1
0
mirror of https://github.com/tmate-io/tmate.git synced 2024-12-30 18:49:06 +01:00
Commit Graph

168 Commits

Author SHA1 Message Date
Tiago Cunha
c643ac4827 Sync OpenBSD patchset 420:
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.
2009-10-23 17:13:10 +00:00
Tiago Cunha
97e02eae7d Sync OpenBSD patchset 419:
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.
2009-10-23 17:11:26 +00:00
Tiago Cunha
2f14ef61ff Sync OpenBSD patchset 416:
Always move the cursor position on !xenl terminals, since there is no invisible
last cursor position.

Also nuke an unused variable.
2009-10-23 17:08:30 +00:00
Tiago Cunha
2e39ab59d7 Sync OpenBSD patchset 415:
Don't print wide characters at screen width - 1. Matches uterm behaviour and
is probably a better idea anyway.
2009-10-23 17:07:18 +00:00
Tiago Cunha
ac4e4a2b6c Sync OpenBSD patchset 414:
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.
2009-10-23 17:06:23 +00:00
Nicholas Marriott
1af09d6330 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.
2009-10-20 19:18:28 +00:00
Nicholas Marriott
62f234ce3b 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.
2009-10-20 17:33:33 +00:00
Nicholas Marriott
387f4d42cc 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.
2009-10-20 16:32:23 +00:00
Nicholas Marriott
daa26079ee Always move the cursor position on !xenl terminals, since there is no invisible
last cursor position.

Also nuke an unused variable.
2009-10-17 08:35:38 +00:00
Nicholas Marriott
fe26b5d25f Don't print wide characters at screen width - 1. Matches uterm behaviour and
is probably a better idea anyway.
2009-10-17 08:32:18 +00:00
Nicholas Marriott
43d62c1ae3 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.
2009-10-17 08:24:46 +00:00
Tiago Cunha
6103628129 Sync OpenBSD patchset 401:
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.
2009-10-15 01:41:14 +00:00
Tiago Cunha
cbd3b1bc9b Sync OpenBSD patchset 400:
When backspace is received at the beginning of a line and the previous line was
wrapped, move the cursor back up to the end of the previous line.

Another one of the forgotten persons requested this quite a while ago (I need
to start noting names on todo items...) when it was quite hard to
implement. Now it is easy and I don't see it can do any harm, so hey presto...
2009-10-15 01:39:30 +00:00
Tiago Cunha
7af01a4afb Sync OpenBSD patchset 397:
Wrap a couple of long lines.
2009-10-15 01:35:35 +00:00
Tiago Cunha
8ab3a3d9aa Sync OpenBSD patchset 394:
Permit attributes to be turned off in #[] by prefixing with "no", for example
"noblink".
2009-10-15 01:31:38 +00:00
Nicholas Marriott
d7626cd9d7 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.
2009-10-12 17:19:47 +00:00
Nicholas Marriott
8608c6970d When backspace is received at the beginning of a line and the previous line was
wrapped, move the cursor back up to the end of the previous line.

Another one of the forgotten persons requested this quite a while ago (I need
to start noting names on todo items...) when it was quite hard to
implement. Now it is easy and I don't see it can do any harm, so hey presto...
2009-10-12 16:59:55 +00:00
Nicholas Marriott
56157444de Wrap a couple of long lines. 2009-10-12 16:33:39 +00:00
Nicholas Marriott
33ae063cae Permit attributes to be turned off in #[] by prefixing with "no", for example
"noblink".
2009-10-12 11:08:02 +00:00
Tiago Cunha
c507bf25de Sync OpenBSD patchset 328:
Stick line length to what is actually used (removing an optimization that
allowed it to be bigger), and use clear line/EOL sequences rather than spaces
in copy/scroll mode.

This fixes xterm copy/paste from tmux which treats trailing spaces differently
from clearing a line with the escape sequences. Reported by martynas@.
2009-09-15 23:54:57 +00:00
Nicholas Marriott
4278199101 Stick line length to what is actually used (removing an optimization that
allowed it to be bigger), and use clear line/EOL sequences rather than spaces
in copy/scroll mode.

