[Catalyst] Which C::View::PDF should I use?

Yeah, they're both basically just wrappers..

PDF::Reuse is supposed to make it easier to reuse PDFs, whereas API2
is more powerful (and more cryptic :)..

They're both being used successfully by people as far as I know, so
pick whichever one appeals most :)

Cheers
t0m

We use simple View::TT and then a Filter:

[%- USE Latex; FILTER latex("pdf") -%]

Its perfekt!

Tomas Doran schrieb:

Good evening,

On 5/03/10 at 11:15 AM +0100, Felix Antonius Wilhelm Ostmann
wrote:
I like that conceptually, but just to be clear, that is latex
code being converted to PDF, correct? Due to time constraints I
can't learn another formatting language for this project. I may
choose to learn it for another project though (which does have
more complex PDF requirements).
So for now I'll follow t0m's advice and choose one of the
C::V::PDF modules.

Thanks to both of you for the feedback.


Charlie

Yes, you must write latex ... this do always my coworkers for me ;)

I am not a very experienced developer, but I was able to learn and use PDF::Reuse. I made a PDF template (11 page report) that I needed to intersperse data throughout. Using Catalyst::View::PDF::Reuse I was able to achieve a good result. I believe it is intended to be a reuse (of a PDF template) rather than a PDF creation tool, but it would work for that as well.  To use with a report was a little tougher as I had to be creative to do wrapping. It is probably best to use where the location to place values on a template remains consistent (invoice, certificate).

I use Debian stable, and there is a package of PDF::Reuse, libpdf-reuse-perl, that can be installed. Catalyst::View::PDF::Reuse can then be install with dh-make-perl.

S. Kiehn


--- On Fri, 3/5/10, Charlie Garrison wrote:

From: Charlie Garrison <garrison@zeta.org.au>
Subject: [Catalyst] Which C::View::PDF should I use?
To: "The elegant MVC web framework" <catalyst@lists.scsys.co.uk>
Date: Friday, March 5, 2010, 5:38 AM

Good afternoon,

Both of these seem to be almost identical.

    Catalyst::View::PDF::Reuse
    Catalyst::View::PDF::API2

Is one better to use than the other? C::V::PDF::API2 has been
updated more recently, but other than that I can't see why I would choose one over the other.

Or is really a choice of PDF::Reuse vs PDF::API2 which then dictates the View to use? It looks like PDF::Reuse might be a wrapper for PDF::API2, but I'm not really sure.

I'm not looking to create PDFs from code, I'll be using an existing PDF and replacing values. And both C::V::P::Reuse & C::V::P::API2 seem to handle that in effectively the same way.

Thanks,
Charlie

--    � Charlie Garrison ♊ <garrison@zeta.org.au>
   〠 PO Box 141, Windsor, NSW 2756, Australia

O< ascii ribbon campaign - stop html mail - www.asciiribbon.org
http://www.ietf.org/rfc/rfc1855.txt

_______________________________________________
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

Good afternoon,
And I apparently mis-read the POD before. I thought there was a
function to *replace* text in an existing PDF with values from
my app. (I should not have stared researching this so late in
the day.)

So how do people insert values into an existing PDF? Do you have
to determine X/Y position for each value to be added to the PDF?

I was expecting to be able to replace a unique string from the
PDF with a value from my app. For some reason I though
Catalyst::View::PDF::Reuse was doing more magic than it really is.

Is there a simpler approach (other than C::V::PDF::Reuse) for
creating PDFs with values from my app?

Once I hand this app to my customer, they need to be able to add
new fields to a dynamically generated form config. My plan was
to have a html 'template' version of the form with unique values
for each field, which the customer can print to PDF and then
re-use in the app.

Is there some way to determine the X/Y position of existing text
in the PDF at runtime?


On 5/03/10 at 11:15 AM +0100, Felix Antonius Wilhelm Ostmann
wrote:
Can anyone advise how long the learning curve is for latex (for
basic usage)? Would I be better off spending my time learning that?


Thanks,
Charlie

Depends how specific your layout requirements are and the kind of document you want to produce. LaTeX is pretty straightforward really, but you don't really have the fine grained control that other document formats have, so if you have exotic layout requirements it's going to be annoying.

http://stp.lingfil.uu.se/~zrajm/nerd/LaTeX/quickguide.html is a good enough start. You don't need to know all of that.

