nodes/mcDandy-more_math
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge pull request #4 from mcDandy/review-fixes

Browse Source 
Review fixes
This commit is contained in:
mcDandy
2026-10-18 14:40:35 +02:00
committed by GitHub
parent 85a7a877c6 6fb7a62699
commit 8d2a18d9b4
36 changed files with 157 additions and 26195 deletions
Download Patch File Download Diff File Expand all files Collapse all files
FUNCTION_CONSISTENCY_REPORT.md
-212
View File
@@ -1,212 +0,0 @@
# Function Consistency Check: README.md vs MathExpr.g4
## Executive Summary
✅ **All documented functions have grammar rules defined**
✅ **All grammar functions have documentation**
⚠️ **Minor discrepancies in aliases and function naming**
## Functions in README but Need Verification in Grammar
### ✅ Verified - All Present
#### Basic Math (All Present)
- ✅ abs (AbsFunc)
- ✅ sqrt (SqrtFunc)
- ✅ ln (LnFunc)
- ✅ log (LogFunc)
- ✅ exp (ExpFunc)
- ✅ pow (PowFunc as POWE)
- ✅ floor (FloorFunc)
- ✅ ceil (CeilFunc)
- ✅ round (RoundFunc)
- ✅ fract (FractFunc)
- ✅ sign (SignFunc)
- ✅ gamma (GammaFunc)
- ✅ dist/distance (DistFunc)
- ✅ clamp (ClampFunc)
- ✅ step (StepFunc)
#### Trigonometric (All Present)
- ✅ sin, cos, tan, asin, acos, atan, atan2 (all present)
#### Hyperbolic (All Present)
- ✅ sinh, cosh, tanh, asinh, acosh, atanh (all present)
#### Machine Learning / Activation (All Present)
- ✅ relu (ReluFunc)
- ✅ gelu (GeluFunc)
- ✅ softplus (SoftplusFunc)
- ✅ sigm (SigmoidFunc as SIGM)
- ✅ softmax (SoftmaxFunc)
- ✅ softmin (SoftminFunc)
#### Interpolation (All Present)
- ✅ smoothstep (SmoothstepFunc)
- ✅ smootherstep (SmootherstepFunc)
- ✅ cubic_ease/cubic (CubicEaseFunc)
- ✅ sine_ease/sine (SineEaseFunc)
- ✅ elastic_ease/elastic (ElasticEaseFunc)
- ✅ lerp (LerpFunc)
#### Aggregates & Tensor Operations (All Present)
- ✅ tmin (TMinFunc)
- ✅ tmax (TMaxFunc)
- ✅ smin (SMinFunc)
- ✅ smax (SMaxFunc)
- ✅ sum (SumFunc)
- ✅ mean (MeanFunc)
- ✅ std (StdFunc)
- ✅ var (VarFunc)
- ✅ median (MedianFunc)
- ✅ mode (ModeFunc)
- ✅ quartile (QuartileFunc)
- ✅ percentile (PercentileFunc)
- ✅ quantile (QuantileFunc)
- ✅ dot (DotFunc)
- ✅ moment (MomentFunc)
- ✅ topk (TopkFunc)
- ✅ botk (BotkFunc)
- ✅ topk_ind/topk_indices (TopkIndFunc)
- ✅ botk_ind/botk_indices (BotkIndFunc)
- ✅ sort (SortFunc)
- ✅ argsort (ArgsortFunc)
- ✅ argmin (ArgminFunc)
- ✅ argmax (ArgmaxFunc)
- ✅ unique (UniqueFunc)
- ✅ tnorm (TNormFunc)
- ✅ snorm (SNormFunc)
- ✅ swap (SwapFunc)
- ✅ cossim (CossimFunc)
- ✅ flip (FlipFunc)
- ✅ cov (CovFunc)
- ✅ append (AppendFunc)
- ✅ any (AnyFunc)
- ✅ all (AllFunc)
- ✅ cumsum (CumsumFunc)
- ✅ cumprod (CumprodFunc)
- ✅ tensor (EmptyTensorFunc)
- ✅ flatten (FlattenFunc)
#### Advanced Tensor Operations (All Present)
- ✅ map (MapFunc)
- ✅ ezconvolution/ezconv (EzConvFunc)
- ✅ convolution/conv (ConvFunc)
- ✅ get_value (GetValueFunc)
- ✅ crop (CropFunc)
- ✅ permute/perm (PermuteFunc)
- ✅ reshape/rshp (ReshapeFunc)
- ✅ blur/gaussian (GaussianFunc)
- ✅ edge (EdgeFunc)
- ✅ batch_shuffle/shuffle/select (BatchShuffleFunc)
- ✅ matmul (MatmulFunc)
- ✅ cross (CrossFunc)
#### Optical Flow (All Present)
- ✅ rife (RifeFunc)
- ✅ motion_mask (MotionMaskFunc)
- ✅ flow_to_image (FlowToImageFunc)
- ✅ flow_apply (FlowApplyFunc)
#### FFT (All Present)
- ✅ fft (SFFT)
- ✅ ifft (SIFFT)
- ✅ angle (ANGL)
#### Utility (All Present)
- ✅ print (PrintFunc as PRNT)
- ✅ print_shape/pshp (PrintShapeFunc)
- ✅ pinv (PinvFunc)
- ✅ range (RangeFunc)
- ✅ nan_to_num/nvl (NvlFunc)
- ✅ remap (RemapFunc)
- ✅ timestamp (TimestampFunc as TIMESTAMP)
- ✅ count/length/cnt (CountFunc)
#### Random Distributions (All Present)
- ✅ random_normal/randn/noise (NoiseFunc)
- ✅ random_uniform/rand (RandFunc)
- ✅ random_exponential/rande (ExponentialFunc)
- ✅ random_cauchy/randc (CauchyFunc)
- ✅ random_log_normal/randln (LogNormalFunc)
- ✅ random_bernoulli/randb (BernoulliFunc)
- ✅ random_poisson/randp (PoissonFunc)
- ✅ random_gamma/randg (GammaDistFunc)
- ✅ random_beta/randbeta (BetaDistFunc)
- ✅ random_laplace/randl (LaplaceDistFunc)
- ✅ random_gumbel/randgumbel (GumbelDistFunc)
- ✅ random_weibull/randw (WeibullDistFunc)
- ✅ random_chi2/randchi2 (Chi2DistFunc)
- ✅ random_studentt/randt (StudentTDistFunc)
#### Bitwise Operations (All Present)
- ✅ band/bitwise_and (BitAndFunc)
- ✅ bor/bitwise_or (BitOrFunc)
- ✅ xor/bitwise_xor (BitXorFunc)
- ✅ bnot/bitwise_not (BitNotFunc)
- ✅ bitcount/popcount/popcnt (BitCountFunc)
- ✅ << (LSHIFT)
- ✅ >> (RSHIFT)
#### Stack (All Present)
- ✅ stack_push (PushFunc)
- ✅ stack_pop (PopFunc)
- ✅ stack_get (GetFunc)
- ✅ stack_clear (ClearFunc)
- ✅ stack_has (HasFunc)
## Minor Naming Discrepancies
### README Mentions These Aliases Not Explicitly in Grammar Lexer
1. `now` - README mentions as alias for `timestamp`, but only TIMESTAMP: 'timestamp' is in grammar
- **Fix Needed**: Add `TIMESTAMP: 'timestamp' | 'now';`
2. `flow_view` - README mentions as alias for `flow_to_image`
- **Fix Needed**: Add to `FLOW_TO_IMAGE: 'flow_to_image' | 'flow_view';`
3. `apply_flow` - README mentions as alias for `flow_apply`
- **Fix Needed**: Add to `FLOW_APPLY: 'flow_apply' | 'apply_flow';`
4. Bitwise aliases - Should update for consistency:
- `BAND: 'band' | 'bitwise_and' | 'bitand';` (currently has 'bitand' as alias)
- `BOR: 'bor' | 'bitwise_or' | 'bitor';` (currently has 'bitor' as alias)
- `XOR: 'xor' | 'bitwise_xor' | 'xor';` (should add 'bitwise_xor')
- `BNOT: 'bnot' | 'bitnot' | 'bitwise_not';` (should add 'bitwise_not')
## Summary of Issues
### Critical (Must Fix)
None - all documented functions work
### High Priority (Should Fix)
1. Add missing aliases for consistency:
- `now` for `timestamp`
- `flow_view` for `flow_to_image`
- `apply_flow` for `flow_apply`
- `bitwise_*` names for bitwise operators
### Low Priority (Nice to Have)
- Documentation in README could list all aliases explicitly
## Recommended Grammar Updates
```antlr
// Update these lexer rules:
TIMESTAMP: 'timestamp' | 'now';
FLOW_APPLY: 'flow_apply' | 'apply_flow';
FLOW_TO_IMAGE: 'flow_to_image' | 'flow_view';
BAND: 'band' | 'bitand' | 'bitwise_and';
BOR: 'bor' | 'bitor' | 'bitwise_or';
XOR: 'xor' | 'bitwise_xor';
BNOT: 'bnot' | 'bitnot' | 'bitwise_not';
BITCOUNT: 'bitcount' | 'popcount' | 'popcnt' | 'bitcount';
```
## Conclusion
✅ **Status: EXCELLENT CONSISTENCY**
- 100% of documented functions have grammar rules
- 100% of grammar functions have documentation
- Only minor alias discrepancies that don't affect functionality
- All core functionality is properly defined and documented
**Recommendation**: Regenerate parser after applying the minor grammar updates above to ensure all documented aliases work correctly.
