diff options
Diffstat (limited to 'ime.c')
| -rw-r--r-- | ime.c | 19 |
1 files changed, 11 insertions, 8 deletions
@@ -194,6 +194,8 @@ done(void *data, struct zwp_text_input_v3 *zwp_text_input_v3, cell_idx += width; } + const size_t byte_len = strlen(seat->ime.preedit.pending.text); + /* Pre-edit cursor - hidden */ if (seat->ime.preedit.pending.cursor_begin == -1 || seat->ime.preedit.pending.cursor_end == -1) @@ -206,6 +208,15 @@ done(void *data, struct zwp_text_input_v3 *zwp_text_input_v3, term->ime.preedit.cursor.end = -1; } + else if (seat->ime.preedit.pending.cursor_begin == byte_len && + seat->ime.preedit.pending.cursor_end == byte_len) + { + /* Cursor is *after* the entire pre-edit string */ + term->ime.preedit.cursor.hidden = false; + term->ime.preedit.cursor.start = cell_count; + term->ime.preedit.cursor.end = cell_count; + } + else { /* * Translate cursor position to cell indices @@ -220,16 +231,8 @@ done(void *data, struct zwp_text_input_v3 *zwp_text_input_v3, * * When we find the matching *byte* index, we at the same * time know both the unicode *and* cell index. - * - * Note that this has only been tested with - * - * cursor_begin == cursor_end == 0 - * - * I haven't found an IME that requests anything else */ - const size_t byte_len = strlen(seat->ime.preedit.pending.text); - int cell_begin = -1, cell_end = -1; for (size_t byte_idx = 0, wc_idx = 0, cell_idx = 0; byte_idx < byte_len && |
