diff --git a/Utilities/cmcompress/CMakeLists.txt b/Utilities/cmcompress/CMakeLists.txt new file mode 100644 index 000000000..f7aacfe00 --- /dev/null +++ b/Utilities/cmcompress/CMakeLists.txt @@ -0,0 +1,3 @@ +PROJECT(CMCompress) + +ADD_LIBRARY(cmcompress cmcompress.c) diff --git a/Utilities/cmcompress/cmcompress.c b/Utilities/cmcompress/cmcompress.c new file mode 100644 index 000000000..6ac4956ec --- /dev/null +++ b/Utilities/cmcompress/cmcompress.c @@ -0,0 +1,545 @@ +/* + * Copyright (c) 1985, 1986 The Regents of the University of California. + * All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * James A. Woods, derived from original work by Spencer Thomas + * and Joseph Orost. + * + * 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. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. All advertising materials mentioning features or use of this software + * must display the following acknowledgement: + * This product includes software developed by the University of + * California, Berkeley and its contributors. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``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 REGENTS OR CONTRIBUTORS 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. + */ + +#include "cmcompress.h" + +#include +#include + +static const char_type magic_header[] = { "\037\235" }; /* 1F 9D */ + +/* Defines for third byte of header */ +#define BIT_MASK 0x1f +#define BLOCK_MASK 0x80 +#define CHECK_GAP 10000 /* ratio check interval */ +/* Masks 0x40 and 0x20 are free. I think 0x20 should mean that there is + a fourth header byte (for expansion). + */ +#define INIT_BITS 9 /* initial number of bits/code */ + +#ifdef COMPATIBLE /* But wrong! */ +# define MAXCODE(n_bits) (1 << (n_bits) - 1) +#else +# define MAXCODE(n_bits) ((1 << (n_bits)) - 1) +#endif /* COMPATIBLE */ + +#define htabof(i) cdata->htab[i] +#define codetabof(i) cdata->codetab[i] + +/* + * the next two codes should not be changed lightly, as they must not + * lie within the contiguous general code space. + */ +#define FIRST 257 /* first free entry */ +#define CLEAR 256 /* table clear output code */ + +static void cl_hash(struct cmcompress_stream* cdata, count_int hsize); /* reset code table */ +static void cl_block (struct cmcompress_stream* cdata); /* table clear for block compress */ +static void prratio( FILE *stream, long int num, long int den); +static void output(struct cmcompress_stream* cdata, code_int code); +static void writeerr(); + +int cmcompress_compress_initialize(struct cmcompress_stream* cdata) +{ + cdata->maxbits = BITS; /* user settable max # bits/code */ + cdata->maxmaxcode = 1 << BITS; /* should NEVER generate this code */ + cdata->hsize = HSIZE; /* for dynamic table sizing */ + cdata->free_ent = 0; /* first unused entry */ + cdata->nomagic = 0; /* Use a 3-byte magic number header, unless old file */ + cdata->block_compress = BLOCK_MASK; + cdata->clear_flg = 0; + cdata->ratio = 0; + cdata->checkpoint = CHECK_GAP; + + cdata->input_stream = 0; + cdata->output_stream = 0; + cdata->client_data = 0; + return 1; +} + +/* + * compress stdin to stdout + * + * Algorithm: use open addressing double hashing (no chaining) on the + * prefix code / next character combination. We do a variant of Knuth's + * algorithm D (vol. 3, sec. 6.4) along with G. Knott's relatively-prime + * secondary probe. Here, the modular division first probe is gives way + * to a faster exclusive-or manipulation. Also do block compression with + * an adaptive reset, whereby the code table is cleared when the compression + * ratio decreases, but after the table fills. The variable-length output + * codes are re-sized at this point, and a special CLEAR code is generated + * for the decompressor. Late addition: construct the table according to + * file size for noticeable speed improvement on small files. Please direct + * questions about this implementation to ames!jaw. + */ + +int cmcompress_compress_start(struct cmcompress_stream* cdata) +{ +#ifndef COMPATIBLE + if (cdata->nomagic == 0) + { + char headLast = (char)(cdata->maxbits | cdata->block_compress); + cdata->output_stream(cdata, magic_header, 2); + cdata->output_stream(cdata, &headLast, 1); + if(ferror(stdout)) + { + printf("Error...\n"); + } + } +#endif /* COMPATIBLE */ + + cdata->offset = 0; + cdata->bytes_out = 3; /* includes 3-byte header mojo */ + cdata->out_count = 0; + cdata->clear_flg = 0; + cdata->ratio = 0; + cdata->in_count = 1; + cdata->checkpoint = CHECK_GAP; + cdata->maxcode = MAXCODE(cdata->n_bits = INIT_BITS); + cdata->free_ent = ((cdata->block_compress) ? FIRST : 256 ); + + cdata->first_pass = 1; + + cdata->hshift = 0; + for ( cdata->fcode = (long) cdata->hsize; cdata->fcode < 65536L; cdata->fcode *= 2L ) + { + cdata->hshift++; + } + cdata->hshift = 8 - cdata->hshift; /* set hash code range bound */ + + cdata->hsize_reg = cdata->hsize; + cl_hash(cdata, (count_int) cdata->hsize_reg); /* clear hash table */ + + return 1; +} + +int cmcompress_compress(struct cmcompress_stream* cdata, void* buff, size_t n) +{ + register code_int i; + register int c; + register int disp; + + unsigned char* input_buffer = (unsigned char*)buff; + + size_t cc; + + /*printf("cmcompress_compress(%p, %p, %d)\n", cdata, buff, n);*/ + + if ( cdata->first_pass ) + { + cdata->ent = input_buffer[0]; + ++ input_buffer; + -- n; + cdata->first_pass = 0; + } + + for ( cc = 0; cc < n; ++ cc ) + { + c = input_buffer[cc]; + cdata->in_count++; + cdata->fcode = (long) (((long) c << cdata->maxbits) + cdata->ent); + i = ((c << cdata->hshift) ^ cdata->ent); /* xor hashing */ + + if ( htabof (i) == cdata->fcode ) + { + cdata->ent = codetabof (i); + continue; + } + else if ( (long)htabof (i) < 0 ) /* empty slot */ + { + goto nomatch; + } + disp = cdata->hsize_reg - i; /* secondary hash (after G. Knott) */ + if ( i == 0 ) + { + disp = 1; + } +probe: + if ( (i -= disp) < 0 ) + { + i += cdata->hsize_reg; + } + + if ( htabof (i) == cdata->fcode ) + { + cdata->ent = codetabof (i); + continue; + } + if ( (long)htabof (i) > 0 ) + { + goto probe; + } +nomatch: + output(cdata, (code_int) cdata->ent ); + cdata->out_count++; + cdata->ent = c; + if ( +#ifdef SIGNED_COMPARE_SLOW + (unsigned) cdata->free_ent < (unsigned) cdata->maxmaxcode +#else + cdata->free_ent < cdata->maxmaxcode +#endif + ) + { + codetabof (i) = cdata->free_ent++; /* code -> hashtable */ + htabof (i) = cdata->fcode; + } + else if ( (count_int)cdata->in_count >= cdata->checkpoint && cdata->block_compress ) + { + cl_block (cdata); + } + } + + return 1; +} + +int cmcompress_compress_finalize(struct cmcompress_stream* cdata) +{ + /* + * Put out the final code. + */ + output(cdata, (code_int)cdata->ent ); + cdata->out_count++; + output(cdata, (code_int)-1 ); + + if(cdata->bytes_out > cdata->in_count) /* exit(2) if no savings */ + { + return 0; + } + return 1; +} + +void cl_block (struct cmcompress_stream* cdata) /* table clear for block compress */ +{ + register long int rat; + + cdata->checkpoint = cdata->in_count + CHECK_GAP; +#ifdef DEBUG + if ( cdata->debug ) + { + fprintf ( stderr, "count: %ld, ratio: ", cdata->in_count ); + prratio ( stderr, cdata->in_count, cdata->bytes_out ); + fprintf ( stderr, "\n"); + } +#endif /* DEBUG */ + + if(cdata->in_count > 0x007fffff) + { /* shift will overflow */ + rat = cdata->bytes_out >> 8; + if(rat == 0) + { /* Don't divide by zero */ + rat = 0x7fffffff; + } + else + { + rat = cdata->in_count / rat; + } + } + else + { + rat = (cdata->in_count << 8) / cdata->bytes_out; /* 8 fractional bits */ + } + if ( rat > cdata->ratio ) + { + cdata->ratio = rat; + } + else + { + cdata->ratio = 0; +#ifdef DEBUG + if(cdata->verbose) + { + dump_tab(); /* dump string table */ + } +#endif + cl_hash (cdata, (count_int) cdata->hsize ); + cdata->free_ent = FIRST; + cdata->clear_flg = 1; + output (cdata, (code_int) CLEAR ); +#ifdef DEBUG + if(cdata->debug) + { + fprintf ( stderr, "clear\n" ); + } +#endif /* DEBUG */ + } +} + +void cl_hash(struct cmcompress_stream* cdata, count_int hsize) /* reset code table */ +{ + register count_int *htab_p = cdata->htab+hsize; + register long i; + register long m1 = -1; + + i = hsize - 16; + do + { /* might use Sys V memset(3) here */ + *(htab_p-16) = m1; + *(htab_p-15) = m1; + *(htab_p-14) = m1; + *(htab_p-13) = m1; + *(htab_p-12) = m1; + *(htab_p-11) = m1; + *(htab_p-10) = m1; + *(htab_p-9) = m1; + *(htab_p-8) = m1; + *(htab_p-7) = m1; + *(htab_p-6) = m1; + *(htab_p-5) = m1; + *(htab_p-4) = m1; + *(htab_p-3) = m1; + *(htab_p-2) = m1; + *(htab_p-1) = m1; + htab_p -= 16; + } + while ((i -= 16) >= 0); + for ( i += 16; i > 0; i-- ) + { + *--htab_p = m1; + } +} + +#if defined(DEBUG) +void prratio(FILE *stream, long int num, long int den) +{ + register int q; /* Doesn't need to be long */ + + if(num > 214748L) + { /* 2147483647/10000 */ + q = num / (den / 10000L); + } + else + { + q = 10000L * num / den; /* Long calculations, though */ + } + if (q < 0) + { + putc('-', stream); + q = -q; + } + fprintf(stream, "%d.%02d%%", q / 100, q % 100); +} +#endif + +/*- + * Output the given code. + * Inputs: + * code: A n_bits-bit integer. If == -1, then EOF. This assumes + * that n_bits =< (long)wordsize - 1. + * Outputs: + * Outputs code to the file. + * Assumptions: + * Chars are 8 bits long. + * Algorithm: + * Maintain a BITS character long buffer (so that 8 codes will + * fit in it exactly). Use the VAX insv instruction to insert each + * code in turn. When the buffer fills up empty it and start over. + */ + +static char buf[BITS]; + +#ifndef vax +char_type lmask[9] = {0xff, 0xfe, 0xfc, 0xf8, 0xf0, 0xe0, 0xc0, 0x80, 0x00}; +char_type rmask[9] = {0x00, 0x01, 0x03, 0x07, 0x0f, 0x1f, 0x3f, 0x7f, 0xff}; +#endif /* vax */ + +void output(struct cmcompress_stream* cdata, code_int code) +{ +#ifdef DEBUG + static int col = 0; +#endif /* DEBUG */ + + /* + * On the VAX, it is important to have the register declarations + * in exactly the order given, or the asm will break. + */ + register int r_off = cdata->offset, bits= cdata->n_bits; + register char * bp = buf; + +#ifdef DEBUG + if ( verbose ) + { + fprintf( stderr, "%5d%c", code, + (col+=6) >= 74 ? (col = 0, '\n') : ' ' ); + } +#endif /* DEBUG */ + if ( code >= 0 ) + { +#if defined(vax) && !defined(__GNUC__) + /* + * VAX and PCC DEPENDENT!! Implementation on other machines is + * below. + * + * Translation: Insert BITS bits from the argument starting at + * cdata->offset bits from the beginning of buf. + */ + 0; /* Work around for pcc -O bug with asm and if stmt */ + asm( "insv 4(ap),r11,r10,(r9)" ); +#else + /* + * byte/bit numbering on the VAX is simulated by the following code + */ + /* + * Get to the first byte. + */ + bp += (r_off >> 3); + r_off &= 7; + /* + * Since code is always >= 8 bits, only need to mask the first + * hunk on the left. + */ + *bp = (*bp & rmask[r_off]) | ((code << r_off) & lmask[r_off]); + bp++; + bits -= (8 - r_off); + code >>= 8 - r_off; + /* Get any 8 bit parts in the middle (<=1 for up to 16 bits). */ + if ( bits >= 8 ) + { + *bp++ = code; + code >>= 8; + bits -= 8; + } + /* Last bits. */ + if(bits) + { + *bp = code; + } +#endif /* vax */ + cdata->offset += cdata->n_bits; + if ( cdata->offset == (cdata->n_bits << 3) ) + { + bp = buf; + bits = cdata->n_bits; + cdata->bytes_out += bits; + do + { + if ( cdata->output_stream(cdata, bp, 1) != 1 ) + //putchar(*bp++); + //if (ferror(stdout)) + { + writeerr(); + } + bp++; + } + while(--bits); + cdata->offset = 0; + } + + /* + * If the next entry is going to be too big for the code size, + * then