From: Don Armstrong <don@donarmstrong.com>
Date: Tue, 27 Nov 2012 02:03:11 +0000 (+0000)
Subject: autodetect r mode
X-Git-Url: https://git.donarmstrong.com/?a=commitdiff_plain;h=a0f50a48f47db3337b33310770e0e3d162398708;p=bin.git

autodetect r mode
---

diff --git a/txt2xls b/txt2xls
index 3ede2f4..147ace9 100755
--- a/txt2xls
+++ b/txt2xls
@@ -61,7 +61,6 @@ use Spreadsheet::WriteExcel;
 my %options = (debug           => 0,
 	       help            => 0,
 	       man             => 0,
-	       rmode           => 1,
 	       remove_name     => [],
 	       );
 
@@ -88,8 +87,6 @@ if (1 < grep {exists $options{$_}} qw(tsv ssv csv)) {
 
 pod2usage(join("\n",@USAGE_ERRORS)) if @USAGE_ERRORS;
 
-my @columns = ('A'..'Z','AA'..'ZZ');
-
 if (not @ARGV) {
     # we'll use this as a special indicator to read stdin
     push @ARGV,undef;
@@ -129,14 +126,29 @@ for my $file (@ARGV) {
     my $row = 1;
     my @header_row;
     my $overflow = 0;
+    my $r_mode = $options{r_mode} // 0;
+    # set to 1 if we have attempted to autodetect R mode
+    my $r_mode_autodetected = 0;
     while (<$fh>) {
 	chomp;
 	# parse the line
-	my @row;
 	die "Unable to parse line $. of $file" unless $csv->parse($_);
-	if ($row==1 and $options{rmode}) {
-	    # R doesn't output headers for rownames
-	    push @row,'';
+	my @row = $csv->fields();
+	if ($row==1 and not $r_mode_autodetected) {
+	    @header_row = @row;
+	    $row++;
+	    next;
+	}
+	if ($row==2 and not $r_mode_autodetected) {
+	    $r_mode_autodetected = 1;
+	    if (@row == (@header_row+1)) {
+		$r_mode = 1 unless exists $options{r_mode} and defined $options{r_mode};
+	    }
+	    if ($r_mode) {
+		# R doesn't output headers for rownames
+		unshift @header_row,'';
+	    }
+	    output_row(\@header_row,1,$ws);
 	}
 	if ($row > 65536) { # ok, we're going to overflow here
 	    my $t_ws_name = $ws_name;
@@ -144,20 +156,23 @@ for my $file (@ARGV) {
 	    $t_ws_name =~ s{^(.{0,$maxlen}).*$}{$1};
 	    $ws = $wb->add_worksheet($ws_name.'.'.$overflow);
 	    $row=1;
-	    for my $i (0..@header_row) {
-		$ws->write($columns[$i].$row,$header_row[$i]);
-	    }
+	    output_row(\@header_row,$row,$ws);
 	    $row++;
 	}
-	push @row,$csv->fields();
 	if ($row==1) {
 	    @header_row = @row;
 	}
-	for my $i (0..$#row) {
-	    $ws->write($columns[$i].$row,$row[$i]);
-	}
+	output_row(\@row,$row,$ws);
 	$row++;
     }
 }
 
+sub output_row{
+    my ($data,$row,$ws) = @_;
+    my @columns = ('A'..'Z','AA'..'ZZ');
+    for my $i (0..$#{$data}) {
+	$ws->write($columns[$i].$row,$data->[$i]);
+    }
+}
+
 __END__