Commit graph

8429 commits

Author SHA1 Message Date
Mahmoud Al-Qudsi
2f2a221c56 Don't spawn keepalive for WSL when only one command
This should speed things up on slower PCs given that the vast majority
of shell commands are simple jobs consisting of a single command without
any pipelines, in which case there's no need for a keepalive process at
all. Applies to WSL only.
2018-03-04 21:54:12 -06:00
Mahmoud Al-Qudsi
000892e315 Use constexpr for is_windows_subsystem_for_linux()
To guarantee that at runtime there will be no branching, using a CMAKE
test/define combined with a constexpr wrapper for code-friendliness.
2018-03-04 21:13:31 -06:00
Mahmoud Al-Qudsi
cf8850a33f Add temporary fix for #4778 (background processes on WSL)
As a temporary workaround for the behavior described in
Microsoft/WSL#2997 wherein WSL does not correctly assign the spawned
child its own PID as its PGID, explicitly set the PGID for the newly
spawned process.
2018-03-04 20:18:38 -06:00
Mahmoud Al-Qudsi
912dbd85d8 Sprinkle in some job control asserts 2018-03-04 20:17:26 -06:00
Mahmoud Al-Qudsi
71329a250b [cmake] Don't create installation directories that already exist
fish's cmake install routines were attempting to create system
directories that already existed, an operation for which the permissions
to do so may not be available (e.g. /usr/local/share/pkgconfig)

This commit first checks if a directory exists before creating it.
2018-03-04 15:46:29 -06:00
Mahmoud Al-Qudsi
9502ea8de3 Add tests for jobs -q 2018-03-04 15:33:36 -06:00
Mahmoud Al-Qudsi
7242c40b5c Document new -q/--quiet flags to jobs builtin 2018-03-04 15:21:31 -06:00
Mahmoud Al-Qudsi
6814f29cd7 Add a -/--quiet option to jobs for silent evaluation 2018-03-04 15:20:14 -06:00
David Adam
99a34fe841 debian packaging: switch to CMake builds 2018-03-04 23:07:33 +08:00
David Adam
5beeb5280f license.hdr: update MIT license for Angular but not MuParser
[ci skip]
2018-03-04 19:41:30 +08:00
Fabian Homborg
0da8022081 [math] Fail without arguments
See #4768.
2018-03-01 22:27:24 +01:00
ridiculousfish
b49e1d7703 Minor cleanup
Mark a function as static and use a std::string instead of malloc.
2018-03-01 21:35:49 +01:00
Fabian Homborg
fd6a1a436b Merge branch 'tinyexpr-rebase'
This replaces muparser with tinyexpr, which

- Saves about 4000 lines
- Removes functionality we do not use, making it so we can document _all_ of it
- Should be more palatable to distributions
2018-03-01 13:10:08 +01:00
Fabian Homborg
84043e425d [math] Keep LC_NUMERIC=C for output as well
Otherwise, we'd have issues with people putting math's output back
into math because of "," vs ".".
2018-03-01 13:09:35 +01:00
Fabian Homborg
f20b6a1e8f [math] Document scientific notation 2018-03-01 13:09:35 +01:00
Fabian Homborg
642a1db4fc [math] Add two error strings as fallback 2018-03-01 13:09:35 +01:00
Fabian Homborg
f60e1549a9 [math] Better error for 2 + 2 4
This now reports "TOO_MANY_ARGS" instead of no error (and triggering
an assertion).

We might want to add a new error type or report the missing operator
before, but this is okay for now.
2018-03-01 13:09:35 +01:00
Fabian Homborg
6a38eb4f7d [math] Better error for 2 -
Instead of the catch-all, report too few arguments.
2018-03-01 13:09:35 +01:00
Fabian Homborg
20af969aba [math] Adjust tests for new error reporting
This shows how it's still weird - "2 - " and "max()" return the same
error.
2018-03-01 13:09:35 +01:00
Fabian Homborg
0d9123ec91 [math] Return STATUS_CMD_ERROR when an error occured 2018-03-01 13:09:35 +01:00
Fabian Homborg
18bc2b0f4a Add tinyexpr to autotools build 2018-03-01 13:09:35 +01:00
Fabian Homborg
d90d0ee08e [tinyexpr] Let specific errors take precedence over generic ones
Fixes the case where `sin()` reported the generic "bogus" error
instead of "too few arguments".

