Annotation of /branches/pure-cfg/TODO

Parent Directory | Revision Log

---

Revision 1049 - (view) (download)

1 :	glk	1049	NOTE: GLK's approximate ranking of 5 most important tagged with
2 :	glk	962	[GLK:1], [GLK:2], ...
3 :
4 :	glk	1049	==============================
5 :			other SHORT TERM ============= (including needed for LIC)
6 :			==============================
7 :	glk	781
8 :	glk	1049	[GLK:1] Add sequence types (needed for evals & evecs)
9 :			syntax
10 :			types: ty '{' INT '}'
11 :			value construction: '{' e1 ',' … ',' en '}'
12 :			indexing: e '{' e '}'
13 :
14 :	jhr	788	IL support for higher-order tensor values (matrices, etc).
15 :	jhr	803	tensor construction [DONE]
16 :			tensor indexing [DONE]
17 :			tensor slicing
18 :	jhr	853	verify that hessians work correctly [DONE]
19 :	jhr	788
20 :			Use ∇⊗ etc. syntax
21 :	jhr	763	syntax [DONE]
22 :			typechecking
23 :			IL and codegen
24 :
25 :	glk	927	test/uninit.diderot:
26 :	glk	941	documents need for better compiler error messages when output variables
27 :	glk	927	are not initialized; the current messages are very cryptic
28 :
29 :	glk	785	determinant ("det") for tensor[3,3]
30 :
31 :	jhr	788	expand trace in mid to low translation
32 :
33 :			value-numbering optimization
34 :
35 :			Add type aliases for color types
36 :			rgb = real{3}
37 :			rgba = real{4}
38 :
39 :	glk	781	==============================
40 :	glk	891	MEDIUM TERM ================== (including needed for streamlines & tractography)
41 :	glk	781	==============================
42 :
43 :	glk	1049	[GLK:1] evals & evecs for symmetric tensor[3,3] (requires sequences)
44 :	glk	962
45 :	glk	1049	[GLK:2] Save Diderot output to nrrd, instead of "mip.txt"
46 :	glk	962	For grid of strands, save to similarly-shaped array
47 :			For list of strands, save to long 1-D (or 2-D for non-scalar output) list
48 :			For ragged things (like tractography output), will need to save both
49 :			complete list of values, as well as list of start indices and lengths
50 :			to index into complete list
51 :
52 :	glk	1049	[GLK:3] Use of Teem's "hest" command-line parser for getting
53 :	glk	962	any input variables that are not defined in the source file
54 :
55 :	glk	1049	[GLK:4] ability to declare a field in such a way so that probe
56 :	glk	962	positions are *always* clamped to the support of "inside";
57 :			there are many cases where this is the sensible behavior.
58 :	glk	1049	(could also have "repeat" declaration, to mimic GL_REPEAT texturing)
59 :	glk	962
60 :	glk	1049	extend norm (|exp|) to all tensor types [DONE for vectors and matrices]
61 :
62 :	glk	962	ability to emit/track/record variables into dynamically re-sized
63 :			runtime buffer
64 :
65 :	glk	945	Want: allow X *= Y, X /= Y, X += Y, X -= Y to mean what they do in C,
66 :			provided that X*Y, X/Y, X+Y, X-Y are already supported.
67 :			Nearly every Diderot program would be simplified by this.
68 :
69 :	glk	962	Want: non-trivial field expressions & functions:
70 :	glk	948	image(2)[2] Vimg = load(...);
71 :	glk	950	field#0(2)[] Vlen = |Vimg ⊛ bspln3|;
72 :	glk	948	to get a scalar field of vector length, or
73 :	glk	950	field#2(2)[] F = Fimg ⊛ bspln3;
74 :			field#0(2)[] Gmag = |∇F|;
75 :	glk	948	to get a scalar field of gradient magnitude, or
76 :	glk	950	field#2(2)[] F = Fimg ⊛ bspln3;
77 :			field#0(2)[] Gmsq = ∇F•∇F;
78 :			to get a scalar field of squared gradient magnitude, which is simpler
79 :	glk	962	to differentiate. However, there is value in having these, even if
80 :			the differentiation of them is not supported (hence the indication
81 :			of "field#0" for these above)
82 :	glk	945