__init__.py
+2 -8
View File
@@ -1,10 +1,4 @@
import traceback
try:
from .more_math.nodes import comfy_entrypoint as comfy_entrypoint
except ImportError:
print("Something is seriously wrong")
traceback.print_exc()
from .more_math.nodes import comfy_entrypoint as comfy_entrypoint
WEB_DIRECTORY = "./web"
__all__ = ["comfy_entrypoint", "WEB_DIRECTORY"]
__all__ = ["comfy_entrypoint", "WEB_DIRECTORY"]
more_math/AudioMathNode.py
+1 -2
View File
@@ -135,7 +135,7 @@ class AudioMathNode(io.ComfyNode):
"batch_count": float(a_w.shape[0]),
} | generate_dim_variables(a_w) | V_norm_waveforms | sample_rates
v_stacked, v_cnt = get_v_variable(V_norm_waveforms, length_mismatch=length_mismatch)
v_stacked, v_cnt = get_v_variable(V_norm_waveforms)
if v_stacked is not None:
variables["V"] = v_stacked
variables["Vcnt"] = float(v_cnt)
@@ -167,6 +167,5 @@ class AudioMathNode(io.ComfyNode):
res_list.append({"waveform": result_chunk, "sample_rate": sample_rate})
return (res_list, stack)
else:
stack = stack if remember_stack else copy.deepcopy(stack)
return ([{"waveform": result, "sample_rate": sample_rate}], stack)
more_math/ConditioningMathNode.py
+2 -3
View File
@@ -147,7 +147,7 @@ class ConditioningMathNode(io.ComfyNode):
"batch_count": a.shape[0],
} | generate_dim_variables(a) | V_norm_tensors
v_stacked, v_cnt = get_v_variable(V_norm_tensors, length_mismatch=length_mismatch)
v_stacked, v_cnt = get_v_variable(V_norm_tensors)
if v_stacked is not None:
variables["V"] = v_stacked
variables["Vcnt"] = float(v_cnt)
@@ -193,7 +193,7 @@ class ConditioningMathNode(io.ComfyNode):
"batch_count": a_p.shape[0] if a_p.numel() > 0 else 0,
} | generate_dim_variables(a_p) | V_norm_pooled
v_stacked, v_cnt = get_v_variable(V_norm_pooled, length_mismatch=length_mismatch)
v_stacked, v_cnt = get_v_variable(V_norm_pooled)
if v_stacked is not None:
variables_pi["V"] = v_stacked
variables_pi["Vcnt"] = float(v_cnt)
@@ -251,5 +251,4 @@ class ConditioningMathNode(io.ComfyNode):
base[0] = (rtensor, new_dict)
res_list = [base]
stack = stack if remember_stack else copy.deepcopy(stack)
return (move_to_device(res_list, original_device),stack)
more_math/GuiderMathNode.py
+10 -5
View File
@@ -92,7 +92,6 @@ class MathGuider:
self.tree1 = expression1
self.inner_model = None # Will be set during sample
self.sigmas = None
self.current_step = 0
self.steps = 0
self.stck = stack
@@ -142,7 +141,6 @@ class MathGuider:
eval_samples, variables = self.setVars(x, sigma, seed, lazy_g_results)
visitor = UnifiedMathVisitor(variables, eval_samples.shape,eval_samples.device,state_storage=self.stck)
result_tensor = visitor.visit(self.tree)
self.current_step = self.current_step + 1;
return as_tensor(result_tensor, eval_samples.shape).to(x.device)
@@ -164,6 +162,13 @@ class MathGuider:
frame_count = eval_samples.shape[time_dim] if time_dim is not None else 1
