🎯 What can I work on ? (Python SDK)

[teolemon]

Simplify developer flow

Add new and missing features

• Document or Increase the maximum amount of results using facets #69
• Add pagination to the category facet #56
• Document image addition #55

GENERAL

• Multilingual (180 languages are supported)
• Create users
• Check if users exist

READ

• Product: you can retrieve information about products
• Product images (get product image by field, size and language, current fields are: Front, Ingredients, Nutrition, Packaging)
• Additives, Allergens, Environment impact, Ingredient analysis, Nutrient levels, Nutrition facts (incl. micro-nutrients)
• Language dependant fields (like ingredients, product name, packaging text)
• Eco-Score, Nutri-Score, NOVA groups
• Translated, high-level and customizable product information (Attributes)
• Product completion status
• Folksonomy Engine (custom properties for products)
• Open Food Facts Events (Gamification)

PERSONALIZATION

• Recommended Daily Intakes
• Translated, high-level and customizable product information (Attributes)
• Autocompleting for the user inputs, by giving suggestions for (Labels, categories, ingredients, additives, traces ...)

WRITE

• Save product to Open Food Facts
• Send images
• User comments and auth
• Anonymous photo upload
• Photo operations (rotate, crop, select, unselect with language awareness)

ML/OCR

• Get the OCR of the ingredients photo (for validation purposes)
• Robotoff support
• Extract ingredients from ingredients image: User can extract the ingredients text if it is missing

Process

• Capture the barcode of the product where you want to perform the OCR.
• The Product Opener server software opens the image (process_image=1)
• Product Opener returns a JSON response. Processing is done using Tesseract or Google Cloud Vision (recommended). The result is often cripped with errors with Tesseract, less with Google Cloud Vision.

Notes:

• The OCR may contain errors. Encourage your users to correct the output using the ingredients WRITE API.
• You can also use your own OCR, especially if to plan to send a high number of queries.

Parameters
Test server: https://world.openfoodfacts.org/cgi/ingredients.pl
code=code
id=imagefield
process_image=1

Roadmap

• List of new APIs to implement: https://github.com/openfoodfacts/api-documentation/issues
• New APIs: https://github.com/openfoodfacts/api-documentation/issues?q=is%3Aissue+is%3Aopen+label%3A%22new+api+to+document%22
• Existing APIs: https://github.com/openfoodfacts/api-documentation/

[Ansh-Sarkar]

Hi ! Is the SDK still being maintained ?

[teolemon]

@Ansh-Sarkar we're trying our best, yes