ada003b346
Split rope implementation off into external lib "Ropey".
...
Ropey can be found at https://github.com/cessen/ropey
2015-02-25 20:05:17 -08:00
82e6fca1dd
Updated code for latest rustc.
2015-02-23 20:50:55 -08:00
d3cd1c4b05
Fixed bug in Editor::move_view_to_cursor().
...
Was causing the view to move unecessarily when the cursor is at the end of
the buffer.
2015-02-22 20:35:41 -08:00
c532b10f22
Fixed bug in Editor::move_view_to_cursor()
...
It thought it was one line shorter than it actually was.
2015-02-22 20:30:33 -08:00
c348f252bb
Fixed bug in buffer drawing code.
...
It was causing the entire last line to be re-scanned to find the
cursor drawing position if the cursor was at the end of the buffer.
This was both slow and resulted in an incorrect position, since
all the rest of the drawing assumes line blocks.
2015-02-22 20:24:19 -08:00
ae26c5f873
Fixed bugs in LineFormatter that was causing weird cursor nav behavior.
...
They were two pretty much identical bugs in both index_offset_vertical_v2d()
and index_set_horizontal_v2d(). Neither method was checking for slipping
off the end of a line block.
2015-02-22 20:09:18 -08:00
1b3acfd29f
Fixed bug in LineFormatter::index_offset_vertical_v2d()
...
The bug was causing very incorrect behavior when a single line was long
enough to be split into multiple blocks.
2015-02-22 19:30:13 -08:00
9c462ee332
Finished converting codebase back to plain Rope's.
...
Everything appears to be working identically to how it was before converting
back.
2015-02-22 18:13:10 -08:00
fee6fc21a3
Removed line and node source files since they are no longer used.
2015-02-22 12:30:02 -08:00
dcfa3fe0d5
Got unit tests compiling again, and added a few more to Rope.
2015-02-22 12:28:31 -08:00
02f24e4e6f
WIP refactor to switch over to Rope from BufferNode.
2015-02-22 11:45:57 -08:00
0ee364f6ce
Added Rope methods for converting between grapheme indices and line/column indices.
2015-02-22 00:15:18 -08:00
759b23db4d
Added methods to Rope for finding the relationship between lines and graphemes.
2015-02-21 23:51:15 -08:00
c276117ac6
Rope now tracks line endings.
2015-02-21 22:55:14 -08:00
0da0b532d0
Forgot to set rope leaf size back to something reasonable.
2015-02-21 13:18:28 -08:00
d87dd9a343
Implemented efficient Rope splitting.
...
Rope::split() should now run in O(log N) time instead of O(N).
2015-02-21 13:12:26 -08:00
e082defad8
Added benchmarks for Rope::append() and Rope::split().
2015-02-21 11:44:32 -08:00
52cb5f9456
Improved Rope's text insertion performance for large insertion strings.
2015-02-21 11:16:21 -08:00
a37bc17289
Even further improved balance unit tests.
...
Now the ropes are directly checked for adhering to the AVL balance
invariants.
2015-02-19 23:39:24 -08:00
b3ba8acdf5
Significantly improved Rope::rebalance() unit tests.
...
Now they directly run rebalance() on hand-constructed trees that test
various situations.
2015-02-19 23:23:36 -08:00
fe0a7e48ca
Fixed bug in Rope::new_from_str().
...
The stragglers appended at the end were being appended at the top of
the tree, making it very unbalanced. Using proper appending now.
2015-02-19 22:52:21 -08:00
2df1302723
Fixed bug in Rope::rebalance(). Stats weren't updated after rebalance.
2015-02-19 22:46:30 -08:00
5500c845f5
Ropes can now generate graphviz of their tree. Useful for debugging.
2015-02-19 22:39:57 -08:00
de668c5ddf
Improved Rope::append()
...
Append no longer depends on a performant rebalance() operation in any
cases. Append is guaranteed O(log N) now.
2015-02-18 23:41:28 -08:00
d95161b88f
Improved Rope::rebalance() to actually rebalance properly in non-AVL cases.
...
The prior version worked fine for the typical single insertion/removal
cases. But when larger imbalances happened due to bulk insertion or
removal it wouldn't work correctly.