# sigma has one entry per batch item, all equal
sigma = float(sigma.flatten()[0]) if isinstance(sigma, torch.Tensor) else sigma
current_step = 0
if self.sigmas is not None:
# Index of the step whose interval contains sigma, so extra model calls within a step (heun, dpmpp_2s, ...) keep the same index
current_step = max((i for i, s in enumerate(self.sigmas.tolist()) if s >= sigma), default=0)
def make_lazy_var(value_fn):
value_fn.is_lazy_var = True
return value_fn
@@ -184,10 +189,10 @@ class MathGuider:
"batch_count": float(eval_samples.shape[0]),
"N": float(eval_samples.shape[channel_dim]) if channel_dim < ndim else 0.0,
"channel_count": float(eval_samples.shape[channel_dim]) if channel_dim < ndim else 0.0,
"sigma": sigma.item() if isinstance(sigma,torch.Tensor) else sigma,
"sigma": sigma,
"seed": seed if seed is not None else 0,
"steps": self.steps,
"current_step": self.current_step,
"current_step": current_step,
"sample": x
})
if g_results is not None:
@@ -226,7 +231,7 @@ class MathGuider:
def sample(self, noise, latent_image, sampler, sigmas, denoise_mask=None, callback=None, disable_pbar=False, seed=None):
self.sigmas = sigmas
self.steps = len(sigmas)
self.steps = len(sigmas) - 1
if sigmas.shape[-1] == 0:
return latent_image
active_guiders = [g for g in self.V.values() if g is not None]
more_math/ImageMathNode.py
+1 -3
View File
@@ -129,7 +129,7 @@ class ImageMathNode(io.ComfyNode):
# Add all dynamic inputs
variables.update(V_norm)
v_stacked, v_cnt = get_v_variable(V_norm, length_mismatch=length_mismatch)
v_stacked, v_cnt = get_v_variable(V_norm)
if v_stacked is not None:
variables["V"] = v_stacked
variables["Vcnt"] = float(v_cnt)
@@ -159,8 +159,6 @@ class ImageMathNode(io.ComfyNode):
res_list = []
for result_chunk in res:
res_list.append(result_chunk)
stack = stack if remember_stack else copy.deepcopy(stack)
return (res_list, stack)
else:
stack = stack if remember_stack else copy.deepcopy(stack)
return ([result], stack)
more_math/LatentMathNode.py
+1 -5
View File
@@ -215,7 +215,7 @@ class LatentMathNode(io.ComfyNode):
# Add all dynamic inputs
variables.update(V_norm_samples)
v_stacked, v_cnt = get_v_variable(V_norm_samples, length_mismatch=length_mismatch)
v_stacked, v_cnt = get_v_variable(V_norm_samples)
if v_stacked is not None:
variables["V"] = v_stacked
variables["Vcnt"] = float(v_cnt)
@@ -242,7 +242,6 @@ class LatentMathNode(io.ComfyNode):
if getattr(result_t, "is_nested", False):
rl = result_latent.copy()
rl["samples"] = result_t
stack = stack if remember_stack else copy.deepcopy(stack)
return (move_to_device([rl], original_device), stack)
# If the visitor produced a list/tuple, emit each element as a separate latent.
@@ -252,7 +251,6 @@ class LatentMathNode(io.ComfyNode):
rl = result_latent.copy()
rl["samples"] = comp
results.append(rl)
stack = stack if remember_stack else copy.deepcopy(stack)
return (move_to_device(results, original_device), stack)
