The user may write for example:
echo foo >&5
and fish would try to output to file descriptor 5, within the fish process
itself. This has unpredictable effects and isn't useful. Make this an
error.
Note that the reverse is "allowed" but ignored:
echo foo 5>&1
this conceptually dup2s stdout to fd 5, but since no builtin writes to fd
5 we ignore it.
The screen output code predates the ENCODE_DIRECT scheme, and
directly-encoded bytes were not being properly output. This manifested as
private-user characters being mangled when printed as part of fish_prompt.
Just use str2wcstring instead.
Fixes#7723
wcs2string_appending is like wcs2string, but appends to a std::string
instead of creating a new one. This will be more efficient when a string
can be reused, or if we want to accumulate multiple wcstrings into a
single std::string.
They are of variable length, taking semicolon-separated ASCII characters
and not single chars/bytes as the parameters. Additionally, the global
maximum size for a CSI is 16 characters (NPAR), even though I believe
the maximum possible mouse-tracking CSI is 12 characters.
fish maintains two tty modes: one for itself and one for external
commands. The external command mode is also used when executing
fish-script key bindings, which was added in 5f16a299a7 (note that
commit had the wrong issue, the correct issue is #2114).
Prior to this fix, when switching to external modes, we would also reset
the tty's foreground color. This bumped tty's timestamp, causing us to
believe that the tty had been modified, and then repainting the prompt. If
the prompt were multi-line, we would repaint the whole prompt starting
from its second line, leaving a trailing line above it.
It would be reasonable to save the tty timestamp after resetting the
color, but given that using external modes for keybindings is new, it's
better to instead not reset the color in this case. So migrate the color
resetting to only when we run external commands.
Fixes#7722
The default case for string literals like `"foo"` is a single trailing
nul, and that's what we have almost everywhere. By checking the
second-to-last index for a non-nul byte, we can skip the recursive
invocation, thus speeding up compilation that teeny, tinsy bit faster.
Rather than making the run-time complexity of the algorithm 𝒪(n) where n
is the length of the string, make it 𝒪(k) where k is the number of
trailing nul bytes.
The second parameter `index` with a default non-value is in lieu of a
helper function that would have had a name like `count_trailing_nuls()`.
e94f86e6d2 removed it in favor of using
fish_wcstod, but this broke the *output* - math currently prints
numbers with "," and then can't read them.
So we partially revert it until we come up with something better.
Maybe set $LC_NUMERIC globally inside fish?
fish_indent used to increment the indentation level whenever we saw an escaped
newline. This broke because of recent changes to parse_util_compute_indents().
Since parse_util_compute_indents() function already indents continuations
there is not much to do for fish_indent - we can simply query the indentation
level of the newline. Reshuffle the code since we need to pass the offset
of the newline. Maybe this can even be simplified further.
Fixes#7720
Bind \cc like normal, since we now no longer use a function, and bind
some important control bindings like \cs and the ever-important emacs \cb/f/p/n.
What really kills the usability here is the up-line vs up-or-search.
Since, unlike e.g. OPOST, this can sometimes be useful, just copy
whatever flow control settings the terminal ends up with.
We still *default* flow control to off (because it's an awful default
and allows us to bind ctrl-s), but if the user decides to enable it so
be it.
Note that it's _possible_ flow control ends up enabled accidentally, I
doubt this happens much and it won't render the shell unusable (and
good terminals might even tell you you've stopped the app).
Fixes#7704
Otherwise this would look weird if you had, say, a tab in there.
See #7716.
(note that this doesn't handle e.g. zero-width-joiners, because those
aren't currently escaped. we might want to add an escape mode for
unprintable characters, but for combining codepoints that's tricky!)
This added a space if only one character was added, e.g.
```fish
cd dev<TAB>
```
would complete to
```fish
cd dev/<SPACE>
```
which makes picking deeper directories awkward.
So just go back to the old behavior of doing it for any length.
This is a regression from e27d97b02e.
cc @krobelus
Similar to what fish_indent does. After typing "echo \" and hitting return,
the cursor will be indented.
A possible annoyance is that when you have multiple indented lines
echo 1 \
2 \
3 \
4 \
If you remove lines in the middle with Control-k, the lines below
the deleted one will start jumping around, as they are disconnected
from and reconnected to "echo".
If a variable is undefined, but it looks like it will be defined by the
current command line, assume the user knows what they are doing.
This should cover most real-world occurrences.
