PhotoBot

A Pillow based layered compositions library. The original version is part of the NodeBox Library.

Original documentation: https://www.nodebox.net/code/index.php/PhotoBot

Landslide

A Photobot demo by Tom de Smedt. https://www.nodebox.net/code/index.php/Landslide

PhotoBot was initially a part of the NodeBox Library.

Since 2013 it has been updated to the current Python 2/3 and Pillow libraries.

This version works standalone with Python2.7, Python3.11 and is identical to the current NodeBox Library version.

Fundamental changes

canvas.draw() has a double functionality. In Nodebox1 it places the result in the content section of the document window.

The image is also written out to a folder "exports" with a name like "photobot_YYYY-MM-DD_HHMMSS.png" using the current datetimestamp.

If run from the Terminal, only the exports part is done.

If the exports folder does not fit your needs use canvas.export() as follows:

cv.export( name="/Full/Path/to/image.png", ext=".png", format="PNG" )

Extension and format follow Pillow Image.save rules

New functions & properties

• property canvas.top adresses the top layer

• property canvas.topindex returns the index of the top layer

• property canvas.dup duplicates the top layer

Installation

python setup.py install

Usage

T.B.D. see examples folder.

Hints

python Kontaktabzug-1.py PATH-TO_IMAGE-FOLDER

creates a contact print of a folder. All other Layer_* examples demonstrate a single function.

Scanning your images with ImageWells

The collage examples make use of the yet unreleased "ImageWells" library of which a preliminary version is included in the examples folder.

Abbreviated example from "Example collage 1.py"

import imagewells
loadImageWell = imagewells.loadImageWell


imagewell = loadImageWell(
				bgsize=(1280, 1024),
				minsize=(256,256),
				pathonly=True,
				maxfilesize=100000000,
				maxpixellength=16000,
				additionals=additionals,
				imagewellfilename="imagewell.txt",
				tabfilename="imagewell.tab",
				ignoreDotFolders=False,
				ignoreFolderNames=('+offline',))

After a script uses loadImageWell() for the first time there should be a file imagewell.txt in the same folder as the script excuting.

The file imagewell.txt looks like this:

/Library/Desktop Pictures
C:\Windows\Web
/usr/share/backgrounds
/usr/share/wallpapers

Each line represents a path for a platform. The file is NL terminated and UTF-8 encoded.

The first line is the "Desktop Pictures" folder on macos.

Line 2 is for windows.

Lines 3+4 are for linux.

If a folder does not exist, it will be ignored. The idea is to put your own image folders into that file.

The parameters for loadImageWell are as follows:

• bgsize=(w,h) - a tuple marking the size at which a image is designated 'background'. Usually the canvas size.

• minsize=(w,h) - at least width>=w or height>=h for a image to be considered. If both are smaller, the image is ignored.

• pathonly=True|False - if False, a tuple with: (path, filesize, lastmodified, mode, islink, w0, h0, proportion, frac) is returned.

• addtionals=('folder',) - a list of folders to be added. If this is active, no caching is used.

• imagewellfilename="imagewell.txt" - the name of the folderlist file to be read.

• tabfilename="imagewell.tab" - the name of the resulting cache file.

• ignoreDotFolders=False - ignore folders starting with a '.'. This is for MacOS '.thumbnails' folders which you usually want to ignore

• ignoreFolderNames=('folder',) - a list of folder names. If a scanned folder STARTS with a name from that list, it will be ignored.

The resulting dictionary contains the following keys:

• allimages - a list of all images

• tiles - a list of images with: minsize <= image <= bgsize

• backgrounds - a list of images considered backgrounds

• landscape - a list of images where WIDTH >= HEIGHT

• portrait - a list of images where WIDTH < HEIGHT

• fractions - dictionary with all the fractions as keys. A fraction key looks like: '1024:575', '4:3' or '16:9'.

• 'WxH largest', 'WxH smallest' and 'WxH median': accumulated sizes

  • 'WxH largest': (8003, 5622),
  • 'WxH median': (1074.9886462882096, 754.4748180494905),
  • 'WxH smallest': (566, 167),

The items of each image list key will depend on the pathonly parameter.

Examples

See "examples/Example collage *.py"

See "examples/Layer_function_*.py"