fd4ca45ae5
clean up luasocket build file
...
Teliva is never intended to be "installed" somewhere. Just work inside
its directory and separately share the .tlv files you create. (Though I
don't yet have a good flow for starting a new .tlv file.)
2021-11-27 06:57:38 -08:00
c558d2c565
.
2021-11-27 00:42:15 -08:00
ff43b5bae1
fix build on Mac
2021-11-27 00:41:17 -08:00
83050ec38e
starting to test compilation on Mac and BSD.
...
For starters, put Linux-specific stuff in a Linux-specific target.
By not resetting MYCFLAGS and MYLDFLAGS, I'm unnecessarily passing in
-DLUA_USE_LINUX. But that'll make it easier to get things running on Mac
and BSD.
2021-11-27 00:34:49 -08:00
1e911f95ef
bugfix: emit legal Lua representation
2021-11-26 22:03:47 -08:00
662d1cf202
support space/backspace in recent changes view
2021-11-26 21:54:45 -08:00
2426fe7a79
standardize indent
2021-11-26 21:50:53 -08:00
7e86f6d123
colorize comments in recent changes
2021-11-26 21:48:31 -08:00
e2a53858a0
support notes on edits
...
Unlike both conventional version control and wiki history, I'm planning
to always allow modifying commit messages.
2021-11-26 21:43:31 -08:00
6b6fe7ce72
periodic cleanup of warnings
2021-11-26 21:22:20 -08:00
db7202ccfd
.
2021-11-26 21:20:23 -08:00
35fe28a36f
.
2021-11-26 21:19:41 -08:00
0c211dec77
don't mislead immediately after undo
2021-11-26 20:00:32 -08:00
b954ab869e
.
2021-11-26 19:52:31 -08:00
0726692546
more obvious serialization of undo events
2021-11-26 19:45:19 -08:00
3cb37a0f39
render undo events
2021-11-26 19:38:51 -08:00
a0171352a1
start processing undo events
2021-11-26 19:33:35 -08:00
576ab1df8d
start recording undo events in timeline
...
We're not using or rendering them yet.
2021-11-26 18:59:31 -08:00
07bf55b460
save timestamp on change; show in recent changes
2021-11-26 18:47:50 -08:00
f3dd541561
.
2021-11-26 18:21:04 -08:00
555159fb9a
clean up traces of an old experiment
2021-11-26 18:18:58 -08:00
720ff19351
add support for metadata in Teliva programs
...
They don't have any semantics yet. We just ignore them for now.
2021-11-26 18:17:35 -08:00
7295ce050d
undo to a specific point
...
Still highly experimental. I'm not persisting state yet.
2021-11-26 18:05:41 -08:00
080ea49d23
standardize screen headings
2021-11-26 18:04:58 -08:00
0800f65eff
start of 'recent changes' screen
...
I'm still unclear on precisely what the experience should be here. We
probably don't need all of a version control system. The goal is just to
be able to answer the question, "what did I change recently that caused
things to break?"
For now let's just start with letting people see past versions.
2021-11-26 18:03:20 -08:00
283cc7ea52
eliminate tail call using goto
2021-11-26 16:54:45 -08:00
0b365d3e4d
clean up terminal in a specific situation
...
The problem: if ever I hit ctrl-e to go to the big picture view and then
hit Esc to go back to running the app, my terminal was messed up after
exiting the app.
Why did I even have this gunk? Perhaps it dates from the time when kilo
was emitting raw escape sequences rather than using ncurses.
2021-11-26 16:39:20 -08:00
e699b5a052
dedup an enum
2021-11-26 16:26:13 -08:00
5bc52a5cba
.
2021-11-26 16:10:02 -08:00
2c80411224
.
2021-11-26 16:09:15 -08:00
f930b9e7c7
extract a function
2021-11-25 21:04:08 -08:00
61ea63adca
more stack hygiene
2021-11-25 21:02:27 -08:00
f5221ba651
one more curses binding
2021-11-25 20:51:38 -08:00
73a22a397e
libraries don't need Lua's `#define`s and whatnot
2021-11-25 20:21:05 -08:00
16a98ea1b3
.
2021-11-25 20:16:41 -08:00
4d7e99c343
.
2021-11-25 20:13:49 -08:00
d6810aba30
dedup definitions in big-picture view
...
We only want to show most recent version of each binding.