increase it, if possible. + */ + if ( cdata->free_ent > cdata->maxcode || (cdata->clear_flg > 0)) + { + /* + * Write the whole buffer, because the input side won't + * discover the size increase until after it has read it. + */ + if ( cdata->offset > 0 ) + { + if ( cdata->output_stream(cdata, buf, cdata->n_bits) != cdata->n_bits ) + //if( fwrite( buf, 1, cdata->n_bits, stdout ) != cdata->n_bits) + { + writeerr(); + } + cdata->bytes_out += cdata->n_bits; + } + cdata->offset = 0; + + if ( cdata->clear_flg ) + { + cdata->maxcode = MAXCODE (cdata->n_bits = INIT_BITS); + cdata->clear_flg = 0; + } + else + { + cdata->n_bits++; + if ( cdata->n_bits == cdata->maxbits ) + { + cdata->maxcode = cdata->maxmaxcode; + } + else + { + cdata->maxcode = MAXCODE(cdata->n_bits); + } + } +#ifdef DEBUG + if ( debug ) + { + fprintf( stderr, "\nChange to %d bits\n", cdata->n_bits ); + col = 0; + } +#endif /* DEBUG */ + } + } + else + { + /* + * At EOF, write the rest of the buffer. + */ + if ( cdata->offset > 0 ) + { + cdata->offset = (cdata->offset + 7) / 8; + if ( cdata->output_stream(cdata, buf, cdata->offset ) != cdata->offset ) + //if( fwrite( buf, 1, cdata->offset, stdout ) != cdata->offset ) + { + writeerr(); + } + cdata->bytes_out += cdata->offset; + } + cdata->offset = 0; + (void)fflush( stdout ); + if( ferror( stdout ) ) + { + writeerr(); + } +#ifdef DEBUG + if ( verbose ) + { + fprintf( stderr, "\n" ); + } +#endif + } +} + +void writeerr() +{ + (void)fprintf(stderr, "compress: %s\n", + strerror(errno)); + exit(1); +} + diff --git a/Utilities/cmcompress/cmcompress.h b/Utilities/cmcompress/cmcompress.h new file mode 100644 index 000000000..a5514659a --- /dev/null +++ b/Utilities/cmcompress/cmcompress.h @@ -0,0 +1,195 @@ +/* + * Copyright (c) 1985, 1986 The Regents of the University of California. + * All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * James A. Woods, derived from original work by Spencer Thomas + * and Joseph Orost. + * + * 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. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. All advertising materials mentioning features or use of this software + * must display the following acknowledgement: + * This product includes software developed by the University of + * California, Berkeley and its contributors. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``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 REGENTS OR CONTRIBUTORS 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. + */ + +#ifndef __cmcompress__h_ +#define __cmcompress__h_ + +#include + +#ifdef __cplusplus +extern "C" +{ +#endif + + /* + * Set USERMEM to the maximum amount of physical user memory available + * in bytes. USERMEM is used to determine the maximum BITS that can be used + * for compression. + * + * SACREDMEM is the amount of physical memory saved for others; compress + * will hog the rest. + */ +#ifndef SACREDMEM +#define SACREDMEM 0 +#endif + +#ifndef USERMEM +# define USERMEM 450000 /* default user memory */ +#endif + +#ifdef pdp11 +# define BITS 12 /* max bits/code for 16-bit machine */ +# define NO_UCHAR /* also if "unsigned char" functions as signed char */ +# undef USERMEM +#endif /* pdp11 */ /* don't forget to compile with -i */ + +#ifdef USERMEM +# if USERMEM >= (433484+SACREDMEM) +# define PBITS 16 +# else +# if USERMEM >= (229600+SACREDMEM) +# define PBITS 15 +# else +# if USERMEM >= (127536+SACREDMEM) +# define PBITS 14 +# else +# if USERMEM >= (73464+SACREDMEM) +# define PBITS 13 +# else +# define PBITS 12 +# endif +# endif +# endif +# endif +# undef USERMEM +#endif /* USERMEM */ + +#ifdef PBITS /* Preferred BITS for this memory size */ +# ifndef BITS +# define BITS PBITS +# endif BITS +#endif /* PBITS */ + +#if BITS == 16 +# define HSIZE 69001 /* 95% occupancy */ +#endif +#if BITS == 15 +# define HSIZE 35023 /* 94% occupancy */ +#endif +#if BITS == 14 +# define HSIZE 18013 /* 91% occupancy */ +#endif +#if BITS == 13 +# define HSIZE 9001 /* 91% occupancy */ +#endif +#if BITS <= 12 +# define HSIZE 5003 /* 80% occupancy */ +#endif + + /* + * a code_int must be able to hold 2**BITS values of type int, and also -1 + */ +#if BITS > 15 + typedef long int code_int; +#else + typedef int code_int; +#endif + +#ifdef SIGNED_COMPARE_SLOW + typedef unsigned long int count_int; + typedef unsigned short int count_short; +#else + typedef long int count_int; +#endif + +#ifdef NO_UCHAR + typedef char char_type; +#else + typedef unsigned char char_type; +#endif /* UCHAR */ + + + + struct cmcompress_stream + { + int n_bits; /* number of bits/code */ + int maxbits; /* user settable max # bits/code */ + code_int maxcode; /* maximum code, given n_bits */ + code_int maxmaxcode; /* should NEVER generate this code */ + + count_int htab [HSIZE]; + unsigned short codetab [HSIZE]; + + code_int hsize; /* for dynamic table sizing */ + code_int free_ent; /* first unused entry */ + int nomagic; /* Use a 3-byte magic number header, unless old file */ + + /* + * block compression parameters -- after all codes are used up, + * and compression rate changes, start over. + */ + int block_compress; + int clear_flg; + long int ratio; + count_int checkpoint; + +#ifdef DEBUG + int debug; + int verbose; +#endif + + /* compress internals */ + int offset; + long int in_count; /* length of input */ + long int bytes_out; /* length of compressed output */ + long int out_count; /* # of codes output (for debugging) */ + + /* internals */ + code_int ent; + code_int hsize_reg; + int hshift; + + long fcode; + int first_pass; + + /* For input and output */ + int (*input_stream)(void*); + int (*output_stream)(void*, const char*,int); + void* client_data; + }; + + int cmcompress_compress_initialize(struct cmcompress_stream* cdata); + int cmcompress_compress_start(struct cmcompress_stream* cdata); + int cmcompress_compress(struct cmcompress_stream* cdata, void* buff, size_t n); + int cmcompress_compress_finalize(struct cmcompress_stream* cdata); + +#ifdef __cplusplus +} +#endif + + +#endif /* __cmcompress__h_ */ diff --git a/Utilities/cmcompress/compress.c.original b/Utilities/cmcompress/compress.c.original new file mode 100644 index 000000000..5062bda98 --- /dev/null +++ b/Utilities/cmcompress/compress.c.original @@ -0,0 +1,1308 @@ +/* + * Copyright (c) 1985, 1986 The Regents of the University of California. + * All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * James A. Woods, derived from original work by Spencer Thomas + * and Joseph Orost. + * + * 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. