Re: Paragraph and documentation

--- In [email protected], "spinningwebz2002"
<spinningwebz2002@...> wrote:
>
> --- In [email protected], "Mr. Shawn H. Corey" 
> <shawnhcorey@> wrote:
> >
> > spinningwebz2002 wrote:
> > > Thanks for the info. In your other message I also saw the posting 
> > > about bold etc. I will just have to write routines that would 
> process 
> > > the paragraph using text_center, left etc. which would give me 
> more 
> > > control anyways.
> > > 
> > > While working with pdf::api2, what variables do I have access to 
> (is 
> > > there a list)?
> > 
> > You will find that $text->text(...) is the workhorse of PDF::API2. 
> > Every routine that prints (via the text object) uses this routine.
> > 
> > I'm not sure what you mean by variables.  PDF::API2 is object-
> oriented. 
> >   It helps if you know something about Perl objects.  There are a 
> lot of 
> > parameters but few variables (and you shouldn't touch them unless 
> you 
> > really, really know what you're doing).
> > 
> > 
> > -- 
> > Just my 0.00000002 million dollars worth,
> >     Shawn
> > 
> > "For the things we have to learn before we can do them, we learn by 
> > doing them."
> >    Aristotle
> >
> 
> Probably Variables is the wrong word. What I was thinking, if I have 
> to write routines to use the text->text for bold, italic etc, then it 
> would be necessary to know where the line ended so that the next 
> block of text would start after it.
> 
> For the time being I am using text->paragraph and setting it up so it 
> is only one line long, this way I get what was overflowed back and 
> then create the next line below it until all text is done displaying. 
> I did this so I can catch when I reach the bottom of the page and add 
> a new page. It will also allow me to do things like wrap text around 
> a photo, have varying line width etc. 
> 
> Ideally what I think I may end up doing is rewriting the text-
> >paragraph as my own routine that would catch simple html markup 
> (<B><I><U><UL> etc) in a line and handle it automatically.
> 
> One thing I notice is that it takes pdf::api2 a bit to get started, 
> is this due to the use strict;? If so, when my code is done can I 
> remove that to speed up the process? Or, will the pdf::api2 run under 
> mod_perl?
> 
> Bill H
>

Some thoughts:
o Investigate HTML::Parser. I use the results to create two stacks:
font and text. Then, in the rendering phase, I pop those stacks in a
coordinated way to match font and text. This lets me embed bold or
italic font in a single paragraph. PDF can do this, the issue is
getting that to work w/ PDF::API2.

o text->paragraph is subject to random breakage. I'd recommend
insulating yourself from "upgrades" by placing a working version into
a local module that declares itself to be part of package
PDF::API2::Content::Text I've been using PDF::API2 for at least 5 yrs.
now. I came up with this solution when text->paragraph "upgraded" a
few years ago in a way that wasn't backwards compatible.

o I've found the following logic to be quite stable
($idt, $base, @overflow) =
$col->jec_paragraph($layout{margin_left}+$layout{gutter_width}
                    , $base
		    , $layout{width_paragraph}
   		    , &$linect($base,$col->lead)
		    , $idt
		    , $cargo);
while ($#overflow != -1)
{	
  if (&$linect($base, $col->lead) <= 4) # Check lines remaining
	{
	&$newpage($col);
	$base -= $layout{size_default};
	}
   ($idt, $base, @overflow) = 
       $col->jec_paragraph(
             $layout{margin_left}+$layout{gutter_width}
             , $base
	     , $layout{width_paragraph}
	     , &$linect($base,$col->lead)
	     , $idt
	     , join(' ', @overflow));
	} # end while handling overflow text
} # end while processing all items with associated commentary
$base -= $layout{size_default};

o I determine end of page by dividing the current base by the current
leading. If the value is less than lines remaining, I create a new
page: $linect = sub { return abs(ceil($_[0] / $_[1]))};