Back to Questions
1.4k
views3
comments
Accented characters appear with question marks when rendered with ui.next text
websharperui.nextunicode

I am trying to render the text á in a text box in a UI.next SPA My code has

1

text "This is á"

but what gets displayed is

1

This is ?

where the question mark is reversed. Inspecting the page source shows that the unicode character is not being sent correctly.

My encoding is correct - other accented characters directly in my template are working.

Using VS 2015 with latest websharper from nuget

After more testing, I've established that: - explicitly using a unicode escape such as "\u00b2" works and sends the expected character - the websharper processor turns all non-ascii literals inside quoted strings into the unicode unknown character

The following code:

1
2

let unistring = "rás\u00e9"
let usmap = unistring |> Seq.map (fun x->sprintf "%d" (int x))

creates the sequence ["114";"225";"115";"233"] as expected when run in fsi or an fsharp console program.

When run in websharper, either client side or server side, it produces ["114";"65533";"115";"233"]

So I think there's a bug in websharper....

Happy to dig further if necessary to establish precisely where the problem is.

.

Thanks for that quick response. I've done some tests and can confirm that if I manually go into the VS File menu, Advanced Save Options, and set Encoding to Unicode ( UTF-8 with Signature) - Codepage 65001 and rebuild the app, it now works as expected.

I can also confirm that the working standard F# program already had this setting by default. So the problem is not with Websharper itself, or with VS itself, it's that the Websharper templates for VS are defaulting to Ansi encoding for the code files rather than Unicode.

I imagine this would be quite easy to fix...unless there was a deliberate good reason to have things the way they are, in whcih case I think the documentation should flag this issue and the workaround somewhere.

By on 4/22/2016 10:54 PM ()

This is a pretty weird issue. Here's what I found:

  • your last example seems to produce the right output on try.websharper,
  • it indeed produces the wrong output when run from VS,
  • if I change the encoding of the file to UTF8 in some other editor the application will correctly output the string "rásé" but it shows up as "rás\u00e9" in the VS editor.

So it seems to be an issue with character encoding rather than WebSharper to me.

By on 4/22/2016 6:21 AM ()

Well, it may be in WebSharper itself actually. If you write "rásé" VS will still save it as an ANSI encoded file but as soon as I insert a unicode character and VS interprets and saves the file as such, it's working. I.e. the string "ráséӘ" will be rendered correctly.

By on 4/22/2016 6:36 AM ()

    Topic tags

    IntelliFactory Offices Copyright (c) 2011-2012 IntelliFactory. All rights reserved.
    Home | Products | Consulting | Trainings | Blogs | Jobs | Contact Us | Terms of Use | Privacy Policy | Cookie Policy     		Built with WebSharper