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. All advertising materials mentioning features or use of this software + * must display the following acknowledgement: + * This product includes software developed by the University of + * California, Berkeley and its contributors. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``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 REGENTS OR CONTRIBUTORS 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. + */ + +#ifndef lint +char copyright[] = +"@(#) Copyright (c) 1985, 1986 The Regents of the University of California.\n\ + All rights reserved.\n"; +#endif /* not lint */ + +#ifndef lint +static char sccsid[] = "@(#)compress.c 5.19 (Berkeley) 3/18/91"; +#endif /* not lint */ + +/* + * compress.c - File compression ala IEEE Computer, June 1984. + * + * Authors: Spencer W. Thomas (decvax!utah-cs!thomas) + * Jim McKie (decvax!mcvax!jim) + * Steve Davies (decvax!vax135!petsd!peora!srd) + * Ken Turkowski (decvax!decwrl!turtlevax!ken) + * James A. Woods (decvax!ihnp4!ames!jaw) + * Joe Orost (decvax!vax135!petsd!joe) + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* + * Set USERMEM to the maximum amount of physical user memory available + * in bytes. USERMEM is used to determine the maximum BITS that can be used + * for compression. + * + * SACREDMEM is the amount of physical memory saved for others; compress + * will hog the rest. + */ +#ifndef SACREDMEM +#define SACREDMEM 0 +#endif + +#ifndef USERMEM +# define USERMEM 450000 /* default user memory */ +#endif + +#ifdef pdp11 +# define BITS 12 /* max bits/code for 16-bit machine */ +# define NO_UCHAR /* also if "unsigned char" functions as signed char */ +# undef USERMEM +#endif /* pdp11 */ /* don't forget to compile with -i */ + +#ifdef USERMEM +# if USERMEM >= (433484+SACREDMEM) +# define PBITS 16 +# else +# if USERMEM >= (229600+SACREDMEM) +# define PBITS 15 +# else +# if USERMEM >= (127536+SACREDMEM) +# define PBITS 14 +# else +# if USERMEM >= (73464+SACREDMEM) +# define PBITS 13 +# else +# define PBITS 12 +# endif +# endif +# endif +# endif +# undef USERMEM +#endif /* USERMEM */ + +#ifdef PBITS /* Preferred BITS for this memory size */ +# ifndef BITS +# define BITS PBITS +# endif BITS +#endif /* PBITS */ + +#if BITS == 16 +# define HSIZE 69001 /* 95% occupancy */ +#endif +#if BITS == 15 +# define HSIZE 35023 /* 94% occupancy */ +#endif +#if BITS == 14 +# define HSIZE 18013 /* 91% occupancy */ +#endif +#if BITS == 13 +# define HSIZE 9001 /* 91% occupancy */ +#endif +#if BITS <= 12 +# define HSIZE 5003 /* 80% occupancy */ +#endif + +/* + * a code_int must be able to hold 2**BITS values of type int, and also -1 + */ +#if BITS > 15 +typedef long int code_int; +#else +typedef int code_int; +#endif + +#ifdef SIGNED_COMPARE_SLOW +typedef unsigned long int count_int; +typedef unsigned short int count_short; +#else +typedef long int count_int; +#endif + +#ifdef NO_UCHAR + typedef char char_type; +#else + typedef unsigned char char_type; +#endif /* UCHAR */ +char_type magic_header[] = { "\037\235" }; /* 1F 9D */ + +/* Defines for third byte of header */ +#define BIT_MASK 0x1f +#define BLOCK_MASK 0x80 +/* Masks 0x40 and 0x20 are free. I think 0x20 should mean that there is + a fourth header byte (for expansion). +*/ +#define INIT_BITS 9 /* initial number of bits/code */ + +int n_bits; /* number of bits/code */ +int maxbits = BITS; /* user settable max # bits/code */ +code_int maxcode; /* maximum code, given n_bits */ +code_int maxmaxcode = 1 << BITS; /* should NEVER generate this code */ +#ifdef COMPATIBLE /* But wrong! */ +# define MAXCODE(n_bits) (1 << (n_bits) - 1) +#else +# define MAXCODE(n_bits) ((1 << (n_bits)) - 1) +#endif /* COMPATIBLE */ + +count_int htab [HSIZE]; +unsigned short codetab [HSIZE]; + +#define htabof(i) htab[i] +#define codetabof(i) codetab[i] +code_int hsize = HSIZE; /* for dynamic table sizing */ +count_int fsize; + +/* + * To save much memory, we overlay the table used by compress() with those + * used by decompress(). The tab_prefix table is the same size and type + * as the codetab. The tab_suffix table needs 2**BITS characters. We + * get this from the beginning of htab. The output stack uses the rest + * of htab, and contains characters. There is plenty of room for any + * possible stack (stack used to be 8000 characters). + */ + +#define tab_prefixof(i) codetabof(i) +# define tab_suffixof(i) ((char_type *)(htab))[i] +# define de_stack ((char_type *)&tab_suffixof(1< maxbits. + * -C => generate output compatible with compress 2.0. + * -c => cat all output to stdout + * -D => debug + * -d => do_decomp + * -f => force overwrite of output file + * -n => no header: useful to uncompress old files + * -V => print Version; debug verbose + * -v => unquiet + */ + + overwrite = 0; +#ifdef DEBUG + while ((ch = getopt(argc, argv, "b:CcDdfnVv")) != EOF) +#else + while ((ch = getopt(argc, argv, "b:Ccdfnv")) != EOF) +#endif + switch(ch) { + case 'b': + maxbits = atoi(optarg); + break; + case 'C': + block_compress = 0; + break; + case 'c': + zcat_flg = 1; + break; +#ifdef DEBUG + case 'D': + debug = 1; + break; +#endif + case 'd': + do_decomp = 1; + break; + case 'f': + overwrite = 1; + force = 1; + break; + case 'n': + nomagic = 1; + break; + case 'q': + quiet = 1; + break; +#ifdef DEBUG + case 'V': + verbose = 1; + break; +#endif + case 'v': + quiet = 0; + break; + case '?': + default: + usage(); + } + argc -= optind; + argv += optind; + + if (maxbits < INIT_BITS) + maxbits = INIT_BITS; + if (maxbits > BITS) + maxbits = BITS; + maxmaxcode = 1 << maxbits; + + /* Build useless input file list. */ + filelist = fileptr = (char **)(malloc(argc * sizeof(*argv))); + while (*argv) + *fileptr++ = *argv++; + *fileptr = NULL; + + if (*filelist != NULL) { + for (fileptr = filelist; *fileptr; fileptr++) { + exit_stat = 0; + if (do_decomp) { /* DECOMPRESSION */ + /* Check for .Z suffix */ + if (strcmp(*fileptr + strlen(*fileptr) - 2, ".Z") != 0) { + /* No .Z: tack one on */ + strcpy(tempname, *fileptr); + strcat(tempname, ".Z"); + *fileptr = tempname; + } + /* Open input file */ + if ((freopen(*fileptr, "r", stdin)) == NULL) { + perror(*fileptr); + perm_stat = 1; + continue; + } + /* Check the magic number */ + if (nomagic == 0) { + if ((getchar() != (magic_header[0] & 0xFF)) + || (getchar() != (magic_header[1] & 0xFF))) { + fprintf(stderr, "%s: not in compressed format\n", + *fileptr); + continue; + } + maxbits = getchar(); /* set -b from file */ + block_compress = maxbits & BLOCK_MASK; + maxbits &= BIT_MASK; + maxmaxcode = 1 << maxbits; + if(maxbits > BITS) { + fprintf(stderr, + "%s: compressed with %d bits, can only handle %d bits\n", + *fileptr, maxbits, BITS); + continue; + } + } + /* Generate output filename */ + strcpy(ofname, *fileptr); + ofname[strlen(*fileptr) - 2] = '\0'; /* Strip off .Z */ + } else { /* COMPRESSION */ + if (strcmp(*fileptr + strlen(*fileptr) - 2, ".Z") == 0) { + fprintf(stderr, "%s: already has .Z suffix -- no change\n", + *fileptr); + continue; + } + /* Open input file */ + if ((freopen(*fileptr, "r", stdin)) == NULL) { + perror(*fileptr); + perm_stat = 1; + continue; + } + stat ( *fileptr, &statbuf ); + fsize = (long) statbuf.st_size; + /* + * tune hash table size for small files -- ad hoc, + * but the sizes match earlier #defines, which + * serve as upper bounds on the number of output codes. + */ + hsize = HSIZE; + if ( fsize < (1 << 12) ) + hsize = MIN ( 5003, HSIZE ); + else if ( fsize < (1 << 13) ) + hsize = MIN ( 9001, HSIZE ); + else if ( fsize < (1 << 14) ) + hsize = MIN ( 18013, HSIZE ); + else if ( fsize < (1 << 15) ) + hsize = MIN ( 35023, HSIZE ); + else if ( fsize < 47000 ) + hsize = MIN ( 50021, HSIZE ); + + /* Generate output filename */ + strcpy(ofname, *fileptr); + strcat(ofname, ".Z"); + } + /* Check for overwrite of existing file */ + if (overwrite == 0 && zcat_flg == 0) { + if (stat(ofname, &statbuf) == 0) { + char response[2]; + response[0] = 'n'; + fprintf(stderr, "%s already exists;", ofname); + if (bgnd_flag == 0 && isatty(2)) { + fprintf(stderr, " do you wish to overwrite %s (y or n)? ", + ofname); + fflush(stderr); + read(2, response, 2); + while (response[1] != '\n') { + if (read(2, response+1, 1) < 0) { /* Ack! */ + perror("stderr"); break; + } + } + } + if (response[0] != 'y') { + fprintf(stderr, "\tnot overwritten\n"); + continue; + } + } + } + if(zcat_flg == 0) { /* Open output file */ + if (freopen(ofname, "w", stdout) == NULL) { + perror(ofname); + perm_stat = 1; + continue; + } + precious = 0; + if(!quiet) + fprintf(stderr, "%s: ", *fileptr); + } + + /* Actually do the compression/decompression */ + if (do_decomp == 0) compress(); +#ifndef DEBUG + else decompress(); +#else + else if (debug == 0) decompress(); + else printcodes(); + if (verbose) dump_tab(); +#endif /* DEBUG */ + if(zcat_flg == 0) { + copystat(*fileptr, ofname); /* Copy stats */ + precious = 1; + if((exit_stat == 1) || (!quiet)) + putc('\n', stderr); + } + } + } else { /* Standard input */ + if (do_decomp == 0) { + compress(); +#ifdef DEBUG + if(verbose) dump_tab(); +#endif /* DEBUG */ + if(!quiet) + putc('\n', stderr); + } else { + /* Check the magic number */ + if (nomagic == 0) { + if ((getchar()!=(magic_header[0] & 0xFF)) + || (getchar()!=(magic_header[1] & 0xFF))) { + fprintf(stderr, "stdin: not in compressed format\n"); + exit(1); + } + maxbits = getchar(); /* set -b from file */ + block_compress = maxbits & BLOCK_MASK; + maxbits &= BIT_MASK; + maxmaxcode = 1 << maxbits; + fsize = 100000; /* assume stdin large for USERMEM */ + if(maxbits > BITS) { + fprintf(stderr, + "stdin: compressed with %d bits, can only handle %d bits\n", + maxbits, BITS); + exit(1); + } + } +#ifndef DEBUG + decompress(); +#else + if (debug == 0) decompress(); + else printcodes(); + if (verbose) dump_tab(); +#endif /* DEBUG */ + } + } + exit(perm_stat ? perm_stat : exit_stat); +} + +static int offset; +long int in_count = 1; /* length of input */ +long int bytes_out; /* length of compressed output */ +long int out_count = 0; /* # of codes output (for debugging) */ + +/* + * compress stdin to stdout + * + * Algorithm: use open addressing double hashing (no chaining) on the + * prefix code / next character combination. We do a variant of Knuth's + * algorithm D (vol. 3, sec. 6.4) along with G. Knott's relatively-prime + * secondary probe. Here, the modular division first probe is gives way + * to a faster exclusive-or manipulation. Also do block compression with + * an adaptive reset, whereby the code table is cleared when the compression + * ratio decreases, but after the table fills. The variable-length output + * codes are re-sized at this point, and a special CLEAR code is generated + * for the decompressor. Late addition: