Today I decided to learn more about the Template-Toolkit. I am the kind of person that can learn the first steps better when I read things on paper. Reading from screen is fine for me for looking up a reference, but reading long texts is a pain. Regardless if it is a CRT or a TFT display.
The docs of Template Toolkit are, as of many Perl modules, splitted in many small chunks, which is also fine for reference, but a pain to print out.
So I decided to put the stuff together in a Word (/me ducks) file and make a pdf out of it, so i can share it with people. I planned to make it in three parts.
- The Manual
- The Tutorial
- The Reference
The manual and the tutorial consist of a few documents so i did them by hand. I copied the relevant parts from the HTML-doc and inserted them into Word, changed headings, inserted pagebreaks, did some pageformatting and printed it to a pdf.
So far so good. But when I came to the reference I realized that there are far too many files to do it conviniently by hand. Solution: script it.
Here is the script I used for the issue. It uses regexes to deal with the html, but since Active States HTML-Docs are computer generated I think that is ok in this case :-D
my @f =
open OUT, ">ref.html";
print OUT <<"HTML";
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w
<title>Template::Config - Factory module for instantiating other TT2 m
<link rel="stylesheet" href="Active.css" type="text/css" />
<link rev="made" href="mailto:" />
<table border="0" width="100%" cellspacing="0" cellpadding="3">
<tr><td class="block" valign="middle">
<big><strong><span class="block"> Template::Config - Factory modu
+le for instantiating other TT2 modules</span></strong></big>
for my $f ( @f )
open IN, $f or die "$f: $!";
while ( <IN> )
$i=1 if /<!-- INDEX END -->/;
$i=0 if /<h1><a name="authors?">/;
#tweak the NAME heading
my $ff = $f;
$ff =~ s/^.+Template/Template/;
$ff =~ s/\.html//;
$ff =~ s/\//::/g;
s/<h1><a name="name">NAME<\/a><\/h1>/[[[PB!]]]<h0><a name="nam
#tweak headings one step smaller
print OUT $_ if $i;
print OUT "</body></html>";
That worked fine. All I had to do manually was replacing the [[[PB!]]]
parts with pagebreaks using the "search and replace" functionality of Word, formatting the pages and adding a autogenerated Table of Contents. Et voila, a full fledged book. :-)
As you may have noticed, the html relies on a external stylesheet, so you can even tweak the formatting to you your liking.
I will put the six Files (TT-Manual-Single, TT-Manual-Double, TT-Tutorial-Single, TT-Tutorial-Double, TT-Reference-Single and TT-Reference-Double) on my webspace at holli.perlmonk.org
as soon as I get home, since I cannot ssh from work.
The difference between "Double" and "Single" here is that one is optimized for double paged print, the other for single page print.Update:
The files are now available at http://holli.perlmonk.org/TTdocs
. There is a 20MB transfer limit per day on jcwren
's server, so please be kind if it does not work for you. Just try the other day.