diff options
| author | Bram Moolenaar <Bram@vim.org> | 2021-02-23 19:32:03 +0100 |
|---|---|---|
| committer | Bram Moolenaar <Bram@vim.org> | 2021-02-23 19:32:03 +0100 |
| commit | 2c6553498e790604f50016d8435403523a2576d6 (patch) | |
| tree | 809b32b0baedaa41e837ca2e6c4eedd6272e623d /src/normal.c | |
| parent | 21a83bd53d1fe21c780316c40fdc14d49167d08b (diff) | |
| download | vim-git-2c6553498e790604f50016d8435403523a2576d6.tar.gz | |
patch 8.2.2547: "%" command not accurate for big filesv8.2.2547
Problem: "%" command not accurate for big files.
Solution: Make it more accurate for files up to 21M lines. (Dominique Pellé,
closes #7889)
Diffstat (limited to 'src/normal.c')
| -rw-r--r-- | src/normal.c | 8 |
1 files changed, 5 insertions, 3 deletions
diff --git a/src/normal.c b/src/normal.c index 9fbfadfbc..d2f41593a 100644 --- a/src/normal.c +++ b/src/normal.c @@ -4769,9 +4769,11 @@ nv_percent(cmdarg_T *cap) { cap->oap->motion_type = MLINE; setpcmark(); - // Round up, so CTRL-G will give same value. Watch out for a - // large line count, the line number must not go negative! - if (curbuf->b_ml.ml_line_count > 1000000) + // Round up, so 'normal 100%' always jumps at the line line. + // Beyond 21474836 lines, (ml_line_count * 100 + 99) would + // overflow on 32-bits, so use a formula with less accuracy + // to avoid overflows. + if (curbuf->b_ml.ml_line_count >= 21474836) curwin->w_cursor.lnum = (curbuf->b_ml.ml_line_count + 99L) / 100L * cap->count0; else |