You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The ingredient analysis should include any label, especially if this is a label by a certification authority. Such a label should overturn the conclusion of the ingredient analysis.
For data quality purposes the (vegan) error could be turned into a warning, just to alert for a possible inconsistency that can be checked and needs to be corrected.
Why
A more correct analysis.
The text was updated successfully, but these errors were encountered:
The current behavior is by design: if we have a no-gluten label, we use it, unless we have ingredients that contradict it...
From Attributes.pm:
# There may be conflicting information on allergens (e.g. a product that claims to be "gluten-free",
# but that also says it may contain traces of cereals containing gluten, or that contains an ingredient
# that usually contains gluten)
# The algorithm below is designed to be conservative: information that indicates the presence
# or the possibility of presence of an allergen prevails on information that indicates its absence
It would be great to review some of the 2638 products that have a "no gluten" label + gluten identified as an allergen: https://world.openfoodfacts.org/label/no-gluten/allergen/gluten
to see if there are systematic issues that we could fix (e.g. wrong parsing of some ingredients)
What
The analysis for gluten free and vegan does not take into account any labels. Sometimes the a label can overturn the ingredient analysis
Steps to reproduce the behavior
see: https://de.openfoodfacts.org/product/4014472277088 for a gluten free example
see: https://world.openfoodfacts.org/product/5030756007775/vegan-no-steak-pie-pukka for a vegan example
Expected behaviour
The ingredient analysis should include any label, especially if this is a label by a certification authority. Such a label should overturn the conclusion of the ingredient analysis.
For data quality purposes the (vegan) error could be turned into a warning, just to alert for a possible inconsistency that can be checked and needs to be corrected.
Why
A more correct analysis.
The text was updated successfully, but these errors were encountered: