1
1
mirror of https://github.com/wez/wezterm.git synced 2024-12-19 11:21:39 +03:00
Commit Graph

477 Commits

Author SHA1 Message Date
Wez Furlong
a0cbea2703 term: ignore first OSC title change sequence on Windows
ConPTY emits a sequence that sets the title to the name of the
program that is initially launched into it.

This commit tries to ignore that sequence in that circumstance,
so that the logic in b5d156c282
can more dynamically set the tab title.
2021-12-25 01:10:37 -07:00
Wez Furlong
ac3cee3fc6 kitty: improve debug when image id is not found
Poking around at an issue, and thinking that some more context
might be nice.

Haven't managed to reproduce the issue so far though :-/

refs: https://github.com/wez/wezterm/issues/1156
2021-12-24 13:19:55 -07:00
Wez Furlong
88b8cb3e21 term: fix overinvalidation of selection on Windows
Not 100% sure why this only really manifested on Windows, but
the symptoms were:

* Run powershell in a tab
* Run `dir`
* Hit enter a couple of times to show a couple of prompts
* Try using the mouse to select across the prompt boundaries

The selection would get invalidated crossing the boundaries.
I traced this down to the lines around those regions having
SEQ_ZERO as their sequence, so this commit ensures that lines
that are created as part of scrolling the screen are correctly
tagged with the current seqno from the terminal display.

Why only windows? Not totally sure; perhaps it is related to
something funky happening in the conpty layer and sending us
unusual escapes (eg: scroll margins?)
2021-12-24 08:53:49 -07:00
Wez Furlong
cde8fe0e85 term: don't force zero-width graphemes into cells
This isn't ideal, as we're discarding information, but in 
where the problem codepoint is a unicode bidi control, we can't and
don't use that information anyway.

We'll need to figure out how to incorporate that when we get to it.

For now, this makes the presentation work correctly.

refs: 
2021-12-23 12:13:38 -07:00
Autumn Lamonte
0a93a58605 New optional argument 'doNotMoveCursor' for iTerm2 images. This
permits iTerm2 images to be drawn anywhere on screen without
scrolling the cursor, including the bottom row.

Also included is a check in fcwrap.rs to_range_set(), without which
was causing a panic at runtime due to subtraction from unsigned
leading to overflow.
2021-12-23 07:07:08 -08:00
Wez Furlong
db3f519a45 term: refine debug logging around image attachment 2021-12-23 07:53:13 -07:00
Wez Furlong
294579fd31 term: introduce right&bottom padding to ImageCell
This helps us correctly set the size of the image cell
for the case where we have a partial cell at the right/bottom
edge of an image being mapped across cells.

refs: 
2021-12-23 07:26:49 -07:00
Wez Furlong
5c222cfcdc term: allow iterm2 file transfer protocol
Adds some plumbing to allow the GUI to implement a download handler
and connect that up for iterm2 image/file transfers that have their
inline property set to false.

Previously we'd just log an error.

Now we will by default download the file to the user's download
directory.

This behavior can be turned off via the new `allow_download_protocols`
configuration setting.

File transfers can be initiated on a remote host via the
https://iterm2.com/utilities/it2dl script.

When the download completes, a toast notification is shown that will
open the file when clicked.

refs: https://github.com/wez/wezterm/discussions/1450
2021-12-22 09:21:39 -07:00
Wez Furlong
b40f65c25e termwiz: s/Fuschia/Fuchsia/g
closes: 
2021-12-21 20:38:59 -07:00
Wez Furlong
23ebcf324d term: revise image geometry calculation
The texture coordinates into the source image weren't quite right;
previously we'd do this math:

num_cells = (image_size / cell_size).ceil()
delta = num_cells / image_size

that would result in the image being stretched to fit across the rounded
up number of cells, leading to a distored image.

