--extensionreplace implemented

A few bugs around {} fixed

parallel

@@ -36,7 +36,7 @@ be used for putting the argument in the command line. It can be
 changed with B<-I>.
 
 
-=item B<{.}> (not implemented)
+=item B<{.}>
 
 Input line without extension. This is a specialized replacement string
 with the extension removed. It will remove from the last B<.> till the
@@ -103,7 +103,9 @@ B<-g> is the default. Can be reversed with B<-u>.
 Use the replacement string I<string> instead of {}.
 
 
-=item B<-U> I<string> (not implemented)
+=item B<-U> I<string>
+
+=item B<--extensionreplace> I<string>
 
 Use the replacement string I<string> instead of {.} for input line without extension.
 
@@ -782,7 +784,7 @@ hard, as all foreground processes get the INT from the shell.
 If there are nomore jobs (STDIN is closed) then make sure to
 distribute the arguments evenly if running -X.
 
-Distibute jobs to computers with different speeds/number-of-cpu-cores using ssh
+Distribute jobs to computers with different speeds/number-of-cpu-cores using ssh
 ask the computers how many cpus they have and spawn appropriately
 according to -j setting. Reuse ssh connection (-M and -S)
 http://www.semicomplete.com/blog/geekery/distributed-xargs.html?source=rss20
@@ -790,9 +792,7 @@ http://code.google.com/p/ppss/wiki/Manual2
 
 http://www.gnu.org/software/pexec/
 
-
+Where will '>' be run? Local or remote? Remote.
-
-Where will '>' be run? Local or remote? Where ever is easier.
 
 
 Parallelize so this can be done:
@@ -870,6 +870,7 @@ GetOptions("debug|D" => \$::opt_D,
 	   "null|0" => \$::opt_0,
 	   "quote|q" => \$::opt_q,
 	   "I=s" => \$::opt_I,
+	   "extensionreplace|U=s" => \$::opt_U,
 	   "jobs|j=s" => \$::opt_P,
 	   "number-of-cpus" => \$::opt_number_of_cpus,
 	   "number-of-cores" => \$::opt_number_of_cores,
@@ -910,6 +911,7 @@ $Global::grouped = 1;
 $Global::keeporder = 0;
 $Global::quoting = 0;
 $Global::replacestring = '{}';
+$Global::replace_no_ext = '{.}';
 $Global::input_is_filename = (@ARGV);
 $/="\n";
 $Global::ignore_empty = 0;
@@ -934,6 +936,7 @@ if(defined $::opt_q) { $Global::quoting = 1; }
 if(defined $::opt_r) { $Global::ignore_empty = 1; }
 if(defined $::opt_t) { $Global::stderr_verbose = 1; }
 if(defined $::opt_I) { $Global::replacestring = $::opt_I; }
+if(defined $::opt_U) { $Global::replace_no_ext = $::opt_U; }
 if(defined $::opt_i and $::opt_i) { $Global::replacestring = $::opt_i; }
 if(defined $::opt_E and $::opt_E) { $Global::end_of_file_string = $::opt_E; }
 if(defined $::opt_n and $::opt_n) { $Global::max_number_of_args = $::opt_n; }
@@ -976,8 +979,9 @@ drain_job_queue();
 sub generate_command_line {
     my $command = shift;
     my ($job_line,$last_good);
-    my ($next_arg,@quoted_args,$arg_length);
+    my ($next_arg,@quoted_args,@quoted_args_no_ext,$arg_length);
     my ($number_of_substitution) = 1;
+    my ($number_of_substitution_no_ext) = 0;
     my ($length_of_context) = 0;
     my ($length_of_command_no_args);
     if($Global::xargs or $Global::Xargs) {
@@ -985,48 +989,76 @@ sub generate_command_line {
 	$number_of_substitution = ($command =~ s/\Q$Global::replacestring\E/$Global::replacestring/go);
 	$number_of_substitution ||= 1;
     }
-    if($Global::xargs) {
+    if($Global::xargs or $Global::Xargs) {
+	# Count number of {.}'s on the command line
+	$number_of_substitution_no_ext =
+	($command =~ s/\Q$Global::replace_no_ext\E/$Global::replace_no_ext/go);
+	$number_of_substitution_no_ext ||= 0;
+    }
+    my $spaces=0;
+    if($Global::xargs or $Global::Xargs) {
 	my $c = $command;
+	# count number of replacements
+	my $no_of_replace = 0;
+	$no_of_replace++ while ($c =~ m/\Q$Global::replacestring\E/g);
+	my $no_of_no_ext = 0;
+	$no_of_no_ext++ while ($c =~ m/\Q$Global::replace_no_ext\E/g);
+
+	if($Global::xargs) {
 	    # remove all {}s
-	$c =~ s/\Q$Global::replacestring\E//go;
+	    $c =~ s/\Q$Global::replacestring\E|\Q$Global::replace_no_ext\E//og;
-	$length_of_command_no_args = length($c);
+	    $length_of_command_no_args = length($c) - $no_of_replace - $no_of_no_ext;
+	    $length_of_context = 0;
+	    $spaces = 1;
 	}
 	if($Global::Xargs) {
-	my $c = $command;
+	    $c =~ s/\S*\Q$Global::replacestring\E\S*//go;
-	while($c =~ s/(\S*\Q$Global::replacestring\E\S*)//o) {
+	    $c =~ s/\S*\Q$Global::replace_no_ext\E\S*//go;
-	    # Length of context minus the {}
+
-	    $length_of_context += length($1) - 2;
+	    $length_of_command_no_args = length($c) - 1;
+	    $length_of_context = length($command) - $length_of_command_no_args
+		- $no_of_replace * length($Global::replacestring)
+		- $no_of_no_ext * length($Global::replace_no_ext);
 	}
-	$length_of_command_no_args = length($c);
     }
 
