Combining Diacritical Marks

[Jonathan El-Amin]

Hey y'all. I've been making a conlang for a little over a year and I've decided to make a Latin-based keyboard for my conlang. I already the letters all figured out, it's just making the keyboard that I have to do - which took a heck of a lot longer than expected! I've been working on this since 2:17 pm and as of writing this sentence, it's 12:56 am, good gosh lol. Anyways, I've been working on creating dead keys to input the tone marks onto my vowels. It was VERY hard to do and understand at first, but I think I have the hang of it now. 

The issue I came to is the "U with a dot above". Unfortunately there doesn't seem to be a unicode for that character. At least, not one that Ukelele supports/has. I read on reddit and in this google group that combining diacritical marks (CDM) is possible, but I'm not entirely sure how to do it or if it would even work in my case. Do I still have to use dead keys to use CDM or are they 2 separate processes? I'm using the latest version of Ukelele btw - 3.5.7. How do you use CDM on Ukelele 3.5.7? Thank you in advance for any help and I hope y'all have a great day.

[Gé van Gasteren]

Hi Jonathan,

There are basically two ways to type accented characters:

• The traditional one is to type an accent key first, then the key for the base letter: this is the "dead key method"

• A newer way is to first type the base character, then a CDM. It’s a more flexible way, because you can easily stack accents, but it’s not universally supported yet.

Which one is better for you depends largely on taste, but also on the number of accents and accented characters you need to incorporate.

As you have chosen the dead-key method already, you may want to keep that for the last combination you need to make: U̇ and u̇.

To include these two letters, the only thing different from precomposed letters that have a Unicode number is that here, instead of the usual typed character or Unicode number, you specify 2 Unicode numbers as the output:

U ̇ = U+0055, U+0307

u ̇ = U+0075, U+0307

The above are hexadecimal numbers, which you can enter as follows in Ukelele (see the PDF manual, section 3.1.3 on page 7):

U̇

u̇

[Sorin Paliga]

On 30 Dec 2022, at 15:51, Gé van Gasteren <gevang...@gmail.com> wrote:

• A newer way is to first type the base character, then a CDM. It’s a more flexible way, because you can easily stack accents, but it’s not universally supported yet.

The good news is that more and more apps do support CDM’s, including usual ones like Pages, Word (last versions), OpenOffice/LibreOffice. Initially, this was possible with only a limited number of apps, e.g. Nisus or Mellel. 

The real problems may occur when such a text is prepared for print, and the DTP app does not support this. So, yes, the user must first analyze the whole process and decide whether CDM’s, which practically cover all the range of diacriticals, are supported or not until the first stage of the process. 

And, of course, the font must also include CDM’s and that they are correctly displayed. 

[Tom]

Perhaps chose another diacritic where the precomposed combo you need is in Unicode?

[Sorin Paliga]

On 30 Dec 2022, at 17:22, Tom <thge...@gmail.com> wrote:

The issue I came to is the "U with a dot above". Unfortunately there doesn't seem to be a unicode for that character. At least, not one that Ukelele supports/has. 

u̇ U̇ ȧ Ȧ ṅ Ṅ etc.

ụ ạ ọ Ụ Ạ ṃ etc..

Using CDM one may practically use any diacritical mark above/below any character. 

I hope this is correctly displayed. I attach a screen shot. 

“...At least, not one that Ukelele supports/has. “

No, UKELELE has nothing to do with the so-called “font inventory”, i.e. the characters included in a given font. If this may happen, the system/the app automatically replaced to font with another font which includes the given char. I do recommend 

Unicode Checker

a very useful tool for accessing various chars with diacritical marks above/below and many other chars.

[Gé van Gasteren]

Thanks Sorin, that UnicodeChecker is great!

I confess that I was put on the wrong leg by their web icon: I thought I had to wait, when the download had long finished ;-)

--
You received this message because you are subscribed to the Google Groups "Ukelele Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ukelele-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ukelele-users/CBE38F22-262F-4B15-AF15-FD2D3533BA72%40gmail.com.

[Jo]

Thank you all so much for the help, and thanks for introducing me to Unicode Checker! I'll check it out. Hope y'all have a great morning/day/night :)