path: root/zgrep.cc

/* Zgrep - search compressed files for a regular expression
   Copyright (C) 2010-2022 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
   the Free Software Foundation, either version 2 of the License, or
   (at your option) any later version.

   This program is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
   GNU General Public License for more details.

   You should have received a copy of the GNU General Public License
   along with this program.  If not, see <http://www.gnu.org/licenses/>.
*/

#define _FILE_OFFSET_BITS 64

#include <cerrno>
#include <climits>
#include <csignal>
#include <cstdio>
#include <cstdlib>
#include <cstring>
#include <list>
#include <string>
#include <vector>
#include <dirent.h>
#include <fcntl.h>
#include <stdint.h>
#include <unistd.h>
#include <sys/stat.h>
#if defined __MSVCRT__ || defined __OS2__
#include <io.h>
#endif

#include "arg_parser.h"
#include "rc.h"
#include "zutils.h"


namespace {

#include "recursive.cc"
#include "zcatgrep.cc"

void show_help()
  {
  std::printf( "zgrep is a front end to the program grep that allows transparent search\n"
               "on any combination of compressed and uncompressed files. If any file\n"
               "given is compressed, its decompressed content is used. If a file given\n"
               "does not exist, and its name does not end with one of the known\n"
               "extensions, zgrep tries the compressed file names corresponding to the\n"
               "formats supported. If a file fails to decompress, zgrep continues\n"
               "searching the rest of the files.\n"
               "\nIf a file is specified as '-', data are read from standard input,\n"
               "decompressed if needed, and fed to grep. Data read from standard input\n"
               "must be of the same type; all uncompressed or all in the same\n"
               "compressed format.\n"
               "\nIf no files are specified, recursive searches examine the current\n"
               "working directory, and nonrecursive searches read standard input.\n"
               "\n'zgrep --verbose -V' prints the version of the grep program used.\n"
               "\nThe formats supported are bzip2, gzip, lzip, xz, and zstd.\n"
               "\nUsage: zgrep [options] <pattern> [files]\n"
               "\nExit status is 0 if match, 1 if no match, 2 if trouble.\n"
               "Some options only work if the grep program used supports them.\n"
               "\nOptions:\n"
               "      --help                   display this help and exit\n"
               "  -V, --version                output version information and exit\n"
               "  -a, --text                   treat all files as text\n"
               "  -A, --after-context=<n>      print <n> lines of trailing context\n"
               "  -b, --byte-offset            print the byte offset of each line\n"
               "  -B, --before-context=<n>     print <n> lines of leading context\n"
               "  -c, --count                  only print a count of matching lines per file\n"
               "  -C, --context=<n>            print <n> lines of output context\n"
               "      --color[=<when>]         show matched strings in color\n"
               "  -e, --regexp=<pattern>       use <pattern> as the pattern to match\n"
               "  -E, --extended-regexp        <pattern> is an extended regular expression\n"
               "  -f, --file=<file>            obtain patterns from <file>\n"
               "  -F, --fixed-strings          <pattern> is a set of newline-separated strings\n"
               "  -G, --basic-regexp           <pattern> is a basic regular expression (default)\n"
               "  -h, --no-filename            suppress the prefixing file name on output\n"
               "  -H, --with-filename          print the file name for each match\n"
               "  -i, --ignore-case            ignore case distinctions\n"
               "  -I                           ignore binary files\n"
               "  -l, --files-with-matches     only print names of files containing matches\n"
               "  -L, --files-without-match    only print names of files containing no matches\n"
               "      --label=<label>          use <label> as file name for standard input\n"
               "      --line-buffered          flush output on every line\n"
               "  -m, --max-count=<n>          stop after <n> matches\n"
               "  -M, --format=<list>          process only the formats in <list>\n"
               "  -n, --line-number            print the line number of each line\n"
               "  -N, --no-rcfile              don't read runtime configuration file\n"
               "  -o, --only-matching          show only the part of a line matching <pattern>\n"
               "  -O, --force-format=<fmt>     force the format given (bz2, gz, lz, xz, zst)\n"
               "  -P, --perl-regexp            <pattern> is a Perl regular expression\n"
               "  -q, --quiet, --silent        suppress all messages\n"
               "  -r, --recursive              operate recursively on directories\n"
               "  -R, --dereference-recursive  recursively follow symbolic links\n"
               "  -s, --no-messages            suppress error messages\n"
               "  -T, --initial-tab            make tabs line up (if needed)\n"
               "  -U, --binary                 don't strip CR characters at EOL (DOS/Windows)\n"
               "  -v, --invert-match           select non-matching lines\n"
               "      --verbose                verbose mode (show error messages)\n"
               "  -w, --word-regexp            match only whole words\n"
               "  -x, --line-regexp            match only whole lines\n"
               "  -Z, --null                   print 0 byte (ASCII NUL) after file name\n"
               "      --bz2=<command>          set compressor and options for bzip2 format\n"
               "      --gz=<command>           set compressor and options for gzip format\n"
               "      --lz=<command>           set compressor and options for lzip format\n"
               "      --xz=<command>           set compressor and options for xz format\n"
               "      --zst=<command>          set compressor and options for zstd format\n"
               "\nNumbers 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();
  }