     my $number_of_args = 0;
     # max number of lines (-L) =
     # number_of_read_lines = 0
     while (defined($next_arg = get_next_arg())) {
-	# if defined max_number_of_lines {
+	my $next_arg_no_ext = $next_arg;
+	$next_arg_no_ext =~ s/\.[^\.]*$//; # Remove .ext from argument
+	# if defined max_number_of_lines
 	# number_of_read_lines++
-	# if $next_arg =~ /\w$/ number_of_read_lines-- (Trailing blanks cause an
+	# if $next_arg =~ /\w$/ then number_of_read_lines--
-	    # input line to be logically continued on the next input line.)
+	    # Trailing blanks cause an
+	    # input line to be logically continued on the next input line.
 	# if number_of_read_lines > max_number_of_lines
 	    # last
 	push (@quoted_args, $next_arg);
+	push (@quoted_args_no_ext, $next_arg_no_ext);
 	$number_of_args++;
 	if(not $Global::xargs and not $Global::Xargs) {
 	    last;
 	} else {
 	    # Emulate xargs if there is a command and -x or -X is set
-	    $arg_length += $number_of_substitution * (1 + length ($next_arg)) 
+	    my $next_arg_len = $number_of_substitution * (length ($next_arg) + $spaces) +
+		+ $number_of_substitution_no_ext * (length ($next_arg_no_ext) + $spaces)
 		+ $length_of_context;
+
+	    $arg_length += $next_arg_len;
 	    # debug("arglen $arg_length = $number_of_substitution * (1 + length ($next_arg)) + $length_of_context\n");
-	    my $job_line_length = $length_of_command_no_args + 1 + $arg_length;
+	    my $job_line_length = $length_of_command_no_args + $arg_length;
 	    # debug("linelen $job_line_length = $length_of_command_no_args + 1 + $arg_length\n");
+#	    print STDERR "1234567890123456789012345678901234567890\n";
+	    #print STDERR "LENcalc $number_of_args CON$length_of_context $length_of_command_no_args ".length ($next_arg)." LL$job_line_length NAL$next_arg_len ",$job_line_length-$next_arg_len-1,"\n";
 	    if($job_line_length >= max_length_of_command_line()) {
 		unget_arg(pop @quoted_args);
-		if($quoted_args[0]) {
+		if(defined $quoted_args[0]) {
 		    last;
 		} else {
-		    die ("Command line too long at: $next_arg");
+		    die ("Command line too long ($job_line_length >= "
+			 . max_length_of_command_line() . ") at number $number_of_args: $next_arg");
 		}
 	    }
 	    if($Global::max_number_of_args and $number_of_args >= $Global::max_number_of_args) {
@@ -1036,27 +1068,18 @@ sub generate_command_line {
     }
     if(@quoted_args) {
 	$job_line = $command;
-	if(defined $job_line and $job_line =~/\Q$Global::replacestring\E/o) {
+	if(defined $job_line and
-	    # substitute {} with args
+	   ($job_line =~/\Q$Global::replacestring\E/o or $job_line =~/\Q$Global::replace_no_ext\E/o)) {
+	    # substitute {} and {.} with args
 	    if($Global::Xargs) {
 		# Context sensitive replace (foo{}bar with fooargsbar)
-		while($job_line =~/\Q$Global::replacestring\E/o) {
+		$job_line = context_replace($job_line, \@quoted_args, \@quoted_args_no_ext);
-		    $job_line =~ /(\S*\Q$Global::replacestring\E\S*)/ or die ("This should never happen"); 
-		    my $wordarg = $1;
-		    my @all_word_arg;
-		    for my $arg (@quoted_args) {
-			my $substituted = $wordarg;
-			$substituted=~s/\Q$Global::replacestring\E/$arg/go;
-			push @all_word_arg, $substituted;
-		    }
-		    my $all_word_arg = join(" ",@all_word_arg);
-		    my ($quoted_wordarg) = shell_quote($wordarg);
-		    $job_line =~ s/$quoted_wordarg/$all_word_arg/;
-		}
 	    } else {
-		# Normal replace {} with args
+		# Normal replace {} with args and {.} with args without extension
 		my $arg=join(" ",@quoted_args);
+		my $arg_no_ext=join(" ",@quoted_args_no_ext);
 		$job_line =~ s/\Q$Global::replacestring\E/$arg/go;
+		$job_line =~ s/\Q$Global::replace_no_ext\E/$arg_no_ext/go;
 	    }
 	} else {
 	    # append args
@@ -1085,6 +1108,28 @@ sub shell_quote {
     return (@strings);
 }
 