This fixes xterm copy/paste from tmux which treats trailing spaces differently
from clearing a line with the escape sequences. Reported by martynas@.
2009-09-15 15:14:09 +00:00
Tiago Cunha
0ec1ce005c Sync OpenBSD patchset 322:
Permit options such as status-bg to be configured using the entire 256 colour
palette by setting "colour0" to "colour255".
2009-09-11 14:13:52 +00:00
Nicholas Marriott
372a8cb1d9 Permit options such as status-bg to be configured using the entire 256 colour
palette by setting "colour0" to "colour255".
2009-09-10 17:16:24 +00:00
Tiago Cunha
5edc465802 Sync OpenBSD patchset 317:
Permit embedded colour and attributes in status-left and status-right using new
#[] special characters, for example #[fg=red,bg=blue,blink].
2009-09-07 23:37:48 +00:00
Nicholas Marriott
e97006b102 Permit embedded colour and attributes in status-left and status-right using new
#[] special characters, for example #[fg=red,bg=blue,blink].
2009-09-07 10:49:32 +00:00
Tiago Cunha
1fa8a8f4a6 Sync OpenBSD patchset 285:
Ugh, committed the wrong version of this change and got both solutions rather
than just the second. Remove unused assignment.
2009-08-21 21:13:20 +00:00
Tiago Cunha
4b883524d8 Sync OpenBSD patchset 282:
A few trivial optimisations: no need to check for zero size if calling
buffer_ensure in buffer.c; expand grid lines by a greater increase than one
each time; and don't read UTF-8 data unless it actually needs to be checked
when overwriting a cell.
2009-08-21 21:09:13 +00:00
Nicholas Marriott
65ac8e9f0c Ugh, committed the wrong version of this change and got both solutions rather
than just the second. Remove unused assignment.
2009-08-21 08:12:05 +00:00
Nicholas Marriott
1501b3fbbd A few trivial optimisations: no need to check for zero size if calling
buffer_ensure in buffer.c; expand grid lines by a greater increase than one
each time; and don't read UTF-8 data unless it actually needs to be checked
when overwriting a cell.
2009-08-20 19:14:42 +00:00
Tiago Cunha
78c789e1cb Sync OpenBSD patchset 252:
It was originally intended that scroll mode would show content that was
currently off-screen due to resize, but somewhere along the way this got
lost. Restore this behaviour to scroll mode by fixing screen_write_copy to read
up to the saved line length rather than the current screen width. Copy mode
remains unaltered for now.
2009-08-14 21:32:38 +00:00
Nicholas Marriott
b02e429788 It was originally intended that scroll mode would show content that was
currently off-screen due to resize, but somewhere along the way this got
lost. Restore this behaviour to scroll mode by fixing screen_write_copy to read
up to the saved line length rather than the current screen width. Copy mode
remains unaltered for now.
2009-08-13 16:24:33 +00:00
Tiago Cunha
97eb537f38 Sync OpenBSD patchset 227:
Add a flags member to the grid_line struct and use it to differentiate lines
wrapped at the screen edge from those terminated by a newline. Then use this
when copying to combine wrapped lines together into one.
2009-08-09 17:32:06 +00:00
Nicholas Marriott
06ddd3dcf8 Add a flags member to the grid_line struct and use it to differentiate lines
wrapped at the screen edge from those terminated by a newline. Then use this
when copying to combine wrapped lines together into one.
2009-08-08 15:57:49 +00:00
Tiago Cunha
09c6432854 Sync OpenBSD patchset 207:
Plug some memory leaks.
2009-07-30 21:14:04 +00:00
Nicholas Marriott
97f105cde2 Plug some memory leaks. 2009-07-30 20:41:48 +00:00
Tiago Cunha
7562781c9d Sync OpenBSD patchset 186:
Draw UTF-8 characters under the selection correctly.
