gfx
|
[. index]
[file:/home/kramm/c/swftools/wx/gfx.so /home/kramm/c/swftools/wx/gfx.so]
|
This module contains a PDF parser (based on xpdf) and a number of rendering backends. In particular, it can extract text from PDF pages, create bitmaps from them, or convert PDF files to SWF.The latter functionality is similar to what is offered by pdf2swf utility, however more powerful.You can also create individual SWF files from single pages of the PDF
or mix pages from different PDF files.</tt>
Classes
|
|
|
|
- [__builtin__.html#object __builtin__.object]
- ;; [gfx.html#Doc Doc]
- [gfx.html#Output Output]
- [gfx.html#Page Page]
class Doc([__builtin__.html#object __builtin__.object])
|
|
|
A [#Doc Doc] [__builtin__.html#object object] is used for storing a document (like a PDF).
doc.pages contains the number of pages in the document,
and doc.filename the name of the file the document was
created (loaded) from. If the document was created from
an image file, the number of pages is always 1
|
|
|
Methods defined here:
- getInfo(...)
- [#Doc-getInfo getInfo](key)
Retrieve some information about a document. For PDF files, key
can have the following values:
"title", "subject", "keywords", "author", "creator", "producer",
"creationdate", "moddate", "linearized", "tagged", "encrypted",
"oktoprint", "oktocopy", "oktochange", "oktoaddnotes", "version".
If the "oktocopy" digital rights management flag is set to "no", then the
pdf parser won't allow you to access the PDF file. Trying to extract pages
from it will raise an exception.
- getPage(...)
- [#Doc-getPage getPage](nr)
Get one page from a document file. The nr parameter specifies
which page to retrieve. Counting starts at 1, so the first page
can be retrieved by
page = doc.[#Doc-getPage getPage](1)
.
You can find out how many pages a document contains by querying
its pages field (doc.pages)
- setparameter(...)
- [#Doc-setparameter setparameter](key, value)
Pass a parameter or setting to the document parser. Unlike
the module level [#Doc-setparameter setparameter]() function, the parameters set
using setparameter will only be valid for the [__builtin__.html#object object] itself
during its lifetime.
|
class Output([__builtin__.html#object __builtin__.object])
|
|
|
An [#Output Output] [__builtin__.html#object object] can be used as parameter to the render()
call of a page. It's not possible to create this type of
[__builtin__.html#object object] directly (i.e., from a class), however you can
use a [#-PassThrough PassThrough]() device to pass things over to Python.
Examples for classes implementing the [#Output Output] class are:
ImageList, SWF, PlainText and PassThrough.
|
|
|
Methods defined here:
- endpage(...)
- [#Output-endpage endpage]()
Ends a page in the output device. This function should be called
once for every [#Output-startpage startpage]()
- fill(...)
- [#Output-fill fill]()
fill a polygon with a color
- fillbitmap(...)
- [#Output-fillbitmap fillbitmap]()
fill a polygon with a bitmap pattern
- save(...)
- [#Output-save save](filename)
Saves the contents of an output device to a file
Depending on what the output device is, the contents
of the file may be plain text, an image, an SWF file,
etc.
For the ImageList device, several files (named
filename.1.png, filename.2.png etc.) might be created)
- setparameter(...)
- [#Output-setparameter setparameter](key, value)
Set a output-device dependent parameter
- startpage(...)
- [#Output-startpage startpage](width, height)
Starts a new page/frame in the output device.
The usual way to render documents is to start a new page in the
device for each page in the document:
for pagenr in range(1,doc.pages+1):
page = doc.getPage(pagenr)
output.[#Output-startpage startpage](page.width, page.height)
page.render(output)
output.[#Output-endpage endpage]()
It is, however, also possible to render more than one document page
to a single output page. E.g. for side-by-side or book views.
- stroke(...)
- [#Output-stroke stroke]()
stroke a polygon with a color
|
class Page([__builtin__.html#object __builtin__.object])
|
|
|
A [#Page Page] [__builtin__.html#object object] contains a single page of a document.
page.width and page.height (or page.size) contain the
page dimensions. page.nr is the number of the page, and
page.doc is the parent document.
|
|
|
Methods defined here:
- asImage(...)
- [#Page-asImage asImage](width, height)
Creates a bitmap from a page. The bitmap will be returned as a string
containing RGB triplets. The bitmap will be rescaled to the specified width and
height. The aspect ratio of width and height doesn't need to be the same
as the page.