+# Replace foo{}bar or foo{.}bar
+sub context_replace {
+    my ($job_line,$quoted,$no_ext) = (@_);
+    while($job_line =~/\Q$Global::replacestring\E|\Q$Global::replace_no_ext\E/o) {
+	$job_line =~ /(\S*(\Q$Global::replacestring\E|\Q$Global::replace_no_ext\E)\S*)/o
+	    or die ("This should never happen");
+	my $wordarg = $1; # This is the context that needs to be substituted
+	my @all_word_arg;
+	for my $n (0 .. $#$quoted) {
+	    my $arg = $quoted->[$n];
+	    my $arg_no_ext = $no_ext->[$n];
+	    my $substituted = $wordarg;
+	    $substituted=~s/\Q$Global::replacestring\E/$arg/go;
+	    $substituted=~s/\Q$Global::replace_no_ext\E/$arg_no_ext/go;
+	    push @all_word_arg, $substituted;
+	}
+	my $all_word_arg = join(" ",@all_word_arg);
+	$job_line =~ s/\Q$wordarg\E/$all_word_arg/;
+    }
+    return $job_line;
+}
+
 #
 # Number of processes, filehandles, max length of command line
 #
@@ -1484,6 +1529,7 @@ sub start_job {
     }
     $Global::running_jobs++;
     debug("$Global::running_jobs processes. Starting: $command\n");
+    #print STDERR "LEN".length($command)."\n";
     $pid = open3(gensym, ">&STDOUT", ">&STDERR", $command) || 
 	die("open3 failed. Report a bug to <parallel\@tange.dk>\n");
     debug("started: $command\n");