Closes#6654
After commit 6dd6a57c60, 3 remaining
builtins were affected by uint8_t overflow: `exit`, `return`, and
`functions --query`.
This commit:
- Moves the overflow check from `builtin_set_query` to `builtin_run`.
- Removes a conflicting int -> uint8_t conversion in `builtin_return`.
- Adds tests for the 3 remaining affected builtins.
- Simplifies the wording for the documentation for `set --query`.
- Does not change documentation for `functions --query`, because it does
not state the exit code in its API.
- Updates the CHANGELOG to reflect the change to all builtins.
This was lost in
6bdbe732e40c2e325aa15fcf0f28ad0dedb3a551..c7160d7cb4970c2a03df34547f357721cb5e88db.
Note that we only print a term-support flog message for now, the
warning seems a bit much.
Fixes#7709.
Prior to this fix, if stdin were explicitly closed, then builtins would
silently fail. For example:
count <&-
would just fail with status 1. Remove this limitation and allow each
builtin to handle a closed stdin how it sees fit.
In an interactive shell, typing "for x in (<RET>" would print an error:
fish: Expected end of the statement, but found a parse_token_type_t::tokenizer_error
Our tokenizer converts "(" into a special error token, hence this message.
Fix two cases by not reporting errors, but only if we allow parsing incomplete
input. I'm not really sure if this is necessary, but it's sufficient.
Fixes#7693
Prior to this change, if you pipe a builtin to another process, it would
be buffered. With this fix the builtin will write directly to the pipe if
safe (that is, if the other end of the pipe is owned by some external
process that has been launched).
Most builtins do not produce a lot of output so this is somewhat tricky to
reproduce, but it can be done like so:
bash -c 'for i in {1..500}; do echo $i ; sleep .5; done' |
string match --regex '[02468]' |
cat
Here 'string match' is filtering out numbers which contain no even digits.
With this change, the numbers are printed as they come, instead of
buffering all the output.
Note that bcfc54fdaa fixed this for the case where the
builtin outputs to stdout directly. This fix extends it to all pipelines
that include only one fish internal process.
Add compile-time checks to ensure list of string subcommands, builtins,
and electric variables are kept in asciibetical order to facilitate
binary search lookups.
This may slightly improve performance by allowing the compiler greater
visibility into what is happing on top of not executing at runtime in
some hot paths, but more importantly, it gets rid of magic constants in a
few different places.
These functions are called in the event queue hot path every time an
input event takes place. If we could guarantee a maximum length of
non-char (i.e. readline) events in the queue, we could use
`event_queue_peeker_t` with a fixed storage size of, e.g., 32 events,
but I'm not sure what a reasonable number would in fact be, so I'm just
changing these to use a thread-local vector that will re-use its
previous heap allocation in subsequent invocations rather than thrashing
the heap.
The lookups are executed on all input events, so they are worth
optimizing.
Cache the list of names, use binary search to get a function code from a
name, and stop enumerating mappings after `has_function` and `has_command`
have been determined.
builtin_set_query returns the number of missing variables. Because the
return value passed to the shell is an 8-bit unsigned integer, if the
number of missing variables is a multiple of 256, it would overflow to 0.
This commit saturates the return value at 255 if there are more than 255
missing variables.
builtin_test stashes some variables in statics, to support
the `test -t` expression. However this will cause conflicts with
concurrent execution, where we may want to run two `test` expressions at
once. Do the grunt work of threading the data into all places it needs
to go.
fish isn't quite sure what to do if the user specifies an fd redirection
for builtins. For example `source <&5` could potentially just read from
an arbitrary file descriptor internal to fish, like the history file.
fish has some lame code that tries to detect these, but got the sense
wrong. Fix it so that fd redirections for builtins are restricted to
range 0 through 2.
This introduces a new variable $fish_color_keyword that will be used
to highlight keywords. If it's not defined, we fall back on
$fish_color_command as before.
An issue here is that most of our keywords have this weird duality of
also being builtins *if* executed without an argument or with
`--help`.
This means that e.g.
if
is highlighted as a command until you start typing
if t
and then it turns keyword.
The iothread pool has a feature where, if the thread is emptied, some
threads will choose to wait around in case new work appears, up to a
certain amount of time (500 msec). This prevents thrashing where new
threads are rapidly created and destroyed as the user types. This is
implemented via `std::condition_variable::wait_for`. However this function
is not properly instrumented under Thread Sanitizer (see
https://github.com/google/sanitizers/issues/1259) so TSan reports false
positives. Just disable this feature under TSan.
fd_monitor_t allows observing a collection of fds. It also has its own
fd, which it uses to awaken itself when there are changes. Switch to
using fd_event_signaller_t instead of a pipe; this reduces the number of
file descriptors and is more efficient under Linux.
queues use std::deque under the hood which is more expensive than a vector.
