Merging upstream version 1.0~rc6.

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

1 files changed, 74 insertions, 123 deletions

diff --git a/zcmp.cc b/zcmp.cc
index b00d40b..c2fdd7a 100644
--- a/zcmp.cc
+++ b/zcmp.cc
@@ -1,5 +1,5 @@
 /*  Zcmp - decompress and compare two files byte by byte
-    Copyright (C) 2010, 2011 Antonio Diaz Diaz.
+    Copyright (C) 2010, 2011, 2012, 2013 Antonio Diaz Diaz.
 
     This program is free software: you can redistribute it and/or modify
     it under the terms of the GNU General Public License as published by
@@ -17,6 +17,7 @@
 
 #define _FILE_OFFSET_BITS 64
 
+#include <algorithm>
 #include <cctype>
 #include <cerrno>
 #include <climits>
@@ -30,9 +31,6 @@
 #include <stdint.h>
 #include <unistd.h>
 #include <sys/stat.h>
-#if defined(__MSVCRT__) || defined(__OS2__)
-#include <io.h>
-#endif
 
 #include "arg_parser.h"
 #include "zutils.h"
@@ -44,73 +42,52 @@
 #ifndef LLONG_MAX
 #define LLONG_MAX  0x7FFFFFFFFFFFFFFFLL
 #endif
-#ifndef LLONG_MIN
-#define LLONG_MIN  (-LLONG_MAX - 1LL)
-#endif
-#ifndef ULLONG_MAX
-#define ULLONG_MAX 0xFFFFFFFFFFFFFFFFULL
-#endif
 
 
 namespace {
 
-#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[] = {
-  { ".bz2",  ""     },
-  { ".tbz",  ".tar" },
-  { ".tbz2", ".tar" },
-  { ".gz",   ""     },
-  { ".tgz",  ".tar" },
-  { ".lz",   ""     },
-  { ".tlz",  ".tar" },
-  { ".xz",   ""     },
-  { ".txz",  ".tar" },
-  { 0,       0      } };
+#include "zcmpdiff.cc"
 
 
-void show_help() throw()
+void show_help()
   {
-  std::printf( "Zcmp compares two files (\"-\" means standard input), and if they\n" );
-  std::printf( "differ, tells the first byte and line number where they differ. Bytes\n" );
-  std::printf( "and lines are numbered starting with 1. If any given file is compressed,\n" );
-  std::printf( "its uncompressed content is used. Compressed files are uncompressed on\n" );
-  std::printf( "the fly; no temporary files are created.\n" );
-  std::printf( "The supported compressors are bzip2, gzip, lzip and xz.\n" );
-  std::printf( "\nUsage: zcmp [options] file1 [file2]\n" );
-  std::printf( "\nCompares <file1> to <file2>. If <file2> is omitted zcmp tries the\n" );
-  std::printf( "following:\n" );
-  std::printf( "If <file1> is compressed, compares <file1> to the file with the\n" );
-  std::printf( "corresponding decompressed file name (removes the extension from\n" );
-  std::printf( "<file1>).\n" );
-  std::printf( "If <file1> is not compressed, compares <file1> to the uncompressed\n" );
-  std::printf( "contents of <file1>.[bz2|gz|lz|xz] (the first one that is found).\n" );
-  std::printf( "If no suitable file is found, compares <file1> to data read from\n" );
-  std::printf( "standard input.\n" );
-  std::printf( "\nExit status is 0 if inputs are identical, 1 if different, 2 if trouble.\n" );
-  std::printf( "\nOptions:\n" );
-  std::printf( "  -h, --help                       display this help and exit\n" );
-  std::printf( "  -V, --version                    output version information and exit\n" );
-  std::printf( "  -b, --print-bytes                print differing bytes\n" );
-  std::printf( "  -i, --ignore-initial=<n>[,<n2>]  ignore differences in the first <n> bytes\n" );
-  std::printf( "  -l, --list                       list position, value of all differing bytes\n" );
-  std::printf( "  -n, --bytes=<n>                  compare at most <n> bytes\n" );
-  std::printf( "  -q, --quiet                      suppress all messages\n" );
-  std::printf( "  -s, --silent                     (same as --quiet)\n" );
-  std::printf( "  -v, --verbose                    verbose mode (same as --list)\n" );
-  std::printf( "Numbers may be followed by a multiplier: k = kB = 10^3 = 1000,\n" );
-  std::printf( "Ki = KiB = 2^10 = 1024, M = 10^6, Mi = 2^20, G = 10^9, Gi = 2^30, etc...\n" );
+  std::printf( "Zcmp compares two files (\"-\" means standard input), and if they\n"
+               "differ, tells the first byte and line number where they differ. Bytes\n"
+               "and lines are numbered starting with 1. If any given file is compressed,\n"
+               "its uncompressed content is used. Compressed files are uncompressed on\n"
+               "the fly; no temporary files are created.\n"
+               "\nThe supported formats are bzip2, gzip, lzip and xz.\n"
+               "\nUsage: zcmp [options] file1 [file2]\n"
+               "\nCompares <file1> to <file2>. If <file2> is omitted zcmp tries the\n"
+               "following:\n"
+               "If <file1> is compressed, compares <file1> to the file with the\n"
+               "corresponding decompressed file name (removes the extension from\n"
+               "<file1>).\n"
+               "If <file1> is not compressed, compares <file1> to the uncompressed\n"
+               "contents of <file1>.[bz2|gz|lz|xz] (the first one that is found).\n"
+               "If no suitable file is found, compares <file1> to data read from\n"
+               "standard input.\n"
+               "\nExit status is 0 if inputs are identical, 1 if different, 2 if trouble.\n"
+               "\nOptions:\n"
+               "  -h, --help                       display this help and exit\n"
+               "  -V, --version                    output version information and exit\n"
+               "  -b, --print-bytes                print differing bytes\n"
+               "      --format=[<fmt1>][,<fmt2>]   force given formats (bz2, gz, lz, xz)\n"
+               "  -i, --ignore-initial=<n>[,<n2>]  ignore differences in the first <n> bytes\n"
+               "  -l, --list                       list position, value of all differing bytes\n"
+               "  -n, --bytes=<n>                  compare at most <n> bytes\n"
+               "  -q, --quiet                      suppress all messages\n"
+               "  -s, --silent                     (same as --quiet)\n"
+               "  -v, --verbose                    verbose mode (same as --list)\n"
+               "Numbers may be followed by a multiplier: k = kB = 10^3 = 1000,\n"
+               "Ki = KiB = 2^10 = 1024, M = 10^6, Mi = 2^20, G = 10^9, Gi = 2^30, etc...\n" );
   show_help_addr();
   }
 
 
 long long getnum( const char * const ptr, const char ** const tailp = 0,
-                  const long long llimit = LLONG_MIN + 1,
-                  const long long ulimit = LLONG_MAX ) throw()
+                  const long long llimit = 0,
+                  const long long ulimit = LLONG_MAX )
   {
   errno = 0;
   char * tail;
@@ -120,6 +97,7 @@ long long getnum( const char * const ptr, const char ** const tailp = 0,
     show_error( "Bad or missing numerical argument.", 0, true );
     std::exit( 2 );
     }
+  if( result < 0 ) errno = ERANGE;
 
   if( !errno && tail[0] && std::isalpha( tail[0] ) )
     {
@@ -148,7 +126,7 @@ long long getnum( const char * const ptr, const char ** const tailp = 0,
       }
     for( int i = 0; i < exponent; ++i )
       {
-      if( LLONG_MAX / factor >= llabs( result ) ) result *= factor;
+      if( ulimit / factor >= result ) result *= factor;
       else { errno = ERANGE; break; }
       }
     }
@@ -163,53 +141,12 @@ long long getnum( const char * const ptr, const char ** const tailp = 0,
   }
 