This commit changes the delta calculation to be based on the remaining
number of pixels in a given dimension relative to the cell size.

refs: 
refs: 
2021-12-11 10:51:13 -07:00
Luis Javier Merino Morán
b262568f0b DECSTR should reset DECLRMM
While the description at DEC STD-070 04-31 does not mention Left Right
Margin Mode (it does mention the left and right margins), the code at
04-34 sets it to FIXED (the reset state, the set state is SETABLE).

From esctest:
DECSETTests.test_DECSET_DECLRMM_ModeResetByDECSTR
2021-12-08 07:29:52 -08:00
Wez Furlong
a61fa57d97 term: Make CUB behave like Backspace, as xterm does
closes: https://github.com/wez/wezterm/issues/1273
2021-12-02 20:50:11 -07:00
Wez Furlong
0bc4347383 term: fix BS in last column in reverse wraparound mode
refs: https://github.com/wez/wezterm/issues/1273
2021-11-25 22:54:27 -07:00
Wez Furlong
e35f567a2e cargo fmt 2021-11-25 12:14:57 -07:00
Wez Furlong
d69df605cd Add escape sequence to control unicode version
As promised in the previous commit, this one implements an escape
sequence to control the unicode version.

Unknown to me in the previous commit, iTerm2 already defines such
an escape sequence, so we simply implement it here with the same
semantics.

refs: 
refs: 
2021-11-25 12:12:52 -07:00
Wez Furlong
225e7a1243 introduce unicode_version config
This is a fairly far-reaching commit. The idea is:

* Introduce a unicode_version config that specifies the default level
  of unicode conformance for each newly created Terminal (each Pane)
* The unicode_version is passed down to the `grapheme_column_width`
  function which interprets the width based on the version
* `Cell` records the width so that later calculations don't need to
  know the unicode version

In a subsequent diff, I will introduce an escape sequence that allows
setting/pushing/popping the unicode version so that it can be overridden
via eg: a shell alias prior to launching an application that uses a
different version of unicode from the default.

This approach allows output from multiple applications with differing
understanding of unicode to coexist on the same screen a little more
sanely.

Note that the default `unicode_version` is set to 9, which means that
emoji presentation selectors are now by-default ignored.  This was
selected to better match the level of support in widely deployed
applications.

I expect to raise that default version in the future.

Also worth noting: there are a number of callers of
`unicode_column_width` in things like overlays and lua helper functions
that pass `None` for the unicode version: these will assume the latest
known-to-wezterm/termwiz version of unicode to be desired. If those
overlays do things with emoji presentation selectors, then there may be
some alignment artifacts. That can be tackled in a follow up commit.

refs: 
refs: 
2021-11-25 09:00:45 -07:00
Luis Javier Merino Morán
65e2c1f914 Fix DECRPTUI response
DECRPTUI, the response to DA3, should be:
 DCS ! | D...D ST
not
 CSI = D...D ST
2021-11-21 07:49:35 -08:00
Wez Furlong
24875004f6 add canonicalize_pasted_newlines config option
refs: 
2021-10-17 13:14:16 -07:00
Wez Furlong
d461c1c0b6 deps: lru 0.6 -> 0.7 2021-10-10 19:29:42 -07:00
Wez Furlong
0ee1f7529b fonts: Fix shaping metrics
This was a bit of a PITA to run down; the essence of the problem
was that the shaper was returning an x_advance of 0 for U+3000,
which caused wezterm's shaping layer to elide that glyph.

I eventually tracked down the x_advance to be the result of
scaling by an x_scale of 0, which in turn is the result of
harfbuzz not knowing the font size.

The critical portion of this diff is the line that advises
harfbuzz that the font has changed after we've applied the
font size.

The rest is just stuff to make it easier to debug and verify.