Good afternoon,
Thanks, that was a good intro. It looks like I could almost get
away with a simple LaTeX table for the form. But I can see
myself getting bogged down by the details; I think latex is
going to be "too hard" for this project. I may still come back
to it as best option though.

And it was good to get the LaTeX basics, I can start thinking
about how I can use it for the next project.

Thanks again for the info.

Charlie

If your needs are pretty simple there's HTMLDOC. I've used it for
simple reports and invoices.
http://www.easysw.com/htmldoc/

It doesn't sound like it would be right in this case (modifying
existing pdf) but it's worth knowing about.

Good morning,
Thanks for that. I've looked at various html 2 pdf tools (I used
html2ps for a project about 10 years ago). But lack of even
basic CSS support makes the them useless for creating "pretty" PDFs.

In reality, this project doesn't really need PDF output (HTML
would work fine), but $customer says it's a requirement, so I've
got to make it work for creating "pretty" PDFs.

Thanks again for all the input.

Charlie

Hi Charlie,
I've found that a simple way to put values into a PDF is to use PDF
forms - position the form fields in the appropriate location in the
template doc, and to generate the output update the form fields and
write a 'flattened' version of the doc. This means you can use visual
tools to layout the doc, and your code doesn't need to know anything
except the field names.

If you're considering non-perl tools, then iText does a pretty good
job of both creating PDF from scratch and updating PDF form fields,
and would be easy to wrap to call from a view.

Chris

Good morning,
Thanks for that tip. And I later realized thats where I got
confused with my first read-through of the docco (thinking I
could do a simple 'replace value'). For this project though,
it's not practical for someone to create a PDF 'form'. I've got
an action generating a blank html form which $customer can print
to pdf (hopefully they can at least do that, I'm still working
through it with them) which PDF::Reuse loads as the template.

I'm still working out how to best specify the X/Y coords for
each value though. I've got it all working; the hard part is
next, explaining it all to the customer.

Thanks,
Charlie

To: "The elegant MVC web framework" <catalyst@lists.scsys.co.uk>
Hi Charlie,

Do you know if there is a way of creating PDFs in the same way the HTML
templates are created, by just putting a variable name in a certain place
and render that pdf file by replacing that variable with its value? (Without
needing to specify the position where those variables will be placed)

Thanks.

Octavian


__________ Information from ESET NOD32 Antivirus, version of virus signature database 4923 (20100307) __________

The message was checked by ESET NOD32 Antivirus.

http://www.eset.com

Good evening,

On 8/03/10 at 8:41 AM +0200, Octavian Rasnita
wrote:
I don't. That's exactly what I was hoping to do, but I couldn't
find a way to do it (without expensive commercial solutions, or
without using hand-crafted PDF 'forms').

I've got a form config which $customer can change (add fields,
etc) and I've included X/Y coords as part of the config. That
was the only way I could get it working in limited time I've got.

I can share code with you if it will help (maybe offlist would
be better).


Charlie

Hi Charlie,

From: "Charlie Garrison" <garrison@zeta.org.au>
Good evening,

On 8/03/10 at 8:41 AM +0200, Octavian Rasnita
wrote:
I don't. That's exactly what I was hoping to do, but I couldn't
find a way to do it (without expensive commercial solutions, or
without using hand-crafted PDF 'forms').

I've got a form config which $customer can change (add fields,
etc) and I've included X/Y coords as part of the config. That
was the only way I could get it working in limited time I've got.

I can share code with you if it will help (maybe offlist would
be better).
Charlie


Unfortunately I don't know if it can help me, because I am blind and I can't
see if the manually configured positions are displayed correctly. That's why
I was hoping that there is a solution that works like the common TT
templating system.

Thank you.

Octavian





__________ Information from ESET NOD32 Antivirus, version of virus signature database 4923 (20100307) __________

The message was checked by ESET NOD32 Antivirus.

http://www.eset.com

Good evening,

On 8/03/10 at 10:27 AM +0200, Octavian Rasnita
wrote:
Wow, that certainly would make it more challenging. And I can't
think of any way around needing to see where the values get
printed. I'm having to position them through trial and error.
(My current solution feels *very* hackish.)

Sorry I can't offer anything more useful.


Charlie

Hi,

Am 08.03.2010 um 09:27 schrieb Octavian Rasnita:
there is a commercial solution called PDFlib PPS ( http://www.pdflib.com/products/pdflib-family/pps/ ) which may help you. With the PPS and the "PDFlib block plugin" you can define text and/or image areas in your PDF within Acrobat Professional. You can drag them everywhere in the PDF and set a name for each. In your Perl code you just write text or place an image into such a named area easliy.

