Conditionally converting tabs to spaces

Anonymous Monk has asked for the wisdom of the Perl Monks concerning the following question:

Hi all,
I have a problem with a text file which I am trying to strip all the tabs out of, and replace with their equivalent space (I know I could probably do this outside of Perl but I'm sure that there must be a way to do it and I not gonna give up till I find it.) The file looks like this :


{some random text} 9.99    9.999   9.999  9.999   9.9999   9.9999 
{more text}        9.99    *9.999  *9.999 9.999   9.9999   9.9999 
{even more text}   9.99    9.999   9.999  *9.999  *9.9999  *9.9999 

..this goes on for 12 or so columns..
[download]

Basically the first two columns, the text and three sig-fig number are fixed, but the following columns sometimes have a *preceding them, and sometimes don't.
All the table is at present tab-delimited.
<nbsp> So, the problem is, how do I strip out the '\t's?
If it weren't for the '*' I could simply use a "s/\t/ \1/g" or something of that nature. But having the *s means in where there is a *, I need one less space than where there isn't one.
So, I tried this:

perl -1 "while (<>) s/\t(\*w+)/      \1/g;s/\t\s(w+)/    \1/g;print}" 
+myfile
[download]

Which did - <bold> absolutely nothing </bold>. The tabs are all still there.
I know I'm probably being a complete retard, but is there an obvious way to do this that I've missed out on?

Thanks,
Mike

Comment on Conditionally converting tabs to spaces Select or Download Code

Replies are listed 'Best First'.
Re: Conditionally converting tabs to spaces by lhoward (Vicar) on Jul 03, 2000 at 22:07 UTC
how about this.... `while(<>){ s/\t/ /g; s/\s\/\/g; print; }` [download]	[reply] [d/l]
Re: Conditionally converting tabs to spaces by cwest (Friar) on Jul 03, 2000 at 21:43 UTC
Here's a solution... `my $tab = 8; while ( <DATA> ) { chomp; my $fixed = join '', map { $_ . ' 'x($tab-length) } split /\t/; print "$fixed\n"; } __DATA__ hello 57 45 78 there 57 93 83 dude 78 23 45` [download] Enjoy! Update: This code will not work of a field is longer than 8 chars, but the following will: ps. I don't think any of the other responses take that for account either, but if I'm wrong, I'll retract that. `my $fixed = join '', map { $_ . ' 'x(((int(length()/$tab)+1)*$tab)-len +gth) } split /\t/;` [download] -- Casey	[reply] [d/l] [select]
RE: Re: Conditionally converting tabs to spaces by chromatic (Archbishop) on Jul 04, 2000 at 03:19 UTC
I like this approach, but it's hard not to use sprintf in this case: `my $tab = 8; while (<DATA>) { chomp; my @fields = split(/\t/, $_); my $cooked = shift (@fields) . join '', sprintf("-$tab%s", @fields +); print $cooked, "\n"; }` [download] lhoward's looks more elegant, though. Make it use variable tabstops like so: `my $tab = 8; s/\t/" " x $tab/ego;` [download]	[reply] [d/l] [select]
Re: Conditionally converting tabs to spaces by nardo (Friar) on Jul 03, 2000 at 22:29 UTC
In addition to lhoward's response, if the text in the first column ends with a tab, you will need to pad the text with spaces so that everything lines up. `s/^([^\t]*)/sprintf('%-10s', $1)/e;` will pad it to 10 characters (and will leave the tab).	[reply] [d/l]