2009-07-28 22:41:38 +00:00
Nicholas Marriott
fc65da1eed Draw UTF-8 characters under the selection correctly. 2009-07-27 11:33:21 +00:00
Tiago Cunha
b72f9bea43 Sync OpenBSD patchset 162:
More tty code tidying: move the saved cursor/region position (from before the
screen was updated) out of struct screen and into struct tty_ctx.
2009-07-23 12:48:18 +00:00
Tiago Cunha
9c6fa90857 Sync OpenBSD patchset 160:
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.
2009-07-23 12:38:01 +00:00
Nicholas Marriott
ddad0be5f7 More tty code tidying: move the saved cursor/region position (from before the
screen was updated) out of struct screen and into struct tty_ctx.
2009-07-22 20:53:38 +00:00
Tiago Cunha
b6afa30c39 Sync OpenBSD patchset 159:
There are relatively few arguments to tty_cmd_* functions now, so tidy them up
by using a struct rather than hiding everything with varargs.
2009-07-22 18:08:56 +00:00
Nicholas Marriott
2ec2837daa 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.
2009-07-22 18:02:23 +00:00
Tiago Cunha
75a44d856e Sync OpenBSD patchset 158:
tty_cmd_raw is only used once, for raw UTF-8 output, so rename it to
tty_cmd_utf8character and eliminate the size argument.
2009-07-22 17:58:42 +00:00
Nicholas Marriott
6a309c53a8 There are relatively few arguments to tty_cmd_* functions now, so tidy them up
by using a struct rather than hiding everything with varargs.
2009-07-22 16:45:31 +00:00
Nicholas Marriott
5bd72ec629 tty_cmd_raw is only used once, for raw UTF-8 output, so rename it to
tty_cmd_utf8character and eliminate the size argument.
2009-07-22 15:55:32 +00:00
Nicholas Marriott
82b45b460e The scroll region cannot be one line only, ignore attempts to make it so. 2009-07-15 17:43:02 +00:00
Nicholas Marriott
975f516f51 The scroll region cannot be one line only, ignore attempts to make it so. 2009-07-14 14:47:32 +00:00
Nicholas Marriott
664a0bd559 Cursor up and down should be limited by the scroll region (cuu should stop at
the scroll region top if starting from below it and cud stop at the bottom if
starting from above). Fixes another vttest test.
2009-07-09 18:08:13 +00:00
Nicholas Marriott
295879cced Change inserting and deleting lines inside the scroll region to properly clear
lines that should be inserted/deleted but not moved. Fixes problems with mutt
reported by Brian Lewis, thanks.
2009-07-09 18:04:53 +00:00
Nicholas Marriott
e073441c36 Tidy by removing unused argument from grid_view_{insert,delete}_line_region
functions (currently don't fully work, this is to make fix easier).
2009-07-09 18:04:17 +00:00
Nicholas Marriott
5f2f07ed8a Cursor up and down should be limited by the scroll region (cuu should stop at
the scroll region top if starting from below it and cud stop at the bottom if
starting from above). Fixes another vttest test.
2009-07-09 17:57:11 +00:00
Nicholas Marriott
86c93c6e34 Change inserting and deleting lines inside the scroll region to properly clear
lines that should be inserted/deleted but not moved. Fixes problems with mutt
reported by Brian Lewis, thanks.
2009-07-09 07:58:14 +00:00
Nicholas Marriott
643c219d18 Tidy by removing unused argument from grid_view_{insert,delete}_line_region
functions (currently don't fully work, this is to make fix easier).
2009-07-09 00:29:32 +00:00
Nicholas Marriott
35092835b0 Fix two errors with character/line insertion and deletion: the maximum number
of characters which may be inserted or deleted is the screen width, not one
less (and similarly for lines and height); and if characters or lines are
deleted by moving the ones that follow, the space at the end needs to be
cleared.

