| Commit message (Collapse) | Author | Age |
| |
|
|
|
|
|
| |
Needs documentation (for the new name), but manages to move . to source,
while preserving compatibility.
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
finally pass.
https://github.com/fish-shell/fish-shell/issues/624
|
| |
|
|
|
|
| |
https://github.com/fish-shell/fish-shell/issues/575
|
|
|
|
| |
builtin_set_color.cpp and taught fish about it.
|
|\
| |
| |
| | |
#537 and #516
|
| |
| |
| |
| |
| |
| |
| | |
unless it fails
https://github.com/fish-shell/fish-shell/issues/547
https://github.com/fish-shell/fish-shell/issues/214
|
| | |
|
|/ |
|
| |
|
| |
|
|
|
|
| |
Stylistic tweaks
|
| |
|
|\
| |
| |
| | |
JanKanis-event-bug-test
|
| | |
|
| | |
|
| |
| |
| |
| |
| |
| |
| | |
https://github.com/fish-shell/fish-shell/issues/384
Word movement should be very similar to fish 1.x
backward-kill-word remains more liberal, but now stops at any of {,'"=}
|
| | |
|
| | |
|
| | |
|
|/ |
|
|
|
|
| |
https://github.com/fish-shell/fish-shell/issues/435
|
| |
|
| |
|
| |
|
|
|
|
| |
https://github.com/fish-shell/fish-shell/issues/29
|
|
|
|
| |
Fixes https://github.com/fish-shell/fish-shell/issues/80
|
| |
|
| |
|
|
|
|
|
|
| |
interpret them
https://github.com/fish-shell/fish-shell/issues/337
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
After updating, fish would immediately die when started in an xterm from
my window manager (wmii) with a message like:
fish: builtin.cpp:3357: int builtin_end(parser_t&, wchar_t**): Assertion `false' failed.
fish: Job 1, “~/code/fish-shell/fish” terminated by signal SIGABRT (Abort)
Starting fish from an existing shell worked, and running it on other
machines that I was only SSHing into worked as well. I did discover that
on one machine starting git from one specific directory
(/home/ian/bml/kernel) would cause this failure, but starting from any
other directory worked as normal. I'm not entirely sure what
circumstances contribute to this failure - starting from a copy of the
kernel directory works fine.
The failure only started with the following commit, which introduced the
assert(false):
commit d788c844408d25eb5f986d8c6e2b609270dbc153
Made type property of block_t constant and private
Further work towards cleaning up block_t hierarchy
Looking at this fail in gdb:
(gdb) r
Starting program: /home/ian/code/fish-shell/fish
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
fish: builtin.cpp:3357: int builtin_end(parser_t&, wchar_t**): Assertion `false' failed.
Program received signal SIGABRT, Aborted.
0x00007ffff6c82475 in *__GI_raise (sig=<optimized out>) at ../nptl/sysdeps/unix/sysv/linux/raise.c:64
64 ../nptl/sysdeps/unix/sysv/linux/raise.c: No such file or directory.
(gdb) bt
function=0x548520 "int builtin_end(parser_t&, wchar_t**)") at assert.c:81
(gdb) up 3
3357 assert(false); //should never get here
(gdb) p parser.current_block->type()
$1 = FAKE
So this happens when we run an end command for a FAKE block.
The below patch adds an empty case for FAKE blocks to avoid hitting the
assert. I would need to study the code in more detail to understand if
we should even be executing this code if we are in a FAKE block, but
this patch seems to solve the issue for me.
Signed-off-by: Ian Munsie <darkstarsword@gmail.com>
|
| |
|
| |
|
| |
|
|
|
|
| |
Removed an auto_ptr (yay)
|
|
|
|
| |
Further work towards cleaning up block_t hierarchy
|
| |
|
|
|
|
|
| |
Rewrite IO chains to be a vector of pointers, instead of a linked list
Removed io_transmogrify
|
| |
|
| |
|
| |
|
|
|
|
| |
mostly in terms of using size_t instead of int
|
|
|
|
| |
Fixes https://github.com/fish-shell/fish-shell/issues/250
|