summaryrefslogtreecommitdiffstats
path: root/zgrep.cc
diff options
context:
space:
mode:
authorDaniel Baumann <daniel.baumann@progress-linux.org>2022-07-26 05:52:19 +0000
committerDaniel Baumann <daniel.baumann@progress-linux.org>2022-07-26 05:52:47 +0000
commit274c86b0397198011756479218dd14262d9421ff (patch)
tree47cd5cedf3cfcbba77d127200852a21d03502e7c /zgrep.cc
parentReleasing debian version 1.11-5. (diff)
downloadzutils-274c86b0397198011756479218dd14262d9421ff.tar.xz
zutils-274c86b0397198011756479218dd14262d9421ff.zip
Merging upstream version 1.12~pre2.
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'zgrep.cc')
-rw-r--r--zgrep.cc154
1 files changed, 83 insertions, 71 deletions
diff --git a/zgrep.cc b/zgrep.cc
index 2ed569b..ef8db6b 100644
--- a/zgrep.cc
+++ b/zgrep.cc
@@ -79,26 +79,33 @@ void show_help()
" -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"
- " -q, --quiet suppress all messages\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"
@@ -110,43 +117,12 @@ void show_help()
}
-int zgrep_stdin( int infd, const int format_index,
- const std::vector< const char * > & grep_args )
- {
- Children children;
- if( !set_data_feeder( "", &infd, children, format_index ) ) return 2;
- const pid_t grep_pid = fork();
- if( grep_pid == 0 ) // child (grep)
- {
- if( dup2( infd, STDIN_FILENO ) >= 0 && close( infd ) == 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; }
-
- int retval = wait_for_child( grep_pid, GREP );
-
- if( !good_status( children, retval == 1 ) ) retval = 2;
-
- if( close( infd ) != 0 )
- { show_close_error(); return 2; }
- return retval;
- }
-
-
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 show_name )
+ 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 ) )
@@ -178,34 +154,48 @@ int zgrep_file( int infd, const int format_index,
enum { buffer_size = 256 };
uint8_t buffer[buffer_size];
bool line_begin = true;
- while( true )
+ bool at_eof = false;
+ while( !at_eof )
{
- const int size = readblock( fda[0], buffer, buffer_size );
- if( size != buffer_size && errno )
- { show_error( "Read error", errno ); return 2; }
+ 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 )
+ if( show_name ) // print the file name for each match
for( int i = 0; i < size; ++i )
{
if( line_begin )
- { line_begin = false; std::printf( "%s:", input_filename.c_str() ); }
- if( buffer[i] == '\n' ) line_begin = true;
+ { 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 );
}
- if( size < buffer_size ) break; // end of grep's output
}
+ 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\n", input_filename.c_str() ); std::fflush( stdout ); }
+ { 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 )
@@ -218,16 +208,21 @@ int zgrep_file( int infd, const int format_index,
int main( const int argc, const char * const argv[] )
{
- enum { help_opt = 256, verbose_opt, color_opt,
+ 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; // needed because of optional arg
+ 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;
@@ -243,6 +238,7 @@ int main( const int argc, const char * const argv[] )
{ '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
@@ -255,17 +251,24 @@ int main( const int argc, const char * const argv[] )
{ '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 },
@@ -286,24 +289,26 @@ int main( const int argc, const char * const argv[] )
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 & arg = parser.argument( argind );
+ 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.c_str() ); break;
+ 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.c_str() ); break;
+ 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.c_str() ); break;
+ grep_args.push_back( arg ); break;
case 'e': grep_args.push_back( "-e" );
- grep_args.push_back( arg.c_str() ); pattern_found = true; break;
+ 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.c_str() ); pattern_found = true; break;
+ 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;
@@ -311,37 +316,46 @@ int main( const int argc, const char * const argv[] )
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.c_str() ); break;
- case 'M': parse_format_list( arg, pn ); break;
+ 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( arg, pn ); 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( !arg.empty() ) { color_option += '='; color_option += arg; }
+ if( !sarg.empty() ) { color_option += '='; color_option += sarg; }
break;
- case bz2_opt: parse_compressor( arg, fmt_bz2 ); break;
- case gz_opt: parse_compressor( arg, fmt_gz ); break;
- case lz_opt: parse_compressor( arg, fmt_lz ); break;
- case xz_opt: parse_compressor( arg, fmt_xz ); break;
- case zst_opt: parse_compressor( arg, fmt_zst ); 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 );
@@ -352,9 +366,9 @@ int main( const int argc, const char * const argv[] )
{
if( argind >= parser.arguments() )
{ show_error( "Pattern not found." ); return 2; }
- const std::string & arg = parser.argument( argind++ );
- if( arg.size() && arg[0] == '-' ) grep_args.push_back( "-e" );
- grep_args.push_back( arg.c_str() );
+ 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 )
@@ -375,7 +389,7 @@ int main( const int argc, const char * const argv[] )
if( input_filename == "." )
{
if( stdin_used ) continue; else stdin_used = true;
- infd = STDIN_FILENO; input_filename = "-";
+ infd = STDIN_FILENO; input_filename = label;
}
else
{
@@ -383,11 +397,9 @@ int main( const int argc, const char * const argv[] )
if( infd < 0 ) { error = true; continue; }
}
- int tmp;
- if( infd == STDIN_FILENO )
- tmp = zgrep_stdin( infd, format_index, grep_args );
- else tmp = zgrep_file( infd, format_index, input_filename, grep_args,
- list_mode, show_name );
+ 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 )