2021-11-25 20:02:10 -08:00
4a8691c3dc
new shortcut: return to big-picture view
2021-11-25 20:01:40 -08:00
f58d5b8354
teliva is now ASan-clean again
...
At least in short runs.
Encouraging that the problem was in a recent commit (5a63a5ca40
2021-11-25 19:27:30 -08:00
05f4482139
make upstream kilo ASan-clean
...
Many thanks to dirkf for https://github.com/antirez/kilo/pull/73 !
However, teliva is still not ASan-clean.
2021-11-25 11:28:05 -08:00
6727045165
temporarily start using the address sanitizer
2021-11-25 11:25:28 -08:00
103b8e17bb
jettison luac
...
I'm starting to see some heap buffer overruns, which means we have too
much C code.
I noticed this because editing life.tlv no longer works after commit
5a63a5ca4
2021-11-25 11:22:26 -08:00
d606e93839
rename
2021-11-24 12:50:05 -08:00
5a63a5ca40
monotonically accumulate versions of definitions
...
One old drawback now has a new look. Before, we loaded definitions in
order, so global definitions had to exist before other global
definitions that used them. See window and grid in life.tlv. Now we load
definitions in reverse order, so initialization needs to change. Worse,
if we update window, we need to edit grid just to fix the order.
This implies that we can't yet optimize away bindings where there are no
new changes.
2021-11-24 10:10:38 -08:00
b40ad26544
more Makefile streamlining
...
Since everything is in my control there's no need to parameterize
include paths.
It's a struggle to get make to run when it should. Lying that something
is phony stops working when it's a dependency. Commands get
unnecessarily run. Just fucking run recursive makes directly in the
target that depends on them.
2021-11-24 09:28:01 -08:00
a788f80a92
.
2021-11-24 07:55:39 -08:00
b670747d9e
consistent file ops
...
Now we're down to 1 real warning and 1 false positive.
2021-11-24 07:54:48 -08:00
aff8afa12d
style
2021-11-24 07:48:49 -08:00
64b609f743
rename
2021-11-24 07:04:04 -08:00
bd9020c5b3
indent
2021-11-24 06:43:32 -08:00
1bf2504a8b
clean up a warning and a bit of duplication
...
Now we have 2 probably-valid warnings caused by my edits, and 1
false-positive.
2021-11-22 19:25:05 -08:00
76329c0206
standardize warning flags everywhere
...
I'd like to enable -Wextra as well, but that creates some false
positives.
I've at least made my changes clean w.r.t. -Wextra.
Now we have 4 remaining warnings with gcc 9.3 that seem genuine. Need to
fix those.
2021-11-22 19:01:07 -08:00
e77f3eb9f9
.
2021-11-22 18:47:56 -08:00
11d370493a
drop a warning implied by -Wall
2021-11-22 18:37:11 -08:00
ef68041137
standardize CFLAGS in luasocket
2021-11-22 18:36:04 -08:00
3b4d7131cb
standardize CFLAGS
...
Adding -Wpedantic creates a new warning. Leaving it alone for now:
https://stackoverflow.com/questions/31526876/casting-when-using-dlsym
2021-11-22 18:11:38 -08:00
97a0254042
standardize CFLAGS in luasec
...
What in the world is up with the $MYCFLAGS convention in these Lua
makefiles? I don't know, but I'm going to leave it undisturbed as far as
possible.
2021-11-22 18:06:20 -08:00
c4ecb9a53d
standardize on gcc
...
cc and gcc are identical on my system. But why rely on that.
2021-11-22 17:56:45 -08:00
c79527b784
delete final vestiges of embedded luasocket
2021-11-22 17:41:08 -08:00
70a2e1db1d
delete most of the embedded luasocket in luasec
...
Hilariously, I wasn't linking against it in the first place.
2021-11-22 17:37:41 -08:00
d40471e9dc
luasec: fix 'make clean'
2021-11-22 00:12:37 -08:00
5a484efe8c
https now working!
...
Still extremely ugly:
- I've inlined all the namespaces under ssl, so you need to know that
context and config are related to ssl.
- luasec comes with its own copy of luasocket. I haven't deduped that
yet.
2021-11-21 15:55:52 -08:00
3b44b9827d
basic http requests starting to work
...
In the process we're starting to load almost all of luasocket by
default. And everything is working as expected, no unpleasant surprises.
