editor/src/main.rs, branch mainA terrible text editorhttps://code.roseh.moe/editor/atom/src/main.rs?h=main2024-01-14T05:05:59ZApply some clippy suggestions.2024-01-14T05:05:59ZRose Hogensonrosehogenson@posteo.net2024-01-14T05:05:59Zurn:sha1:c92c5a78e6333b6025131c031c6466413e60e442Escape control characters.2024-01-14T01:45:37ZRose Hogensonrosehogenson@posteo.net2024-01-14T01:45:37Zurn:sha1:d3f52cf5dc1e06a81fb75a8631ac00e8adc04ac7Reject files with invalid utf-8.2024-01-13T20:31:42ZRose Hogensonrosehogenson@posteo.net2024-01-13T20:31:42Zurn:sha1:1e32f5d480ecefa3a1c8fd1c43ac6b4260459266
It's easier to only support files with valid utf-8.
Start tuning the performance of rope.2024-01-13T05:03:25ZRose Hogensonrosehogenson@posteo.net2024-01-13T05:03:25Zurn:sha1:a41476ac28083a2d8a54161d4177b3ee5c028525Fix a bug where cursor_pos would return old data.2024-01-12T16:31:12ZRose Hogensonrosehogenson@posteo.net2024-01-12T16:31:12Zurn:sha1:d333343b15aab5747ccf74c495cb085b4e32b029
For now I'm just counting the failures, and aborting if cursor_pos
immediately failed before. We have to skip the incoming status reports
with read_key before cursor_pos will start working again. This should
flush any old status reports, and prevent reading them incorrectly.
Implement Home, End, PgUp, and PgDn.2024-01-12T05:31:39ZRose Hogensonrosehogenson@posteo.net2024-01-12T00:38:04Zurn:sha1:e23b1c7254aa21c0eb2fd18f720df262725c6d81Keep the cursor position on error.2024-01-12T03:04:27ZRose Hogensonrosehogenson@posteo.net2024-01-12T03:04:27Zurn:sha1:28dffed640adf8815294faa7a1f4b4d8a50c9707
Now it handles errors so well! You can barely tell that nearly all of
the calls to cursor_pos are failing.
Do a little more to handle errors.2024-01-12T02:53:20ZRose Hogensonrosehogenson@posteo.net2024-01-12T02:53:20Zurn:sha1:965c0aa3446da87b84c5db656b1cc3ce6e4a691bOptimize for ASCII case in floor/ceil_grapheme.2024-01-12T02:47:06ZRose Hogensonrosehogenson@posteo.net2024-01-12T02:47:06Zurn:sha1:781478d73dd3dfe42596e1dd370339ca6ad30e31Optimise line_offset for the ASCII case.2024-01-12T02:01:10ZRose Hogensonrosehogenson@posteo.net2024-01-12T02:01:10Zurn:sha1:7c66a5c1216f334bfa29bddd7e4c3a451900efbb
In most cases we really won't need the full binary search. We can check
quickly for ASCII without ever printing anything, and fall back to
binary search if we find any Unicode characters.