in reply to Re: Re: Text::Wrap - having problems with long words
in thread Text::Wrap - having problems with long words

Like sauoq, I have tried but failed to reproduce the results you reported -- either of them -- using the four paragraphs you cited. I am assuming that given this text, @CONTENT should have 7 elements (maybe 8, at most 9, if you have blank lines at the beginning and/or end of the array): 
1: "This program ... details.\n"
2: "\n"
3: "You should have ... provide one.\n"
4: "\n"
5: "You should also ... /gpl.html.\n"
6: "\n"
7: "For those ... to use that.\n"

[download]
(update: added "\n" at the end of each text line, just to be clear about that; also fixed the summary of output below.) Given data like that, and using these lines of code: 
$Text::Wrap::columns = '40';
print wrap( '','',@CONTENT );

[download]
I get the following sequence of paragraphs (summarized): 
This program is distributed in the hope
... ## 5 lines ## ...
Artistic License for more details.

You should have received a copy of the
... ## 2 lines ## ...
glad to provide one.

You should also have received a copy of
... ## 6 lines ## ...
on the internet at
http://www.gnu.org/copyleft/hhhhhhhhhhh
hh/hhhhhhhhhhhhhhhhhh/gpl.html.

For those of you that choose to use the
... ## 37 lines ## ...
protection, so you may prefer to use
that.

[download]
So it looks like the URL is being rendered where it should be. Also, when I pipe the output through "od -a", I see that the paragraph boundaries (the blank lines) are rendered as two consecutive "nl" (\n) characters, with no extra spaces. (But frankly, if there were extra spaces, I don't know why this would be a problem.)

The only thing I would consider fixing is to have the URL be broken after a slash character, rather than in the middle of a "word" token, but that probably doesn't matter.

My Perl installation is v5.8.0 built for i686-linux, and I'm using the "2001.0929" version of Text::Wrap. What are you using, and how, exactly, are you filling the @CONTENT array?

Replies are listed 'Best First'.
RE: Text::Wrap - having problems with long words
by Hagbone (Monk) on Jan 18, 2003 at 14:51 UTC
    First, thanks for you efforts ... this really is puzzling.

    OK - I fooled around, and found an approach that eliminates the spaces, but not the "over the limit" word.

    FWIW, I'm using Perl 5.00503 (OS solaris), and don't know the Text::Wrap version (I've no root access, but could find out the version if it ends up being critical).

    When I read the file into a variable (rather then an array), the spaces do not appear, but the long URL still gets moved to the end of the output.

    If I use: 
    open (FILE,"<$script_path/$file") || die "NFG1 $script_path/$file!\n";
@CONTENT = <FILE>;
close(FILE);
$Text::Wrap::columns = '40';
print wrap("","",@CONTENT);

    [download]
    The spaces show up. If I use: 
    open (FILE,"<$script_path/$file") || die "NFG1 $script_path/$file!\n";
while (<FILE>){
    $CONTENT .= $_;
}
close(FILE);
$Text::Wrap::columns = '40';
print wrap("","",$CONTENT);

    [download]
    The spaces no longer appear, but again, the long URL is tossed in at the bottom. Here's how the output appears:

    This program is distributed in the hope
    ... ## 7 lines ## ...
    Artistic License for more details.

    You should have received a copy of the
    ... ## 4 lines THIS PARAGRAPH CONTAINS THE LONG URL ## ...
    glad to provide one.

    You should also have received a copy of
    ... ## 8 lines ## ...
    on the internet at

    For those of you that choose to use the
    ... ## 40 lines ## ...
    that.

    ##and tossed in at the end is the (split)long URL:

    http://www.gnu.org/copyleft/hhhhhhhhhhh
    hh/hhhhhhhhhhhhhhhhhh/gpl.html.

    Very strange, all suggestions welcome

[reply]
[d/l]
[select]

In Section Seekers of Perl Wisdom