Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Showing content from http://pyopengl.sourceforge.net/documentation/pydoc/OpenGL.images.html below:

OpenGL.images

• Home
• Docs
• Install
• OpenGL

Image/texture implementation code

This module provides the Pan-OpenGL operations required to support OpenGL image handling. Most of this code is simply boilerplate code that sets OpenGL parameters such that normal Pythonic assumptions about data-ordering are met to allow easier interaction with other projects (such as PIL or Numpy).

Generally speaking, there are 3 pieces of information which control how an image is processed in the system:

format

this is the pixel format, such as GL_RGB/GL_RED/GL_ABGR_EXT

dims

tuple of dimensions for the image, (width,height,depth) order

type

the storage data-type for the image, normally GL_UNSIGNED_BYTE when working in Python, but all of the standard OpenGL types for images can be used if you happen to have your data in some exotic format.

OpenGL.UNSIGNED_BYTE_IMAGES_AS_STRING -- if this global value is set, then read of unsigned byte images using glReadPixels and glGetTexImage produces a string instead of the default array format.

Attributes of Note:

COMPONENT_COUNTS

used to lookup how many units of a given storage type are required to store a unit in a given format

TYPE_TO_ARRAYTYPE

maps Image storage types to their array data-type constants, i.e. maps GL_UNSIGNED_SHORT_4_4_4_4 to GL_UNSIGNED_SHORT so that we can use the standard array types for manipulating image arrays.

RANK_PACKINGS

commands required to set up default array-transfer operations for an array of the specified rank.

New image formats and types will need to be registered here to be supported, this means that extension modules which add image types/formats need to alter the tables described above!

XXX Should be an API to handle that instead of direct modification.

(

)

Create storage array for given parameters

If storage type requires > 1 unit per format pixel, then dims will be extended by 1, so in the common case of RGB and GL_UNSIGNED_BYTE you will wind up with an array of dims + (3,) dimensions. See COMPONENT_COUNTS for table which controls which formats produce larger dimensions. The secondary table TIGHT_PACK_FORMATS overrides this case, so that image formats registered as TIGHT_PACK_FORMATS only ever return a dims-shaped value. TIGHT_PACK_FORMATS will raise ValueErrors if they are used with a format that does not have the same number of components as they define.

Note that the base storage type must provide a zeros method. The zeros method relies on their being a registered default array-implementation for the storage type. The default installation of OpenGL-ctypes will use Numpy arrays for returning the result.

(

)

Given an OpenGL image format specification, get components/pixel

(

)

Set the pixel-transfer modes for a given image "rank" (# of dims)

Uses RANK_PACKINGS table to issue calls to glPixelStorei

(

)

Perform compatibility conversion for PyOpenGL 2.x image-as string results

Uses OpenGL.UNSIGNED_BYTE_IMAGES_AS_STRING to control whether to perform the conversions.

(

)

Set pixel transfer mode to assumed internal structure of arrays

Basically OpenGL-ctypes (and PyOpenGL) assume that your image data is in non-byte-swapped order, with big-endian ordering of bytes (though that seldom matters in image data). These assumptions are normally correct when dealing with Python libraries which expose byte-arrays.

(

)

Setup transfer mode for a read into a numpy array return the array

Calls setupDefaultTransferMode, sets rankPacking and then returns a createTargetArray for the parameters.

RetroSearch is an open source project built by @garambo | Open a GitHub Issue

Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo

HTML: 3.2 | Encoding: UTF-8 | Version: 0.7.4