New problem with Neko/.CPP and strings and utf8, extended ascii and all that jazz

[Keith Weatherby II]

So some of you probably already know that I am working on a ascii/extended ascii text renderer in haxeflixel.  My original problems stemmed with the fact that neko/.cpp uses essentially an array of bytes for strings, instead of the regular unicode strings in other platforms.  So my solution was to use the UTF8 Class' "iter" function, to iterate over each character and store their character codes in an array.

So I was able to get everything pretty much working by using the UTF8 class however I've run into a new problem.  When I want to check for a character I use String.fromCharCode() -- only problem is the same problem I had before, fromCharCode doesn't work correctly in neko or cpp targets.  Also I can't use the Utf8.decode function because it's reporting invalid character in the character string, my guess is that an ISO string is only the lower 127 characters so that means I can't use that.

Does anyone have any other suggestions for how to return a character from a string in the neko/windows targets?

The only thing I can think of is actually using UTF8.iter just like I did on the whole character string and instead do it on a single character, then I can compare character codes.

Any suggestions?

[azrafe7]

Hey Keith,

not sure I understand exactly what you're trying to do, so if you can post a little example would surely help clarify things.

In the meantime I'll point you to this lib https://github.com/mandel59/unifill which used to cover unicode in pretty much all targets (I still use it from time to time) and comes from a pre haxe.Utf8 era.

Cheers, azrafe7

[Keith Weatherby II]

Essentially I want to convert Utf8 characters into standard haxe string that works on windows/neko targets (as well as flash and html5), that will work for extended ascii codes.  I have a string of 256 characters that corresponds to the 256 ascii+extended codes (ibm code page 437) -- I can't convert them to the proper characters on cpp/neko targets.  String.fromCharCode() doesn't work because it's a utf8 string, however, apparently the haxe.utf8 class doesn't really have any functions I can use there either.  I can't run decode on it, because it says the characters in my character string are invalid and throws an exception.  I'm guessing this is because an ISO string is essentially regular ascii ie < 128 - The type of characters I need to grab/convert are like the shaded block ones or the ones to make little text windows (in other words the extended set or those 128 and above)

Did that do a better job at explaining what I want?  (I have trouble explaining things clearly)

Keith

[Keith Weatherby II]

btw i just used haxelib to install unifill so if I don't come up with something satisfactory soon, I'll probably convert my code to use it.

On Sunday, July 9, 2017 at 3:12:11 AM UTC-4, azrafe7 wrote:

[Ashiq A.]

Unifill looks awesome, thanks for sharing.

Keith: please let us know if you get this working in HaxeFlixel. I struggled with non-ACII characters in HaxeFlixel a couple of year ago and gave up.

--
To post to this group haxe...@googlegroups.com
http://groups.google.com/group/haxelang?hl=en
---
You received this message because you are subscribed to the Google Groups "Haxe" group.
For more options, visit https://groups.google.com/d/optout.

[azrafe7]

Keith, what I meant, is if you can write up a quick code snippet showing the issue (on try.haxe would be great, or on haxe.mrcdk.com).

Also, we're talking about writing to the console/terminal, into a textfield with openfl/flixel or what exactly?

On Saturday, July 8, 2017 at 5:20:27 PM UTC+2, Keith Weatherby II wrote: