Prior to this fix, if you exported a variable in one scope
and then unexported it in the next, it would remain exported.
Example:
set -gx VAR 1
function foo; set -l VAR; env; end
foo
Here 'VAR' would be exported to 'env' because we failed to
notice that the env var is shadowed by an unexported variable.
This occurred at env var computation time, not in env_set!
Fixes#2132
su does not reset XDG_RUNTIME_DIR, which means that XDG_RUNTIME_DIR
may point to directories that the user does not have permission
to access. Similarly there is no guarantee that XDG_RUNTIME_DIR
points to a directory that actually exists. Rather than try to
handle these issues, we simply ignore them, effectively disabling
realtime uvar notifications. Fixes#1955.
In FAQ:
> I'm seeing weird output before each prompt when using screen. What's wrong?
The command provided is
echo 'function fish_title;end' > ~/.config/fish/config.fish
Using `>` will overwrite current config.fish.
We should use `>>` instead.
With the fix for #365, fish_command_not_found event handlers
receive the command and all of its arguments. But commands
like /usr/lib/command-not-found expect only the command name.
So when invoking an external command, just pass the command
name, not all of the arguments.
Before running a command, we add the command to history, so
that if the command causes us to exit it's still captured in
history. But that command should not be considered part of
history when expanding the history within the command itself.
For example, `echo $history[1]` should be the previously
run command, not `echo $history[1]` itself.
Fixes#2028
For the case
```
bind \et "commandline -i 1" "commandline -i 2"
```
the order of execution of the commands is now in-order.
Note that functions codes are prepended to the queue in reverse order, so they
will be executed in-order. This should allow all bindings of the form
```
bind \et beginning-of-line force-repaint
```
to remain unchanged.
Using builtin `commandline -f`, one would expect to have commands executed in
the order that they were given. This motivates the change to a queue.
Unfortunately, fish internals still need lookahead_list to act as a stack. Add
and rename functions to support both cases and have lookahead_list as
a std::deque internally.
This code is delicate, and we should probably dog-food this in nightly for
a while before the next-minor release.
Fixes#1567
Examples that work as expected (even completions don't get confused):
$ begin true; end;
$ begin if true; end; end
$ begin if true; echo hi; end
The last example correctly expects another 'end' to match 'begin'.
Fixes#1248.
Do not tombstone a function when it is evicted normally from the LRU cache.
This broke changing `fish_function_path`, since that would evict all nodes,
resulting in accidental tombstones, which caused autoloaded functions to
never be reloaded.
See #213.
In 73f344f41b, we allowed autoloaded functions to be deleted.
For some reason, funcsave immediately deletes the function it
creates. This previously did very little, since the function would
immediately be re-autoloaded, but with the fix for 73f344f41b
the function gets tombstoned. So the effect is that funcsave
makes the function disappear! This simply removes the erase call,
which dates back to fish 1.x.