if batching > 0:
@@ -262,9 +260,7 @@ class LatentMathNode(io.ComfyNode):
rl = result_latent.copy()
rl["samples"] = result_tensor
results.append(rl)
stack = stack if remember_stack else copy.deepcopy(stack)
return (move_to_device(results, original_device), stack)
rl = result_latent.copy()
rl["samples"] = result_t
stack = stack if remember_stack else copy.deepcopy(stack)
return (move_to_device([rl], original_device), stack)
more_math/MaskMathNode.py
+1 -3
View File
@@ -108,7 +108,7 @@ class MaskMathNode(io.ComfyNode):
"batch_count": float(ae.shape[0]),
} | generate_dim_variables(ae)
v_stacked, v_cnt = get_v_variable(V_norm, length_mismatch=length_mismatch)
v_stacked, v_cnt = get_v_variable(V_norm)
if v_stacked is not None:
variables["V"] = v_stacked
variables["Vcnt"] = float(v_cnt)
@@ -140,8 +140,6 @@ class MaskMathNode(io.ComfyNode):
res_list = []
for result_chunk in res:
res_list.append(result_chunk)
stack = stack if remember_stack else copy.deepcopy(stack)
return (res_list, stack)
else:
stack = stack if remember_stack else copy.deepcopy(stack)
return ([result], stack)
more_math/NoiseMathNode.py
+2 -2
View File
@@ -73,8 +73,8 @@ class NoiseExecutor:
else:
self.tree = expr
self.stack = stack
seed = -1
# Samplers seed their own noise (ancestral/SDE) from noise.seed, so pass the first input's seed through
self.seed = next((v.seed for v in V.values() if v is not None), 0)
def generate_noise(self, input_latent: torch.Tensor) -> torch.Tensor:
samples = input_latent["samples"]
more_math/Parser/MathExpr.g4
+1 -1
View File
@@ -1383,7 +1383,7 @@ func3:
/**
text_image(text, font, size, [max_width], [weight], [rotation_angle], [line_spacing], [italic], [underline]) - renders text to a 2D tensor with values 0-1
- text: text to render; line breaks start new lines
- font: font name to look up in the system fonts, or a path to a font file
- font: font name to look up in the system fonts, or a path to a font file in a system font folder or the ComfyUI input folder
- size: font size in pixels
- max_width: wrap lines wider than this many pixels (default: 0, no wrapping)
- weight: font weight 100-900 as in CSS (default: 400)
more_math/Parser/MathExpr.g4~RF43e401.TMP
-1212
View File
File diff suppressed because it is too large Load Diff
more_math/Parser/MathExpr.interp
-540
View File
File diff suppressed because one or more lines are too long
more_math/Parser/MathExpr.tokens
-423
View File
@@ -1,423 +0,0 @@
SIN=1
COS=2
TAN=3
ASIN=4
ACOS=5
ATAN=6
ATAN2=7
SINH=8
COSH=9
TANH=10
ASINH=11
ACOSH=12
ATANH=13
ABS=14
SQRT=15
LN=16
LOG=17
EXP=18
SMIN=19
SMAX=20
TMIN=21
TMAX=22
TNORM=23
SNORM=24
FLOOR=25
CEIL=26
ROUND=27
GAMMA=28
POWE=29
SIGM=30
CLAMP=31
SFFT=32
SIFFT=33
ANGL=34
PRNT=35
PRINT_SHAPE=36
NVL=37
LERP=38
STEP=39
SMOOTHSTEP=40
FRACT=41
RELU=42
SOFTPLUS=43
GELU=44
SIGN=45
MAP=46
EZCONV=47
CONV=48
SWAP=49
TRANSPOSE=50
PERM=51
RESHAPE=52
RANGE=53