2021-11-21 15:07:42 -08:00
fdd87c5eda
now. we. have. JSON.
...
Completely unmodified from upstream.
2021-11-21 14:22:20 -08:00
15b43d3092
drop some redundant `require`s
2021-11-20 23:35:22 -08:00
548d59f918
luasocket now loading properly
...
I still haven't tried actually running it.
2021-11-20 23:34:29 -08:00
02acfa7c9c
rename
2021-11-20 21:03:38 -08:00
fa6faaa700
extract a helper
2021-11-20 20:35:37 -08:00
6bdc3d17a7
inline luasocket
...
Just builds for now, isn't available yet to Lua code.
2021-11-20 20:26:17 -08:00
38b21ef79b
bugfix: support running from top-level
...
Unfortunately we can't currently run teliva from anywhere else :/
2021-11-20 19:59:32 -08:00
d6ff198e78
get rid of a distracting name
...
No distinction now between the C and Lua versions of the curses library.
We build them all together in one place.
2021-11-20 16:33:41 -08:00
b618bfc7cf
port changes from minimal to maximal version
...
From lcurseslib.c to lcurses/ directory.
2021-11-20 16:33:41 -08:00
8a3e81b4b0
report errors when calling non-existent functions
2021-11-20 10:13:58 -08:00
e693448b65
inline in C is not worth the trouble
...
https://merveilles.town/@akkartik/107310347838372198
2021-11-20 10:10:26 -08:00
b43747ecef
inline lcurses maximally rather than minimally
...
Until now we had just the bare minimum bindings needed for the demos
built so far. Now we have all of lcurses building in place with minimal
changes.
The changes in this commit can run hanoi.lua when inlined into Lua 5.1,
but don't work with Teliva.
2021-11-19 19:11:03 -08:00
720d728a8d
tested on Mac OS
2021-11-14 12:33:25 -08:00
f40bc31cc3
import life.teliva into image format
2021-11-14 11:10:18 -08:00
58d2d596ea
cleaner error handling
2021-11-14 09:55:13 -08:00
a3ca7877da
.
2021-11-14 09:48:41 -08:00
7b91691b14
remain in editor on error
2021-11-14 09:48:41 -08:00
b3b844c5f7
.
2021-11-14 09:48:41 -08:00
2cc6912e85
find _some_ way to show error on initial load
2021-11-14 09:48:41 -08:00
4ab3ccebb5
more menu cleanup
2021-11-14 09:48:40 -08:00
af4fbe97bf
.
2021-11-14 09:48:40 -08:00
33331a7c50
drop support for non-image Lua scripts
...
We still have no story for error messages. We'll work on that next.
2021-11-14 07:59:15 -08:00
b50e393770
draw the browse dialog the same way
2021-11-14 07:53:20 -08:00
12d738d5ce
dialogs are not status messages
2021-11-14 07:34:23 -08:00
b4e3936170
smarter image browsing
...
- distinguish between unused functions and data
- don't hardcode a specific convention for the curses window object
2021-11-14 07:11:30 -08:00
5be743324c
slightly more robust on-disk format
...
Looks like Lua supports a little bit of programmability in its
multi-line string literals. Even though I can't find this documented
anywhere.
2021-11-14 00:52:25 -08:00
b64a21771e
janky way to cancel browsing
2021-11-14 00:38:00 -08:00
d9aac11889
slightly more obvious browse dialog
2021-11-14 00:27:59 -08:00
9db1cd174f
cleaner dialogs
2021-11-14 00:23:20 -08:00
e451206e06
no, use Esc to cancel
...
It inserts an ugly pause for ghastly historical reasons having to do
with the origins of terminals. But hopefully this isn't a common case.
2021-11-14 00:07:28 -08:00
c5261224df
use word at cursor when it's not at start of line
2021-11-14 00:02:04 -08:00
8e9534ed4f
editor: use ctrl-g to cancel dialogs
2021-11-13 23:59:05 -08:00
711d764b37
jump to word at cursor by default
2021-11-13 23:55:08 -08:00
812b930815
ctrl-u to clear response in go menu
2021-11-13 23:27:51 -08:00
d79e394a30
reset cursor position when switching definitions
2021-11-13 23:27:06 -08:00
3c2a5e26ae
.