This appears to solve long-standing redraw issues most visible when using the
force-width option then scrolling in view(1) or unwrapping lines in emacs.
2009-07-01 19:14:33 +00:00
Nicholas Marriott
68e06fb6b7 Fix two errors with character/line insertion and deletion: the maximum number
of characters which may be inserted or deleted is the screen width, not one
less (and similarly for lines and height); and if characters or lines are
deleted by moving the ones that follow, the space at the end needs to be
cleared.

This appears to solve long-standing redraw issues most visible when using the
force-width option then scrolling in view(1) or unwrapping lines in emacs.
2009-06-29 21:30:50 +00:00
Nicholas Marriott
1fb3948ff1 Whoops, revert. 2009-06-26 22:18:36 +00:00
Nicholas Marriott
1496aa5dbc Remove some unused function declarations; no binary change. 2009-06-26 22:18:03 +00:00
Nicholas Marriott
de2ea2d178 Status line fixes: don't truncate status-right now the length calculation is
done for UTF-8, limit to the maximum length correctly when printing, and always
print a space even if the left string is longer than the width available.
2009-06-26 15:34:12 +00:00
Nicholas Marriott
1ec736a3dc Status line fixes: don't truncate status-right now the length calculation is
done for UTF-8, limit to the maximum length correctly when printing, and always
print a space even if the left string is longer than the width available.
2009-06-26 15:13:39 +00:00
Nicholas Marriott
bb459beb03 Whitespace and more syncing. 2009-06-25 16:34:50 +00:00
Nicholas Marriott
a5830b9603 Restore $Id$ and add script to do so. 2009-06-25 16:21:32 +00:00
Nicholas Marriott
b8ec0c713f Support insert mode by using insert character to shift the cells before writing
as normal.
2009-06-25 15:52:12 +00:00
Nicholas Marriott
93631b3ca7 Implement the DEC alignment test. With the last change this is enough for the
first cursor test in vttest (in ports) to pass; it still shops a few more
problems though.
2009-06-25 15:51:54 +00:00
Nicholas Marriott
49477de55c Fix some miscalculations when clearing to start of screen: the number of lines
to the cursor is cy not cy - 1, and the current cursor cell should be included.
2009-06-25 15:50:48 +00:00
Nicholas Marriott
28bcf774e5 New session option, status-utf8, to control the interpretation of top-bit-set
characters in status-left and status-right (if on, they are treated as UTF-8;
otherwise passed through).
2009-06-25 15:48:25 +00:00
Nicholas Marriott
1e06ec41dc Add a UTF-8 aware string length function and make UTF-8 in
status-left/status-right work properly. At the moment any top-bit-set
characters are assumed to be UTF-8: a status-utf8 option to configure this will
come shortly.
2009-06-25 15:47:07 +00:00
Ray Lai
74749e8705 Remove trailing newlines, spaces, and tabs.
No binary change.
2009-06-05 03:13:16 +00:00
Nicholas Marriott
c1a4c0186a Support insert mode by using insert character to shift the cells before writing
as normal.
2009-06-03 23:37:30 +00:00
Nicholas Marriott
52ec9b9ec4 Implement the DEC alignment test. With the last change this is enough for the
first cursor test in vttest (in ports) to pass; it still shops a few more
problems though.
2009-06-03 23:30:40 +00:00
Nicholas Marriott
ebe07c2726 Fix some miscalculations when clearing to start of screen: the number of lines
to the cursor is cy not cy - 1, and the current cursor cell should be included.
2009-06-03 23:26:56 +00:00
Nicholas Marriott
6521427a45 New session option, status-utf8, to control the interpretation of top-bit-set
characters in status-left and status-right (if on, they are treated as UTF-8;
otherwise passed through).
2009-06-03 16:54:26 +00:00
Nicholas Marriott
7d45e29683 Add a UTF-8 aware string length function and make UTF-8 in
status-left/status-right work properly. At the moment any top-bit-set
characters are assumed to be UTF-8: a status-utf8 option to configure this will
come shortly.
2009-06-03 16:05:46 +00:00
Nicholas Marriott
35876eaab9 Import tmux, a terminal multiplexor allowing (among other things) a single
terminal to be switched between several different windows and programs
displayed on one terminal be detached from one terminal and moved to another.