We always consume the entire queue so there is no advantage to use deque here.
Just use a vector.
Replace the complicated implementation which shared a condition variable, with
one which just uses std::future<void>. This may allocate more condition
variables but is much simpler.
Fish was previously oblivious to the existence of mouse-tracking ANSI
escapes; this was mostly OK because they're disabled by default and we
don't enable them, but if a TUI application that turned on mouse
reporting crashed or exited without turning mouse reporting off, fish
would be left in an unusable state as all mouse reporting CSI sequences
would be posted to the prompt.
This can be tested by executing `printf '\x1b[?1003h'` at the prompt,
then clicking with any mouse button anywhere within the terminal window.
Previously, this would have resulted in seeming garbage being spewed to
the prompt; now, fish detects the mouse tracking CSIs posted to stdin by
the terminal emulator and a) ignores them to prevent invalid input, as
well as b) posts the CSI needed to disable future mouse tracking events
from being emitted on subsequent mouse interactions (until re-enabled).
Note that since we respond to a mouse tracking CSI rather than
pre-emptively disable mouse reporting, we do not need to do any sort of
feature detection to determine whether or not the terminal supports
mouse reporting (otherwise, if it didn't support it and we posted the
CSI anyway, we'd end up with exactly the kind of cruft posted to the
prompt that we're trying to avoid).
Fixes#4873
This is a stack-allocating utility class to peek up to N
characters/events out of an `event_queue_t` object. The need for a
hard-coded maximum peek length N at each call site is to avoid any heap
allocation, as this would be called in a hot path on every input event.
This allows directly inserting multiple characters/events in one go at
the front of the input queue, instead of needing to add them one-by-one
in reverse order.
In addition to improving performance in case of fragmented dequeue
allocation, this also is less error prone since a dev need not remember
to use reverse iterators when looping over a vector of peeked events.
Under non-Linux builds, binary_semaphore is implemented with a
self-pipe. When TSan is active we mark the pipe as non-blocking as TSan
cannot interrupt read (but can interrupt select). However we weren't
properly testing for EAGAIN leading to an assertion failure.
Allow looping on EAGAIN.
io_buffer_t is a buffer that fills itself by reading from a file
descriptor (typically a pipe). When the file descriptor is widowed, the
operation completes, and it reports completion by marking a
`std::promise<void>`. The "main thread" waits for this by waiting on the
promise's future. However TSan was reporting that the future's destructor
races with its promise's wait method. It's not obvious if this is valid,
but we can fix it by keeping the promise alive until the io_buffer_t is
deallocated.
This fixes the TSan issues reported under
`complete_background_fillthread_and_take_buffer` for #7681 (but there
are other unresolved issues).
This concerns how fish prevents its own fds from interfering with
user-defined fd redirections, like `echo hi >&5`. fish has historically
done this by tracking all user defined redirections when running a job,
and ensuring that pipes are not assigned the same fds. However this is
annoying to pass around - it means that we have to thread user-defined
redirections into pipe creation.
Take a page from zsh and just ensure that all pipes we create have fds in
the "high range," which here means at least 10. The primary way to do this
is via the F_DUPFD_CLOEXEC syscall, which also sets CLOEXEC, so we aren't
invoking additional syscalls in the common case. This will free us from
having to track which fds are in user-defined redirections.
fds.h will centralize logic around working with file descriptors. In
particular it will be the new home for logic around moving fds to high
unused values, replacing the "avoid conflicts" logic.
This needs to be rewritten, I'm pretty sure we have like 6 of these
kinds of ad-hoc "is this quoted" things lying around.
But for now, at least don't just check if the *previous* character was
a backslash.
Fixes#7685.
Previously we sometimes wanted to access an io_buffer_t to append to it
directly, but that's no longer true; all we really care about is its
separated_buffer_t. Make io_bufferfill_t::finish return the
separated_buffer directly, simplifying call sites. No user visible changes
expected here.
This concerns builtins writing to an io_buffer_t. io_buffer_t is how fish
captures output, especially in command substitutions:
set STUFF (string upper stuff)
Recall that io_buffer_t fills itself by reading from an fd (typically
connected to stdout of the command). However if our command is a builtin,
then we can write to the buffer directly.
