + + my $k; + ok(($k, $err) = new Compress::Raw::Bunzip2( {-Bufsize => 1, -AppendOutput =>1})); + ok $k ; + cmp_ok $err, '==', BZ_OK ; + + my $Z; + foreach (@Answer) + { + $status = $k->bzinflate($_, $Z) ; + last if $status == BZ_STREAM_END or $status != BZ_OK ; + + } + + cmp_ok $status, '==', BZ_STREAM_END ; + is $Z, $hello ; +} + + +{ + + title "bzdeflate/bzinflate - larger buffer"; + # ============================== + + # generate a long random string + my $contents = '' ; + foreach (1 .. 50000) + { $contents .= chr int rand 255 } + + + ok my ($x, $err) = new Compress::Raw::Bzip2(0) ; + ok $x ; + cmp_ok $err, '==', BZ_OK ; + + my (%X, $Y, %Z, $X, $Z); + #cmp_ok $x->bzdeflate($contents, $X{key}), '==', BZ_RUN_OK ; + cmp_ok $x->bzdeflate($contents, $X), '==', BZ_RUN_OK ; + + #$Y = $X{key} ; + $Y = $X ; + + + #cmp_ok $x->bzflush($X{key}), '==', BZ_RUN_OK ; + #$Y .= $X{key} ; + cmp_ok $x->bzclose($X), '==', BZ_STREAM_END ; + $Y .= $X ; + + + + my $keep = $Y ; + + my $k; + ok(($k, $err) = new Compress::Raw::Bunzip2(0, 0) ); + ok $k ; + cmp_ok $err, '==', BZ_OK ; + + #cmp_ok $k->bzinflate($Y, $Z{key}), '==', BZ_STREAM_END ; + #ok $contents eq $Z{key} ; + cmp_ok $k->bzinflate($Y, $Z), '==', BZ_STREAM_END ; + ok $contents eq $Z ; + + # redo bzdeflate with AppendOutput + + ok (($k, $err) = new Compress::Raw::Bunzip2(1, 0)) ; + ok $k ; + cmp_ok $err, '==', BZ_OK ; + + my $s ; + my $out ; + my @bits = split('', $keep) ; + foreach my $bit (@bits) { + $s = $k->bzinflate($bit, $out) ; + } + + cmp_ok $s, '==', BZ_STREAM_END ; + + ok $contents eq $out ; + + +} + + +for my $consume ( 0 .. 1) +{ + title "bzinflate - check remaining buffer after BZ_STREAM_END, Consume $consume"; + + ok my $x = new Compress::Raw::Bzip2(0) ; + + my ($X, $Y, $Z); + cmp_ok $x->bzdeflate($hello, $X), '==', BZ_RUN_OK; + cmp_ok $x->bzclose($Y), '==', BZ_STREAM_END; + $X .= $Y ; + + ok my $k = new Compress::Raw::Bunzip2(0, $consume) ; + + my $first = substr($X, 0, 2) ; + my $remember_first = $first ; + my $last = substr($X, 2) ; + cmp_ok $k->bzinflate($first, $Z), '==', BZ_OK; + if ($consume) { + ok $first eq "" ; + } + else { + ok $first eq $remember_first ; + } + + my $T ; + $last .= "appendage" ; + my $remember_last = $last ; + cmp_ok $k->bzinflate($last, $T), '==', BZ_STREAM_END; + is $hello, $Z . $T ; + if ($consume) { + is $last, "appendage" ; + } + else { + is $last, $remember_last ; + } + +} + + +{ + title "ConsumeInput and a read-only buffer trapped" ; + + ok my $k = new Compress::Raw::Bunzip2(0, 1) ; + + my $Z; + eval { $k->bzinflate("abc", $Z) ; }; + like $@, mkErr("Compress::Raw::Bunzip2::bzinflate input parameter cannot be read-only when ConsumeInput is specified"); + +} + +foreach (1 .. 2) +{ + next if $[ < 5.005 ; + + title 'test bzinflate/bzdeflate with a substr'; + + my $contents = '' ; + foreach (1 .. 5000) + { $contents .