Datepicker: removed arrow symbols from next and previous month hovertext in swedish

[adalid-1]

Fixes #2048

[linux-foundation-easycla]

• ❌ The commit (2baed25) is missing the User's ID, preventing the EasyCLA check. Consult GitHub Help to resolve.For further assistance with EasyCLA, please submit a support request ticket.

[linux-foundation-easycla]

• ❌ The commit (2baed25) is missing the User's ID, preventing the EasyCLA check. Consult GitHub Help to resolve.For further assistance with EasyCLA, please submit a support request ticket.

[adalid-1]

Commit is made from work git account and PR is made with private git account.

[mgol]

@adalid-1 can you amend your commit so that the commit author info matches an account with which you can sign a CLA? A CLA signature is a hard requirement for us to accept contributions.

[mgol]

@fnagel any idea why these symbols were included in some language versions, leading to issues like #2048, and not in others, like en-GB?

[adalid-1]

@mgol I rolled back the local branch and recomitted the change with correct author and did a forced push. I am not great with git so please be wary if you see something wierd. If so I can remake the pullrequest from scratch after the weekend.

[fnagel]

Afaics we have the following prefixes for prevText and nextText:

• < «
• « «
• < <
• ← ←
• ← and → actual UTF symbols

@mgol

any idea why these symbols were included in some language versions, leading to issues like #2048, and not in others, like en-GB?

No, not really. And there is no consistency. Which is odd. Why would we have those in some languages only. Makes no sense to me. So either remove or at least using the same char.

Looks like the majority is with one of those prefixes, but according to #2048 those chars could cause issues so removal might be the better option. I don't think its a good thing for a11y.

What do you guys think?

[mgol]

Yeah, I'd also opt into removing all of those prefixes/suffixes. @adalid-1 would you like to do it for all the locales?

[adalid-1]

@mgol Sure thing!

[adalid-1]

A few languages turned up empty when removing arrow. Maybe that is ok?

[adalid-1]

Do I need to do anything more here or can I just wait for someone to review and merge this?

[mgol]

Hmm, I wonder if this is really behaving in the same way as 1.12.1 did. As prevText & similar were documented to accept text, not HTML but were in fact accepting HTML, we switched that to only accept text as a security fix. It looks like this is something that broke those arrows... @adalid-1 are you sure the behavior in 1.12.1 was the same?

Making the texts empty may be bad for accessibility, potentially worse than at least having an arrow (as long as it works, of course). Maybe we should leave the arrows - but fixed, inserted as Unicode, not HTML entities - in those languages that would otherwise have empty prevText, etc.? What do you think, @fnagel?

[adalid-1]

You are right it works in the older version. I was misstaken, I must have made some changes that did not come through when testing earlier. Now that i tested downloading the older version from here:
https://jqueryui.com/download/
And setting nextText to contain the arrow it came out as intended.

[fnagel]

As prevText & similar were documented to accept text, not HTML but were in fact accepting HTML, we switched that to only accept text as a security fix.

Sounds reasonable. Anyway, a fix would be good.

A few languages turned up empty when removing arrow. Maybe that is ok?
Making the texts empty may be bad for accessibility, potentially worse than at least having an arrow (as long as it works, of course).

We should definitely not have those texts empty. I'm ok with using the arrows but having a real text would be better. I could add a commit introducing those missing texts translated by Google Translate (and verified by another). @mgol What do you think?

[mgol]

I could add a commit introducing those missing texts translated by Google Translate (and verified by another). @mgol What do you think?

@fnagel If you could do that, that'd be great! We just won't be able to verify those easily and I had an impression you prefer to be more cautious here. Please let me know when you do it so that we can proceed with the rest of the changes.

[fnagel]

We might want to merge #2067 before working on this. This one seems related: #1900

[mgol]

This landed in 3126e12 as part of #2100; your author info has been preserved. Thanks!