Description: When editing text, it is important to be able to delete diacritics separate of the characters. In Writer, "Back space" button works as intended, removing only a single diacritic. But, "Delete" button has a different behavior and removes both the character and the diacritics. Steps to Reproduce: 1. Write some text with diacritics, like عَرَبِيَّة and make sure that the paragraph is right to left 2. Go to the start of the text 3. Press "Delete" multiple times to remove parts of the text from right to left عَرَبِيَّة<- 4. Undo, and the try removing using "Back space" from left to right ->عَرَبِيَّة Actual Results: "Back space" button removes only a single diacritic. But, "Delete" button removes both the character and the diacritics. Expected Results: "Delete" button should also remove a single diacritic. Considering some text like ٱللُّغَةُ ٱلْعَرَبِيَّة please note that: * It is not intended to remove integrated diacritics like the one the top of alef ٱ separately. * If there are multiple independent diacritics on some character, they should be removed one bye one. Reproducible: Always User Profile Reset: No Additional Info: Version: 25.8.0.0.alpha0+ (X86_64) / LibreOffice Community Build ID: 8fdef548702ef240980b52e4076af36122534fed CPU threads: 12; OS: Linux 6.2; UI render: default; VCL: gtk3 Locale: en-US (en_US.UTF-8); UI: en-US Calc: threaded
Wait, if you position edit cursor within the word and toggle to Unicode (<Alt>+X) you get the base and the combining diacritic. Can make selection/edit then, and (<Alt>+X) to convert back to text span. Same for bug 164252
(In reply to V Stuart Foote from comment #1) > Wait, if you position edit cursor within the word and toggle to Unicode > (<Alt>+X) you get the base and the combining diacritic. Can make > selection/edit then, and (<Alt>+X) to convert back to text span. > > Same for bug 164252 Thanks for providing this method, it can be helpful for debugging. But, it does not help much here, as even re-typing would be faster. Or, one may use back space. This is in itself an issue for the people who type a lot with diacritics, like Arabic (and in some cases Persian).
Does it really happen often, that you want to keep the base character and only change the diacritics? It is really much effort for the user to delete a combination and then type the desired one? The keyboards I have seen in Wikipedia seem to have the diacritics directly on the keys, so that using the Unicode-Dialog is not needed.
It is definitely needed in Complex Fonts like Arabic, Devanagari, etc. Most of the times I have to proofread a text in Devanagari which has been extracted after OCR. Instead of having to type some complex ligatures, it would be easy if we could just replace the problematic character. At least the option for the cursor to go through every character could be provided in Overwrite mode. The (<Alt>+X) method fails with combined characters in Devanagari.