Try to reduce the UTF-8 mess.
Get rid of passing around u_char[4]s and define a struct utf8_data which has
character data, size (sequence length) and width. Move UTF-8 character
collection into two functions utf8_open/utf8_append in utf8.c which fill in
this struct and use these functions from input.c and the various functions in
screen-write.c.
Space for rather more data than is necessary for one UTF-8 sequence is in the
utf8_data struct because screen_write_copy is still nasty and needs to reinject
the character (after combining) into screen_write_cell.
UTF-8 combined character fixes.
Thai can have treble combinations (1 x width=1 then 2 x width=0) so bump the
UTF-8 cell data size to 9 and alter the code to allow this.
Also break off the combining code into a separate function, handle any further
combining beyond the buffer size by replacing the character with _s, and when
redrawing the UTF-8 character don't assume the first part has just been
printed, redraw the entire line.
Move the check for whether to force a line wrapper lower down into the tty code
where it has access to the tty width, which is what should have been checked.
Instead of having a complicated check to see if the cursor is in the last
position to avoid an explicit wrap, actually move it there.
Some UTF-8 fixes to come.
When checking whether the region will scroll and the cursor position is thus
unsuitable for using CUD/CUU, check the current cursor position not the target
position.
When drawing lines that have wrapped naturally, don't force a newline but
permit them to wrap naturally again. This allows terminals that use this to
guess where lines start and end for eg mouse selecting (like xterm) to work
correctly.
This was another long-standing issue raised by several people over the last
while.
Thanks to martynas@ for much testing. This was not trivial to get right so
bringing it in for wider testing and adn to fix any further glitches in-tree.
Don't run through the column unchanged case if the row was unchanged but there
were no suitable optimisations, instead make it an else to fall through to
absolute addressing.
Use relative cursor movement instead of absolute when possible and when
supported by the terminal to reduce the size of the output data (generally
about 10-20%).
Like linefeed, don't set the scroll region for reverse index unless it will be
needed.
While here, also tidy up a couple of long lines and remove an extraneous blank.
Be less aggressive about turning the cursor off, only explicitly turn it off
when tmux is redrawing, otherwise leave in the state set by the application.
Support -c like sh(1) to execute a command, useful when tmux is a login
shell. Suggested by halex@.
This includes another protocol version increase (the last for now) so again
restart the tmux server before upgrading.
Remove the internal tmux locking and instead detach each client and run the
command specified by a new option "lock-command" (by default "lock -np") in
each client.
This means each terminal has to be unlocked individually but simplifies the
code and allows the system password to be used to unlock.
Note that the set-password command is gone, so it will need to be removed from
configuration files, and the -U command line flag has been removed.
This is the third protocol version change so again it is best to stop the tmux
server before upgrading.
Trim some code by moving the ioctl(TIOCGWINSZ) after SIGWINCH from the client
into the server.
This is another (the second of four) protocol version changes coming this
morning, so again the server should be killed before upgrading.
Don't attempt to open() the tty path, rely on the client sending its stdin fd
with imsg and fatal if it doesn't, then set the FD_CLOEXEC flag in tty_init
instead of tty_open to prevent them leaking into child processes if any are
created between the two calls.
This bumps the protocol version, so the tmux server should be killed before
upgrading.
Be more careful about what flags are cleared when opening the terminal,
otherwise the opened/started flags are cleared and the terminal never released.
Add a new display-panes command, with two options (display-panes-colour and
display-panes-time), which displays a visual indication of the number of each
pane.
Send SGR0 when initialising the screen. Fixes problems on terminals with BCE
(like putty) if the background colours is non-default when tmux starts. May
also fix problems when resuming a suspended tmux.
A tty context must not be modified as it may be reused to update multiple
clients, so make it const.
Also fix an actual modification which caused a hang when a session was
connected to multiple terminals at least one of which was missing ich/ich1.
Initialise log_fd to -1, prevents spurious disconnection of the client when it
ends up as fd 0 (likely if the server is started with "tmux start").
Also add some extra debugging messages to server.c.
Have the client pass its stdin fd to the server when identifying itself and
have the server use that rather than reopening the tty. If the fd isn't given,
use the old behaviour (so no need for a version change).
This allows tmux to be used as the shell, so also change so that when working
out the command to execute if default-command is empty (the default), tmux will
try not execute itself.
Change the way the grid is stored, previously it was:
- a two-dimensional array of cells;
- a two-dimensional array of utf8 data;
- an array of line lengths.
Now it is a single array of a new struct grid_line each of which represents a
line and contains the length and an array of cells and an array of utf8 data.
This will make it easier to add additional per-line members, such as flags.
If colours are not supported by the terminal, try to emulate a coloured
background by setting or clearing the reverse attribute.
This makes a few applications which don't use the reverse attribute themselves
a little happier, and allows the status, message and mode options to have
default attributes and fg/bg options that work as expected when set as reverse.
Add a terminal-overrides session option allowing individual terminfo(5) entries
to be overridden. The 88col/256col checks are now moved into the default
setting and out of the code.
Also remove a couple of old workarounds for xterm and rxvt which are no longer
necessary (tmux can emulate them if missing).
enum tty_cmd is only used as an index into the array of command function
pointers, so remove it and use the function pointers directly to represent
themselves.
normal eight-bit output is wrong, separate it into a different function. Fixes
spacing when mixing UTF-8 with some escape sequences, notably the way w3m does
it.
redraw the entire window. Also add a flag to skip updating the window any
further if it is scheduled to be redrawn. This has the effect of batching
multiple redraws together.
issues - particularly, don't mix with manual pane resizing and be careful when
viewing from multiple clients; generally cycling the layout a few times will
fix most problems. Getting this in for testing while I think about how to deal
with manual mode.
Split window as normal and cycle the layouts with C-b space. Some of the
layouts will work better when swap-pane comes along.
as UTF-8 in a separate array, the code does a lookup into this every time it
gets to a UTF-8 cell. Zero width characters are just appended onto the UTF-8
data for the previous cell. This also means that almost no bytes extra are
wasted non-Unicode data (yay).
Still some oddities, such as copy mode skips over wide characters in a strange
way, and the code could do with some tidying.
Split grid into two arrays, one containing grid attributes/flags/colours (keeps
the name grid_cell for now) and a separate with the character data (called
text). The text is stored as a u_short but is treated as a uint64_t elsewhere;
eventually the grid will have two arrays.
I'm not happy with the naming so that might change.
Still need to decide where to go from here. I'm not sure whether to combine
the peek/set functions together, and also whether to continue to treat the
text as a uint64_t (and convert to/from Unicode) or make it a char array
(of size one when UTF-8 disabled, eight when enabled) and keep everything
as UTF-8.
Also since UTF-8 will eventually become an attribute of the grid itself it
might be nice to move all the padding crap into grid.c.
which may involve changing the cursor position; however, the old (before
redraw) cursor position is necessary for writing to the tty. So, save it before
doing the redraw then update the internal screen then update the tty.
Not sure I like this solution but it does the job for now.
Tiago Cunha