Merging upstream version 1.5~rc1.

Signed-off-by: Daniel Baumann <mail@daniel-baumann.ch>

1 files changed, 118 insertions, 22 deletions

diff --git a/main.c b/main.c
index 1bf2fab..1862ffd 100644
--- a/main.c
+++ b/main.c
@@ -54,6 +54,10 @@
 #include "lzip.h"
 #include "decoder.h"
 
+#ifndef O_BINARY
+#define O_BINARY 0
+#endif
+
 #if CHAR_BIT != 8
 #error "Environments where CHAR_BIT != 8 are not supported."
 #endif
@@ -64,12 +68,6 @@ const char * const program_name = "lunzip";
 const char * const program_year = "2013";
 const char * invocation_name = 0;
 
-#ifdef O_BINARY
-const int o_binary = O_BINARY;
-#else
-const int o_binary = 0;
-#endif
-
 struct { const char * from; const char * to; } const known_extensions[] = {
   { ".lz",  ""     },
   { ".tlz", ".tar" },
@@ -86,7 +84,18 @@ bool delete_output_on_interrupt = false;
 
 static void show_help( void )
   {
-  printf( "%s - Decompressor for lzip files.\n", Program_name );
+  printf( "Lunzip is a decompressor for lzip files. It is written in C and its\n"
+          "small size makes it well suited for embedded devices or software\n"
+          "installers that need to decompress files but do not need compression\n"
+          "capabilities. Lunzip is fully compatible with lzip-1.4 or newer.\n"
+          "\nIf the size of the output buffer is specified with the '--buffer-size'\n"
+          "option, lunzip uses the decompressed file as dictionary for distances\n"
+          "beyond the buffer size and is able to decompress any file using as\n"
+          "little memory as 50 kB, irrespective of the dictionary size used to\n"
+          "compress the file. Of course, the smaller the output buffer size used in\n"
+          "relation to the dictionary size, the more accesses to disk are needed\n"
+          "and the slower the decompression is. This 'low memory' mode only works\n"
+          "when decompressing to a regular file.\n" );
   printf( "\nUsage: %s [options] [files]\n", invocation_name );
   printf( "\nOptions:\n"
           "  -h, --help                 display this help and exit\n"
@@ -98,6 +107,7 @@ static void show_help( void )
           "  -o, --output=<file>        if reading stdin, place the output into <file>\n"
           "  -q, --quiet                suppress all messages\n"
           "  -t, --test                 test compressed file integrity\n"
+          "  -u, --buffer-size=<bytes>  set output buffer size in bytes\n"
           "  -v, --verbose              be verbose (a 2nd -v gives more)\n"
           "If no file names are given, lunzip decompresses from standard input to\n"
           "standard output.\n"
@@ -137,6 +147,73 @@ static void show_header( const File_header header )
   }
 
 
+static unsigned long getnum( const char * const ptr,
+                             const unsigned long llimit,
+                             const unsigned long ulimit )
+  {
+  unsigned long result;
+  char * tail;
+  errno = 0;
+  result = strtoul( ptr, &tail, 0 );
+  if( tail == ptr )
+    {
+    show_error( "Bad or missing numerical argument.", 0, true );
+    exit( 1 );
+    }
+
+  if( !errno && tail[0] )
+    {
+    int factor = ( tail[1] == 'i' ) ? 1024 : 1000;
+    int exponent = 0, i;
+    bool bad_multiplier = false;
+    switch( tail[0] )
+      {
+      case ' ': break;
+      case 'Y': exponent = 8; break;
+      case 'Z': exponent = 7; break;
+      case 'E': exponent = 6; break;
+      case 'P': exponent = 5; break;
+      case 'T': exponent = 4; break;
+      case 'G': exponent = 3; break;
+      case 'M': exponent = 2; break;
+      case 'K': if( factor == 1024 ) exponent = 1; else bad_multiplier = true;
+                break;
+      case 'k': if( factor == 1000 ) exponent = 1; else bad_multiplier = true;
+                break;
+      default : bad_multiplier = true;
+      }
+    if( bad_multiplier )
+      {
+      show_error( "Bad multiplier in numerical argument.", 0, true );
+      exit( 1 );
+      }
+    for( i = 0; i < exponent; ++i )
+      {
+      if( ulimit / factor >= result ) result *= factor;
+      else { errno = ERANGE; break; }
+      }
+    }
+  if( !errno && ( result < llimit || result > ulimit ) ) errno = ERANGE;
+  if( errno )
+    {
+    show_error( "Numerical argument out of limits.", 0, false );
+    exit( 1 );
+    }
+  return result;
+  }
+
+
+static int get_dict_size( const char * const arg )
+  {
+  char * tail;
+  int bits = strtol( arg, &tail, 0 );
+  if( bits >= min_dictionary_bits &&
+      bits <= max_dictionary_bits && *tail == 0 )
+    return ( 1 << bits );
+  return getnum( arg, min_dictionary_size, max_dictionary_size );
+  }
+
+
 static int extension_index( const char * const name )
   {
   int i;
@@ -152,9 +229,9 @@ static int extension_index( const char * const name )
 