 
-int open_instream( const std::string & input_filename ) throw()
-  {
-  int infd = open( input_filename.c_str(), O_RDONLY | o_binary );
-  if( infd < 0 )
-    show_error2( "Can't open input file", input_filename.c_str() );
-  return infd;
-  }
-
-
-int open_other_instream( std::string & name ) throw()
-  {
-  for( int i = 0; known_extensions[i].from; ++i )
-    {					// search uncompressed version
-    const std::string from( known_extensions[i].from );
-    if( name.size() > from.size() &&
-        name.compare( name.size() - from.size(), from.size(), from ) == 0 )
-      {
-      name.resize( name.size() - from.size() );
-      name += known_extensions[i].to;
-      return open( name.c_str(), O_RDONLY | o_binary );
-      }
-    }
-  for( int i = 0; simple_extensions[i]; ++i )
-    {					// search compressed version
-    const std::string s( name + simple_extensions[i] );
-    const int infd = open( s.c_str(), O_RDONLY | o_binary );
-    if( infd >= 0 ) { name = s; return infd; }
-    }
-  return -1;
-  }
-
-
-bool check_identical( const char * const name1, const char * const name2 ) throw()
-  {
-  if( !std::strcmp( name1, name2 ) ) return true;
-  struct stat stat1, stat2;
-  if( stat( name1, &stat1 ) || stat( name2, &stat2 ) ) return false;
-  return ( stat1.st_ino == stat2.st_ino && stat1.st_dev == stat2.st_dev );
-  }
-
-
 void parse_ignore_initial( const char * const arg, long long ignore_initial[2] )
   {
   const char * tail;
-  ignore_initial[0] = getnum( arg, &tail, 0 );
+  ignore_initial[0] = getnum( arg, &tail );
   if( *tail == ',' || *tail == ':' )
-    ignore_initial[1] = getnum( ++tail, 0, 0 );
+    ignore_initial[1] = getnum( ++tail );
   else ignore_initial[1] = ignore_initial[0];
   }
 
@@ -253,10 +190,11 @@ void sprintc( char * const buf, unsigned char c )
 
 int block_compare( const uint8_t * const buffer0,
                    const uint8_t * const buffer1,
-                   long long * line_numberp )
+                   unsigned long long * const line_numberp )
   {
   const uint8_t * p0 = buffer0;
   const uint8_t * p1 = buffer1;
+
   if( verbosity == 0 )
     {
     int nl_count = 0;
@@ -272,10 +210,11 @@ int block_compare( const uint8_t * const buffer0,
 int cmp( const long long max_size, const int infd[2],
          const std::string filenames[2], const bool print_bytes )
   {
-  enum { buffer_size = 4096 };
-  long long byte_number = 1;
-  long long line_number = 1;
-  long long rest = max_size;	// remaining number of bytes to compare
+  const int buffer_size = 4096;
+  unsigned long long byte_number = 1;
+  unsigned long long line_number = 1;
+  // remaining number of bytes to compare
+  long long rest = ( ( max_size >= 0 ) ? max_size : buffer_size );
   // buffers with space for sentinels at the end
   uint8_t * const buffer0 = new uint8_t[2*(buffer_size+1)];
   uint8_t * const buffer1 = buffer0 + buffer_size + 1;
@@ -285,7 +224,8 @@ int cmp( const long long max_size, const int infd[2],
 
   while( rest > 0 )
     {
-    const int size = std::min( rest, (long long)buffer_size );
+    const int size = std::min( (long long)buffer_size, rest );
+    if( max_size >= 0 ) rest -= size;
     int rd[2];			// number of bytes read from each file
     for( int i = 0; i < 2; ++i )
       {
@@ -296,7 +236,6 @@ int cmp( const long long max_size, const int infd[2],
         return 2;
         }
       }
-    rest -= size;
 
     buffer0[rd[0]] = ~buffer1[rd[0]];	// sentinels for the block compare
     buffer1[rd[1]] = ~buffer0[rd[1]];
@@ -311,7 +250,7 @@ int cmp( const long long max_size, const int infd[2],
       if( verbosity == 0 )			// show first difference
         {
         if( !print_bytes )
-          std::printf( "%s %s differ: byte %lld, line %lld\n",
+          std::printf( "%s %s differ: byte %llu, line %llu\n",
                        filenames[0].c_str(), filenames[1].c_str(),
                        byte_number, line_number );
         else
@@ -320,7 +259,7 @@ int cmp( const long long max_size, const int infd[2],
           const unsigned char c1 = buffer1[first_diff];
           char buf0[5], buf1[5];
           sprintc( buf0, c0 ); sprintc( buf1, c1 );
-          std::printf( "%s %s differ: byte %lld, line %lld is %3o %s %3o %s\n",
+          std::printf( "%s %s differ: byte %llu, line %llu is %3o %s %3o %s\n",
                        filenames[0].c_str(), filenames[1].c_str(),
                        byte_number, line_number, c0, buf0, c1, buf1 );
           }
@@ -336,12 +275,12 @@ int cmp( const long long max_size, const int infd[2],
           if( c0 != c1 )
             {
             if( !print_bytes )
-              std::printf( "%lld %3o %3o\n", byte_number, c0, c1 );
+              std::printf( "%llu %3o %3o\n", byte_number, c0, c1 );
             else
               {
               char buf0[5], buf1[5];
               sprintc( buf0, c0 ); sprintc( buf1, c1 );
-              std::printf( "%lld %3o %-4s %3o %s\n",
+              std::printf( "%llu %3o %-4s %3o %s\n",
                            byte_number, c0, buf0, c1, buf1 );
               }
             }
@@ -368,9 +307,11 @@ int cmp( const long long max_size, const int infd[2],
 