parallel.1

@@ -155,8 +155,8 @@ If \fBcommand\fR is given, \fBparallel\fR will behave similar to \fBxargs\fR. If
 Input line. This is the default replacement string and will normally
 be used for putting the argument in the command line. It can be
 changed with \fB\-I\fR.
-.IP "\fB{.}\fR (not implemented)" 9
+.IP "\fB{.}\fR" 9
-.IX Item "{.} (not implemented)"
+.IX Item "{.}"
 Input line without extension. This is a specialized replacement string
 with the extension removed. It will remove from the last \fB.\fR till the
 end of line of each input line and replace {.} with the
@@ -219,8 +219,12 @@ printed when the command is finished. \s-1STDERR\s0 first followed by \s-1STDOUT
 .IP "\fB\-I\fR \fIstring\fR" 9
 .IX Item "-I string"
 Use the replacement string \fIstring\fR instead of {}.
-.IP "\fB\-U\fR \fIstring\fR (not implemented)" 9
+.IP "\fB\-U\fR \fIstring\fR" 9
-.IX Item "-U string (not implemented)"
+.IX Item "-U string"
+.PD 0
+.IP "\fB\-\-extensionreplace\fR \fIstring\fR" 9
+.IX Item "--extensionreplace string"
+.PD
 Use the replacement string \fIstring\fR instead of {.} for input line without extension.
 .IP "\fB\-\-jobs\fR \fIN\fR" 9
 .IX Item "--jobs N"
@@ -896,7 +900,7 @@ hard, as all foreground processes get the \s-1INT\s0 from the shell.
 If there are nomore jobs (\s-1STDIN\s0 is closed) then make sure to
 distribute the arguments evenly if running \-X.
 .PP
-Distibute jobs to computers with different speeds/number\-of\-cpu\-cores using ssh
+Distribute jobs to computers with different speeds/number\-of\-cpu\-cores using ssh
 ask the computers how many cpus they have and spawn appropriately
 according to \-j setting. Reuse ssh connection (\-M and \-S)
 http://www.semicomplete.com/blog/geekery/distributed\-xargs.html?source=rss20
@@ -904,7 +908,7 @@ http://code.google.com/p/ppss/wiki/Manual2
 .PP
 http://www.gnu.org/software/pexec/
 .PP
-Where will '>' be run? Local or remote? Where ever is easier.
+Where will '>' be run? Local or remote? Remote.
 .PP
 Parallelize so this can be done:
 mdm.screen find dir \-execdir mdm-run cmd {} \e;

unittest/actual-results/test10

@@ -4,9 +4,9 @@ b35d8e49be8d94899b719c40d3f1f4bb  -
       3   60000  348894
 1foo bar2foo bar3 Afoo barBfoo barC
 1foo2foo3 1bar2bar3 AfooBfooC AbarBbarC
-51736abdee4738369ce04b354d40c887  -
+31d9274be5fdc2de59487cb05ba57776  -
       6  119994  697800
-51736abdee4738369ce04b354d40c887  -
+31d9274be5fdc2de59487cb05ba57776  -
 Chars per line: 116300
 'a' 
 'a'

unittest/actual-results/test14

@@ -73,7 +73,7 @@ a7 b1 2 3 4 5 6 7
 a8 b1 2 3 4 5 6 7 8
 a9 b1 2 3 4 5 6 7 8 9
 a10 b1 2 3 4 5 6 7 8 9 10
-51736abdee4738369ce04b354d40c887  -
+31d9274be5fdc2de59487cb05ba57776  -
-Chars per line: 116300
+Chars per line (697800/6): 116300
-1471045299517233a8dc29b1c3227f2e  -
+22074f9acada52462defb18ba912d744  -
-Chars per line: 102223
+Chars per line (817788/7): 116826

unittest/actual-results/test15

@@ -53,26 +53,26 @@ line 1 line 1
 line 2
 line 1 line 1
 line 2
-1
+max proc 1
-2
+max proc 2
-3
+max proc 3
-4
+max proc 4
-5
+max proc 5
-6
+max proc 6
-7
+max proc 7
-8
+max proc 8
-9
+max proc 9
-10
+max proc 10
-1
+200% proc 1
-2
+200% proc 2
-3
+200% proc 3
-4
+200% proc 4
-5
+200% proc 5
-6
+200% proc 6
-7
+200% proc 7
-8
+200% proc 8
-9
+200% proc 9
-10
+200% proc 10
 This is line 1
 This is line 2
 This is line 3
@@ -81,10 +81,10 @@ This is line 1
 This is line 2
 This is line 3
 