LINSPACE=54
LOGSPACE=55
TOPK=56
BOTK=57
PINV=58
SUM=59
MEAN=60
STD=61
VAR=62
QUARTILE=63
PERCENTILE=64
QUANTILE=65
DOT=66
MOMENT=67
ERF=68
ERFINV=69
ANY=70
ALL=71
EDGE=72
GAUSSIAN=73
MEDIAN=74
MODE=75
CUMSUM=76
CUMPROD=77
TOPK_IND=78
BOTK_IND=79
CUBIC_EASE=80
ELASTIC_EASE=81
SINE_EASE=82
SMOOTHERSTEP=83
DIST=84
REMAP=85
COSSIM=86
COUNT=87
KEYS=88
ENABLE_TRACKING=89
DISABLE_TRACKING=90
RESET_TRACKING=91
GET_CHANGED_KEYS=92
GET_ADDED_KEYS=93
GET_REMOVED_KEYS=94
GET_UNCHANGED_KEYS=95
REPEAT=96
FLATTEN=97
APPEND=98
GET_VALUE=99
FLOW_APPLY=100
BATCH_SHUFFLE=101
MOTION_MASK=102
FLOW_TO_IMAGE=103
FLOW_MAG=104
FLOW_ANG=105
WHERE=106
HISTOGRAM=107
OVERLAY=108
PAD=109
CROSS=110
MATMUL=111
RIFE=112
BNOT=113
BITCOUNT=114
SHAPE=115
BAND=116
XOR=117
BOR=118
TENSOR=119
ADD_KEY=120
REMOVE_KEY=121
PUSH=122
POP=123
CLEAR=124
HAS=125
GET=126
CONCAT=127
STACK=128
NONZERO=129
INT=130
FLOAT=131
UPPER=132
LOWER=133
TRIM=134
SPLIT=135
JOIN=136
SUBSTRING=137
FIND=138
REPLACE=139
DILATE=140
ERODE=141
MORPH_OPEN=142
MORPH_CLOSE=143
RGB_TO_OKLAB=144
RGB_TO_CIELAB=145
OKLAB_TO_RGB=146
CIELAB_TO_RGB=147
RGB_TO_HSV=148
HSV_TO_RGB=149
INT_TO_RGB=150
RGB_TO_INT=151
INTERPOLATE_LINEAR=152
INTERPOLATE_AREA=153
INTERPOLATE_NEAREST=154
TEXT_IMAGE=155
AS_NESTED=156
SVD=157
DIAG=158
IF=159
ELSE=160
WHILE=161
FOR=162
IN=163
BREAK=164
CONTINUE=165
RETURN=166
TIMESTAMP=167
SORT=168
ARGSORT=169
ARGMIN=170
ARGMAX=171
SOFTMAX=172
SOFTMIN=173
UNIQUE=174
FLIP=175
STARTSWITH=176
ENDSWITH=177
SQUEEZE=178
UNSQUEEZE=179
ROLL=180
COV=181
CORR=182
ENTROPY=183
CROP=184
NONE=185
COORDS=186
GATHER=187
SCATTER=188
SCATTER_ADD=189
NOISE=190
RAND=191
CAUCHY=192
EXPONENTIAL=193
LOGNORMAL=194
BERNOULLI=195
POISSON=196
GAMMADIST=197
BETADIST=198
LAPLACEDIST=199
GUMBELDIST=200
WEIBULLDIST=201
CHI2DIST=202
STUDENTTDIST=203
PERLIN=204
CELLULAR=205
PLASMA=206
RIDGED=207
DOMAIN_WARP=208
PLUS=209
MINUS=210
MULT=211
DIV=212
MOD=213
POW=214
LSHIFT=215
RSHIFT=216
GE=217
GT=218
LE=219
LT=220
EQ=221
EQUEALS=222
PLUS_EQ=223
MINUS_EQ=224
MULT_EQ=225
DIV_EQ=226
MOD_EQ=227
NE=228
PIPE=229
LPAREN=230
RPAREN=231
COMMA=232
SEMICOLON=233
ARROW=234
LBRACKET=235
RBRACKET=236
QUESTION=237
COLON=238
LBRACE=239
RBRACE=240
NUMBER=241
CONSTANT=242
STRING=243
VARIABLE=244
SL_COMMENT=245
ML_COMMENT=246
WS=247
'sin'=1
'cos'=2
'tan'=3
'asin'=4
'acos'=5
'atan'=6
'atan2'=7
'sinh'=8
'cosh'=9
'tanh'=10
'asinh'=11
'acosh'=12
'atanh'=13
'abs'=14
'sqrt'=15
'ln'=16
'log'=17
'exp'=18
'smin'=19
'smax'=20
'tmin'=21
'tmax'=22
'tnorm'=23
'snorm'=24
'floor'=25
'ceil'=26
'round'=27
'gamma'=28
'pow'=29
'sigm'=30
'clamp'=31
'fft'=32
'ifft'=33
'angle'=34
'print'=35
'lerp'=38
'step'=39
'smoothstep'=40
'fract'=41
'relu'=42
'softplus'=43
'gelu'=44
'sign'=45
'map'=46
'swap'=49
'range'=53
'linspace'=54
'logspace'=55
'topk'=56
'botk'=57
'pinv'=58
'sum'=59
'mean'=60
'std'=61
'var'=62
'quantile'=65
'dot'=66
'moment'=67
'erf'=68
'erfinv'=69
'any'=70
'all'=71
'edge'=72
'median'=74
'mode'=75
'cumsum'=76
'cumprod'=77
'smootherstep'=83
'remap'=85
'keys'=88
'repeat'=96
'flatten'=97
'append'=98
'get_value'=99
'flow_apply'=100
'motion_mask'=102
'flow_to_image'=103
'where'=106
'overlay'=108
'pad'=109
'cross'=110
'matmul'=111
'rife'=112
'shape'=115
'tensor'=119
'add_key'=120
'stack_push'=122
'stack_pop'=123
'stack_clear'=124
'stack_has'=125
'stack_get'=126
'stack'=128
'nonzero'=129
'int'=130
'float'=131
'upper'=132
'lower'=133
'trim'=134
'split'=135
'join'=136
'find'=138
'replace'=139
'dilate'=140
'erode'=141
'morph_open'=142
'morph_close'=143
'rgb_to_oklab'=144
'rgb_to_cielab'=145
'oklab_to_rgb'=146
'cielab_to_rgb'=147
'rgb_to_hsv'=148
'hsv_to_rgb'=149
'int_to_rgb'=150
'rgb_to_int'=151
'interpolate_linear'=152
'interpolate_area'=153
'text_image'=155
'if'=159
'else'=160
'while'=161
'for'=162
'in'=163
'break'=164
'continue'=165
'return'=166
'sort'=168
'argsort'=169
'argmin'=170
'argmax'=171
'softmax'=172
'softmin'=173
'unique'=174
'flip'=175
'startswith'=176
'endswith'=177
'squeeze'=178
'unsqueeze'=179
'roll'=180
'cov'=181
'entropy'=183
'crop'=184
'gather'=187
'scatter'=188
'scatter_add'=189
'+'=209
'-'=210
'*'=211
'/'=212
'%'=213
'^'=214
'<<'=215
'>>'=216
'>='=217
'>'=218
'<='=219
'<'=220
'=='=221
'='=222
'+='=223
'-='=224
'*='=225
'/='=226
'%='=227
'!='=228
'|'=229
'('=230
')'=231
','=232
';'=233
'->'=234
'['=235
']'=236
'?'=237
':'=238
'{'=239
'}'=240
more_math/Parser/MathExprLexer.interp
-758
View File
File diff suppressed because one or more lines are too long
more_math/Parser/MathExprLexer.py
-1601
View File
File diff suppressed because it is too large Load Diff
more_math/Parser/MathExprLexer.tokens
-423
View File
@@ -1,423 +0,0 @@
SIN=1
COS=2
TAN=3
ASIN=4
ACOS=5
ATAN=6
ATAN2=7
SINH=8
COSH=9
TANH=10
ASINH=11
ACOSH=12
ATANH=13
ABS=14
SQRT=15
LN=16
LOG=17
EXP=18
SMIN=19
SMAX=20
TMIN=21
TMAX=22
TNORM=23
SNORM=24
FLOOR=25
CEIL=26
ROUND=27
GAMMA=28
POWE=29
SIGM=30
CLAMP=31
SFFT=32
SIFFT=33
ANGL=34
PRNT=35
PRINT_SHAPE=36
NVL=37
LERP=38
STEP=39
SMOOTHSTEP=40
FRACT=41
RELU=42
SOFTPLUS=43
GELU=44
SIGN=45
MAP=46
EZCONV=47
CONV=48
SWAP=49
TRANSPOSE=50
PERM=51
RESHAPE=52
RANGE=53
LINSPACE=54
LOGSPACE=55
TOPK=56
BOTK=57
PINV=58
SUM=59
MEAN=60
STD=61
VAR=62
QUARTILE=63
PERCENTILE=64
QUANTILE=65
DOT=66
MOMENT=67
ERF=68
ERFINV=69
ANY=70
ALL=71
EDGE=72
GAUSSIAN=73
MEDIAN=74
MODE=75
CUMSUM=76
CUMPROD=77
TOPK_IND=78
BOTK_IND=79
CUBIC_EASE=80
ELASTIC_EASE=81
SINE_EASE=82
SMOOTHERSTEP=83
DIST=84
REMAP=85
COSSIM=86
COUNT=87
KEYS=88
ENABLE_TRACKING=89
DISABLE_TRACKING=90
RESET_TRACKING=91
GET_CHANGED_KEYS=92
GET_ADDED_KEYS=93
GET_REMOVED_KEYS=94
GET_UNCHANGED_KEYS=95
REPEAT=96
FLATTEN=97
APPEND=98
GET_VALUE=99
FLOW_APPLY=100
BATCH_SHUFFLE=101
MOTION_MASK=102
FLOW_TO_IMAGE=103
FLOW_MAG=104
FLOW_ANG=105
WHERE=106
HISTOGRAM=107
OVERLAY=108
PAD=109
CROSS=110
MATMUL=111
RIFE=112
BNOT=113