This:

```
printf "x\u3000x."
```

Now correctly renders on screen as "x  x".

fixes: 
2021-09-22 08:34:40 -07:00
Wez Furlong
0e5d0d8117 term: fix encoding F1-F4
This was accidentally broken by 8da1f676a8

refs: 
2021-09-12 09:30:14 -07:00
Greg V
ce5615082c gui: cancel mouse click streak when the cursor has moved to a different cell (fixes )
Fast-clicking users generally expect to be able to rapidly do regular selections
or otherwise cancel double/triple/quad clicks, so significant mouse movement should end the streak.

Allowing tiny pixel movements to account for touchpads is necessary.
Fortunately, the position here is already in character grid cells, which provides enough margin for this.
Other code editors like gedit also seem to do this based on the character grid.
2021-09-08 17:38:59 -07:00
Wez Furlong
545a8fc8a7 deps: ordered-float -> 2.8 2021-09-05 23:25:37 -07:00
Wez Furlong
311bd78902 Fix SU (scroll up) using default-blank cells
closes: https://github.com/wez/wezterm/issues/1102#
2021-09-04 17:35:33 -07:00
Wez Furlong
67e8bdc5c2 term: fix DCH removing cells instead of setting to current bg color
refs: 
2021-09-03 16:21:15 -07:00
Wez Furlong
c5d1f67853 term/termwiz: move key encoding to termwiz
This will enable eg: a lua helper function to serialize keycodes to
assist in some key rebinding scenarios (see:
https://github.com/wez/wezterm/pull/1091#issuecomment-910940833 for the
gist of it) but also makes it a bit easier to write unit tests for key
encoding so that situations like those in  are potentially less
likely to occur in the future.
2021-09-03 11:12:42 -07:00
Wez Furlong
24d971c87d Defend against crazy big images sent via iterm image protocol
This also handles a Nan case and allows us to survive term-crash-tests as of fb6fa3192a65e8862c3c743ba102f7e2af45b519

refs: https://github.com/wez/wezterm/issues/1031
2021-08-16 23:44:31 -07:00
Wez Furlong
ab21910d80 deps: update bitflags -> 1.3 2021-08-15 18:21:17 -07:00
Wez Furlong
aa9cbd4552 refactor: update a couple of callers to Cell::blank 2021-08-15 15:42:35 -07:00
Wez Furlong
6f78ee4f14 deps: update unicode-segmentation to 1.8 2021-08-11 23:06:17 -07:00
Wez Furlong
6c2f9f2d18 term: restore line length pre-allocation
A while back I made the line lengths lazily grown; the reduction
in memory was nice, and it helped with render performance for
really wide screens.

Unfortunately, it puts a bunch of reallocation into the hot path
of the parser and updating the terminal model when people run
the inevitable `cat giant-file.txt` benchmark.

This commit reinstates pre-allocating lines to match the physical
terminal width, and tweaks the code a bit to take advantage of
const Cell allocation and to avoid some clones (a really micro
optimization).
2021-08-11 23:06:17 -07:00
Wez Furlong
9134da8c07 term: reuse string buffer for flush_print
Avoids allocating and discarding on separate passes
2021-08-11 23:06:17 -07:00
Wez Furlong
c90ccec102 micro-optimize TeenyString and cell width
For simple graphemes, we can avoid subsequently calling
grapheme_column_width and cache that information in TeenyString.

Make blank TeenyString and Cell initializations const.
2021-08-11 23:06:17 -07:00
Wez Furlong
1e7c7e7059 term: micro optimize flush_print: directly call grapheme_column_width
Saves an additional call through unicode segmentation
2021-08-11 23:06:16 -07:00
Wez Furlong
80d261977f term: avoid dirtying lines when the cursor moves
Tag CursorPosition with the seqno of the cursor move instead.
This should avoid over-invalidating lines and the selection
if it was just the cursor that moved.
2021-08-08 22:47:10 -07:00
Wez Furlong
eae327efcc model: replace LineBits::DIRTY with a sequence number
Terminal now maintains a sequence number that increments
for each Action that is applied to it.

Changes to lines are tagged with the current sequence number.

