Require ModeSymlink when checking the SymlinkToDir marker as it shares a
bit with the other-execute permission, use a single helper for the check,
avoid stat() on hidden entries that will be skipped and treat symlinks to
directories as directories in save file path completion.
Recognize the existing directory-symlink marker when scanning, filtering
and deciding which results can be selected. Resolve symlinks before file
filters and cover directory selection and filtered navigation.
- Only restore entry sizes in the Splits layout resize mode was opened in;
otherwise fall back to the default reset rather than refusing to reset
- Do not take a snapshot lazily from already resized sizes when resize mode
was opened in another layout
- Running the resize kitten while its overlay is open replaces the overlay so
the new options take effect, keeping the original-size snapshot
- Newer edge statuses replace older local messages, and edge failures beep
- Switching strategy no longer discards a --fraction command-line override
- Compute divider room without deep copying the pair subtree
- Fractional resize in other layouts adopts the probed state instead of
replaying every step
- Parse in-band resize reports robustly: require the 48; prefix, accept
extra parameters and reject negative values instead of panicking
- Handle bracketed paste in parse_input_from_terminal with a dedicated
scan for the end of paste marker so pasted escape bytes are never lost
- Guard against NaN in scroll_by_pixels and scroll_to_absolute, clamp
negative absolute scroll targets, let infinite fractional scrolls clamp
and propagate overflow errors from PyLong_AsDouble
- Empty disk cache entries no longer claim offset 0 in the cache file and
defrag only moves entries actually written to disk
- Reject zero and too large block sizes in rsync signature headers
- Only replace the encryptor tag once it has been read successfully
Use Unicode word boundaries with the stdlib regexp engine
Go's regexp only treats ASCII characters as word characters for \b, while
regexp2 uses Unicode ones, so switching the builtin path pattern to the
stdlib engine changed which paths were marked next to non-ASCII text, for
example config.yaml文件 marked config.yaml. Match against a proxy of the
text in which every non-ASCII rune is replaced by a single ASCII byte that
is a word character iff the rune is one, then map the offsets back.
find_matches_in_lines was passing the exclusive end offset of a match to
find_pos, so a match ending at the end of the joined screen lines mapped
to no line and was dropped. It was not highlighted, not counted and n/N
skipped it. Since screen lines are not padded, this hit any match on the
last character of a line, for example searching for foo in "x := foo".
Look up the line holding the last matched byte instead.
regexp2 backtracks on the path pattern, retrying it at every position
of a long line that has no spaces, so the path and linenum types took
seconds on a screen full of such text (about 36s for a 380x100 window
of hex). The builtin patterns dont need lookarounds, so use the linear
time stdlib engine for them. User supplied --regex patterns stay on
regexp2 so lookarounds keep working (#6265).
word_diff_center filled an (m+1)*(n+1) table of ints for the differing
middle of every changed line pair. For a 100 KB line of minified
JavaScript with a word inserted near each end that is 2.6 GB and over
two seconds, and compute_centers runs the line pairs in parallel.
Changed words are only highlighted when they pair up by position, which
needs the same number of words on both sides, so return changed_center
right away when the counts differ. The output is the same as before in
that case. Lines with equal word counts but more than 2048 words between
the first and last changed word now get the central highlight too
instead of a huge table, and the table uses int32.
Reconcile the three ported kittens with 529 upstream commits. Only conflicts
were in the shrunk resize_window/broadcast/remote_file main.py stubs (upstream
reformatted their OPTIONS strings); resolved by keeping upstream's formatting
plus the Go-port stubs (shrunk main, remote_file handle_result shim, dropped
-h alias). Go build + kitten tests pass; dev build regenerates CLI cleanly;
Python suite at baseline (1 pre-existing unrelated failure).
main()'s error path after handle_action fails (e.g. malformed
--ssh-connection-data, or a ControlMaster connect failure) was missing the
reset_terminal() call Python's equivalent except-path makes
(kittens/remote_file/main.py pre-shrink: `except Exception: print(reset_terminal(),
...); traceback.print_exc(); show_error(...)`). Without it, the error rendered
over the stale ask-menu screen instead of a clean one. Added reset_terminal()
before show_error(), matching the other four call sites already added for
this kitten.
Verified: launched `kitten remote-file --path /nonexistent --hostname x
--ssh-connection-data 'not-json'` in a disposable kitty instance, selected an
action to trigger the parse failure, and confirmed via `get-text`/`get-text
--extent=all` that only the error message is present -- no ask-menu leftovers.
Adds kittens/remote_file/main.go (ask menu, hostname-mismatch prompt,
save-path prompt with tab completion, edit/open/save actions) on top of the
ControlMaster/ssh helpers from the previous two commits, registers it in
tools/cmd/tool/main.go, and adds remote_file to shell-integration/ssh/kitty's
wrapped_kittens list so both `kitty +kitten remote_file` and ssh-session
invocations exec the Go binary directly. main.py is shrunk to option_text
(kept importable by kitty/cli_stub.py and used for --doc/--help generation),
is_ssh_kitten_sentinel (kept importable by kitty/window.py), and the
handle_result shim boss.py runs in-process to open the returned file.
Also drops the dead `-h` alias on --hostname: Python's CLI layer always
special-cases -h/--help ahead of user options, so it never worked as a
hostname shorthand, but the Go CLI layer hard-errors on the collision; removing
it is a no-op for existing behavior (boss.py only ever passes --hostname) and
unblocks `kitten remote-file --help`. Also adds the same reset_terminal()
calls Python makes between UI phases (ask menu -> action, hostname-mismatch
prompt, overwrite prompt, $EDITOR handoff) so the Go and Python screens match.
Manually verified parity against Python (ask menu, cancel, open/edit/save
download+upload round-trips, hostname-mismatch Y/N, save-path prompt and
overwrite O/A/R/N) using a fake-ssh test double, and confirmed post-wrap that
`kitty +kitten remote_file` execs the Go kitten binary.