construct the table according to + * file size for noticeable speed improvement on small files. Please direct + * questions about this implementation to ames!jaw. + */ + +compress() +{ + register long fcode; + register code_int i = 0; + register int c; + register code_int ent; + register int disp; + register code_int hsize_reg; + register int hshift; + +#ifndef COMPATIBLE + if (nomagic == 0) { + putchar(magic_header[0]); + putchar(magic_header[1]); + putchar((char)(maxbits | block_compress)); + if(ferror(stdout)) + writeerr(); + } +#endif /* COMPATIBLE */ + + offset = 0; + bytes_out = 3; /* includes 3-byte header mojo */ + out_count = 0; + clear_flg = 0; + ratio = 0; + in_count = 1; + checkpoint = CHECK_GAP; + maxcode = MAXCODE(n_bits = INIT_BITS); + free_ent = ((block_compress) ? FIRST : 256 ); + + ent = getchar (); + + hshift = 0; + for ( fcode = (long) hsize; fcode < 65536L; fcode *= 2L ) + hshift++; + hshift = 8 - hshift; /* set hash code range bound */ + + hsize_reg = hsize; + cl_hash( (count_int) hsize_reg); /* clear hash table */ + +#ifdef SIGNED_COMPARE_SLOW + while ( (c = getchar()) != (unsigned) EOF ) { +#else + while ( (c = getchar()) != EOF ) { +#endif + in_count++; + fcode = (long) (((long) c << maxbits) + ent); + i = ((c << hshift) ^ ent); /* xor hashing */ + + if ( htabof (i) == fcode ) { + ent = codetabof (i); + continue; + } else if ( (long)htabof (i) < 0 ) /* empty slot */ + goto nomatch; + disp = hsize_reg - i; /* secondary hash (after G. Knott) */ + if ( i == 0 ) + disp = 1; +probe: + if ( (i -= disp) < 0 ) + i += hsize_reg; + + if ( htabof (i) == fcode ) { + ent = codetabof (i); + continue; + } + if ( (long)htabof (i) > 0 ) + goto probe; +nomatch: + output ( (code_int) ent ); + out_count++; + ent = c; +#ifdef SIGNED_COMPARE_SLOW + if ( (unsigned) free_ent < (unsigned) maxmaxcode) { +#else + if ( free_ent < maxmaxcode ) { +#endif + codetabof (i) = free_ent++; /* code -> hashtable */ + htabof (i) = fcode; + } + else if ( (count_int)in_count >= checkpoint && block_compress ) + cl_block (); + } + /* + * Put out the final code. + */ + output( (code_int)ent ); + out_count++; + output( (code_int)-1 ); + + /* + * Print out stats on stderr + */ + if(zcat_flg == 0 && !quiet) { +#ifdef DEBUG + fprintf( stderr, + "%ld chars in, %ld codes (%ld bytes) out, compression factor: ", + in_count, out_count, bytes_out ); + prratio( stderr, in_count, bytes_out ); + fprintf( stderr, "\n"); + fprintf( stderr, "\tCompression as in compact: " ); + prratio( stderr, in_count-bytes_out, in_count ); + fprintf( stderr, "\n"); + fprintf( stderr, "\tLargest code (of last block) was %d (%d bits)\n", + free_ent - 1, n_bits ); +#else /* !DEBUG */ + fprintf( stderr, "Compression: " ); + prratio( stderr, in_count-bytes_out, in_count ); +#endif /* DEBUG */ + } + if(bytes_out > in_count) /* exit(2) if no savings */ + exit_stat = 2; + return; +} + +/*- + * Output the given code. + * Inputs: + * code: A n_bits-bit integer. If == -1, then EOF. This assumes + * that n_bits =< (long)wordsize - 1. + * Outputs: + * Outputs code to the file. + * Assumptions: + * Chars are 8 bits long. + * Algorithm: + * Maintain a BITS character long buffer (so that 8 codes will + * fit in it exactly). Use the VAX insv instruction to insert each + * code in turn. When the buffer fills up empty it and start over. + */ + +static char buf[BITS]; + +#ifndef vax +char_type lmask[9] = {0xff, 0xfe, 0xfc, 0xf8, 0xf0, 0xe0, 0xc0, 0x80, 0x00}; +char_type rmask[9] = {0x00, 0x01, 0x03, 0x07, 0x0f, 0x1f, 0x3f, 0x7f, 0xff}; +#endif /* vax */ + +output( code ) +code_int code; +{ +#ifdef DEBUG + static int col = 0; +#endif /* DEBUG */ + + /* + * On the VAX, it is important to have the register declarations + * in exactly the order given, or the asm will break. + */ + register int r_off = offset, bits= n_bits; + register char * bp = buf; + +#ifdef DEBUG + if ( verbose ) + fprintf( stderr, "%5d%c", code, + (col+=6) >= 74 ? (col = 0, '\n') : ' ' ); +#endif /* DEBUG */ + if ( code >= 0 ) { +#if defined(vax) && !defined(__GNUC__) + /* + * VAX and PCC DEPENDENT!! Implementation on other machines is + * below. + * + * Translation: Insert BITS bits from the argument starting at + * offset bits from the beginning of buf. + */ + 0; /* Work around for pcc -O bug with asm and if stmt */ + asm( "insv 4(ap),r11,r10,(r9)" ); +#else +/* + * byte/bit numbering on the VAX is simulated by the following code + */ + /* + * Get to the first byte. + */ + bp += (r_off >> 3); + r_off &= 7; + /* + * Since code is always >= 8 bits, only need to mask the first + * hunk on the left. + */ + *bp = (*bp & rmask[r_off]) | (code << r_off) & lmask[r_off]; + bp++; + bits -= (8 - r_off); + code >>= 8 - r_off; + /* Get any 8 bit parts in the middle (<=1 for up to 16 bits). */ + if ( bits >= 8 ) { + *bp++ = code; + code >>= 8; + bits -= 8; + } + /* Last bits. */ + if(bits) + *bp = code; +#endif /* vax */ + offset += n_bits; + if ( offset == (n_bits << 3) ) { + bp = buf; + bits = n_bits; + bytes_out += bits; + do { + putchar(*bp++); + if (ferror(stdout)) + writeerr(); + } while(--bits); + offset = 0; + } + + /* + * If the next entry is going to be too big for the code size, + * then increase it, if possible. + */ + if ( free_ent > maxcode || (clear_flg > 0)) + { + /* + * Write the whole buffer, because the input side won't + * discover the size increase until after it has read it. + */ + if ( offset > 0 ) { + if( fwrite( buf, 1, n_bits, stdout ) != n_bits) + writeerr(); + bytes_out += n_bits; + } + offset = 0; + + if ( clear_flg ) { + maxcode = MAXCODE (n_bits = INIT_BITS); + clear_flg = 0; + } + else { + n_bits++; + if ( n_bits == maxbits ) + maxcode = maxmaxcode; + else + maxcode = MAXCODE(n_bits); + } +#ifdef DEBUG + if ( debug ) { + fprintf( stderr, "\nChange