The new improved version works on any tree whose children are themselves
balanced, even if they are extremely different heights when compared to
each other. It works in O(log N) time.
2015-02-18 23:11:10 -08:00
535f219c41
Improved algorithm of Rope::append()
...
It doesn't use an intermediate conversion to a string anymore, and should
execute in O(log N) as long as the right-hand rope is equal length or
smaller than the left.
It may also be O(log N) for the left-hand rope being smaller, but that
depends on whether the rebalance code executes in O(log N) time when
the left and right hand side are individually balanced. I haven't
analysed it yet, so dunno.
2015-02-17 23:10:38 -08:00
d26c95ab1d
Fixed some bugs in the display code.
...
Everything seems to display correctly now.
2015-02-16 22:46:38 -08:00
311f8aca67
First attempt at making display code really take advantage of long line support.
...
It's not quite working right, but it does at least prevent performance
degradation.
2015-02-16 22:29:41 -08:00
86f3ada8ea
Small tweak to indentation style auto-detection.
...
Very slightly favor larger indentations, instead of very slight favor
smaller indentations. This is more likely to do the right thing on
very small files.
2015-02-16 22:06:48 -08:00
d3cbc7644f
Updates for latest rustc.
2015-02-16 22:01:22 -08:00
0bc809781d
WIP: fixes for long line support.
2015-02-15 23:23:17 -08:00
ecf91607ac
WIP long line support.
...
Got index_offset_vertical_v2d() partially working. Still seems to be a tad
buggy, though.
2015-02-15 23:09:13 -08:00
a1d636a4d8
WIP: handling extremely long lines with good performance.
...
The formatters now work on grapheme iterators instead of directly on
lines, which frees up the LineFormatter to break up long lines into
smaller blocks of text. This is partially taken advantage of right
now in various parts of the code, but more work is still needed to
get it both working properly and fast.
2015-02-15 22:41:11 -08:00
f4a70ba1ad
Made screen drawing a tad faster for long lines.
...
In preparation for much more major speed-ups with long lines.
2015-02-15 20:35:17 -08:00
6eed3fcc74
Improved file load times.
...
There was a regression after switching to ropes. Now it's pretty much
back to pre-rope load times.
2015-02-15 20:15:19 -08:00
de58f01b2f
Added some benchmarks for Rope, and improved new_from_str() for Rope.
...
The new new_from_str() uses an algorithm that should be linear time
complexity compared to the length of the string.
2015-02-15 19:42:07 -08:00
e1c9a6955d
Lines now internally use ropes to store their text.
...
The rope uses a large leaf-node text length, so in the vast majority of
cases this ends up being the same as directly storing the string data.
But in the case that the line becomes extremely long, this will allow
for reasonable performance
2015-02-15 14:56:56 -08:00
a89e505fb0
Beginnings of a Rope data structure for strings.
...
This will be used to hold the text of an individual line, for better
long-line handling.
2015-02-14 14:19:46 -08:00
902d93e957
Line numbers are now displayed in left-hand gutter.
2015-02-08 19:17:48 -08:00
9b24b2b4f2
Added rust-encoding as a dependency, in prep for text encoding support.
2015-02-08 00:19:37 -08:00
ed888939bf
Updated todo file.
2015-02-07 23:38:32 -08:00
2d50906f43
Moved the rest of file loading/saving into Buffer.
2015-02-07 23:35:25 -08:00
8a077bf4d0
Updated todo file.
2015-02-07 23:27:38 -08:00
8d76e3fb17
View scrolling seems to work properly now.
2015-02-07 23:14:56 -08:00
8f94b32953
Have view scrolling nearly working again.
...
Just doesn't *quite* work with wrapped lines. Need to investigate.
2015-02-07 22:57:25 -08:00
870818fec8
Cursor movement appears to be working properly again.
2015-02-07 22:07:48 -08:00
27959dadf8
WIP fixing cursor movement code.
2015-02-07 21:38:16 -08:00
6fc2f34053
Finished converting the term_ui drawing code over to the new formatting code.
2015-02-07 21:37:11 -08:00
e9462e4400
Silencing some rustc warnings.
2015-02-07 19:55:25 -08:00
26965417f3
WOrking on getting cursor movement working again.
2015-02-07 19:46:08 -08:00
