-
Hi, I’ve recently tried to move from ibus to fcitx5 and thus I tried to migrate my Cyrillic ibus-table input method to m17n so that I could use it in fcitx. The first attempt can be seen in my dotfiles on gitlab. But I have a problem. I use Polish keyboard as my main default keyboard layout – and Polish layout lets me input characters like ⟨ż⟩ via AltGr + z, or ⟨ę⟩ via AltGr + e – and I did use those in the input sequences to transform them into Cyrillic ⟨ж⟩ and ⟨ѧ⟩ respectively. So I added rules like these to my mapping: ("Ż" ?Ж)
("ż" ?ж) but they don’t work – when I switch the input method to my Cyrillic, hitting AltGr + z produced unchanged Polish ⟨ż⟩. The m17n data format spec says about
so I’d expect this to work. Is this a bug in fcitx5, a consciously enforced limitation? Is there a way around it? I also tried making rules with modifier keys explicitly myself, eg: ((A-z) ?ж) and this lets me use the left Alt but not the right one, I also tried: ((G-z) ?ж) ( |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 5 replies
-
While it's totally up to you that whether to use m17n to implement it, have you checked the table engine in fcitx5? (It's a part of fcitx5-chinese-addons). Use libime_tabledict to convert it
And save it to Save this file to |
Beta Was this translation helpful? Give feedback.
While it's totally up to you that whether to use m17n to implement it, have you checked the table engine in fcitx5? (It's a part of fcitx5-chinese-addons).
Use libime_tabledict to convert it
interslavic-cyrillic.txt
And save it to
~/.local/share/fcitx5/table/
Save this file to
~/.local/share/fcitx5/inputmethod/
you may want to update the informational field to some other value like lang code & name.interslavic-cyrillic.conf