pisa 3.0.32
XHTML/HTML/CSS to PDF converter
(C)opyright by Dirk Holtwick, Germany
dirk.holtwick@gmail.com
http://www.xhtml2pdf.com
pisa is a HTML/XHTML/CSS to PDF converter written in Python and based on Reportlab Toolkit, pyPDF, TechGame Networks CSS Library and HTML5lib. The primary focus is not on generating perfect printable webpages but to use HTML and CSS as commonly known tools to generate PDF files within Applications. For example generating documentations (like this one), generating invoices or other office documents etc.
As pisa is a Python pakage an installed version of Python <http://www.python.org> is needed. For the moment Python 2.3 to 2.5 is supported. For Python 3000 a special version will be needed, because it is not compatible with the 2.x series. A proper version will be made available as soon as Python 3000 becomes stable.
The easiest way to install pisa is to use easy_install:
$ easy_install pisa
But you may also download the source code of pisa, then enter the main directory and execute this command (on Linux and MacOS you may prepend a sudo command):
$ python setup.py install
pisa needs also some additional Python packages to be installed to work. Please follow the setup instruction for each package:
For Windows a precompiled version exists that includes Python and all needed libraries. The package contains the file xhtml2pdf.exe. Please add the directory where xhtml2pdf.exe is placed to the Windows PATH variable.
The Windows version is distributed via the Website <http://www.xhtml2pdf.com> in the "Download" section.
If you do not want to integrate pisa in your own application, you may use the command line tool that gives you a simple interface to the features of pisa. Just call xhtml2pdf --help to get the following help informations:
To generate a PDF from an HTML file called test.html call:
$ xhtml2pdf -s test.html
The resulting PDF will be called test.pdf (if this file is locked e.g. by the Adobe Reader it will be called test-0.pdf and so on). The -s option takes care that the PDF will be opened directly in the Operating Systems default viewer.
To convert more than one file you may use wildcard patterns like * and ?:
$ xhtml2pdf "test/test-*.html"
You may also directly access pages from the internet:
$ xhtml2pdf -s http://www.xhtml2pdf.com/
If the conversion doesn't work as expected some more informations may be usefull. You may turn on the output of warnings adding -w or even the debugging output by using -d.
Another reason could be, that the parsing failed. Consider trying the -xhtml and -html options. pisa uses the HTMLT5lib parser that offers two internal parsing modes: one for HTML and one for XHTML.
When generating the HTML output pisa uses an internal default CSS definition (otherwise all tags would appear with no diffences). To get an impression of how this one looks like start pisa like this:
$ xhtml2pdf --css-dump > xhtml2pdf-default.css
The CSS will be dumped into the file pisa-default.css. You may modify this or even take a totaly self defined one and hand it in by using the -css option, e.g.:
$ xhtml2pdf --css=xhtml2pdf-default.css test.html
XXX TO BE COMPLETED
The integration into a Python program is quite easy. We will start with a simple "Hello World" example:
import ho.pisa as pisa (1)
def helloWorld():
filename = __file__ + ".pdf" (2)
pdf = pisa.CreatePDF( (3)
"Hello <strong>World</strong>",
file(filename, "wb"))
if not pdf.err: (4)
pisa.startViewer(filename) (5)
if __name__=="__main__":
pisa.showLogging() (6)
helloWorld()
Comments:
(1) Import the pisa Python module
(2) Calculate a sample filename. If your demo is saved under test.py the filename will be test.py.pdf.
(3) The function CreatePDF is called with the source and the destination. In this case the source is a string and the destination is a fileobject. Other values will be discussed later (XXX to do!). An object will be returned as result and saved in pdf.
(4) The property pdf.err is checked to find out if errors occured
(5) If no errors occured a helper function will open a PDF Reader with the resulting file
(6) Errors and warnings are written as log entries by using the Python standard module logging. This helper enables printing warnings on the console.
The main function of pisa is called CreatePDF(). It offers the following arguments in this order:
String - or even better - a Unicode object.CreatePDF. (XXX allow file name?) String. If set to None the predefined CSS of pisa is used. Images, backgrounds and stylesheets are loaded form an HTML document. Normaly pisa expects these files to be found on the local drive. They may also be referenced relative to the original document. But the programmer might want to load form different kind of sources like the Internet via HTTP requests or from a database or anything else. Therefore you may define a link_callback that handles these reuests.
XXX
XXX
Some notes on some default values:
body, but you better leave the name empty for defining the default template (XXX May be changed in the future!) pisa supports a lot of Cascading Style Sheet (CSS). The following styles are supported:
background-color
border-bottom-color
border-bottom-style
border-bottom-width
border-left-color
border-left-style
border-left-width
border-right-color
border-right-style
border-right-width
border-top-color
border-top-style
border-top-width
color
display
font-family
font-size
font-style
font-weight
height
line-height
list-style-type
margin-bottom
margin-left
margin-right
margin-top
padding-bottom
padding-left
padding-right
padding-top
page-break-after
page-break-before
size
text-align
text-decoration
text-indent
vertical-align
white-space
width
zoom
And it adds some vendor specific styles:
-pdf-frame-border
-pdf-frame-break
-pdf-frame-content
-pdf-keep-with-next
-pdf-next-page
-pdf-outline
-pdf-outline-level
-pdf-outline-open
-pdf-page-break
Pages can be layouted by using some special CSS at-keywords and properties. All special properties start with -pdf- to mark them as vendor specific as defined by CSS 2.1. Layouts may be defined by page using the @page keyword. Then text flows in one or more frames which can be defined within the @page block by using @frame. Example:
@page {
@frame {
margin: 1cm;
}
}
In the example we define an unnamed page template - though it will be used as the default template - having one frame with 1cm margin to the page borders. The first frame of the page may also be defined within the @page block itself. See the equivalent example:
@page {
margin: 1cm;
}
To define more frames just add some more @frame blocks. You may use the following properties to define the dimensions of the frame:
marignmargin-topmargin-leftmargin-rightmargin-bottomtopleftrightbottomwidthheightHere is a more complex example:
@page lastPage {
top: 1cm;
left: 2cm;
right: 2cm;
height: 2cm;
@frame middle {
margin: 3cm;
}
@frame footer {
bottom: 2cm;
margin-left: 1cm;
margin-right: 1cm;
height: 1cm;
}
}
Layout scheme:
top
+--------------------------+ ---
| margin-top | /|\
| +---------------+ | |
| | | |
| | | | height
| | | |
By default the Frame uses the whole page and is defined to begin in the upper left corner and end in the lower right corner. Now you can add the position of the frame using top, left, bottom and right. If you now add height and you have a value other than zero in top the bottom will be modified. (XXX If you had not defined top but bottom the height will be ...)
A page layout may also define the page size and the orientation of the paper using the size property as defined in CSS 3. Here is an example defining page size "DIN A5" with "landscape" orientation (default orientation is "portrait"):
@page {
size: a5 landscape;
margin: 1cm;
}
Here is the complete list of valid page size identifiers:
a0 ... a6b0 ... b6letterlegalelevenseventeenFor the use of PDF backgrounds specify the source file in the background-image property, like this:
@page {
background-image: url(bg.pdf);
}
Some frames should be static like headers and footers that means they are on every page but do not change content. The only information that may change is the page number. Here is a simple example that show how to make an element named by ID the content of a static frame. In this case it is the ID footer.
<html>
<style>
@page {
margin: 1cm;
margin-bottom: 2.5cm;
@frame footer {
-pdf-frame-content: footerContent;
bottom: 2cm;
margin-left: 1cm;
margin-right: 1cm;
height: 1cm;
}
}
</style>
<body>
Some text
<div id="footerContent">
This is a footer on page #<pdf:pagenumber>
</div>
</body>
</html>
For better debugging you may want to add this property for each frame definition: -pdf-frame-border: 1. It will paint a border around the frame.
By default there is just a certain set of fonts available for PDF. Here is the complete list - and their repective alias names - pisa knows by default (the names are not case sensitive):
But you may also embed new font faces by using the @font-face keyword in CSS like this:
@font-face {
font-family: Example, "Example Font";
src: url(example.ttf);
}
The font-family property defines the names under which the embedded font will be known. src defines the place of the fonts source file. This can be a TrueType font or a Postscript font. The file name of the first has to end with .ttf the latter with one of .pfb or .afm. For Postscript font pass just one filename like <name>.afm or <name>.pfb, the missing one will be calculated automatically.
