Many Paratext supporters are aware of how Paratext can help you manage Unicode characters:
- You can type a Unicode character code value, then press Alt-X to switch to the character. Or if you press ALT-X when the cursor is to the right of a character, Paratext will change that character into its Unicode code value. (If you cannot edit the text, it will show the code in a popup).
Example: type 0257, press ALT-x, you will see ɗ in the text. Move the cursor to the right of a ɗ and press ALT-x, you will see 0257. (This also works in the language properties lists, the wordlist filter box, and other places, not just in a text window). - The characters inventory lists all the characters in a project and shows their Unicode code values. You can sort by the Unicode code values.
- You can specify a Unicode character in a autocorrect.txt file with a code \uXXXX where the 4 Xs represent the Unicode value. For example d/-->\u0257 will generate a ɗ when you type d then /.
- In RegExPal you can specify a Unicode character by its code value with \uXXXX, for example \u0257 for ɗ.
But what about Unicode characters off the Basic Multilingual Plane (BMP), whose code value is more than four hex digits? Do these tools still work?
- To work with ALT-X, enter 8 digits for any code greater than 4 digits. For example, U+11010 Brahmi Letter Ai you would enter the code as 00011010. (Despite having only 1's and 0's, this is not a binary number but a hexadecimal number)
- In the character inventory, any Unicode value greater than four digits will also appear as 8 digits. So the character inventory will show 00011010 as the Unicode value for Brahmi Letter AI.
- For an autocorrect file, you have to encode these characters as a pair of four digit characters called a surrogate pair. Our Brahmi Letter AI can be written like this: ai-->\ud804\udc10. You can find the surrogate pair values for Unicode characters by Googling the code value ("U+11010"), then looking on the entry on compart.com or fileformat.info for the UTF16 value.
- In RegExPal, you specify the surrogate pair to specify a non-BMP character. \ud804\udc10 will match the Brahmi Letter Ai.