ok deraadt pirofti
2009-06-01 22:58:49 +00:00
Nicholas Marriott
143aa718e5 Space trimmage mega-diff. 2009-05-04 17:58:27 +00:00
Nicholas Marriott
a6f2d82335 Use a (pre)randomised binary tree for UTF-8 character widths. Probably overkill
but meh.
2009-04-30 20:54:53 +00:00
Nicholas Marriott
8a160f2d58 - Unbreak selection colouring.
- Get UTF-8 check the right way round...
2009-03-30 20:14:50 +00:00
Nicholas Marriott
23fef99009 8 -> UTF8_SIZE. 2009-03-29 11:18:28 +00:00
Nicholas Marriott
4cbf7d489e - Ignore padding when writing so copying the screen for copy-mode doesn't include extra characters.
- Handle combined characters properly when the previous character is not UTF-8.
2009-03-29 11:06:22 +00:00
Nicholas Marriott
cf7b384c43 Better UTF-8 support, including combined characters. Unicode data is now stored
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.
2009-03-28 20:17:29 +00:00
Nicholas Marriott
6c0728fe07 Step 2 of the Grand Plan To Make UTF-8 Better.
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.
2009-03-28 16:30:05 +00:00
Nicholas Marriott
989d67ccd7 struct grid_data -> struct grid. Stage 1 of the Grand Plan To Make UTF-8 Better. 2009-03-28 15:43:41 +00:00
Nicholas Marriott
8a6a7e74da CSR resets the cursor to the top-left. Reported by merdely. 2009-03-05 20:27:51 +00:00
Nicholas Marriott
eae026f2f7 Redraw selection properly. 2009-02-25 21:56:46 +00:00
Nicholas Marriott
c45eb2713b Check the right variable. Doh. 2009-02-24 21:49:05 +00:00
Nicholas Marriott
3969b59ae5 Don't turn cursor off unless actually drawing something. 2009-02-21 19:25:58 +00:00
Nicholas Marriott
514fe11727 We are going to need the updated screen data for emulating the scroll region,
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.
2009-02-11 17:04:39 +00:00
Nicholas Marriott
c6be7d3ee6 Simplify screen/tty write code to eliminate function pointer which is now
unnecessary (it always does the same thing).
2009-02-11 06:31:09 +00:00
Nicholas Marriott
4428987e95 * Better support for at least the most common variant of mouse input: parse it and adjust for different panes. Also support mouse in window/session choice mode.
* Bring back the fancy window titles with session/window names: it is easy to work around problems with elinks (see FAQ).
2009-01-28 19:52:21 +00:00
Nicholas Marriott
c1726281c9 Handle cursor on/off better. 2009-01-27 21:39:15 +00:00
Nicholas Marriott
4d7e555a48 Don't screw up UTF-8 combined characters. 2009-01-26 20:57:45 +00:00
Nicholas Marriott
95323a6a36 Leftover label. 2009-01-19 17:15:19 +00:00
Nicholas Marriott
4cdc228353 Unbreak UTF-8. 2009-01-18 21:46:30 +00:00
Nicholas Marriott
1fa36b2419 Use the right size. 2009-01-18 21:21:53 +00:00
Nicholas Marriott
0e197b417a Tidy a few warnings. 2009-01-17 18:47:37 +00:00
Nicholas Marriott
162bacdcd9 Window splitting. Two vertical panes fixed 50% each. This is a huge diff, still a couple of bugs (notably heap corruption somewhere causing segfault on exit). 2009-01-11 23:31:46 +00:00
Nicholas Marriott
a1b43faa43 Trim spaces. 2009-01-10 01:51:22 +00:00
Nicholas Marriott
3ed5aa3e72 Build array of codes, stop using ncurses global variables and push ncurses crap into tty-term.c. 2009-01-09 23:57:42 +00:00
Nicholas Marriott
cd8f85c2bd When I wrote this I somehow forgot that the cursor could helpfully sit at sx
(one character off the right of the screen), so there are several out-of-bounds issues. Add some additional checking.
2009-01-08 21:55:12 +00:00
Nicholas Marriott
4ebb85aae8 Don't return NULL when retrieving a string that is too long. 2009-01-08 21:22:01 +00:00
Nicholas Marriott
d7e0ea5cc6 Don't die when backspacing off the edge of the screen, reported by David Chisnall.. 2009-01-06 18:07:57 +00:00