Fall back to mul labels
#70
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Wikidata recently introduced “default values for labels” (https://www.wikidata.org/wiki/WD:mul) to reduce size of items. One particular example where these are used is humans, since humans’ names are written the same in all Latin-script languages. Well, mostly. For example, Leonardo da Vinci (https://www.wikidata.org/wiki/Q762) is called “Léonard de Vinci” in French, and because all labels that use the Italian spelling have been removed (including Dutch, English and German), now the Dutch interface falls back to the only label we extracted, saying that the namesake of Leonardo da Vincistraat is “Léonard de Vinci” rather than “Leonardo da Vinci”. (I couldn’t find such an example in Brussels, but it’s possible that *all* labels are removed in favor of `mul`, in which case we wouldn’t extract any label, and the frontend would display `null`. That’d be even worse than showing a French label on the Dutch interface.) This patch fixes this by falling back to the `mul` label if there is no label in the given language. This is only a fallback, so the French interface still says that the rue Léonard de Vinci is named after “Léonard de Vinci”; and it affects only labels, not descriptions, as there are no `mul` descriptions.
Contributor
Author
|
Thanks for triggering CI! The PHPStan errors are all unrelated, most of them fixed in #71. |
jbelien
approved these changes
Sep 25, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Wikidata recently introduced “default values for labels” to reduce size of items. One particular example where these are used is humans, since humans’ names are written the same in all Latin-script languages. Well, mostly. For example, Leonardo da Vinci is called “Léonard de Vinci” in French, and because all labels that use the Italian spelling have been removed (including Dutch, English and German), now the Dutch interface falls back to the only label we extracted, saying that the namesake of Leonardo da Vincistraat is “Léonard de Vinci” rather than “Leonardo da Vinci”.
(I couldn’t find such an example in Brussels, but it’s possible that all labels are removed in favor of
mul, in which case we wouldn’t extract any label, and the frontend would displaynull. That’d be even worse than showing a French label on the Dutch interface.)This patch fixes this by falling back to the
mullabel if there is no label in the given language. This is only a fallback, so the French interface still says that the rue Léonard de Vinci is named after “Léonard de Vinci”; and it affects only labels, not descriptions, as there are nomuldescriptions.