db8fa8ae80
add getPrefixList to bro-tree (just folds over the tree in O(n) time, which is the best we can do for that data structure). Will add a benchmark comparison of this.
Humza Shahid2024-09-11 11:25:33 +01:00
af504350c7
further improvements to conv-words.sml, escaping chars when necessary, as we may generate an invalid .sml file otherwise
Humza Shahid2024-09-11 10:01:30 +01:00
442f9f88d2
better word tokenisation algorithm in bench/conv-words.sml
Humza Shahid2024-09-11 09:50:07 +01:00
662be1da75
a bit of progress moving insert functions into zip-string-set.sml
Humza Shahid2024-09-10 09:05:41 +01:00
f97d092996
implement getPrefixSubtrie to ZipStringSet
Humza Shahid2024-09-09 21:42:56 +01:00
ede0da6643
alternative isEmpty implementation in string-set.sml
Humza Shahid2024-09-09 18:22:56 +01:00
6228a259bc
extract more duplicated code into reusable functions (in src/string-set.sml)
Humza Shahid2024-09-09 18:17:07 +01:00
272ccf1b03
amend compilation errors in zip-string-set.sml
Humza Shahid2024-09-09 18:04:11 +01:00
dbc0fdc178
progress with zip-string-set.sml (refactored up to the exists function), and reduce code duplication for one case in string-set.sml
Humza Shahid2024-09-09 17:39:45 +01:00
b7a365c6b8
implement 'exists' function on bro-tree as well, and benchmark it compared to string-set
Humza Shahid2024-09-09 15:57:58 +01:00
fc0fd6923b
use larger set of words in words.txt dataset
Humza Shahid2024-09-09 15:19:18 +01:00
7d707c9af4
add insert benchmark comparing StringSet to BroTree (StringSet is just a few ms faster for this)
Humza Shahid2024-09-09 14:54:25 +01:00
5d121f3864
done testing string-set.sml, with all tests passing
Humza Shahid2024-09-08 04:57:41 +01:00
36c9b45c1a
handle duplicates in insertion into trie properly (return same trie if it already has a FOUND/FOUND_WITH_CHILDREN node awith that key, or change tag of node from CHILDREN to FOUND_WITH_CHILDREN to mark it as inserted
Humza Shahid2024-09-08 04:41:14 +01:00
88c48b0592
add 1-line toList function to StringSet, which is not useful to me but very convenient to add
Humza Shahid2024-09-07 21:57:01 +01:00
738e7d93d1
fix all bugs in remove5; now, all remove tests have been ported and they all pass fully; should port other tests from KTrie next too
Humza Shahid2024-09-07 21:54:11 +01:00
01d79cd801
address a bug where Vector.sub (keys, ~1) was used, solving subscript error, but revealing another error in a subsequent test case
Humza Shahid2024-09-07 21:35:31 +01:00
352e4ff611
identify error (subscript raised) with a comment by the offending line stating that it needs to be fixed
Humza Shahid2024-09-07 20:56:21 +01:00
ab1b7e4557
I think I'm done coding removal functionality; next thing to do is test it
Humza Shahid2024-09-06 10:31:52 +01:00
9154fe5b3f
added empty value for trie (which is different from an empty type constructor), and check against empty at the start of insert and search/exists function
Humza Shahid2024-09-05 03:06:00 +01:00
7cfa393e05
tested getPrefixList, fixed errors, and now it seems to work fine
Humza Shahid2024-09-04 19:30:29 +01:00
dd41c971e3
seem to have fixed error with 'exists' function previously mentioned in todo, without introducing regressions
Humza Shahid2024-09-04 18:57:39 +01:00
b2b0837524
just formatting, and removing part of the todo list
Humza Shahid2024-09-04 16:39:13 +01:00
20de15b349