Also rename the constant to "TE_ERROR_UNKNOWN".
2018-03-01 13:09:35 +01:00
Fabian Homborg
4f39cc4d82 [tinyexpr] Remove closures
These are only available as a customization point. Since we don't use
that, it's dead code.
2018-03-01 13:09:35 +01:00
Fabian Homborg
4f56ce6d33 [math] Improve error formatting
This looked like

    math: Error in expression
    'sin(5,4)'
          ^
    Too many arguments

Now it looks like

    math: Too many arguments
    'sin(5,4)'
          ^
2018-03-01 13:09:35 +01:00
Fabian Homborg
0fa0b620b3 [tinyexpr] Remove list()
This just read comma-separated expressions and returned the last of
them, which is useless and confusing.
2018-03-01 13:09:35 +01:00
Fabian Homborg
ea2934e644 [tinyexpr] Make unary functions fail with too many arguments
As it turns out, this was, for some reason, actual wanted behavior.

We _don't_ want it, so just remove the code.
2018-03-01 13:09:35 +01:00
Fabian Homborg
7b9c75094c [tinyexpr] Add error handling
This turns a bunch of ifs on their heads.

We often see this pattern in te:

```c
if (s->type != SOME_TYPE) {
   // error handling
}  else {
   // normal code
}
```

Only, since we want to return the first error, we do

```c
if (s->type == SOME_TYPE) {
    // normal code
} else if (s->type != TOK_ERROR) {
    // Add a new error - if it already has type error
    // this should already be handled.
}
```

One big issue is the comma operator, that means arity-1 functions can
take an arbitrary number of arguments. E.g.

```fish
math "sin(5,9)"
```

will return the value of sin for _9_, since this is read as "5 COMMA
9".
2018-03-01 13:09:35 +01:00
Fabian Homborg
b9a3938a2b [tinyexpr] Add change notice 2018-03-01 13:09:35 +01:00
Fabian Homborg
2ab1466c8f [tinyexpr] Remove #ifdefs
We don't use them, so we don't need them.
2018-03-01 13:09:35 +01:00
Fabian Homborg
ce28891c76 [math] Set LC_NUMERIC to C
This allows us to always use "." as radix character, so e.g.

    math 2.5 - 2

is always valid, regardless of locale.
2018-03-01 13:09:35 +01:00
Fabian Homborg
234ebb5d7b [math] Fix docs
We no longer use muparser, but tinyexpr.

tinyexpr does not have:

- "Statistical functions" like min, max, avg

- Multiple expressions separated with ","

[ci skip]
2018-03-01 13:09:35 +01:00
Fabian Homborg
b75c3b968c Replace muparser with tinyexpr 2018-03-01 13:09:35 +01:00
George Christou
6f1ae79a13 alias: Fix string args being parsed as options 2018-02-28 11:17:00 +01:00
ridiculousfish
973533e374 Teach alias about wrap argument injection
Update the alias function to pass arguments to 'wraps'. For example
alias gco='git checkout' now works like it ought to.
2018-02-27 14:12:44 -08:00
ridiculousfish
bb7b649132 Wrapping completions to allow injecting arguments
This enables some limited use of arguments for wrapping completions. The
simplest example is that complete gco -w 'git checkout' now works like
you would want: `gco <tab>` now invokes git's completions with the
`checkout` argument prepended.

Fixes #1976
2018-02-27 14:12:44 -08:00
Fabian Homborg
c793570f2c Fix punctuation movement with one char tokens
Previously, in

    ls ^a bcd

(with "^" as the cursor), kill-word would delete the "a" and then go
on, remove the space and the "bcd".

With this, it will only kill the "a".