2021-11-13 23:22:00 -08:00
902ce6009d
cleaner layout for a function's code and data
2021-11-13 23:14:52 -08:00
0e74056cbb
.
2021-11-13 22:48:59 -08:00
e990682fb2
.
2021-11-13 22:48:14 -08:00
55302fd877
.
2021-11-13 22:45:44 -08:00
a0dbc38c34
generalize kilo's hard-coded multiline C syntax
2021-11-13 22:39:28 -08:00
308cb7acbe
.
2021-11-13 22:02:30 -08:00
485cd25739
fix for editor highlighting
...
Bug introduced during the transition to ncurses.
2021-11-13 22:00:50 -08:00
a7139a6b14
clean up after the editor
...
Even though we don't need to leave ncurses mode, there's still a process
boundary to clean up before.
2021-11-13 21:33:12 -08:00
da76ec44f1
.
2021-11-13 21:25:52 -08:00
88f4b1bb2b
.
2021-11-13 21:23:14 -08:00
90f86daf1c
kill highlighting flags
...
Looks like they were never used in original Kilo.
2021-11-13 21:20:30 -08:00
a655059c8d
replace Kilo's raw terminal handling with ncurses
...
This was on my todo list. What made it urgent was finding that calling
getch() even once while in ncurses caused Kilo to stop detecting arrow
keys. No need to debug that sort of nonsense.
2021-11-13 21:20:30 -08:00
625a66e196
avoid side-effects on the Lua stack
2021-11-13 17:48:04 -08:00
12b2a44cf6
stop leaking memory
2021-11-13 17:46:08 -08:00
398112f756
memory corruption bug
...
I was saving an address on the stack to a global, and it was getting
clobbered later. This is the sort of thing I completely eliminated in
https://github.com/akkartik/mu :/
Now I'm taking a leaf out of the Mu playbook and leaking a little bit of
memory every time I switch definitions.
2021-11-13 17:42:39 -08:00
51378afb2a
.
2021-11-13 16:53:39 -08:00
c34db20756
very rudimentary definition browser
...
Never shows definitions that were never called. Including non-functions.
2021-11-13 16:36:12 -08:00
7a24bed432
.
2021-11-13 16:28:13 -08:00
998e666958
.
2021-11-13 16:13:05 -08:00
5bf5e47077
.
2021-11-13 16:13:05 -08:00
647a2520d6
shield kilo.c from teliva_editbuffer detail
2021-11-13 16:13:05 -08:00
ee0e7c36f5
.
2021-11-13 15:11:03 -08:00
4e62aa95b0
.
2021-11-13 15:07:28 -08:00
cbd5e7d552
.
2021-11-13 15:07:14 -08:00
e06a7b1a7b
.
2021-11-13 15:00:49 -08:00
33b705282d
.
2021-11-13 14:59:22 -08:00
7623e7069a
clean up experiment from previous commit
2021-11-13 10:10:16 -08:00
5a44605143
instrumenting function calls with their depth
2021-11-13 08:16:41 -08:00
0f4faf0e53
start using static linkage like the rest of Lua
2021-11-13 05:14:41 -08:00
d72eae5978
create empty file when navigating to empty definition
2021-11-11 17:41:37 -08:00
e7f33de81e
fix incorrect hotkey in menu
2021-11-11 17:32:50 -08:00
f6c7bf6fd0
ctrl-g to edit a different definition
2021-11-11 17:29:40 -08:00
4d20e71c80
gracefully handle missing definition
2021-11-11 17:04:29 -08:00
8398f8e2d1
.
2021-11-11 17:00:18 -08:00
a76cd9067a
global for current definition being edited
2021-11-11 16:47:09 -08:00
d303c6efb0
pass lua_State into editor
2021-11-11 16:42:02 -08:00
0d6378aa51
delete a helper that hides stack side effects
2021-11-11 16:32:08 -08:00
7b389ce41e
extract a few helper functions
2021-11-11 16:16:55 -08:00
2d05f5090f
.
2021-11-11 16:15:06 -08:00
15e0ac70f9
.
2021-11-11 16:09:46 -08:00
3e7ff252af
reorg
2021-11-11 15:50:52 -08:00
f43a1c7da0
edit a single hard-coded definition in the image
...
src/teliva counter.tlv
C-e # switch to editor
C-e # save and quit
C-x # exit
counter.tlv now has the same logical contents, though the whitespace has
changed, and the order of keys is different.