- render(...)
- [#Page-render render](output, move=(0,0), clip=None)
Renders a page to the rendering backend specified by the output
parameter. Rendering consists of calling a number of functions on the
output device, see the description of the "PassThrough" device.
The page may be shifted to a given position using the move parameter,
and may also be clipped to a specific size using the clip parameter.
The clipping operation is applied after the move operation.
|
|
Functions
|
|
|
|
- ImageList(...)
- [#-ImageList ImageList]()
Creates a device which renders documents to bitmaps.
Each page that is rendered will create new bitmap.
Using save(), you can save the images to a number
of files
- OCR(...)
- [#-OCR OCR]()
Creates a device which processes documents using OCR (optical
character recognition).
This is handy for e.g. extracting fulltext from PDF documents
which have broken fonts, and where hence the "PlainText"
device doesn't work.
- OpenGL(...)
- [#-OpenGL OpenGL]()
Creates a device which renders everything to OpenGL.
Can be used for desktop display and debugging.
This device is not available on all systems.
- PassThrough(...)
- [#-PassThrough PassThrough](device)
Creates a PassThrough device, which can be used as parameter in calls
to page.render().
device needs to be a class implementing at least the following functions:
[#-setparameter setparameter](key,value)
startclip(outline)
endclip()
stroke(outline, width, color, capstyle, jointstyle, miterLimit)
fill(outline, color)
fillbitmap(outline, image, matrix, colortransform)
fillgradient(outline, gradient, gradienttype, matrix)
[#-addfont addfont](font)
drawchar(font, glyph, color, matrix)
drawlink(outline, url)
If any of these functions are not defined, a error message will be printed,
however the rendering process will *not* be aborted.
- PlainText(...)
- [#-PlainText PlainText]()
Creates a device which can be used to extract text from documents,
by passing it as parameter to page.render().
The extracted text can be saved by plaintext.save(filename).
- SWF(...)
- [#-SWF SWF]()
Creates a device which renders documents to SWF (Flash) files.
Depending on the way the document parser behaves (see the poly2bitmap
and bitmap parameters), the resulting SWF might use vector operations
and Flash Texts to display the document, or just a single bitmap.
- addfont(...)
- [#-addfont addfont](filename)
Passes an additional font file to the PDF parser. If a PDF contains
external fonts (i.e. fonts which are not contained in the PDF itself)
then the files added by [#-addfont addfont]() will be searched.
- addfontdir(...)
- [#-addfontdir addfontdir](dirname)
Passes a complete directory containing fonts to the PDF parser. Any
font file within this directory might be used to resolve external fonts
in PDF files
- open(...)
- [#-open open](type, filename) -> [__builtin__.html#object object]
Open a PDF, SWF or image file. The type argument should be "pdf",
"swf" or "image" accordingly. It returns a doc [__builtin__.html#object object] which can be
used to process the file contents.
E.g.
doc = [#-open open]("pdf", "document.pdf")
doc = [#-open open]("swf", "flashfile.swf")
doc = [#-open open]("image", "image.png")
If the file could not be loaded, or is a encrypted PDF file without
a proper password specified, an exception is being raised.
If the filename argument contains a '|' char, everything behind
the '|' is treated as password used for opening the file.
E.g.
doc = [#-open open]("pdf", "document.pdf|mysecretpassword")
.
Notice that for image files, the only supported file formats right now
are jpeg and png.
- setparameter(...)
- [#-setparameter setparameter](key,value)
Set a parameter in the gfx module (which might affect the PDF
parser or any of the rendering backends). This is a parameter
which would usually be passed with the "-s" option to pdf2swf.
For a list of all parameters, see the output of
pdf2swf -s help
and
pdf2swf somefile.pdf -s help
.
- verbose(...)
- [#-verbose verbose](level)
Set the logging verbosity of the gfx module. Log levels are:
level=-1 Log nothing
level=0 (fatal) Log only fatal errors
level=1 (error) Log only fatal errors and errors
level=2 (warn) Log all errors and warnings
level=3 (notice) Log also some rudimentary data about the parsing/conversion
level=4 (verbose) Log some additional parsing information
level=5 (debug) Log debug statements
level=6 (trace) Log extended debug statements
All logging messages are written to stdout.
|