Fixes #4747.
2018-02-27 22:56:15 +01:00
David Adam
2747945c55 [cmake] don't try too hard to create extra vendor directories
Homebrew and other systems set the path for the extra completion,
function and configuration directories outside the writeable prefix.

Mirror the autotools build in trying to create these directories, but
not causing the whole install to fail if this operation in unsuccessful.
2018-02-27 20:42:58 +08:00
ridiculousfish
5282d3e711 Add fish_emoji_width variable to control computed emoji width
This is part of an effort to improve fish's Unicode handling. This commit
attempts to grapple with the fact that, certain characters (principally
emoji) were considered to have a wcwidth of 1 in Unicode 8, but a width of
2 in Unicode 9.

The system wcwidth() here cannot be trusted; terminal emulators do not
respect it. iTerm2 even allows this to be set in preferences.

This commit introduces a new function is_width_2_in_Uni9_but_1_in_Uni8() to
detect characters of version-ambiguous width. For these characters, it
returns a width guessed based on the value of TERM_PROGRAM and
TERM_VERSION, defaulting to 1. This value can be overridden by setting the
value of a new variable fish_emoji_width (presumably either to 1 or 2).

Fixes #4539, #2652.
2018-02-25 23:38:10 -08:00
ridiculousfish
7bd4af51a1 Switch to Unicode 9 savvy wcwidth
Previously fish used the venerable wcwidth implementation from Markus Kuhn.
This switches to wcwidth9() from https://github.com/joshuarubin/wcwidth9
2018-02-25 23:12:37 -08:00
Fabian Homborg
69f68d31df Reserve some builtin names
`argparse`, `read`, `set`, `status`, `test` and `[` now can't be used
as function names anymore.

This is because (except for `test` and `[`) there is no way to wrap these properly, so any
function called that will be broken anyway.

For `test` (and `[`), there is nothing that can be added and there
have been confused users who created a function that then broke
everything.

Fixes #3000.
2018-02-25 21:29:24 +01:00
Xavier Lepaul
5648322993 Add required argument to history delete (#4740) 2018-02-25 16:46:25 +01:00
Kevin Konrad
8536a6825f add completion for MacPorts 2018-02-24 23:32:25 +08:00
ridiculousfish
99200d3bfb Attempt to fix the Linux Travis build 2018-02-23 21:38:48 -08:00
ridiculousfish
c4d903ff98 Rationalize how the parser reports tokenizer errors
Remove the unnecessary SQUASH_ERROR flag and correctly report errors
generated from the tokenizer.
2018-02-23 17:28:12 -08:00
ridiculousfish
0950c35eb2 Reduce the amount of copying when the parser drives the tokenizer 2018-02-23 15:58:13 -08:00
ridiculousfish
99fb7bb6aa Refactor how redirections are represented by the tokenizer
Prior to this fix, each redirection type was a separate token_type.
Unify these under a single type TOK_REDIRECT and break the redirection
type out into a new sub-type redirection_type_t.
2018-02-23 15:19:58 -08:00
ridiculousfish
6673fe5457 Clean up tokenizer implementation
Rather than storing a bunch of "next_foo" fields, simply populate the
tok_t directly.
2018-02-23 14:31:13 -08:00
Steely Wing
e9a4875a6b Remove the duplicate "(" (#4748) 2018-02-23 18:19:19 +01:00
Mahmoud Al-Qudsi
1b27c97bfb [cmake] Fix dependency path for FBVF for fish.pc rule
The custom command for fish.pc had a dependency on FBVF, but it appears
that the relative path to FBVF was incorrect and with CMake 3.10.1 under
FreeBSD this was consistently causing the build to fail if
../build_tools/git_version_gen.sh hadn't (coincidentally, I think?)
already run.

Explicitly set the dependency path for FBVF to the binary directory.
2018-02-21 13:58:56 -06:00
Mahmoud Al-Qudsi
9e1576bdc4 Revert "[cmake] Add rule to generate FBVF"
This reverts commit e4c59ac60a.

I hadn't seen the FBVF rule in the `Version` CMake file. There's
something else going on here.
2018-02-21 13:50:05 -06:00