 
 static int open_instream( const char * const name, struct stat * const in_statsp,
-                          const bool testing, const bool to_stdout )
+                          const bool no_ofile )
   {
-  int infd = open( name, O_RDONLY | o_binary );
+  int infd = open( name, O_RDONLY | O_BINARY );
   if( infd < 0 )
     {
     if( verbosity >= 0 )
@@ -168,7 +245,6 @@ static int open_instream( const char * const name, struct stat * const in_statsp
     const bool can_read = ( i == 0 &&
                             ( S_ISBLK( mode ) || S_ISCHR( mode ) ||
                               S_ISFIFO( mode ) || S_ISSOCK( mode ) ) );
-    const bool no_ofile = to_stdout || testing;
     if( i != 0 || ( !S_ISREG( mode ) && ( !can_read || !no_ofile ) ) )
       {
       if( verbosity >= 0 )
@@ -224,7 +300,7 @@ static void set_d_outname( const char * const name, const int i )
 
 static bool open_outstream( const bool force )
   {
-  int flags = O_CREAT | O_WRONLY | o_binary;
+  int flags = O_APPEND | O_CREAT | O_RDWR | O_BINARY;
   if( force ) flags |= O_TRUNC; else flags |= O_EXCL;
 
   outfd = open( output_filename, flags, outfd_mode );
@@ -263,10 +339,14 @@ static void close_and_set_permissions( const struct stat * const in_statsp )
   bool warning = false;
   if( in_statsp )
     {
+    const mode_t mode = in_statsp->st_mode;
     /* fchown will in many cases return with EPERM, which can be safely ignored. */
-    if( ( fchown( outfd, in_statsp->st_uid, in_statsp->st_gid ) != 0 &&
-          errno != EPERM ) ||
-        fchmod( outfd, in_statsp->st_mode ) != 0 ) warning = true;
+    if( fchown( outfd, in_statsp->st_uid, in_statsp->st_gid ) == 0 )
+      { if( fchmod( outfd, mode ) != 0 ) warning = true; }
+    else
+      if( errno != EPERM ||
+          fchmod( outfd, mode & ~( S_ISUID | S_ISGID | S_ISVTX ) ) != 0 )
+        warning = true;
     }
   if( close( outfd ) != 0 ) cleanup_and_fail( 1 );
   outfd = -1;
@@ -283,8 +363,8 @@ static void close_and_set_permissions( const struct stat * const in_statsp )
   }
 
 
-static int decompress( const int infd, struct Pretty_print * const pp,
-                       const bool testing )
+static int decompress( const int buffer_size, const int infd,
+                       struct Pretty_print * const pp, const bool testing )
   {
   unsigned long long partial_file_pos = 0;
   struct Range_decoder rdec;
@@ -292,7 +372,7 @@ static int decompress( const int infd, struct Pretty_print * const pp,
   bool first_member;
   if( !Rd_init( &rdec, infd ) )
     {
-    show_error( "Not enough memory. Find a machine with more memory.", 0, false );
+    show_error( "Not enough memory.", 0, false );
     cleanup_and_fail( 1 );
     }
 