int zgrep_file( int infd, const int format_index,
                const std::string & input_filename,
                const std::vector< const char * > & grep_args,
                const int list_mode, const bool initial_tab,
                const bool line_buffered, const bool show_name,
                const bool z_null )
  {
  Children children;
  if( !set_data_feeder( input_filename, &infd, children, format_index ) )
    return 2;
  int fda[2];				// pipe from grep
  if( pipe( fda ) < 0 )
    { show_error( "Can't create pipe", errno ); return 2; }
  const pid_t grep_pid = fork();
  if( grep_pid == 0 )			// child (grep)
    {
    if( dup2( infd, STDIN_FILENO ) >= 0 &&
        dup2( fda[1], STDOUT_FILENO ) >= 0 &&
        close( infd ) == 0 && close( fda[0] ) == 0 && close( fda[1] ) == 0 )
      {
      const char ** const argv = new const char *[grep_args.size()+2];
      argv[0] = GREP;
      for( unsigned i = 0; i < grep_args.size(); ++i )
        argv[i+1] = grep_args[i];
      argv[grep_args.size()+1] = 0;
      execvp( argv[0], (char **)argv );
      }
    show_exec_error( GREP );
    _exit( 2 );
    }
  if( grep_pid < 0 )			// parent
    { show_fork_error( GREP ); return 2; }

  close( fda[1] );
  enum { buffer_size = 256 };
  uint8_t buffer[buffer_size];
  bool line_begin = true;
  bool at_eof = false;
  while( !at_eof )
    {
    int size;
    bool error = false;
    if( line_buffered )
      for( size = 0; size < buffer_size; )
        { if( readblock( fda[0], buffer + size, 1 ) == 1 )
            { ++size; if( buffer[size-1] == '\n' ) break; }
          else { at_eof = true; if( errno ) { error = true; } break; } }
    else
      { size = readblock( fda[0], buffer, buffer_size );
        if( size < buffer_size ) { at_eof = true; if( errno ) error = true; } }
    if( error )
      { std::fflush( stdout ); show_error( "Read error", errno ); return 2; }
    if( size > 0 && !list_mode )
      {
      if( show_name )			// print the file name for each match
        for( int i = 0; i < size; ++i )
          {
          if( line_begin )
            { line_begin = false;
              const int len = std::printf( "%s%c", input_filename.c_str(),
                                           z_null ? 0 : ':' );
              if( initial_tab && len > 0 && len % 8 ) putchar( '\t' ); }
          putchar( buffer[i] );
          if( buffer[i] == '\n' )
            { line_begin = true; if( line_buffered ) std::fflush( stdout ); }
          }
      else if( std::fwrite( buffer, 1, size, stdout ) != (unsigned)size )
        { std::fflush( stdout ); show_error( "Write error", errno ); return 2; }
      }
    }
  std::fflush( stdout );

  int retval = wait_for_child( grep_pid, GREP );

  if( !good_status( children, retval == 1 ) ) retval = 2;

  if( list_mode && (retval == 0) == (list_mode == 1) )
    { std::printf( "%s%c", input_filename.c_str(), z_null ? 0 : '\n' );
      std::fflush( stdout ); }
  if( close( infd ) != 0 )
    { show_close_error(); return 2; }
  if( close( fda[0] ) != 0 )
    { show_close_error( GREP ); return 2; }
  return retval;
  }

} // end namespace


int main( const int argc, const char * const argv[] )
  {
  enum { help_opt = 256, verbose_opt, color_opt, label_opt, linebuf_opt,
         bz2_opt, gz_opt, lz_opt, xz_opt, zst_opt };
  int format_index = -1;
  int list_mode = 0;		// 1 = list matches, -1 = list non-matches
  int recursive = 0;		// 1 = '-r', 2 = '-R'
  int show_name = -1;		// tri-state bool
  bool initial_tab = false;
  bool line_buffered = false;
  bool no_messages = false;
  bool z_null = false;				// for '-Z, --null'
  std::list< std::string > filenames;
  std::vector< const char * > grep_args;	// args to grep, maybe empty
  std::string color_option;			// additional args to grep
  std::string label_option;
  std::string label = "(standard input)";	// prefix for standard input
  program_name = "zgrep";
  invocation_name = ( argc > 0 ) ? argv[0] : program_name;