Prior to this change, when a builtin anticipated writing to an
io_buffer_t, it would first write into an internal buffer, and then after
the builtin was finished, we would copy it to the io_buffer_t. This was
because we didn't have a polymorphic receiver for builtin output: we
always buffered it and then directed it to the io_buffer_t or file
descriptor or stdout or whatever.
Now that we have polymorphpic io_streams_t, we can notice ahead of time
that the builtin output is destined for an internal buffer and have it
just write directly to that buffer. This saves a buffering step, which is
a nice simplification.
While the user waits at the prompt, fish is waiting in select(), on stdin.
The sigio based universal notifier interrupts select() by arranging for a
signal to be delivered, which causes select() to return with EINTR.
However we weren't polling the notifier at that point so we would not
notice uvar changes, until we got some real input.
I didn't notice this when testing, because my testing was changing fish
prompt colors which updated the prompt for other reasons.
Fixes#7671.
Now that we have multiple clients of count_preceding_backslashes, factor
it out from fish_indent into wcstringutil.h, and then use the shared
implementation.
This goes to a separate file because that makes option parsing easier
and allows profiling both at the same time.
The "normal" profile now contains only the profile data of the actual
run, which is much more useful - you can now profile a function by
running
fish -C 'source /path/to/thing' --profile /tmp/thefunction.prof -c 'thefunction'
and won't need to filter out extraneous information.
Expansion parses slices like "$PATH[1..2]", but so does "set" when assigning
"set PATH[1..2] . .". Commit be06f842a ("Allow to omit indices in index
range expansions") forgot the latter.
Just like OPOST this just breaks output for anything not prepared for
it. Fish itself might work with it (and #4505 recommends it), but external commands are broken.
You'll see output like
foo
⏎
from `echo foo`.
Fixes#4873.
Continuation of #7133.
If given a windows path like `F:\foo`, this currently ends up
assert()ing in path_normalize_for_cd.
Instead, since these paths violate a bunch of assumptions we make, we
reject them and fall back on getting $PWD via getcwd() (which should
give us a nice proper unixy path).
Fixes#7636.
This isn't tested because it would require a system where a windowsy
path passes paths_are_same_file, and on the unix systems we run our
tests that's impossible as far as I can tell?
This used to print a literal DEL character in the output for `bind`,
which wouldn't actually show up and made it hard to figure out what
the key was.
So we just escape it back to how we actually used it - `\x7f`.
Fixes#7631.
This sometimes fails on github actions with ASAN. I am assuming that's
because the ctrl-c happens *before* the process has had a chance to
start.
So we do what we do and increase the delay.
These are a foreground and a background color. Now I see the point in
not naming them "foreground_color" and "background_color", but at
least "fg" and "bg" should do, right?
Prior to this change, histories were immortal and allocated with either
unique_ptr or just leaked via new. But this can result in races in the
path detection test, as the destructor races with the pointer-captured
history. Switch to using shared_ptr.
A weird interaction between grouped short options and our weird option
parsing that puts unknown options back:
```
echo "-n foo"
```
would see the `-n`, turn off printing newlines, interpret the " " as
another grouped short option, see that there is no short option for
space and put the entire token back on the arguments pile.
So it would print "-n foo" *without a newline*.
Fix this by keeping an old state of the options around and reverting
it when putting options back.
The alternative is *probably* to forbid the " " short option in
wgetopt, then check if an option group contains it and error out, but
this should only really be a problem in `echo` because that is,
AFAICT, the only thing that puts the options back.
Fixes#7614
When adding a command to history, we first expand its arguments to see
if any arguments are paths which refer to files. If so, we will only
autosuggest that command from history if the files are still valid. For
example, if the user runs `rm ./file.txt` then we will remember that
`./file.txt` referred to a file, and then only autosuggest that if the file
is present again.
Prior to this change we only performed simple expansion relative to the
working directory. This change extends it to variables and tilde
expansion. For example we will now apply the same hinting for
`rm ~/file.txt`
Fixes#7582
This removes the 100 msec timeout from io_buffer_t. We no longer need to
periodically wake up to check if a command substitution is finished,
because we get explicitly poked when that happens.
io_buffer_t is used to buffer output from a command substitution, so we
can split it into arguments. Typically io_buffer_t reads from its pipe
until it gets EOF and then stops reading. However it may be that the
cmdsub ends but EOF is not delivered because the stdout of the cmdsub
escaped with a background process.