83 :	glk	962	Want: ability to apply "normalize" to a field itself, e.g.
84 :			field#0(2)[2] V = normalize(Vimg ⊛ ctmr);
85 :			so that V(x) = normalize((Vimg ⊛ ctmr)(x)).
86 :			Having this would simplify expression of standard LIC method, and
87 :			would also help express other vector field expressions that arise
88 :			in vector field feature exraction.
89 :	glk	941
90 :	glk	781	tensor fields: convolution on general tensor images
91 :
92 :			==============================
93 :			other MEDIUM TERM ============ (needed for particles)
94 :			==============================
95 :
96 :	glk	1049	[GLK:5] run-time death of strands; test/iso2d.diderot provides
97 :			great initial test for "die" command.
98 :	glk	781
99 :	glk	1049	run-time birth of strands
100 :
101 :	glk	781	"initially" supports lists
102 :
103 :			"initially" supports lists of positions output from
104 :			different initalization Diderot program
105 :
106 :			spatial data structure that permits strands' queries of neighbors
107 :
108 :	jhr	784	proper handling of stabilize method
109 :
110 :	glk	891	test/vr-kcomp2.diderot: Add support for code like
111 :	jhr	796
112 :			(F1 if x else F2)@pos
113 :
114 :	glk	962	This will require duplication of the continuation of the conditional
115 :			(but we should only duplicate over the live-range of the result of the
116 :			conditional.
117 :	jhr	796
118 :	glk	1049	add ":" for tensor dot product (contracts out two indices
119 :			instead of one like •), valid for all pairs of tensors with
120 :			at least two indices
121 :
122 :	glk	781	==============================
123 :	jhr	797	other MEDIUM TERM ============
124 :			==============================
125 :
126 :	glk	927	want: warnings when "D" (reserved for differentiation) is declared as
127 :			a variable name (get confusing error messages now)
128 :	jhr	797
129 :	jhr	824	support for Python interop and GUI
130 :
131 :	glk	1049	Python/ctypes interface to run-time
132 :
133 :	jhr	797	==============================
134 :	glk	781	LONG TERM ====================
135 :			==============================
136 :
137 :	glk	927	Better handling of variables that determines the scope of a variable
138 :			based on its actual use, instead of where the user defined it. So,
139 :			for example, we should lift strand-invariant variables to global
140 :			scope. Also prune out useless variables, which should include field
141 :			variables after the translation to mid-il.
142 :	jhr	796
143 :	glk	781	co- vs contra- index distinction
144 :
145 :			some indication of tensor symmetry
146 :			(have to identify the group of index permutations that are symmetries)
147 :
148 :			dot works on all tensors
149 :
150 :			outer works on all tensors
151 :
152 :			Einstein summation notation
153 :
154 :			"tensor comprehension" (like list comprehension)
155 :	jhr	784
156 :			======================
157 :			BUGS =================
158 :			======================
159 :
160 :	glk	1049	test/fields.diderot:
161 :			Can't assign on field to another.
162 :	glk	962
163 :			test/read2vecs.diderot:
164 :			// HEY (BUG?) shouldn't it be a type error to load this 2-D array of
165 :			// 2-vectors into a 2-D *scalar* field? Instead, get:
166 :			// uncaught exception Fail [Fail: Error in compiling lic.diderot]
167 :			// raised at driver/main.sml:31.39-31.76
168 :			image(2)[] Vimg = load("../data/vorttest.nrrd");
169 :
170 :	jhr	784	test/zslice2.diderot:
171 :			// HEY (bug) bspln5 leads to problems ...
172 :			// uncaught exception Size [size]
173 :			// raised at c-target/c-target.sml:47.15-47.19
174 :			//field#4(3)[] F = img ⊛ bspln5;

---

root@smlnj-gforge.cs.uchicago.edu	ViewVC Help
Powered by ViewVC 1.0.0