  const Arg_parser::Option options[] =
    {
    { 'a', "text",                  Arg_parser::no  },	// grep GNU
    { 'A', "after-context",         Arg_parser::yes },	// grep GNU
    { 'b', "byte-offset",           Arg_parser::no  },	// grep GNU
    { 'B', "before-context",        Arg_parser::yes },	// grep GNU
    { 'c', "count",                 Arg_parser::no  },	// grep
    { 'C', "context",               Arg_parser::yes },	// grep GNU
    { 'e', "regexp",                Arg_parser::yes },	// grep
    { 'E', "extended-regexp",       Arg_parser::no  },	// grep
    { 'f', "file ",                 Arg_parser::yes },	// grep
    { 'F', "fixed-strings",         Arg_parser::no  },	// grep
    { 'G', "basic-regexp",          Arg_parser::no  },	// grep GNU
    { 'h', "no-filename",           Arg_parser::no  },	// grep GNU
    { 'H', "with-filename",         Arg_parser::no  },	// grep GNU
    { 'i', "ignore-case",           Arg_parser::no  },	// grep
    { 'I',  0,                      Arg_parser::no  },	// grep GNU
    { 'l', "files-with-matches",    Arg_parser::no  },	// grep
    { 'L', "files-without-match",   Arg_parser::no  },	// grep GNU
    { 'm', "max-count",             Arg_parser::yes },	// grep GNU
    { 'M', "format",                Arg_parser::yes },
    { 'n', "line-number",           Arg_parser::no  },	// grep
    { 'N', "no-rcfile",             Arg_parser::no  },
    { 'o', "only-matching",         Arg_parser::no  },	// grep
    { 'O', "force-format",          Arg_parser::yes },
    { 'P', "perl-regexp",           Arg_parser::no  },	// grep GNU
    { 'q', "quiet",                 Arg_parser::no  },
    { 'q', "silent",                Arg_parser::no  },
    { 'r', "recursive",             Arg_parser::no  },
    { 'R', "dereference-recursive", Arg_parser::no  },
    { 's', "no-messages",           Arg_parser::no  },	// grep
    { 'T', "initial-tab",           Arg_parser::no  },	// grep GNU
    { 'U', "binary",                Arg_parser::no  },	// grep GNU
    { 'v', "invert-match",          Arg_parser::no  },	// grep
    { 'V', "version",               Arg_parser::no  },
    { 'w', "word-regexp",           Arg_parser::no  },	// grep GNU
    { 'x', "line-regexp",           Arg_parser::no  },	// grep
    { 'Z', "null",                  Arg_parser::no  },	// grep GNU
    { help_opt,    "help",          Arg_parser::no  },
    { verbose_opt, "verbose",       Arg_parser::no  },
    { color_opt,   "color",         Arg_parser::maybe },
    { label_opt,   "label",         Arg_parser::yes },
    { linebuf_opt, "line-buffered", Arg_parser::no  },
    { bz2_opt, "bz2",               Arg_parser::yes },
    { gz_opt,  "gz",                Arg_parser::yes },
    { lz_opt,  "lz",                Arg_parser::yes },
    { xz_opt,  "xz",                Arg_parser::yes },
    { zst_opt, "zst",               Arg_parser::yes },
    {  0,   0,                      Arg_parser::no  } };

  const Arg_parser parser( argc, argv, options );
  if( parser.error().size() )				// bad option
    { show_error( parser.error().c_str(), 0, true ); return 2; }

  maybe_process_config_file( parser );