@@ -332,9 +412,9 @@ static int decompress( const int infd, struct Pretty_print * const pp,
     if( verbosity >= 2 || ( verbosity == 1 && first_member ) )
       { Pp_show_msg( pp, 0 ); if( verbosity >= 3 ) show_header( header ); }
 
-    if( !LZd_init( &decoder, header, &rdec, outfd ) )
+    if( !LZd_init( &decoder, header, &rdec, buffer_size, outfd ) )
       {
-      show_error( "Not enough memory. Find a machine with more memory.", 0, false );
+      show_error( "Not enough memory. Try a smaller output buffer size.", 0, false );
       cleanup_and_fail( 1 );
       }
     result = LZd_decode_member( &decoder, pp );
@@ -431,6 +511,7 @@ int main( const int argc, const char * const argv[] )
   const char * default_output_filename = "";
   const char ** filenames = 0;
   int num_filenames = 0;
+  int buffer_size = max_dictionary_size;
   int infd = -1;
   int argind = 0;
   int retval = 0;
@@ -453,6 +534,7 @@ int main( const int argc, const char * const argv[] )
     { 'o', "output",          ap_yes },
     { 'q', "quiet",           ap_no  },
     { 't', "test",            ap_no  },
+    { 'u', "buffer-size",     ap_yes },
     { 'v', "verbose",         ap_no  },
     { 'V', "version",         ap_no  },
     {  0 ,  0,                ap_no  } };
@@ -463,7 +545,7 @@ int main( const int argc, const char * const argv[] )
   CRC32_init();
 
   if( !ap_init( &parser, argc, argv, options, 0 ) )
-    { show_error( "Memory exhausted.", 0, false ); return 1; }
+    { show_error( "Not enough memory.", 0, false ); return 1; }
   if( ap_error( &parser ) )				/* bad option */
     { show_error( ap_error( &parser ), 0, true ); return 1; }
 
@@ -483,6 +565,7 @@ int main( const int argc, const char * const argv[] )
       case 'o': default_output_filename = arg; break;
       case 'q': verbosity = -1; break;
       case 't': testing = true; break;
+      case 'u': buffer_size = get_dict_size( arg ); break;
       case 'v': if( verbosity < 4 ) ++verbosity; break;
       case 'V': show_version(); return 0;
       default : internal_error( "uncaught option" );
@@ -507,6 +590,19 @@ int main( const int argc, const char * const argv[] )
     if( strcmp( filenames[i], "-" ) != 0 ) filenames_given = true;
     }
 
+  if( buffer_size < max_dictionary_size )
+    {
+    if( to_stdout || testing )
+      { show_error( "'--buffer-size' is incompatible with '--stdout' and '--test'.", 0, false );
+        return 1; }
+    if( !default_output_filename[0] )
+      for( i = 0; i < num_filenames; ++i )
+        if( !filenames[i][0] || strcmp( filenames[i], "-" ) == 0 )
+          { show_error( "Output file must be specified when decompressing from stdin with a\n"
+                        "        reduced buffer size.", 0, false );
+            return 1; }
+    }
+
   if( !to_stdout && !testing &&
       ( filenames_given || default_output_filename[0] ) )
     set_signals();
@@ -547,7 +643,7 @@ int main( const int argc, const char * const argv[] )
     else
       {
       input_filename = filenames[i];
-      infd = open_instream( input_filename, &in_stats, testing, to_stdout );
+      infd = open_instream( input_filename, &in_stats, to_stdout || testing );
       if( infd < 0 ) { if( retval < 1 ) retval = 1; continue; }
       if( !testing )
         {
@@ -576,7 +672,7 @@ int main( const int argc, const char * const argv[] )
       delete_output_on_interrupt = true;
     in_statsp = input_filename[0] ? &in_stats : 0;
     Pp_set_name( &pp, input_filename );
-    tmp = decompress( infd, &pp, testing );
+    tmp = decompress( buffer_size, infd, &pp, testing );
     if( tmp > retval ) retval = tmp;
     if( tmp && !testing ) cleanup_and_fail( retval );