9f4b41d829
exclude left margin from my word-split heuristic
...
Gratifying how few tests need changing. Recent commits seem on the right
track.
2022-07-20 07:15:46 -07:00
a472d218f2
allow Text.nearest_pos_less_than to return 0
...
This eliminates another case of overflowing margins.
2022-07-20 07:08:28 -07:00
7dc47edde8
drop an unnecessary level of indent
2022-07-20 07:03:33 -07:00
39f71784c9
clean up some prints
...
(Looking at prints in Text.draw, Text.compute_fragments and
Text.nearest_pos_less_than)
2022-07-20 07:00:39 -07:00
b5451e4351
move a var closer to its use
2022-07-20 06:56:19 -07:00
d41301c1b7
more clearly skip prints before screen top
2022-07-20 06:54:03 -07:00
26ba6e4e5a
unify two similar functions
...
The cost is just having to tweak a few more brittle tests. I can't
actually perceive any difference in how the cursor moves when I click on
text.
2022-07-19 17:13:16 -07:00
efd4a8a883
keep text from overflowing right margin
...
I've been sloppy about this so far, and outside of tests I can't find
any examples where it matters, but it matters in a potential fork where
I'm rendering multiple columns of text.
It's unfortunate that my tests have this level of brittleness. What I'd
really like to assert in many of these changed lines is that the text
stays inside the margins and that more text would overflow margins.
2022-07-19 16:49:08 -07:00
cc5ab51c53
delete some duplicate initialization
2022-07-19 08:04:21 -07:00
8abd986054
clean up a cross-test leakage
2022-07-19 07:50:53 -07:00
82cdd9ddd1
bugfix: couple of margin-relative computations
2022-07-17 22:29:37 -07:00
29dac6a6ec
separate data structure for each line's cache data
...
I have no idea what the performance implications of this are..
2022-07-17 21:05:01 -07:00
64eed74980
keep all text cache writes inside text.lua
2022-07-17 10:06:39 -07:00
4ff6de9951
switch to line index in a function
...
- Text.screen_line_width
2022-07-17 09:37:50 -07:00
7e7f5b9256
switch to line index in a function
...
- Text.compute_fragments
2022-07-17 09:30:50 -07:00
860cd49f67
make a function oblivious to line data structure
...
- Text.screen_line_index
2022-07-17 09:21:57 -07:00
2859063d9d
switch to line index in a function
...
- Text.to_pos_on_line
2022-07-17 09:19:53 -07:00
6b3ddc816f
switch to line index in a function
...
- Text.populate_screen_line_starting_pos
2022-07-17 09:14:10 -07:00
48b7de4fde
switch to line index in a function
...
- Text.in_line
2022-07-17 09:14:10 -07:00
c0ea369607
drop some extra args
2022-07-17 09:14:10 -07:00
b95206fd0d
drop some redundant args when clearing the cache
2022-07-17 08:18:48 -07:00
2fc5bb9a61
fold together two largely similar cases
2022-07-17 08:05:22 -07:00
038f06a342
drop some redundant args from Text.draw
2022-07-17 08:00:56 -07:00
0ac44f0756
formatting
2022-07-16 19:27:58 -07:00
f61d4269f5
bring back a set of constants
...
It's starting to sink in that I don't want hard-coded constants inside
objects.
2022-07-16 08:48:02 -07:00
c26b377ab3
more decoupling editor tests from App
2022-07-16 08:33:05 -07:00
a2b654b534
more decoupling editor tests from App
2022-07-16 08:27:11 -07:00
51b4efa59a
more decoupling editor tests from App
2022-07-16 08:20:47 -07:00
a1da18c107
more precise name
2022-07-15 22:15:07 -07:00
2884f872d8
make test initializations a little more obvious
2022-07-15 22:12:25 -07:00
fc6b602342
bugfix: zoom in/out hotkeys
...
Broken since commit 3b36093553
2022-07-13 23:31:52 -07:00
119abbd52e
bugfix: resize
2022-07-13 15:40:14 -07:00
e4f9d50a55
a more radical attempt at ignoring nil y's
2022-07-13 09:31:54 -07:00
d231f4d37d
no, bring back that defense
...
Scenario where I ran into it:
start a stroke inside a drawing
press 'o' to switch to circle mode (but any mode switch will do)
The problem is that Text.textinput unconditionally blows away all .y
values. Sometimes we have to wait for a draw() to fill them back in.
2022-07-13 09:21:19 -07:00
22ad5373c5
a bug I've never run into
...
The previous commit was failing inside a test that I can't reproduce
manually. Perhaps it's something about how often draw/update run in
practice. Anyways, it's definitely uncovered a real issue.
2022-07-13 08:40:04 -07:00
6659de60f3
hoist couple of variables out
...
This seems to uncover a corner case I'd never run into until now.
2022-07-13 08:24:16 -07:00
3b1f2cf9de
use available variables
2022-07-13 08:19:58 -07:00
5017659bf4
more idiomatic variable names
2022-07-13 08:19:06 -07:00
457136a986
chunking by simple local variable
2022-07-13 08:14:01 -07:00
9f962b7cac
make local functions look different
2022-07-13 08:04:06 -07:00
cfa04c3ca1
more chunks, same approach
2022-07-13 08:02:41 -07:00
5a04d0c221
chunk up some long lines
...
Hopefully this is more approachable. Though now I'm creating two new
temporary functions on every draw. Whether I need them or not.
2022-07-13 07:50:50 -07:00
75ec5c9087
.
2022-07-12 22:13:06 -07:00
9e9e8ed35c
drop final mention of state global beyond main.lua
...
This is all unfortunate in several ways
- lots of functions have extra args
- table lookups where we used to have a simple variable
- program initialization is a lot more delicate
Let's see if it was worthwhile. Can we now actually build around the
editor component in forks?
2022-07-12 22:05:53 -07:00
70a15b3bc9
replace globals with args in a few functions
...
- Drawing.draw_shape
- Drawing.draw_pending_shape
- Drawing.in_drawing
- Drawing.find_or_insert_point
- Drawing.near
- Drawing.pixels
2022-07-12 21:58:46 -07:00
307e875c1c
correct a mis-named threshold
2022-07-12 21:41:09 -07:00
3dd5413dbf
drop heavyweight near check on file load/store
2022-07-12 21:29:30 -07:00
2ae9cacd97
deduce left/right from state where possible
2022-07-12 20:59:24 -07:00
3b36093553
left/right margin -> left/right coordinates
...
Editor state initialization now depends on window dimensions, so we have
to more carefully orchestrate startup.
2022-07-12 20:54:50 -07:00
1ede1c3c6d
add state arg to a few functions
...
- Drawing.current_drawing
- Drawing.select_shape_at_mouse
- Drawing.select_point_at_mouse
- Drawing.select_drawing_at_mouse
2022-07-12 17:27:00 -07:00
Kartik K. Agaram