= chr int rand 255 } + ok my $x = new Compress::Raw::Bzip2(1) ; + + my $X ; + my $status = $x->bzdeflate(substr($contents,0), $X); + cmp_ok $status, '==', BZ_RUN_OK ; + + cmp_ok $x->bzflush($X), '==', BZ_RUN_OK ; + + my $append = "Appended" ; + $X .= $append ; + + ok my $k = new Compress::Raw::Bunzip2(1, 0) ; + + my $Z; + my $keep = $X ; + $status = $k->bzinflate(substr($X, 0), $Z) ; + + cmp_ok $status, '==', BZ_STREAM_END ; + #print "status $status X [$X]\n" ; + is $contents, $Z ; + ok $X eq $append; + #is length($X), length($append); + #ok $X eq $keep; + #is length($X), length($keep); +} + +title 'Looping Append test - checks that deRef_l resets the output buffer'; +foreach (1 .. 2) +{ + + my $hello = "I am a HAL 9000 computer" ; + my @hello = split('', $hello) ; + my ($err, $x, $X, $status); + + ok( ($x, $err) = new Compress::Raw::Bzip2 (0) ); + ok $x ; + cmp_ok $err, '==', BZ_OK ; + + $X = "" ; + my $Answer = ''; + foreach (@hello) + { + $status = $x->bzdeflate($_, $X) ; + last unless $status == BZ_RUN_OK ; + + $Answer .= $X ; + } + + cmp_ok $status, '==', BZ_RUN_OK ; + + cmp_ok $x->bzclose($X), '==', BZ_STREAM_END ; + $Answer .= $X ; + + my @Answer = split('', $Answer) ; + + my $k; + ok(($k, $err) = new Compress::Raw::Bunzip2(1, 0) ); + ok $k ; + cmp_ok $err, '==', BZ_OK ; + + my $GOT ; + my $Z; + $Z = 1 ;#x 2000 ; + foreach (@Answer) + { + $status = $k->bzinflate($_, $GOT) ; + last if $status == BZ_STREAM_END or $status != BZ_OK ; + } + + cmp_ok $status, '==', BZ_STREAM_END ; + is $GOT, $hello ; + +} + +if ($] >= 5.005) +{ + title 'test bzinflate input parameter via substr'; + + my $hello = "I am a HAL 9000 computer" ; + my $data = $hello ; + + my($X, $Z); + + ok my $x = new Compress::Raw::Bzip2 (1); + + cmp_ok $x->bzdeflate($data, $X), '==', BZ_RUN_OK ; + + cmp_ok $x->bzclose($X), '==', BZ_STREAM_END ; + + my $append = "Appended" ; + $X .= $append ; + my $keep = $X ; + + ok my $k = new Compress::Raw::Bunzip2 ( 1, 1); + +# cmp_ok $k->bzinflate(substr($X, 0, -1), $Z), '==', BZ_STREAM_END ; ; + cmp_ok $k->bzinflate(substr($X, 0), $Z), '==', BZ_STREAM_END ; ; + + ok $hello eq $Z ; + is $X, $append; + + $X = $keep ; + $Z = ''; + ok $k = new Compress::Raw::Bunzip2 ( 1, 0); + + cmp_ok $k->bzinflate(substr($X, 0, -1), $Z), '==', BZ_STREAM_END ; ; + #cmp_ok $k->bzinflate(substr($X, 0), $Z), '==', BZ_STREAM_END ; ; + + ok $hello eq $Z ; + is $X, $keep; + +} + +exit if $] < 5.006 ; + +title 'Looping Append test with substr output - substr the end of the string'; +foreach (1 .. 2) +{ + + my $hello = "I am a HAL 9000 computer" ; + my @hello = split('', $hello) ; + my ($err, $x, $X, $status); + + ok( ($x, $err) = new Compress::Raw::Bzip2 (1) ); + ok $x ; + cmp_ok $err, '==', BZ_OK ; + + $X = "" ; + my $Answer = ''; + foreach (@hello) + { + $status = $x->bzdeflate($_, substr($Answer, length($Answer))) ; + last unless $status == BZ_RUN_OK ; + + } + + cmp_ok $status, '==', BZ_RUN_OK ; + + cmp_ok $x->bzclose(substr($Answer, length($Answer))), '==', BZ_STREAM_END ; + + my @Answer = split('', $Answer) ; + + my $k; + ok(($k, $err) = new Compress::Raw::Bunzip2(1, 0) ); + ok $k ; + cmp_ok $err, '==', BZ_OK ; + + my $GOT = ''; + my $Z; + $Z = 1 ;#x 2000 ; + foreach (@Answer) + { + $status = $k->bzinflate($_, substr($GOT, length($GOT))) ; + last if $status == BZ_STREAM_END or $status != BZ_OK ; + } + + cmp_ok $status, '==', BZ_STREAM_END ; + is $GOT, $hello ; + +} + +title 'Looping Append test with substr output - substr the complete string'; +foreach (1 .. 2) +{ + + my $hello = "I am a HAL 9000 computer" ; + my @hello = split('', $hello) ; + my ($err, $x, $X, $status); + + ok( ($x, $err) = new Compress::Raw::Bzip2 (1) ); + ok $x ; + cmp_ok $err, '==', BZ_OK ; + + $X = "" ; + my $Answer = ''; + foreach (@hello) + { + $status = $x->bzdeflate($_, substr($Answer, 0)) ; + last unless $status == BZ_RUN_OK ; + + } + + cmp_ok $status, '==', BZ_RUN_OK ; + + cmp_ok $x->bzclose(substr($Answer, 0)), '==', BZ_STREAM_END ; + + my @Answer = split('', $Answer) ; + + my $k; + ok(($k, $err) = new Compress::Raw::Bunzip2(1, 0) ); + ok $k ; + cmp_ok $err, '==', BZ_OK ; + + my $GOT = ''; + my $Z; + $Z = 1 ;#x 2000 ; + foreach (@Answer) + { + $status = $k->bzinflate($_, substr($GOT, 0)) ; + last if $status == BZ_STREAM_END or $status != BZ_OK ; + } + + cmp_ok $status, '==', BZ_STREAM_END ; + is $GOT, $hello ; +} + diff --git a/ext/Compress-Raw-Bzip2/t/99pod.t b/ext/Compress-Raw-Bzip2/t/99pod.t new file mode 100644 index 0000000..760f737 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/t/99pod.t @@ -0,0 +1,16 @@ +BEGIN { + if ($ENV{PERL_CORE}) { + chdir 't' if -d 't'; + @INC = ("../lib", "lib/compress"); + } +} + +use lib qw(t t/compress); +use Test::More; + +eval "use Test::Pod 1.00"; + +plan skip_all => "Test::Pod 1.00 required for testing POD" if $@; + +all_pod_files_ok(); + diff --git a/ext/Compress-Raw-Bzip2/typemap b/ext/Compress-Raw-Bzip2/typemap new file mode 100644 index 0000000..8736816 --- /dev/null +++ b/ext/Compress-Raw-Bzip2/typemap @@ -0,0 +1,54 @@ +Compress::Raw::Bzip2 T_PTROBJ +Compress::Raw::Bunzip2 T_PTROBJ + +const char * T_PV +char * T_PV +uLong T_UV +z_off_t T_UV +DualType T_DUAL +int_undef T_IV_undef + + +############################################################################# +INPUT +T_UV + $var = (unsigned long)SvUV($arg) +T_IV_undef + if (SvOK($arg)) + $var = SvIV($arg); + else + $var = 0 ; +T_PV + if (SvOK($arg)) + $var = ($type)SvPVbyte_nolen($arg); + else + $var = NULL ; + +T_PTROBJ_AV + if ($arg == &PL_sv_undef || $arg == NULL) + $var = NULL ; + else if (sv_derived_from($arg, \"${ntype}\")) { + IV tmp = SvIV(getInnerObject($arg)) ; + $var = INT2PTR($type, tmp); + + } + else + croak(\"$var is not of type ${ntype}\") + + + +############################################################################# +OUTPUT +T_UV + sv_setuv($arg, (IV)$var); + + +T_DUAL + setDUALstatus($arg, $var) ; + + +T_PV + sv_setpv((SV*)$arg, $var); + + +