 int main( const int argc, const char * const argv[] )
   {
+  enum { format_opt = 256 };
   // number of initial bytes ignored for each file
   long long ignore_initial[2] = { 0, 0 };
-  long long max_size = LLONG_MAX;
+  long long max_size = -1;			// < 0 means unlimited size
+  int format_types[2] = { -1, -1 };
   bool print_bytes = false;
   invocation_name = argv[0];
   util_name = "zcmp";
@@ -386,6 +327,7 @@ int main( const int argc, const char * const argv[] )
     { 's', "silent",         Arg_parser::no  },
     { 'v', "verbose",        Arg_parser::no  },
     { 'V', "version",        Arg_parser::no  },
+    { format_opt, "format",  Arg_parser::yes },
     {  0 ,  0,               Arg_parser::no  } };
 
   const Arg_parser parser( argc, argv, options );
@@ -404,18 +346,19 @@ int main( const int argc, const char * const argv[] )
       case 'h': show_help(); return 0;
       case 'i': parse_ignore_initial( arg, ignore_initial ); break;
       case 'l': verbosity = 1; break;
-      case 'n': max_size = getnum( arg, 0, 0 ); break;
+      case 'n': max_size = getnum( arg ); break;
       case 'q':
       case 's': verbosity = -1; break;
       case 'v': verbosity = 1; break;
       case 'V': show_version( "Zcmp" ); return 0;
+      case format_opt: get_format_types( arg, format_types ); break;
       default : internal_error( "uncaught option" );
       }
     } // end process options
 
 #if defined(__MSVCRT__) || defined(__OS2__)
-  _setmode( STDIN_FILENO, O_BINARY );
-  _setmode( STDOUT_FILENO, O_BINARY );
+  _fsetmode( stdin, "b" );
+  _fsetmode( stdout, "b" );
 #endif
 
   if( argind >= parser.arguments() )
@@ -449,6 +392,9 @@ int main( const int argc, const char * const argv[] )
     }
   else
     {
+    if( format_types[0] >= 0 || format_types[1] >= 0 )
+      { show_error( "Two files must be given when format is specified.", 0, true );
+        return 2; }
     filenames[1] = filenames[0];
     infd[1] = open_other_instream( filenames[1] );
     if( infd[1] < 0 ) { infd[1] = STDIN_FILENO; filenames[1] = "-"; }
@@ -457,8 +403,8 @@ int main( const int argc, const char * const argv[] )
   int old_infd[2];		// copy of file descriptors of the two files
   old_infd[0] = infd[0]; old_infd[1] = infd[1];
   pid_t pid[2];
-  if( !set_data_feeder( &infd[0], &pid[0] ) ||
-      !set_data_feeder( &infd[1], &pid[1] ) )
+  if( !set_data_feeder( &infd[0], &pid[0], format_types[0] ) ||
+      !set_data_feeder( &infd[1], &pid[1], format_types[1] ) )
     return 2;
 
   for( int i = 0; i < 2; ++i )
@@ -470,10 +416,15 @@ int main( const int argc, const char * const argv[] )
 
   int retval = cmp( max_size, infd, filenames, print_bytes );
 
-  if( retval != 0 )
+  if( retval != 0 || max_size >= 0 )
     {
-    if( pid[0] ) kill( pid[0], SIGTERM );
-    if( pid[1] ) kill( pid[1], SIGTERM );
+    for( int i = 0; i < 2; ++i )
+      if( pid[i] )
+        {
+        const int tmp = child_status( pid[i], "data feeder" );
+        if( tmp < 0 ) kill( pid[i], SIGTERM );	// child not terminated
+        else if( tmp != 0 ) retval = 2;		// child status != 0
+        }
     }
   else
     if( ( pid[0] && wait_for_child( pid[0], "data feeder" ) != 0 ) ||