The implementation is utterly ghastly. For one, I'm unnecessarily
interfacing with kilo through the file system.
2021-11-10 23:20:59 -08:00
9a49b92703
.
2021-11-10 22:15:13 -08:00
efbb57d339
new .tlv image format
...
Plan is for this to be the default representation for Teliva programs.
Text-friendly but not meant to be edited directly as text. Will
eventually include both code and data definitions, both current snapshot
and past revision history.
Right now .tlv files seem to run. Error checking is non-existent,
because I don't understand Lua's idioms around 'status' yet. Opening the
editor expectedly segfaults.
This commit is the most mind-bending bit of code I've written in a long
time.
2021-11-10 22:09:12 -08:00
dc05a10e23
life: move starting pattern around on the screen
...
This assumes we're doing it early soon after opening a new pattern, when
it hasn't yet reached the margins. Quick and dirty, but seems good
enough.
2021-11-10 14:26:38 -08:00
b397edcfbd
.
2021-11-09 20:52:53 -08:00
ef9c42084d
switch menu bars to reverse video
2021-11-07 10:29:29 -08:00
2b7ffea316
fix menu colors
...
I had it switching to a dark background on me.
2021-11-07 08:06:42 -08:00
22f1624728
better support light backgrounds
...
Thanks nicolas decoster for reporting this issue.
2021-11-07 08:04:33 -08:00
40fdaf2a2a
more obvious phrasing
...
Early returns are only worthwhile if they're utterly obvious.
2021-11-07 06:30:15 -08:00
51d6ae77d2
fix a regression in commit ee85ad384f
2021-11-07 06:28:24 -08:00
d79c83f209
.
2021-11-07 06:05:55 -08:00
a8668eaf9f
game of life
...
This required me to figure out some unicode-related nuances, but no new
primitives.
2021-11-06 21:49:23 -07:00
8d1e5f6475
.
2021-11-06 19:21:10 -07:00
5619f0117a
window:nodelay() for non-blocking input
2021-11-06 19:21:01 -07:00
62ecf17870
escape hatch to quit with a pending error
2021-11-06 14:41:11 -07:00
cba53ec9f6
reorg editor transitions
2021-11-06 13:22:05 -07:00
3305ac0b90
start showing error messages in editor
...
Before we'd end up in cryptic situations where error messages would get
hidden when the program got out of ncurses mode.
Now it's a little nicer with error messages showing up at the bottom of
the editor.
But there's still a problem: there's no way to abort without fixing an
error.
2021-11-06 13:10:22 -07:00
ee85ad384f
simple interface for adding to app menu
...
We're not going to enforce that the menu items actually do what they
advertise. It's just a way to draw on the bottom line of screen,
something apps aren't otherwise allowed to do.
2021-11-06 12:44:47 -07:00
2964a5e74a
refactor menu drawing
2021-11-06 12:41:14 -07:00
9b99cd2a56
.
2021-11-06 12:15:41 -07:00
3e94073a88
auto-save
...
We'll add version control soon.
2021-11-06 10:37:49 -07:00
38d202d68d
replace initial help message with a menu
2021-11-06 10:35:50 -07:00
0de8ee3911
no, more consistent to toggle run/edit with ctrl-e
2021-11-06 10:18:22 -07:00
29bb6403eb
exit editor with ctrl-x for consistency
2021-11-06 10:13:43 -07:00
e0c0dea5b7
.
2021-11-06 10:12:50 -07:00
567ee60c5a
.
2021-11-06 10:12:50 -07:00
491f9a378f
clean up when leaving editor
...
kilo.c is still calling exit() in several places that we'll need to
gradually clean up.
2021-11-06 08:06:31 -07:00
9275c954c8
readme and docs
2021-11-05 23:52:58 -07:00
52d73ff930
clean up first paint of editor
2021-11-05 22:40:12 -07:00
6fc0e72cb3
utterly ghastly way to rerun script after edit
2021-11-05 22:39:36 -07:00
6b8da095b1
stitch editor in
2021-11-05 22:12:24 -07:00
dd434b2e1c
drop test array data structure
2021-11-05 22:09:15 -07:00
4ad3390ce3
.
2021-11-05 22:07:46 -07:00
95841b8a43
select an editor to bundle: kilo
...
