not attached, the server process asks it to open the file, similar to
how works for stdin, stdout, stderr. This makes special files like
/dev/fd/X work (used by some shells). stdin, stdout and stderr and
control mode are now just special cases of the same mechanism. This will
also make it easier to use for other commands that read files such as
source-file.
but there is also now a global command queue. Instead of command queues
being dispatched on demand from wherever the command happens to be
added, they are now all dispatched from the top level server
loop. Command queues may now also include callbacks as well as commands,
and items may be inserted after the current command as well as at the end.
This all makes command queues significantly more predictable and easier
to use, and avoids the complex multiple nested command queues used by
source-file, if-shell and friends.
A mass rename of struct cmdq to a better name (cmdq_item probably) is
coming.
the state (client, session, winlink, pane) for it it before entering the
command. Each command provides some flags that tell the prepare step
what it is expecting.
This is a requirement for having hooks on commands (for example, if you
hook "select-window -t1:2", the hook command should to operate on window
1:2 not whatever it thinks is the current window), and should allow some
other target improvements.
The old cmd_find_* functions remain for the moment but that layer will
be dropped later.
Joint work with Thomas Adam.
around, we can't use file descriptors for the working directory because
we will be unable to pass it to a privileged process to tell it where to
read or write files or spawn children. So move tmux back to using
strings for the current working directory. We try to check it exists
with access() when it is set but ultimately fall back to ~ if it fails
at time of use (or / if that fails too).
use event_once to queue a callback to deal with them. Also dead clients
with references would never actually be freed because the wrap-up
functions (the callback for stdin, or status_prompt_clear) would never
be called. So call them in server_client_lost.
consistent but with much less duplication, but keeping the same internal
API. Also adds more readable aliases for some of the special tokens used
in targets (eg "{start}" instead of "^"). Some behaviours may have
changed, for example prefix matches now happen before fnmatch.
directly with a helper function in the cmd_entry, include a table of
bind-key commands and pass them through the command parser and a
temporary cmd_q.
As well as being smaller, this will allow default bindings to be command
sequences which will probably be needed soon.
much as before - buffers are automatically named "buffer0000",
"buffer0001" and so on and ordered as a stack. Buffers can be named
explicitly when creating ("loadb -b foo" etc) or renamed ("setb -b
buffer0000 -n foo"). If buffers are named explicitly, they are not
deleted when buffer-limit is reached. Diff from J Raynor.
descriptors rather than strings.
- Each session still has a current working directory.
- New sessions still get their working directory from the client that
created them or its attached session if any.
- New windows are created by default in the session working directory.
- The -c flag to new, neww, splitw allows the working directory to be
overridden.
- The -c flag to attach let's the session working directory be changed.
- The default-path option has been removed.
To get the equivalent to default-path '.', do:
bind c neww -c $PWD
To get the equivalent of default-path '~', do:
bind c neww -c ~
This also changes the client identify protocol to be a set of messages rather
than one as well as some other changes that should make it easier to make
backwards-compatible protocol changes in future.
commands and allow a command to block execution of subsequent
commands. This allows run-shell and if-shell to be synchronous which has
been much requested.
Each client has a default command queue and commands are consumed one at
a time from it. A command may suspend execution from the queue by
returning CMD_RETURN_WAIT and then resume it by calling cmd_continue() -
for example run-shell does this from the callback that is fired after
the job is freed.
When the command queue becomes empty, command clients are automatically
exited (unless attaching). A callback is also fired - this is used for
nested commands in, for example, if-shell which can block execution of
the client's cmdq until a new cmdq becomes empty.
Also merge all the old error/info/print functions together and lose the
old curclient/cmdclient distinction - a cmdq is bound to one client (or
none if in the configuration file), this is a command client if
c->session is NULL otherwise an attached client.
add a new value to mean "leave client running but don't attach" to fix
problems with using some commands in a command sequence. Most of the
work by Thomas Adam, problem reported by "jspenguin" on SF bug 3535531.
handling them in the server, handle them in the client and pass buffers
over imsg. This is much tidier for some upcoming changes and the
performance hit isn't critical.
The tty fd is still passed to the server as before.
This bumps the tmux protocol version so new clients and old servers are
incompatible.
Originally, tmux commands were parsed in the client process into a
struct with the command data which was then serialised and sent to the
server to be executed. The parsing was later moved into the server (an
argv was sent from the client), but the parse step and intermediate
struct was kept.
This change removes that struct and the separate parse step. Argument
parsing and printing is now common to all commands (in arguments.c) with
each command left with just an optional check function (to validate the
arguments at parse time), the exec function and a function to set up any
key bindings (renamed from the old init function).
This is overall more simple and consistent.
There should be no changes to any commands behaviour or syntax although
as this touches every command please watch for any unexpected changes.
stdin up to the server, I forgot one essential point - the tmux server
could now be both the producer and consumer. This happens when tmux is
run inside tmux, as well as when piping tmux commands together.
So, using stdio(3) was a bad idea - if sufficient data was written, this
could block in write(2). When that happened and the server was both
producer and consumer, it deadlocks.
Change to use libevent bufferevents for the client stdin, stdout and
stderr instead. This is trivial enough for output but requires a
callback mechanism to trigger when stdin is finished.
This relies on the underlying polling mechanism for libevent to work
with whatever devices to which the user could redirect stdin, stdout or
stderr, hence the change to use poll(2) over kqueue(2) for tmux.