fixed 'helpExists' function; error was in the FOUND case which checked 'keyPos = String.size searchKey' instead of 'keyPos = String.size searchKey - 1'. The latter is correct, because it checks that the keyPos ended at the last character, which is the intended meaning
Humza Shahid2024-09-04 16:38:27 +01:00
86f171cd08
try implementing getPrefixList function, but it seems insert function has an error, so made a note of it to fix this error in a comment; fixing this error is the first priority
Humza Shahid2024-09-04 15:26:21 +01:00
ccd0ee42d5
add function to retrieve prefix subtrie (but will still want a function that returns a list of prefixes, and a vector of prefixes
Humza Shahid2024-09-04 14:56:20 +01:00
65eff83c0b
fix error with 'helpExists' function; the error was that we recursed in 'helpExists' function by calling keyPos + 1 (nextKeyPos) which is the wrong way to search when the trie key has a string size larger than 1
Humza Shahid2024-09-04 13:32:45 +01:00
6696480a39
amend a couple of errors (wrong sorting for some strings), and add todo for bug discovered
Humza Shahid2024-09-04 13:14:54 +01:00
ec7ff7ede6
some refactoring (and progress filling in the FOUND_WITH_CHILDREN case in the 'helpInsert' function)
Humza Shahid2024-09-04 12:17:06 +01:00
0d25b88caa
change 'helpInsert' function to return trie instead of trie option (I would guess there are fewer cases where the same string is inserted into the trie repeatedly)
Humza Shahid2024-09-04 12:05:43 +01:00
90019de17f
amend bug in 'insertBinSearch' function which caused subscriptt error, thereby addressing one of the todo items
Humza Shahid2024-09-04 11:53:00 +01:00
f518ebb4b3
progress with insert function, adding note for next todo
Humza Shahid2024-09-03 20:16:39 +01:00
2f14c99952
init with working 'exists' function for string set
Humza Shahid2024-09-03 10:15:55 +01:00
2d6ee874af
improve A (add space for other fonts to have tail, make the width a little narrower), and create new directory for fonts
Humza Shahid2024-08-30 22:01:24 +01:00
42c4b92105
address edge case where Real32.toString 1.0 seems to give the string '1' instead of '1.0', and improve the drawing for upper case A
Humza Shahid2024-08-30 11:13:08 +01:00
c16fa58bf6
fix bug caused by the calculation change in the previous commit
Humza Shahid2024-08-30 09:15:33 +01:00
f3a0eb3e30
minor change to generated export code (avoid redundant maths operations)
Humza Shahid2024-08-30 05:20:56 +01:00
db92b95c41
tick another item off from the to-do list in the readme
Humza Shahid2024-08-30 03:57:13 +01:00
f5b0c8e705
tested export functionality, and fixed it so that it works as intended (previous error was that the drawn object was located on the wrong part of the Y-axis)
Humza Shahid2024-08-30 03:46:05 +01:00
acd68a055d
code shortcut to trigger export (with <Ctrl-E>); it outputs syntatically valid .sml, but I am not sure if the output is correct as intended so it needs testing
Humza Shahid2024-08-30 02:34:24 +01:00
7ab7b8931d
code function to export triangle to .sml file (linear interpolation only; untested and also not currently executable by shortcut
Humza Shahid2024-08-30 02:22:02 +01:00
da2a652506
improve parsing code slightly, with successful parsing when saved .dsc file has inconsistent spacing
Humza Shahid2024-08-29 11:18:17 +01:00
24bc45142d
slight refactoring (move pure parsing functions to functional core in a separate file)
Humza Shahid2024-08-29 09:55:08 +01:00
3bc43aec3a
update readme to mark another to-do task as done
Humza Shahid2024-08-29 05:41:26 +01:00
e69c6b42ef
done with loading file into app (loaded triangles are visible, etc.)