From https://github.com/antirez/kilo
Source code description: https://viewsourcecode.org/snaptoken/kilo
BSD 2-clause license seems identical to the MIT in the current codebase.
2021-11-05 21:41:39 -07:00
a2aa431c1e
slightly improve hanoi rendering
2021-11-05 19:53:16 -07:00
5da9f27079
menu entry: cleanly exit
2021-11-05 19:50:12 -07:00
ee3f6e8a22
colors: init_pair/color_pair
2021-11-05 19:43:40 -07:00
6c9e26e07c
.
2021-11-05 19:26:05 -07:00
643deca2e2
slightly clearer rendering in the example app
2021-11-05 18:44:16 -07:00
cab996b194
make some space for the global menu
...
We'll eventually need some interface to add entries to it.
2021-11-05 18:43:37 -07:00
78898d2716
move getch out of window scope
...
The window only matters for output, which seems like a stupid interface.
2021-11-05 17:54:45 -07:00
18f06a5169
rename 'screen' to 'window'
...
We're going to set aside a portion of the screen soon that apps can't
touch.
2021-11-05 17:48:09 -07:00
c0307fae88
resist the temptation to add to the Lua API
...
Instead we'll include code in the Lua app itself, to minimize the
differences between what runs on regular Lua and what runs on Teliva.
2021-11-05 16:41:31 -07:00
216e56effc
hanoi.lua now working
...
There's something strange in the combination of Lua 5.1 and lcurses:
window.getch() returns a char but curses.getch() returns an int.
2021-11-05 16:25:45 -07:00
d04ecfeaf3
hanoi.lua _almost_ working
2021-11-05 13:34:38 -07:00
936f8f916b
vimrc
2021-11-05 13:23:58 -07:00
44258f7657
window:getch()
...
But how do we get curses.getch() to work?
I don't see it implemented in lcurses.
2021-11-05 13:23:14 -07:00
14b16003b1
hanoi.lua now rendering
2021-11-05 13:16:39 -07:00
b882d405ca
curses print constants
...
Very satisfying to debug the difference between lcurses putting the
module table in an upvalue. Since I implicitly call initstr() rather
than define it as a primitive, I don't need to bother with that. I am
awesome. Lua is awesome for giving me that sense.
2021-11-05 13:13:25 -07:00
a7351c723e
grouping
2021-11-05 13:08:12 -07:00
13784eb757
mvaddch/mvaddstr
...
I think we now have all the output functions/methods we need. Just some
constants remaining.
2021-11-05 12:57:13 -07:00
6f96d30641
attron/attroff
2021-11-05 12:48:27 -07:00
a4ce7ffbd0
window:getmaxyx()
2021-11-05 12:25:31 -07:00
595a963671
.
2021-11-05 12:21:16 -07:00
debe292277
window:clear()
2021-11-05 12:16:31 -07:00
048abcfbc2
ok, what do we need next for hanoi.lua?
2021-11-05 12:12:16 -07:00
d1c9bff737
reindent
...
I'm trying to follow the style of lua sources even when they're not my
preference. lcurses code is a bit different.
2021-11-05 12:06:14 -07:00
ffaf456654
ohh, that word 'index' was key
2021-11-05 12:03:27 -07:00
ac6c9319f9
oh, that's just a cosmetic thing
...
So why isn't this working?
a = curses:stdscr()
a:addstr(abc)
The error is "attempt to index global 'a' (a userdata value)"
2021-11-05 12:00:37 -07:00
f2d61064fb
copy metatable name from lcurses
...
Makes no difference to the results of:
print(curses:stdscr())
2021-11-05 11:54:55 -07:00
bf7baffa30
metatables seem to be a separate namespace from globals
2021-11-05 11:52:59 -07:00
c4c3b93bf5
snapshot
...
Not quite working. curses.stdscr() is returning userdata, not a window.
This is true even of the raw array example from the book. So we need to
learn something new here. How does lcurses's Pinitscr return a special
window object? From what I can tell it's just putting the results of
lc_newwin() on the stack. Which is the same as my curses_newwin() here.
2021-11-05 11:46:47 -07:00
f761d0ae36
stdscr binding
...
print(curses.stdscr())
print(curses:stdscr())
2021-11-05 10:51:47 -07:00
8552ad4ced
starting on curses library
...