-line 1
+delimiter NUL line 1
 line 2
 line 3
-line 1
+delimiter TAB line 1
 line 2
 line 3
 line 1 line 1

unittest/tests-to-run/test14.sh

@@ -9,14 +9,14 @@ seq 1 10 | parallel -k 'seq 1 {} | parallel -m -k -I :: echo a{} b::'
 
 seq 1 60000 | parallel -I :: -m echo a::b::c | \
   mop -q "|sort |md5sum" :parallel
-echo -n "Chars per line: "
 CHAR=$(cat ~/.mop/:parallel | wc -c)
 LINES=$(cat ~/.mop/:parallel | wc -l)
+echo -n "Chars per line ($CHAR/$LINES): "
 echo "$CHAR/$LINES" | bc
 
 seq 1 60000 | parallel -I :: -X echo a::b::c | \
   mop -q "|sort |md5sum" :parallel
-echo -n "Chars per line: "
 CHAR=$(cat ~/.mop/:parallel | wc -c)
 LINES=$(cat ~/.mop/:parallel | wc -l)
+echo -n "Chars per line ($CHAR/$LINES): "
 echo "$CHAR/$LINES" | bc

unittest/tests-to-run/test15.sh

@@ -40,15 +40,15 @@ parallel --arg-file /tmp/$$ echo
 (echo line 1;echo line 1;echo line 2) | parallel -k --max-args 2 -X echo
 
 # Test --max-procs and -P: Number of processes
-seq 1 10 | parallel -k --max-procs +0 echo 
+seq 1 10 | parallel -k --max-procs +0 echo max proc
-seq 1 10 | parallel -k -P 200% echo 
+seq 1 10 | parallel -k -P 200% echo 200% proc
 
 # Test --delimiter and -d: Delimiter instead of newline
 # Yes there is supposed to be an extra newline for -d N
 echo line 1Nline 2Nline 3 | parallel -k -d N echo This is
 echo line 1Nline 2Nline 3 | parallel -k --delimiter N echo This is
-printf "line 1\0line 2\0line 3" | parallel -d '\0' echo
+printf "delimiter NUL line 1\0line 2\0line 3" | parallel -k -d '\0' echo
-printf "line 1\tline 2\tline 3" | parallel --delimiter '\t' echo
+printf "delimiter TAB line 1\tline 2\tline 3" | parallel -k --delimiter '\t' echo
 
 # Test --max-chars and -s: Max number of chars in a line
 (echo line 1;echo line 1;echo line 2) | parallel -k --max-chars 25 -X echo

unittest/wanted-results/test10

@@ -4,9 +4,9 @@ b35d8e49be8d94899b719c40d3f1f4bb  -
       3   60000  348894
 1foo bar2foo bar3 Afoo barBfoo barC
 1foo2foo3 1bar2bar3 AfooBfooC AbarBbarC
-51736abdee4738369ce04b354d40c887  -
+31d9274be5fdc2de59487cb05ba57776  -
       6  119994  697800
-51736abdee4738369ce04b354d40c887  -
+31d9274be5fdc2de59487cb05ba57776  -
 Chars per line: 116300
 'a' 
 'a'

unittest/wanted-results/test14

@@ -73,7 +73,7 @@ a7 b1 2 3 4 5 6 7
 a8 b1 2 3 4 5 6 7 8
 a9 b1 2 3 4 5 6 7 8 9
 a10 b1 2 3 4 5 6 7 8 9 10
-51736abdee4738369ce04b354d40c887  -
+31d9274be5fdc2de59487cb05ba57776  -
-Chars per line: 116300
+Chars per line (697800/6): 116300
-1471045299517233a8dc29b1c3227f2e  -
+22074f9acada52462defb18ba912d744  -
-Chars per line: 102223
+Chars per line (817788/7): 116826

unittest/wanted-results/test15

@@ -53,26 +53,26 @@ line 1 line 1
 line 2
 line 1 line 1
 line 2
-1
+max proc 1
-2
+max proc 2
-3
+max proc 3
-4
+max proc 4
-5
+max proc 5
-6
+max proc 6
-7
+max proc 7
-8
+max proc 8
-9
+max proc 9
-10
+max proc 10
-1
+200% proc 1
-2
+200% proc 2
-3
+200% proc 3
-4
+200% proc 4
-5
+200% proc 5
-6
+200% proc 6
-7
+200% proc 7
-8
+200% proc 8
-9
+200% proc 9
-10
+200% proc 10
 This is line 1
 This is line 2
 This is line 3
@@ -81,10 +81,10 @@ This is line 1
 This is line 2
 This is line 3
 
-line 1
+delimiter NUL line 1
 line 2
 line 3
-line 1
+delimiter TAB line 1
 line 2
 line 3
 line 1 line 1