To define other shapes you may do like this:
/* Normal */
@font-face {
font-family: DejaMono;
src: url(font/DejaVuSansMono.ttf);
}
/* Bold */
@font-face {
font-family: DejaMono;
src: url(font/DejaVuSansMono-Bold.ttf);
font-weight: bold;
}
/* Italic */
@font-face {
font-family: DejaMono;
src: url(font/DejaVuSansMono-Oblique.ttf);
font-style: italic;
}
/* Bold and italic */
@font-face {
font-family: DejaMono;
src: url(font/DejaVuSansMono-BoldOblique.ttf);
font-weight: bold;
font-style: italic;
}
PDF supports outlines (Adobe calls them "bookmarks"). By default pisa defines the <h1> to <h6> tags to be shown in the outline. But you can specify exactly for every tag which outline behaviour it should have. Therefore you may want to use the following vendor specific styles:
-pdf-outline -pdf-outline-level-pdf-outline-openExample:
h1 {
-pdf-outline: true;
-pdf-level: 0;
-pdf-open: false;
}
It is possible to automatically generate a Table of Contents (TOC) with pisa. By default all headings from <h1> to <h6> will be inserted into that TOC. But you may change that behaviour by setting the CSS property -pdf-outline to true or false. To generate the TOC simply insert <pdf:toc /> into your document. You then may modify the look of it by defining styles for the pdf:toc tag and the classes pdftoc.pdftoclevel0 to pdftoc.pdftoclevel5. Here is a simple example for a nice looking CSS:
pdftoc {
color: #666;
}
pdftoc.pdftoclevel0 {
font-weight: bold;
margin-top: 0.5em;
}
pdftoc.pdftoclevel1 {
margin-left: 1em;
}
pdftoc.pdftoclevel2 {
margin-left: 2em;
font-style: italic;
}
Tables are supported but may behave a little different to the way you might expect them to do. These restriction are due to the underlying table mechanism of ReportLab.
Pisa is not able to split table cells that are larger than the available space. To work around it you may define what should happen in this case. The -pdf-keep-in-frame-mode can be one of: "error", "overflow", "shrink", "truncate", where "shrink" is the default value.
table {
-pdf-keep-in-frame-mode: shrink;
}
The table renderer is not able to adjust the width of the table automatically. Therefore you should explicitly set the width of the table and to the table rows or cells.
It is possible to repeat table rows if a page break occurs within a table. The number of repeated rows is passed in the attribute repeat. Example:
<table repeat="1"> <tr><th>Column 1</th><th>...</th></tr> ... </table>
Borders are supported. Use corresponding CSS styles.
By default JPG images are supported. If the Python Imaging Library (PIL) is installed the file types supported by it are available too. As mapping pixels to points is not trivial the images may appear bigger in the PDF as in the browser. To adjust this you may want to use the zoom style. Here is a small example:
img { zoom: 80%; }
Since Reportlab Toolkit does not yet support the use of images within paragraphs, images are always rendered in a seperate paragraph. Therefore floating is not available yet.
XXX TO BE WRITTEN
<pdf:barcode>
pisa provides some custom tags. They are all prefixed by the namespace identifier pdf:. As the HTML5 parser used by pisa does not know about these specific tags it may be confused if they are without a block. To avoid problems you may condsider sourrounding them by <div> tags, like this:
<div> <pdf:toc /> </div>
Creates a barcode.
Defines the template to be used on the next page.
Create a new page after this position.
Jump to next unused frame on the same page or to the first on a new page. You may not jump to a named frame.
Creates an object of a specific size.
Creates a Table of Contents.
pisa is copyrighted by Dirk Holtwick, Germany.
pisa is distributed by Dirk Holtwick, Schreiberstra�e 2, 47058 Duisburg, Germany.
pisa is licensed under the GNU Gerneral Public License version 2.
For commercial usage of pisa a developer license can be purchased!