First piece of working new vocabulary:
print(curses:cols())
Just pulling in code from lcurses for the most part. But I'm trying to
understand its internals as I gradually add them in, after my more blunt
first approach of packaging up lcurses/ext failed.
Overall plan for Teliva's API:
- start out with a 'curses' library that does what people who are used
to ncurses/lcurses expect.
- over time create a more opinionated library called 'screen' or
'window' or something.
2021-11-05 10:30:07 -07:00
37b05c2957
https://www.lua.org/pil/28.3.html
...
a = array.new(1000)
for i=1,1000 do
a:set(i, 1/i)
end
print(a:get(10)) -- 0.1
2021-11-05 09:17:54 -07:00
b4d86665b8
https://www.lua.org/pil/28.2.html
2021-11-05 08:54:19 -07:00
f53ca6f804
confirmed that this is the same
...
And it seems simpler to me.
2021-11-05 08:31:43 -07:00
712dc16b23
going through chapter 28 of https://www.lua.org/pil
...
User-defined C data.
I think I have some understanding of the Lua stack now. It's a different
kind of verbose, error-prone syntax than Mu that requires me to play
computer in my head. But I don't fully grok metatables yet. At least not
well enough to grok everything that's going on in lcurses/ext.
2021-11-05 08:31:43 -07:00
e35454c9a8
I don't yet understand the stack
2021-10-24 14:20:37 -07:00
d8fcd648d7
ok, starting to make sense now
...
Putting together two resources:
https://lucasklassmann.com/blog/2019-02-02-how-to-embeddeding-lua-in-c/#exposing-a-simple-variable
https://www.lua.org/manual/5.3/manual.html , section 2.1, "Values and Types", particularly the description of light user data.
And lo, I see lua_pushlightuserdata in lapi.c
2021-10-24 10:32:22 -07:00
4a0822929b
done reading lua_newstate
2021-10-24 10:08:15 -07:00
e3b1ac2c83
mildly less confusing
2021-10-24 09:58:47 -07:00
7b285449c3
back to making sense of lua_newstate
2021-10-24 09:57:19 -07:00
4a28aace9a
get rid of userstate hooks
2021-10-24 09:52:07 -07:00
67e4fb8c8e
get rid of fromstate/tostate
2021-10-24 09:46:56 -07:00
a1d3225305
get rid of state_size
2021-10-24 09:39:30 -07:00
a3c090ce09
get rid of LUAI_EXTRASPACE
...
Lua's power may come from extensibility, but the indirections are
currently in the way
2021-10-24 09:35:12 -07:00
c80bafed74
trying to make sense of how bindings are created
...
lua_State contains these StkId fields (stack, stack_last, base, top)
that expand to a pointer of a struct containing a Lua value and an int.
Unclear how it's used, or how you build a stack out of it.
2021-10-24 09:23:54 -07:00
8d8580089f
print's newline now returns to column 0
...
At this point I'm done making this repo ncurses-ready. Remaining files
that allude to stdin/stdout/stderr:
lauxlib.c - unclear how these primitives should work; may kill them
ldblib.c - unclear what debug experience should be
liolib.c - might kill or simulate these
luac.c - let the compiler continue to be a terminal program
2021-10-22 21:22:27 -07:00
05fa5124c2
drop lua_stdin_is_tty
...
luaconf.h now no longer refers to stdin/stdout/stderr.
2021-10-22 21:02:47 -07:00
9792ac1e09
drop support for '-' filename
...
lua.c now no longer refers to stdin/stdout/stderr.
2021-10-22 21:00:26 -07:00
1445cbc5b1
ncurses I/O in lua.c
...
Currently working:
> print(1)
1
> print(2)
2
Print's newline doesn't return to column 0 yet.
Ctrl-d no longer works. Ctrl-c exits cleanly.
2021-10-22 20:54:34 -07:00
d92a484b91
kill lua_readline abstraction
2021-10-22 20:37:56 -07:00
9eb951608b
purge notion of commandline history
2021-10-22 20:30:13 -07:00
0ab2c77e6c
delete readline support
...
We're going to be using full-on ncurses.
2021-10-22 20:29:31 -07:00
c03ee20559
clean up a few warnings with gcc 9.3.0
2021-10-22 19:25:59 -07:00
74f8cd15bb
new fork of Lua 5.1
...
https://www.lua.org
2021-10-22 19:24:44 -07:00
Kartik K. Agaram