Prior to this change we would wake up every 100 msec (select timeout) to
check if the cmdsub is finished. However this 100 msec adds latency if a
background process is launched from e.g. fish_prompt.
Switch to the new poke() function. Now when the cmdsub is finished, it
pokes its item, which explicitly wakes it up. This removes the extra
latency.
Fixes#7559
In preparation for fixing #7559, add a function poke_item to fd_monitor.
fd_monitor has a list of file descriptors, and invokes a callback when an
fd becomes readable. With this change, we assign each item a unique ID and
return it when the item is added; the ID may then be used to invoke the
callback explicitly.
The idea is that we can stop reading from the pipe associated with the
cmdsub when the job is finished, even if the pipe is still open.
This allows for multiple edits to be undone/redone in one go, as if they
were one edit.
Useful when a function is editing the commandline buffer via scripted
changes or via a keybinding so the internal changes to the buffer can be
abstracted away.
(Having extreme difficulty getting pexpect to play nice with the concept
of undo/redo...)
Currently binding `exit` to a key checks too late that it's exitted,
so it leaves the shell hanging around until the user does an execute
or similar.
As I understand it, the `exit` builtin is supposed to only exit the
current "thread" (once that actually becomes a thing), and the
bindings would probably run in a dedicated one, so the simplest
solution here is to just add an `exit` bind function.
Fixes#7604.
Prior to this change, `fish_private_mode` worked by just suppressing
history outright. With this change, `fish_private_mode` can be toggled on
and off. Commands entered while `fish_private_mode` is set are stored but
in memory only; they are not written to disk.
Fixes#7590Fixes#7589
Commands that start with a space should not be written to the history
file. Prior to this change, that was implemented by simply not adding them
to history. Items with leading spaces were simply dropped.
With this change, we add a 'history_persistence_mode_t' to
history_item_t, which tracks how the item persists. Items with leading
spaces are now marked as "ephemeral": they can be recovered via up arrow,
until the user runs another command, or types a space and hits return.
This matches zsh's HIST_IGNORE_SPACE feature.
Fixes#1383
Don't go into implicit interactive mode without ever executing
anything - not even `exit` or reacting to ctrl-d. That just renders
the shell useless and unquittable.
It was always a bit ridiculous that argparse required `X-longflag` if
that "X" short flag was never actually used anywhere.
Since the short letter is for getopt's benefit, we can hack around
this with our old friend: Unicode Private Use Areas.
We have a counter, starting at 0xE000 and going to 0xF8FF, that counts
up for all options that don't have a short flag and provides one. This
gives us up to 6400 long-only options.
6.4K should be enough for everybody.
A mildly interesting one is the call to test_wchar2utf8 with a non-null
pointer ("u1"/"dst") but 0 length. In this case we relied on malloc(0)
returning non-null which is not guaranteed.
src/fish_tests.cpp:1619:23: warning: Call to 'malloc' has an allocation
size of 0 bytes [clang-analyzer-optin.portability.UnixAPI]
mem = (char *)malloc(dlen);
^
test_wchar2utf8(w1, sizeof(w1) / sizeof(*w1), u1, 0, 0, 0,
"invalid params, dst is not NULL");
Prior to this change, a glob like `**/file.txt` would only match
`file.txt` in subdirectories; the `**` must match at least one directory.
This is historical behavior.
With this change we move a little closer to bash's implementation by
allowing a literal `**` segment to match in the current directory. That
is, `**/foo` will match both `foo` and `bar/foo`, while `b**/foo` will
only match `bar/foo`.
Fixes#7222.
Before running a command, or before importing a command from bash history,
we perform error checking. As part of error checking we expand commands
including variables and globs. If the glob is very large, like `/**`, then
we could hang expanding it.
One fix would be to limit the amount of expansion from the glob, but
instead let's just not expand command globs when performing error checking.
Fixes#7407
If the user types something like `/**`, prior to this change we would
attempt to expand it in the background for both highlighting and
autosuggestions. This could thrash your disk and also consume a lot of
memory.
Add a a field to operation_context_t to allow specifying a limit, and add
a "default background" limit of 512 items.
Historically fish has not supported tab completing or autosuggesting
wildcards with **. Prior to this fix, we would test every file match,
discover the ** wildcard, and then ignore it. Instead look for **
wildcards at the top level.
This prevents autosuggesting with /** from chewing up your disk.