Humza Shahid2024-08-29 05:38:58 +01:00
2549703dc7
verify that parsing saved file is successful
Humza Shahid2024-08-29 05:21:04 +01:00
f0958686b7
implement parser for loading (but this parser is untested)
Humza Shahid2024-08-29 04:39:23 +01:00
1fdd4f33ec
remember to close file descriptor in file thread
Humza Shahid2024-08-29 00:43:51 +01:00
eb13636993
very basic saving (hardcoded file name, which is fine fow now)
Humza Shahid2024-08-29 00:05:30 +01:00
c40e8b98e0
slight renaming of constructors in FileMessage.t datatype, to provide more clarity (SAVE_TRIANGLES is for saving to a custom format, LOAD_TRIANGLES is for loading from a file in that custom format, and export is for generating code from custom format)
Humza Shahid2024-08-28 20:47:40 +01:00
48c721e29b
done with scaffolding for file mailbox; next, send message to file mailbox to export and import, and handle these messages properly from the file mailbox
Humza Shahid2024-08-28 20:42:52 +01:00
e587c6cf73
preparation for adding new additional threads/mailboxes (namely, the file mailbox)
Humza Shahid2024-08-28 20:11:00 +01:00
fb9202248c
a little refactoring (different CML loops have their own files now)
Humza Shahid2024-08-28 19:34:47 +01:00
0a4de92862
in C files where GLFW is included, #define GLFW_INCLUDE_NONE before GLFW is included in that file, because GLFW tries looking for <GL/gl.h> header otherwise, which is not on my system
Humza Shahid2024-08-28 00:58:12 +01:00
1049f2b561
scaffolding in preparation for adding simple load/export/save functionality
Humza Shahid2024-08-17 09:46:53 +01:00
3f4c52e0d8
upload new dots to GPU when there is a window resize too
Humza Shahid2024-08-14 21:35:01 +01:00
30dd9de17a
rename 'button' (the coloured square that appears over a clickpoint) to 'dot' throughout codebase, because I may want actual buttons later
Humza Shahid2024-08-14 21:24:46 +01:00
abddddbc9e
fix minor bug when resizing (resizing caused graph to be drawn and resized, even when graph was toggled off)
Humza Shahid2024-08-14 03:54:52 +01:00
a3290337a0
done adding ability to toggle graph by pressing 'g'
Humza Shahid2024-08-14 02:35:44 +01:00
575b8d009e
amend order of Glfw.swapBuffers and Glfw.pollEvents in imperative shell (previous order polled events before swapping buffers which is not the intended order)
Humza Shahid2024-08-14 02:00:07 +01:00
dd9ecfcbe5
delete shaders after linking them in program, which reduces memory by like 0.1 MB (very minor)
Humza Shahid2024-08-14 01:44:30 +01:00
c740e5d751
delete shaders after linking them in program, which reduces memory by like 0.1 MB (very minor)
Humza Shahid2024-08-14 01:43:53 +01:00
0c76edfde2
remove graph lines vector from app type, so the vector can be garbage collected (it is not really needed)
Humza Shahid2024-08-13 23:49:34 +01:00
6878f7fa58
a little clean up (make the ClickPoints.getClickPosition easier for callers to use, because fewer arguments need to be passed)
Humza Shahid2024-08-13 23:38:04 +01:00
c475a06341
fix minor bug with undo (was meant to call TriangleStage.secondToVector with an initial empty vector in one case, for drawing, but mistakenly used a non-empty vector as initial)
Humza Shahid2024-08-11 15:02:04 +01:00
d9c7b753f8
100/home/humza/Downloads/sml/dotscape/README.md done with undo/redo functionality (tested and both of them work as expected)
Humza Shahid2024-08-08 23:52:49 +01:00
4d0d8ef0a1
done with functional-core logic for redo (now just need to send REDO_ACTION on key callback when shortcut is pressed)
Humza Shahid2024-08-08 23:43:38 +01:00
10b3617689
add redo stack, and adjust functions to clear it when new triangle or triangle_stage is added; when an undo function is called, the undoHd becomes the redoHd, and undo becomes undoTl, which is expected behaviour. Still need to implement REDO_ACTION though.
Humza Shahid2024-08-08 23:10:38 +01:00
4243e4c679
refactor a little bit, making signature of app-update.sml simpler to use by storing mouseX and mouseY inside the app_type
Humza Shahid2024-08-08 21:58:50 +01:00
1264e20dc8
remove instances of 'local ... in ... end', replacing it with opauque structures with signatures, because I like the lower indentation level and additional whitespace in the latter
Humza Shahid2024-08-08 01:37:46 +01:00
Humza Shahid