ncplane_destroy() needs to call ncplane_reparent_family(), not
ncplane_reparent() as it was doing (closes#1291). ->absy and
->absx actually are absolute; return them directly for an O(1)
ncplane_abs_yx() (down from O(N), huzzah). Add some unit tests
related to destroying and reparenting (#1286). Add ncplane_abs_y()
and ncplane_abs_x(), document them, etc.
We've been using #ifdefs and such to check for libc
Unicode 13 support at compile time. Instead, execute
a wcwidth() using a sextant. If it comes back -1, the
libc lacks support; disable sextants. Otherwise, let
it go. This ought allow us to support sextants on BSD
as soon as it has support; it furthermore allows us to
properly disable them on Linuxes lacking such support.
Closes#1289.
Use 74 columns for infoplane, handling longer model names.
Match /etc/os-release fields whether delimited by tabs or
spaces. Properly extract only the names from /etc/os-release.
/proc/cpuinfo doesn't always declare a Model Name, but
it seems to have a vendor_id everywhere I've been able
to check (s390x etc.). fall back to vendor_id in the
absence of a model name. count cores based off instances
of core id while we're at it (likewise fixes s390x).
I've created the new files src/compat/compat.{hc}. These
are available to all binaries by adding src to the include
directories, and src/compat/compat.c to the sources. Several
functions are implemented here which one or more target
operating systems are missing, right now all related to time.
This includes clock_nanosleep(), which is missing on OS X and
DragonFly BSD. Eliminate the other three definitions of
timespec_to_ns() and friends. Standardize on NANOSECS_IN_SEC
rather than the more opaque GIG. Progress on #1121.
Don't bias the nccell width by 1, or 0-length EGCs become
255 columns. We weren't actually using the width to drive
much anything until now, so this wasn't a problem, but it
is exposed as an error once we got rid of CELL_WIDE_ASIAN
and start looping through the actual egc column width.
Closes#1278, closes#1277
postpaint_cell sometimes needs to skip over the following
cells, due to a multicolumn EGC. postpaint() was only
passing x as a value parameter, though, and thus any
changes made weren't escaping out to postpaint(). replace
it with a value-result parameter. might fix#1276.
* Use NCBLIT_DEFAULT rather than explicit 2x2, so that we use
NCBLIT_3x2 on terminals which support it.
* Use NCSCALE_SCALE_HIRES rather than NCSCALE_SCALE.
* Support LOGO in /etc/os-release, with or without quotes.
Closes#1246.
Improve the documentation for ncplane_set_base_cell() and
ncplane_set_base(). Go by render.c's authority (all three
channels are independently checked for transparency, and
replaced with the base cell if so--egc, fg, and bg). Add a
BASE CELL section to notcurses_plane.3. Check for, and reject,
an nccell which is secondary among a multicolumn EGC.
Closes#1238.
Add a new member 'sextants' to the terminfo cache (both
notcurses and ncvisual contain one of these, and both
initialize it the same way -- interrogate_terminfo()).
Add a new function, 'notcurses_media_defblitter()', and
deprecate 'ncvisual_default_blitter()' (the latter didn't
receive enough information to return NCBLIT_3x2). Update
all callers. Add new *internal* function rgba_default_blitter(),
so this logic can be freely changed in the future. If
sextants are available, and we're scaling, return NCBLIT_3x2.
Once we detect sextant availability, we'll have sexblitter
as a default -- stay tuned! #1114
Kitty conflates an RGB background specification that aligns
with the default background color with that actual background
color. This can result in translucent background when we're
expecting opaque ones. Detect kitty (strstr check of terminal
name for "kitty"), and if it's active, mark `bg_collides_default`
with the rgb of the default background color. For now, we assume
this to be (0,0,0), but we ought improve it by determining (or
setting, if that's impossible) the precise default bg color.
Closes#1117.
When the -t switch is provided to notcurses-view, its
floating-point argument is the number of seconds for
which we delay after showing an image. When this is used,
the standard "press any key to continue" prompt is not
displayed, and indeed keybindings are not honored. This
can be used for a slideshow, or just for fullscreen display
involving no user interaction. It does not apply in the
case of -L for looping. Closes#1229.
Thread out ncls to perform the media decode in different threads, in parallel. Only the display needs be locked. On a directory of 200 files on my 39070X, this speeds ncls from ~5s to ~1s. On 75 files, we go from ~.5s to ~.2s. On a single file, we lose about 5%. To facilitate this, ncdirect_render_image() has been split into two helpers, ncdirect_render_frame() and ncdirect_raster_frame().
Progress bars now support four 32-bit channels in ncprogbar_options, and uses them to color a gradient. The redraw procedure has been changed from O(n) to O(1). ASCII progress bars have been repaired. Closes#1213.
Tremendous thanks to @kaniini, @Ikke and everyone else over
at Alpine Linux for helping to chase this down. I blundered
in ncplane_reparent() when reparenting a root plane with
children (the children need to come into the root plane
list). This will hopefully fix our amd64 failures on Alpine
since late 2.0.x. Closes#1197.
nick black
Ariadne Conill