Hello,
i've found a few times books that have a wrong (non existing) language, that contains special characters, such as semicolon ;
If a user has read a book in this language, in their Stats page they will see this language listed, but when clicking on it, to see which are the books with that language, it will not load. This would be nice to have it fixed in order to be able to correct the languages in those books.
Some examples i found of these languages are:
Spanish; Castil
Italian ; latin
When clicking on the url for seeing the books in the stat page, it will try to open an url like this (i removed the user_id, but i can pass it privately if needed):
?user_id=&chart_type=Languages&label=italian ; latin&year=&read_status=Read
This url doesn't work and it won't load any book.
I tried modifying it in the browser by replacing the semicolon with %3b and it works:
?user_id=&chart_type=Languages&label=italian %3B latin&year=&read_status=Read
I think it would be nice to have this replacement for all common special characters so people can easily find the books with the wrong info in their read list
Thank you for this post. There's a lot we need to do to improve our language data. For now, please report individual cases to support@thestorygraph.com.
Hi Nadia , thanks for your reply.
i'm not sure i've explained myself clearly, the issue is not with language data, the issue is with how the website encodes urls from the Stats page: they don't work if they contain special characters.
I just tested with Authors, and the same problem happens (video attached). In this specific example, the author has in the name the character ";", when clicking on it it gets passed as query param without being encoded, and it tries to load only the first part of the query param. ";" should be encoded as "%3b" in the query param, if i replace it manually in the url it works.
In the query the website should send "chart_type=Most Read Authors&label=test %3b test" instead of "chart_type=Most Read Authors&label=test ; test"
Same thing for languages, it should send query params "chart_type=Languages&label=italian %3b latin" instead of "chart_type=Languages&label=italian ; latin".
I'm a librarian so i'll fix the data, finding the wrong books by adjusting the url manually, but i think most users wouldn't guess that they have to modify the url manually to see the actual affected book, so they wouldn't be able to report them.
i add also a video from the web browser, so I can show better what i mean by saying that we have to replace characters in the url
bianco_c Hello! Thank you for the further info. I did understand you, but in general, the language info needs to be sanitised more broadly, such that special characters don't appear in them. That would be a higher priority for us, independently of us addressing the URL encoding!