fix(cli): map kitty CSI-u lock-bit variants so key combos survive NumLock - #89676
fix(cli): map kitty CSI-u lock-bit variants so key combos survive NumLock#89676liuhao1024 wants to merge 1 commit into
Conversation
…Lock kitty and ghostty OR the CapsLock (64) / NumLock (128) state into the CSI-u modifier parameter. With NumLock on, Ctrl+C arrives as ESC[99;133u (5 + 128) instead of ESC[99;5u; the alias table had no entry for it, so every key combo leaked as literal text like [127;133u (NousResearch#89651). Install every CSI-u alias with the lock-bit variants (+64/+128/+192); the xterm modifyOtherKeys encoding never carries lock bits, so the ESC[27;N;CP~ form is left untouched. The Esc-key registration now covers modifier 1 as well (1+128=129 is a lone Esc with NumLock on).
alt-glitch
added
type/bug
…s, and PUA functional keys Follow-up to the salvaged NousResearch#89676 + NousResearch#90291 lock-bit fixes: extract a shared _lock_variants() helper and cover the sites both PRs missed - install_shift_enter_alias / install_ctrl_enter_alias / install_cmd_backspace_alias CSI-u spellings, legacy CSI-letter and CSI-tilde navigation twins derived from the existing table for ALL modifiers 1-16 (not just plain/shift), plain F1-F4 SS3 fallback, unmodified CSI-u keys (Tab/Enter/Space/Backspace), and kitty PUA functional keys (keypad, F13-F24, Ignore range) under lock bits. 8 new tests.
…s, and PUA functional keys Follow-up to the salvaged #89676 + #90291 lock-bit fixes: extract a shared _lock_variants() helper and cover the sites both PRs missed - install_shift_enter_alias / install_ctrl_enter_alias / install_cmd_backspace_alias CSI-u spellings, legacy CSI-letter and CSI-tilde navigation twins derived from the existing table for ALL modifiers 1-16 (not just plain/shift), plain F1-F4 SS3 fallback, unmodified CSI-u keys (Tab/Enter/Space/Backspace), and kitty PUA functional keys (keypad, F13-F24, Ignore range) under lock bits. 8 new tests.
|
Merged via #90561 with your commit cherry-picked onto current main — your authorship is preserved in the git history ( Your producer-contract insight — that the xterm modifyOtherKeys tilde form never carries lock bits, so only the CSI-u form needs the twins — shaped the final design and is negative-tested in the merged suite. Thanks! |
|
Thank you for the merge path via #90561 and for preserving authorship in the git history — and for extending the coverage well beyond the original slice. Combining the lock-bit mapping with #90291's legacy CSI-letter/tilde navigation, widening to the alias installers (Shift/Ctrl+Enter, Cmd+Backspace), covering modified navigation keys for all modifiers 1–16, and adding the kitty PUA function keys turns two narrow fixes into a coherent terminal-input layer. That's the right shape for this family of bugs — glad the NumLock lock-bit diagnosis could serve as the seed. |
…s, and PUA functional keys Follow-up to the salvaged NousResearch#89676 + NousResearch#90291 lock-bit fixes: extract a shared _lock_variants() helper and cover the sites both PRs missed - install_shift_enter_alias / install_ctrl_enter_alias / install_cmd_backspace_alias CSI-u spellings, legacy CSI-letter and CSI-tilde navigation twins derived from the existing table for ALL modifiers 1-16 (not just plain/shift), plain F1-F4 SS3 fallback, unmodified CSI-u keys (Tab/Enter/Space/Backspace), and kitty PUA functional keys (keypad, F13-F24, Ignore range) under lock bits. 8 new tests.
3 participants
What does this PR do?
When the CLI pushes the kitty keyboard protocol (disambiguate mode) to allowlisted terminals, kitty and ghostty encode the CapsLock/NumLock state as extra bits in the CSI-u modifier parameter: with NumLock on, Ctrl+C arrives as
ESC[99;133u(5 + 128) instead ofESC[99;5u. The alias table installed byinstall_modify_other_keys_aliases()only mapped the exact base modifiers, so with a lock enabled every key combo was unmapped and leaked as literal text ([127;133u[127;133u...), breaking Ctrl+C, Ctrl+Backspace word-movement, and all other combos — while gnome terminal (no kitty protocol) kept working.This installs every CSI-u alias with its lock-bit variants (base + 64 CapsLock, +128 NumLock, +192 both), mapped to the same
Keysvalue as the base form. The xterm modifyOtherKeys encoding (ESC[27;N;CP~) never carries lock bits, so that form is deliberately left untouched. The Esc-key registration also now covers modifier 1 (a lone Esc with NumLock on arrives asESC[27;129u).Related Issue
Fixes #89651
Type of Change
Changes Made
hermes_cli/pt_input_extras.py:_install_paired()now installs each CSI-u sequence with the four lock-bit variants (0/64/128/192); the modifyOtherKeys tilde form stays base-only. The Esc-key loop registers modifier 1..16 with the same variants. Docstring documents the new variants.How to Test
pytest tests/cli/test_modify_other_keys_aliases.py -q— 206 passed (34 new lock-bit tests).pytest tests/cli/ -qon branch vs stashed-main: identical failure sets apart from the new tests (remaining failures are pre-existing environment-dependent ones in both runs)._parse("\x1b[127;133u")now yields[Escape, ControlH](Ctrl+Backspace) instead of leaking[,1,2,7... as text.Checklist
Code
fix(scope):,feat(scope):, etc.)pytest tests/ -qand all tests passDocumentation & Housekeeping
docs/, docstrings) — or N/Acli-config.yaml.exampleif I added/changed config keys — or N/ACONTRIBUTING.mdorAGENTS.mdif I changed architecture or workflows — or N/A