Support

There is nothing we can or should do because this is between your browser and your server. The URL your server sees is something like this:

conditions/4-conditions-g%C3%A9n%C3%A9rales-d-adh%C3%A9sion-au-programme-d-affiliation

This is the URL you need to redirect. I tried that on two different test sites and it works.

IMPORTANT!

If you have a multilingual site which adds the language slug in the URL you need to also include it in the URL you are redirecting from. So, if your site adds /fr/ for all French pages you'd have to redirect fr/conditions/4-conditions-g%C3%A9n%C3%A9rales-d-adh%C3%A9sion-au-programme-d-affiliation instead. I don't see your site doing that but it's good to know.

Remember that browsers usually URL-encode any character which is invalid in a URL, including accented Latin-1 characters and several special characters. It is very strongly recommended that your URL aliases do not contain accented characters. Ideally, your aliases should consist entirely of the lowercase characters a-z, numbers 0-9, dashes and underscores. Anything else will be trouble.

As to why we shouldn't do anything about it. We could try to URL-decode the path. But… URL decode with which character set? UTF-8, ISO-8859-1, Windows-1253, something else? Choosing the wrong character set could result in an invalid path which could have unexpected results in redirections. To give you an idea, URL-encoding the character é in ISO-8859-1 codepage results in %E9. URL-encoding the same character as UTF-8 results in %C3%A9. URL-decoding the latter as ISO-8859-1 results in Ã©. So, if we decoded the URL conditions/4-conditions-g%C3%A9n%C3%A9rales-d-adh%C3%A9sion-au-programme-d-affiliation as ISO-8859-1 we'd get conditions/4-conditions-générales-d-adhésion-au-programme-d-affiliation and we'd still be unable to figure it out. Trying to URL decode the URL conditions/4-conditions-g%E9n%E9rales-d-adh%E9sion-au-programme-d-affiliation as UTF-8 would fail to decode %E9 leading (in the best case) to conditions/4-conditions-gnrales-d-adhsion-au-programme-d-affiliation (missing characters). It gets worse when you have to take into account code-pages other than Western European such as Greek, Cyrillic and so on. Since the encoded URL does not contain information about the character set we do not have a well-defined problem to solve. Browsers encode URLs based on what the server tells them it supports, information we do not have access to when we need to do URL decoding.