  int argind = 0;
  bool pattern_found = false;
  for( ; argind < parser.arguments(); ++argind )
    {
    const int code = parser.code( argind );
    if( !code ) break;					// no more options
    const char * const pn = parser.parsed_name( argind ).c_str();
    const std::string & sarg = parser.argument( argind );
    const char * const arg = sarg.c_str();
    switch( code )
      {
      case 'a': grep_args.push_back( "-a" ); break;
      case 'A': grep_args.push_back( "-A" );
                grep_args.push_back( arg ); break;
      case 'b': grep_args.push_back( "-b" ); break;
      case 'B': grep_args.push_back( "-B" );
                grep_args.push_back( arg ); break;
      case 'c': grep_args.push_back( "-c" ); break;
      case 'C': grep_args.push_back( "-C" );
                grep_args.push_back( arg ); break;
      case 'e': grep_args.push_back( "-e" );
                grep_args.push_back( arg ); pattern_found = true; break;
      case 'E': grep_args.push_back( "-E" ); break;
      case 'f': grep_args.push_back( "-f" );
                grep_args.push_back( arg ); pattern_found = true; break;
      case 'F': grep_args.push_back( "-F" ); break;
      case 'G': grep_args.push_back( "-G" ); break;
      case 'h': show_name = false; break;
      case 'H': show_name = true; break;
      case 'i': grep_args.push_back( "-i" ); break;
      case 'I': grep_args.push_back( "-I" ); break;
      case 'l': grep_args.push_back( "-l" ); list_mode = 1; break;
      case 'L': grep_args.push_back( "-L" ); list_mode = -1; break;
      case 'm': grep_args.push_back( "-m" );
                grep_args.push_back( arg ); break;
      case 'M': parse_format_list( sarg, pn ); break;
      case 'n': grep_args.push_back( "-n" ); break;
      case 'N': break;
      case 'o': grep_args.push_back( "-o" ); break;
      case 'O': format_index = parse_format_type( sarg, pn ); break;
      case 'P': grep_args.push_back( "-P" ); break;
      case 'q': grep_args.push_back( "-q" ); verbosity = -1; break;
      case 'r': recursive = 1; break;
      case 'R': recursive = 2; break;
      case 's': grep_args.push_back( "-s" ); no_messages = true; break;
      case 'T': grep_args.push_back( "-T" ); initial_tab = true; break;
      case 'U': grep_args.push_back( "-U" ); break;
      case 'v': grep_args.push_back( "-v" ); break;
      case 'V': show_version( GREP " --version" ); return 0;
      case 'w': grep_args.push_back( "-w" ); break;
      case 'x': grep_args.push_back( "-x" ); break;
      case 'Z': z_null = true; break;
      case help_opt: show_help(); return 0;
      case verbose_opt: no_messages = false; if( verbosity < 4 ) ++verbosity;
        break;
      case color_opt: color_option = "--color";
        if( !sarg.empty() ) { color_option += '='; color_option += sarg; }
        break;
      case label_opt: label_option = label = sarg; break;
      case linebuf_opt: grep_args.push_back( "--line-buffered" );
        line_buffered = true; break;
      case bz2_opt: parse_compressor( sarg, fmt_bz2 ); break;
      case gz_opt: parse_compressor( sarg, fmt_gz ); break;
      case lz_opt: parse_compressor( sarg, fmt_lz ); break;
      case xz_opt: parse_compressor( sarg, fmt_xz ); break;
      case zst_opt: parse_compressor( sarg, fmt_zst ); break;
      default : internal_error( "uncaught option." );
      }
    } // end process options

  if( !color_option.empty() )		// push the last value set
    grep_args.push_back( color_option.c_str() );
  if( !label_option.empty() )		// for "Binary file <label> matches"
    grep_args.push_back( label_option.insert( 0, "--label=" ).c_str() );

#if defined __MSVCRT__ || defined __OS2__
  setmode( STDIN_FILENO, O_BINARY );
  setmode( STDOUT_FILENO, O_BINARY );
#endif

  if( !pattern_found )
    {
    if( argind >= parser.arguments() )
      { show_error( "Pattern not found." ); return 2; }
    const std::string & pat = parser.argument( argind++ );
    if( pat.size() && pat[0] == '-' ) grep_args.push_back( "-e" );
    grep_args.push_back( pat.c_str() );
    }

  for( ; argind < parser.arguments(); ++argind )
    filenames.push_back( parser.argument( argind ) );

  if( filenames.empty() ) filenames.push_back( recursive ? "." : "-" );

  if( show_name < 0 ) show_name = ( filenames.size() != 1 || recursive );

  std::string input_filename;
  int retval = 1;
  bool error = false;
  bool stdin_used = false;
  while( next_filename( filenames, input_filename, error, recursive,
                        false, no_messages ) )
    {
    int infd;
    if( input_filename == "." )
      {
      if( stdin_used ) continue; else stdin_used = true;
      infd = STDIN_FILENO; input_filename = label;
      }
    else
      {
      infd = open_instream( input_filename, format_index < 0, no_messages );
      if( infd < 0 ) { error = true; continue; }
      }

    const int tmp = zgrep_file( infd, format_index, input_filename, grep_args,
                                list_mode, initial_tab, line_buffered,
                                show_name, z_null );
    if( tmp == 0 || ( tmp == 2 && retval == 1 ) ) retval = tmp;

    if( close( infd ) != 0 )
      { show_file_error( input_filename.c_str(), "Error closing input file",
                         errno ); error = true; }
    if( retval == 0 && verbosity < 0 ) break;
    }

  if( std::fclose( stdout ) != 0 )
    {
    show_error( "Error closing stdout", errno );
    error = true;
    }
  if( error && ( retval != 0 || verbosity >= 0 ) ) retval = 2;
  return retval;
  }