I used the PDFlib in a previous job and it worked really nice and had lots of features! The bad thing is that it is a bit expensive (EUR 2390 / USD 3345) and a license is limited to one server/cpu. But you can download, develop and test the PDFlib with the PPS addon for free if you can accept having a stamp on each PDF you produce.

Regards,
matt

I've been using iText (http://itextpdf.com/) after using PDF::API2 for years. API2 hasn't grown and even once you know it, it's tricky to use. The community is inactive.

iText is Java and the community is vibrant. I'll have my Catalyst app assemble a whole bunch of data and ship it to iText as a JSON object (iText reads XML, too). Then I have a program I wrote in Java suck in the JSON and render the PDF. It's way faster than Perl and there's all sorts of handy features that make doc generation easier.

-steve

I confirm that iText works well. I've been using the C#/.NET version
which is API-identical to the Java version, and it's very good for
producing templated PDFs. Plenty of docs and examples too, plus the
'iText in Action' book.

- Chris

On our project, we dropped some cash coin to buy a licence for
PrinceXML <http://princexml.com/>. That thing is totally awesome,
you give it pretty much any old XHTML with CSS and it spits out
a PDF that looks the way you?d expect. It also supports a bunch
of custom CSS rules for even more control. It?s not cheap, but
for our purposes the price was completely worth it ? I spent some
time integrating it, but it has completely eliminated all further
work related to producing PDFs. (In contrast, our Excel exports
f.ex. have consistently consumed non-trivial amounts of work.)

As a rule I don?t like closed-source, binary stuff, but PrinceXML
has proven to be an exception.

Regards,

JJ Allen gave MiltonKeynes.pm a good talk on what he uses to generate PDF
invoices etc. for Penny's Arcade website
http://www.pennysarcade.co.uk/files/Creating_PDF_files_with_Catalyst.pdf

To use it

1. Install OpenOffice, use Writer to create a PDF form, export as PDF
http://danilop.wordpress.com/2008/07/02/how-to-create-pdf-forms-with-openoffice/

2. Use http://search.cpan.org/perldoc?Catalyst::View::PDF::Reuse to
mail-merge the PDF document to output


Regards, Peter
http://perl.dragonstaff.co.uk

Holy crap. After going through samples on their site, I'm very impressed. And trying to pass Acid2 in a PDF renderer? Wow.

Cheers,
Ovid

Yeah, it was one of the first renderers to pass the test suite,
beating out several of the big browsers. A magical piece of work.

Regards,

There's also an open-source solution based on Webkit rendering engine,
and if you have no excuses to use a binary you should give it a try.
I'm pretty sure it passes Acid2, does CSS3, Javascript and won't hurt
your pocket. : )

http://code.google.com/p/wkhtmltopdf/

Try it and see:
$ wkhtmltopdf http://www.nytimes.com/ nyt.pdf

And thanks to maluco++ we have a perl interface for it:

http://search.cpan.org/~tbr/WKHTMLTOPDF-0.02/lib/WKHTMLTOPDF.pm

Cheers

That looks interesting, thanks!


* Adam Sj?gren [2010-03-16 18:15]:
It requires an X11 server, so it isn?t.

The wkhtmltopdf doesn?t make it easy either: it requires
a patched Qt for that. But at least it?s possible at all.

Regards,

-Wade

And how is Xvfb (which is an X11 server) a solution the problem
of requiring an X11 server?

Regards,

It is a virtual frame buffer that allows x11 requiring apps to run
without a full display/head. What besides that is your issue around
the x11 requirement on servers?

On Mar 25, 2010, at 12:01 PM, Aristotle Pagaltzis <pagaltzis@gmx.de>
I've used it before to run unit tests from batch under Solaris where no
interactive X display is attached for code targeted to run using WxWidgets
on Windows.
Unless you actually need a display to look at it's useful to be able to run
libraries that expect a display.

Regards, Peter
http://perl.dragonstaff.co.uk

Either you start one per CutyCapt instance which can get quite
expensive quickly, or you have a new d?mon to manage. And either
way you need to install a metric ton of X11 crap on the server.

Regards,

An alternative could perhaps be CutyCapt:

* http://cutycapt.sourceforge.net/

which is based on WebKit, free software, and seems to generate quite
nice output.


Best regards,

Adam