This makes it a bit easier to reason about when an individual
line has changed relative to some point in "time"; the consumer
of the terminal can sample the current sequence number and then
can later determine which lines have changed since that point
in time.

refs: https://github.com/wez/wezterm/issues/867
2021-08-08 12:45:08 -07:00
Wez Furlong
6d9957e6e3 term, gui: allow tracking multiple buttons
We were previously only remembering the most recently pressed
button, but that's a lossy thing to do.

This commit remembers the button presses so that we can correctly
report all press/release events.

refs: 
2021-08-07 18:37:38 -07:00
Wez Furlong
6a7e6596b9 kitty img: make allowances for permanence of image attachments
An implementation detail in wezterm is that it doesn't model
image placements as a separate entity; they are all bound to
the image cells in the terminal model.

The semantics of the kitty image protocol are that placements
are "permanent" wrt. overwriting a cell with text, except for
the explicit EraseInLine/EraseInDisplay sequences that are
used for clearing.

This commit takes a pass at implementing that semantic in
the wezterm data model.

refs: 
2021-08-06 09:01:18 -07:00
Wez Furlong
3ea13fb530 kitty img: tidy up blit and bounds check, skip 0 duration root frame 2021-08-05 23:59:35 -07:00
Wez Furlong
815e24c24d kitty img: slightly improve protocol responses
Make a helper function for these, but there is still room
for improvement.

refs: 
2021-08-04 13:53:54 -07:00
Wez Furlong
f3d4b59139 kitty img: flesh out some todos with anim composition
refs: 
2021-08-04 13:31:18 -07:00
Wez Furlong
a726a170ff kitty img: support for removing all placements
and do that when the terminal is reset

refs: 
2021-08-04 09:15:09 -07:00
Wez Furlong
6d702275df kitty img: don't move the cursor when requested not to
We were actually moving it during placement, and then restoring
the original placement.  That could potentially lead to the
screen being scrolled, so we want to avoid that.

refs: 
2021-08-04 09:14:58 -07:00
Wez Furlong
02b5defac1 kitty img: parse and implement basic anim frame composition
This is just enough for notcurses-demo to run, and is definitely
missing various cases.

refs: 
2021-08-03 21:57:27 -07:00
Wez Furlong
0c5753eb58 kitty img: round out support for animation frame parameters
As of this commit, `kitty +kitten icat ~/Downloads/fast-parallax.gif`
(wherein the icat kitten decodes the gif into frames and sends them
to the terminal to animate) behaves equivalently in wezterm and kitty.

(There appears to be an issue with the background color/deltas in
the icat kitten in kitty-0.21.1-1.fc33.x86_64 which I have installed,
so both wezterm and kitty have a funky black background for this
particular gif).

refs: 
2021-08-03 19:27:14 -07:00
Wez Furlong
07bea8c269 gui: change image frame caching scheme to use frame hash
Since we can now mutate individual frames, we need to avoid
falsely caching across a change; switch from using (image_id, frame_idx)
to frame_hash.

refs: 
2021-08-03 08:42:39 -07:00
Wez Furlong
e85e94248b basic kitty img animation support
this is the bare minimum to squeak by with notcurses; it currently
only supports editing single frame images.

refs: 
2021-08-02 21:35:57 -07:00
Wez Furlong
29a64e10b8 allow locking ImageData's ImageDataType for mutation
refs: 
2021-08-02 21:35:57 -07:00
Wez Furlong
90b16b9518 move image decoding into termwiz
Adds a use_image feature to termwiz that enables an optional
dep on the image crate.  This in turn allows decoding of animation
formats (gif, apng) from file data, but more crucially, allows
modeling animation frames at the termwiz layer, which is a pre-req
for enabling kitty img protocol animation support.

refs: 
2021-08-02 21:35:50 -07:00
Wez Furlong
b862c8d111 termwiz: add ImageData::hash
Moves the localized hashing logic from term -> termwiz
where it can be re-used.

refs: 
2021-08-02 09:18:25 -07:00