searchcolor 2.1.3

Image color extraction from web image search

This module uses imagecolor with PIL(Pillow) to extract colors from web image searches.

Available functions

average_image_url(url, name)

Averages a single image from a url into RGB color values. Returns a dictionary with the following keys: name, red, green, blue

• url - image url.

• name - name to return. Generally passed from the function that generates the url.

• timeout - requests timeout in seconds.

• max_size - maximum size of image to fetch in MB.

_image_search_average(url_list, max_threads=20)

Averages all urls in a list into a singular RGB average.

• url_list - path to directory

• max_threads - max processes to spawn.

• timeout - requests timeout in seconds. This gets passed to average_image_url

• max_size - maximum size of image to fetch in MB. This gets passed to average_image_url

google_average(search_term, num_results, api_key, cse_id, max_threads=20)

Does a Google image search and averages all the images into a singular RGB search average. Returns a dictionary with the following keys: name, red, green, blue

• search_term - Google image search term.

• num_results - Number of results to include.

• api_key - Google API key.

• cse_id - Google CSE ID.

• max_threads - max processes to spawn. This gets passed to \_image_search_average

• timeout - requests timeout in seconds. This gets passed to average_image_url

• max_size - maximum size of image to fetch in MB. This gets passed to average_image_url

Future work

• add more information to readme

• build offline tests

Tests

Testing is done with pytest

Currently tests only the google api. Requires the enviroment variables GOOGLE_SEARCH_API and GOOGLE_SEARCH_CSE to be set to a valid google custom search api and id

Run with python3 setup.py test