to %d bits\n", n_bits ); + col = 0; + } +#endif /* DEBUG */ + } + } else { + /* + * At EOF, write the rest of the buffer. + */ + if ( offset > 0 ) { + offset = (offset + 7) / 8; + if( fwrite( buf, 1, offset, stdout ) != offset ) + writeerr(); + bytes_out += offset; + } + offset = 0; + (void)fflush( stdout ); + if( ferror( stdout ) ) + writeerr(); +#ifdef DEBUG + if ( verbose ) + fprintf( stderr, "\n" ); +#endif + } +} + +/* + * Decompress stdin to stdout. This routine adapts to the codes in the + * file building the "string" table on-the-fly; requiring no table to + * be stored in the compressed file. The tables used herein are shared + * with those of the compress() routine. See the definitions above. + */ + +decompress() { + register char_type *stackp; + register int finchar; + register code_int code, oldcode, incode; + int n, nwritten, offset; /* Variables for buffered write */ + char buff[BUFSIZ]; /* Buffer for buffered write */ + + + /* + * As above, initialize the first 256 entries in the table. + */ + maxcode = MAXCODE(n_bits = INIT_BITS); + for ( code = 255; code >= 0; code-- ) { + tab_prefixof(code) = 0; + tab_suffixof(code) = (char_type)code; + } + free_ent = ((block_compress) ? FIRST : 256 ); + + finchar = oldcode = getcode(); + if(oldcode == -1) /* EOF already? */ + return; /* Get out of here */ + + /* first code must be 8 bits = char */ + n=0; + buff[n++] = (char)finchar; + + stackp = de_stack; + + while ( (code = getcode()) > -1 ) { + + if ( (code == CLEAR) && block_compress ) { + for ( code = 255; code >= 0; code-- ) + tab_prefixof(code) = 0; + clear_flg = 1; + free_ent = FIRST - 1; + if ( (code = getcode ()) == -1 ) /* O, untimely death! */ + break; + } + incode = code; + /* + * Special case for KwKwK string. + */ + if ( code >= free_ent ) { + *stackp++ = finchar; + code = oldcode; + } + + /* + * Generate output characters in reverse order + */ +#ifdef SIGNED_COMPARE_SLOW + while ( ((unsigned long)code) >= ((unsigned long)256) ) { +#else + while ( code >= 256 ) { +#endif + *stackp++ = tab_suffixof(code); + code = tab_prefixof(code); + } + *stackp++ = finchar = tab_suffixof(code); + + /* + * And put them out in forward order + */ + do { + /* + * About 60% of the time is spent in the putchar() call + * that appeared here. It was originally + * putchar ( *--stackp ); + * If we buffer the writes ourselves, we can go faster (about + * 30%). + * + * At this point, the next line is the next *big* time + * sink in the code. It takes up about 10% of the time. + */ + buff[n++] = *--stackp; + if (n == BUFSIZ) { + offset = 0; + do { + nwritten = write(fileno(stdout), &buff[offset], n); + if (nwritten < 0) + writeerr(); + offset += nwritten; + } while ((n -= nwritten) > 0); + } + } while ( stackp > de_stack ); + + /* + * Generate the new entry. + */ + if ( (code=free_ent) < maxmaxcode ) { + tab_prefixof(code) = (unsigned short)oldcode; + tab_suffixof(code) = finchar; + free_ent = code+1; + } + /* + * Remember previous code. + */ + oldcode = incode; + } + /* + * Flush the stuff remaining in our buffer... + */ + offset = 0; + while (n > 0) { + nwritten = write(fileno(stdout), &buff[offset], n); + if (nwritten < 0) + writeerr(); + offset += nwritten; + n -= nwritten; + } +} + +/*- + * Read one code from the standard input. If EOF, return -1. + * Inputs: + * stdin + * Outputs: + * code or -1 is returned. + */ +code_int +getcode() { + /* + * On the VAX, it is important to have the register declarations + * in exactly the order given, or the asm will break. + */ + register code_int code; + static int offset = 0, size = 0; + static char_type buf[BITS]; + register int r_off, bits; + register char_type *bp = buf; + + if ( clear_flg > 0 || offset >= size || free_ent > maxcode ) { + /* + * If the next entry will be too big for the current code + * size, then we must increase the size. This implies reading + * a new buffer full, too. + */ + if ( free_ent > maxcode ) { + n_bits++; + if ( n_bits == maxbits ) + maxcode = maxmaxcode; /* won't get any bigger now */ + else + maxcode = MAXCODE(n_bits); + } + if ( clear_flg > 0) { + maxcode = MAXCODE (n_bits = INIT_BITS); + clear_flg = 0; + } + size = fread( buf, 1, n_bits, stdin ); + if ( size <= 0 ) + return -1; /* end of file */ + offset = 0; + /* Round size down to integral number of codes */ + size = (size << 3) - (n_bits - 1); + } + r_off = offset; + bits = n_bits; +#ifdef vax + asm( "extzv r10,r9,(r8),r11" ); +#else /* not a vax */ + /* + * Get to the first byte. + */ + bp += (r_off >> 3); + r_off &= 7; + /* Get first part (low order bits) */ +#ifdef NO_UCHAR + code = ((*bp++ >> r_off) & rmask[8 - r_off]) & 0xff; +#else + code = (*bp++ >> r_off); +#endif /* NO_UCHAR */ + bits -= (8 - r_off); + r_off = 8 - r_off; /* now, offset into code word */ + /* Get any 8 bit parts in the middle (<=1 for up to 16 bits). */ + if ( bits >= 8 ) { +#ifdef NO_UCHAR + code |= (*bp++ & 0xff) << r_off; +#else + code |= *bp++ << r_off; +#endif /* NO_UCHAR */ + r_off += 8; + bits -= 8; + } + /* high order bits. */ + code |= (*bp & rmask[bits]) << r_off; +#endif /* vax */ + offset += n_bits; + + return code; +} + +#ifdef DEBUG +printcodes() +{ + /* + * Just print out codes from input file. For debugging. + */ + code_int code; + int col = 0, bits; + + bits = n_bits = INIT_BITS; + maxcode = MAXCODE(n_bits); + free_ent = ((block_compress) ? FIRST : 256 ); + while ( ( code = getcode() ) >= 0 ) { + if ( (code == CLEAR) && block_compress ) { + free_ent = FIRST - 1; + clear_flg = 1; + } + else if ( free_ent < maxmaxcode ) + free_ent++; + if ( bits != n_bits ) { + fprintf(stderr, "\nChange to %d bits\n", n_bits ); + bits = n_bits; + col = 0; + } + fprintf(stderr, "%5d%c", code, (col+=6) >= 74 ? (col = 0, '\n') : ' ' ); + } + putc( '\n', stderr ); + exit( 0 ); +} + +code_int sorttab[1<= 0) { + sorttab[codetabof(i)] = i; + } + } + first = block_compress ? FIRST : 256; + for(i = first; i < free_ent; i++) { + fprintf(stderr, "%5d: \"", i); + de_stack[--stack_top] = '\n'; + de_stack[--stack_top] = '"'; + stack_top = in_stack((htabof(sorttab[i])>>maxbits)&0xff, + stack_top); + for(ent=htabof(sorttab[i]) & ((1< 256; + ent=htabof(sorttab[ent]) & ((1<> maxbits, + stack_top); + } + stack_top = in_stack(ent, stack_top); + fwrite( &de_stack[stack_top], 1, STACK_SIZE-stack_top, stderr); + stack_top = STACK_SIZE; + } + } else if(!debug) { /* decompressing */ + + for ( i = 0; i < free_ent; i++ ) { + ent = i; + c = tab_suffixof(ent); + if ( isascii(c) && isprint(c) ) + fprintf( stderr, "%5d: %5d/'%c' \"", + ent, tab_prefixof(ent), c ); + else + fprintf( stderr, "%5d: %5d/\\%03o \"", + ent, tab_prefixof(ent), c ); + de_stack[--stack_top] = '\n'; + de_stack[--stack_top] = '"'; + for ( ; ent != NULL; + ent = (ent >= FIRST ? tab_prefixof(ent) : NULL) ) { + stack_top = in_stack(tab_suffixof(ent), stack_top); + } + fwrite( &de_stack[stack_top], 1, STACK_SIZE - stack_top, stderr ); + stack_top = STACK_SIZE; + } + } +} + +int +in_stack(c, stack_top) + register c, stack_top; +{ + if ( (isascii(c) && isprint(c) && c != '\\') || c == ' ' ) { + de_stack[--stack_top] = c; + } else { + switch( c ) { + case '\n': de_stack[--stack_top] = 'n'; break; + case '\t': de_stack[--stack_top] = 't'; break; + case '\b': de_stack[--stack_top] = 'b'; break; + case '\f': de_stack[--stack_top] = 'f'; break; + case '\r': de_stack[--stack_top] = 'r'; break; + case '\\': de_stack[--stack_top] = '\\'; break; + default: + de_stack[--stack_top] = '0' + c % 8; + de_stack[--stack_top] = '0' + (c / 8) % 8; + de_stack[--stack_top] = '0' + c / 64; + break; + } + de_stack[--stack_top] = '\\'; + } + return stack_top; +} +#endif /* DEBUG */ + +writeerr() +{ + (void)fprintf(stderr, "compress: %s: %s\n", + ofname[0] ? ofname : "stdout", strerror(errno)); + (void)unlink(ofname); + exit(1); +} + +copystat(ifname, ofname) +char *ifname, *ofname; +{ + struct stat statbuf; + int mode; + struct utimbuf tp; + + fclose(stdout); + if (stat(ifname, &statbuf)) { /* Get stat on input file */ + perror(ifname); + return; + } + if ((statbuf.st_mode & S_IFMT/*0170000*/) != S_IFREG/*0100000*/) { + if(quiet) + fprintf(stderr, "%s: ", ifname); + fprintf(stderr, " -- not a regular file: unchanged"); + exit_stat = 1; + perm_stat = 1; + } else if (statbuf.st_nlink > 1) { + if(quiet) + fprintf(stderr, "%s: ", ifname); + fprintf(stderr, " -- has %d other links: unchanged", + statbuf.st_nlink - 1); + exit_stat = 1; + perm_stat = 1; + } else if (exit_stat == 2 && (!force)) { /* No compression: remove file.Z */ + if(!quiet) + fprintf(stderr, " -- file unchanged"); + } else { /* ***** Successful Compression ***** */ + exit_stat = 0; + mode = statbuf.st_mode & 07777; + if (chmod(ofname, mode)) /* Copy modes */ + perror(ofname); + chown(ofname, statbuf.st_uid, statbuf.st_gid); /* Copy ownership */ + tp.actime = statbuf.st_atime; + tp.modtime = statbuf.st_mtime; + utime(ofname, &tp); /* Update last accessed and modified times */ + if (unlink(ifname)) /* Remove input file */ + perror(ifname); + if(!quiet) + fprintf(stderr, " -- replaced with %s", ofname); + return; /* Successful return */ + } + + /* Unsuccessful return -- one of the tests failed */ + if (unlink(ofname)) + perror(ofname); +} + +void +onintr ( ) +{ + if (!precious) + unlink ( ofname ); + exit ( 1 ); +} + +void +oops ( ) /* wild pointer -- assume bad input */ +{ + if ( do_decomp ) + fprintf ( stderr, "uncompress: corrupt input\n" ); + unlink ( ofname ); + exit ( 1 ); +} + +cl_block () /* table clear for block compress */ +{ + register long int rat; + + checkpoint = in_count + CHECK_GAP; +#ifdef DEBUG + if ( debug ) { + fprintf ( stderr, "count: %ld, ratio: ", in_count ); + prratio ( stderr, in_count, bytes_out ); + fprintf ( stderr, "\n"); + } +#endif /* DEBUG */ + + if(in_count > 0x007fffff) { /* shift will overflow */ + rat = bytes_out >> 8; + if(rat == 0) { /* Don't divide by zero */ + rat = 0x7fffffff; + } else { + rat = in_count / rat; + } + } else { + rat = (in_count << 8) / bytes_out; /* 8 fractional bits */ + } + if ( rat > ratio ) { + ratio = rat; + } else { + ratio = 0; +#ifdef DEBUG + if(verbose) + dump_tab(); /* dump string table */ +#endif + cl_hash ( (count_int) hsize ); + free_ent = FIRST; + clear_flg = 1; + output ( (code_int) CLEAR ); +#ifdef DEBUG + if(debug) + fprintf ( stderr, "clear\n" ); +#endif /* DEBUG */ + } +} + +cl_hash(hsize) /* reset code table */ + register count_int hsize; +{ + register count_int *htab_p = htab+hsize; + register long i; + register long m1 = -1; + + i = hsize - 16; + do { /* might use Sys V memset(3) here */ + *(htab_p-16) = m1; + *(htab_p-15) = m1; + *(htab_p-14) = m1; + *(htab_p-13) = m1; + *(htab_p-12) = m1; + *(htab_p-11) = m1; + *(htab_p-10) = m1; + *(htab_p-9) = m1; + *(htab_p-8) = m1; + *(htab_p-7) = m1; + *(htab_p-6) = m1; + *(htab_p-5) = m1; + *(htab_p-4) = m1; + *(htab_p-3) = m1; + *(htab_p-2) = m1; + *(htab_p-1) = m1; + htab_p -= 16; + } while ((i -= 16) >= 0); + for ( i += 16; i > 0; i-- ) + *--htab_p = m1; +} + +prratio(stream, num, den) +FILE *stream; +long int num, den; +{ + register int q; /* Doesn't need to be long */ + + if(num > 214748L) { /* 2147483647/10000 */ + q = num / (den / 10000L); + } else { + q = 10000L * num / den; /* Long calculations, though */ + } + if (q < 0) { + putc('-', stream); + q = -q; + } + fprintf(stream, "%d.%02d%%", q / 100, q % 100); +} + +usage() +{ + (void)fprintf(stderr, +#ifdef DEBUG + "compress [-CDVcdfnv] [-b maxbits] [file ...]\n"); +#else + "compress [-Ccdfnv] [-b maxbits] [file ...]\n"); +#endif + exit(1); +}