BITCOUNT=114
SHAPE=115
BAND=116
XOR=117
BOR=118
TENSOR=119
ADD_KEY=120
REMOVE_KEY=121
PUSH=122
POP=123
CLEAR=124
HAS=125
GET=126
CONCAT=127
STACK=128
NONZERO=129
INT=130
FLOAT=131
UPPER=132
LOWER=133
TRIM=134
SPLIT=135
JOIN=136
SUBSTRING=137
FIND=138
REPLACE=139
DILATE=140
ERODE=141
MORPH_OPEN=142
MORPH_CLOSE=143
RGB_TO_OKLAB=144
RGB_TO_CIELAB=145
OKLAB_TO_RGB=146
CIELAB_TO_RGB=147
RGB_TO_HSV=148
HSV_TO_RGB=149
INT_TO_RGB=150
RGB_TO_INT=151
INTERPOLATE_LINEAR=152
INTERPOLATE_AREA=153
INTERPOLATE_NEAREST=154
TEXT_IMAGE=155
AS_NESTED=156
SVD=157
DIAG=158
IF=159
ELSE=160
WHILE=161
FOR=162
IN=163
BREAK=164
CONTINUE=165
RETURN=166
TIMESTAMP=167
SORT=168
ARGSORT=169
ARGMIN=170
ARGMAX=171
SOFTMAX=172
SOFTMIN=173
UNIQUE=174
FLIP=175
STARTSWITH=176
ENDSWITH=177
SQUEEZE=178
UNSQUEEZE=179
ROLL=180
COV=181
CORR=182
ENTROPY=183
CROP=184
NONE=185
COORDS=186
GATHER=187
SCATTER=188
SCATTER_ADD=189
NOISE=190
RAND=191
CAUCHY=192
EXPONENTIAL=193
LOGNORMAL=194
BERNOULLI=195
POISSON=196
GAMMADIST=197
BETADIST=198
LAPLACEDIST=199
GUMBELDIST=200
WEIBULLDIST=201
CHI2DIST=202
STUDENTTDIST=203
PERLIN=204
CELLULAR=205
PLASMA=206
RIDGED=207
DOMAIN_WARP=208
PLUS=209
MINUS=210
MULT=211
DIV=212
MOD=213
POW=214
LSHIFT=215
RSHIFT=216
GE=217
GT=218
LE=219
LT=220
EQ=221
EQUEALS=222
PLUS_EQ=223
MINUS_EQ=224
MULT_EQ=225
DIV_EQ=226
MOD_EQ=227
NE=228
PIPE=229
LPAREN=230
RPAREN=231
COMMA=232
SEMICOLON=233
ARROW=234
LBRACKET=235
RBRACKET=236
QUESTION=237
COLON=238
LBRACE=239
RBRACE=240
NUMBER=241
CONSTANT=242
STRING=243
VARIABLE=244
SL_COMMENT=245
ML_COMMENT=246
WS=247
'sin'=1
'cos'=2
'tan'=3
'asin'=4
'acos'=5
'atan'=6
'atan2'=7
'sinh'=8
'cosh'=9
'tanh'=10
'asinh'=11
'acosh'=12
'atanh'=13
'abs'=14
'sqrt'=15
'ln'=16
'log'=17
'exp'=18
'smin'=19
'smax'=20
'tmin'=21
'tmax'=22
'tnorm'=23
'snorm'=24
'floor'=25
'ceil'=26
'round'=27
'gamma'=28
'pow'=29
'sigm'=30
'clamp'=31
'fft'=32
'ifft'=33
'angle'=34
'print'=35
'lerp'=38
'step'=39
'smoothstep'=40
'fract'=41
'relu'=42
'softplus'=43
'gelu'=44
'sign'=45
'map'=46
'swap'=49
'range'=53
'linspace'=54
'logspace'=55
'topk'=56
'botk'=57
'pinv'=58
'sum'=59
'mean'=60
'std'=61
'var'=62
'quantile'=65
'dot'=66
'moment'=67
'erf'=68
'erfinv'=69
'any'=70
'all'=71
'edge'=72
'median'=74
'mode'=75
'cumsum'=76
'cumprod'=77
'smootherstep'=83
'remap'=85
'keys'=88
'repeat'=96
'flatten'=97
'append'=98
'get_value'=99
'flow_apply'=100
'motion_mask'=102
'flow_to_image'=103
'where'=106
'overlay'=108
'pad'=109
'cross'=110
'matmul'=111
'rife'=112
'shape'=115
'tensor'=119
'add_key'=120
'stack_push'=122
'stack_pop'=123
'stack_clear'=124
'stack_has'=125
'stack_get'=126
'stack'=128
'nonzero'=129
'int'=130
'float'=131
'upper'=132
'lower'=133
'trim'=134
'split'=135
'join'=136
'find'=138
'replace'=139
'dilate'=140
'erode'=141
'morph_open'=142
'morph_close'=143
'rgb_to_oklab'=144
'rgb_to_cielab'=145
'oklab_to_rgb'=146
'cielab_to_rgb'=147
'rgb_to_hsv'=148
'hsv_to_rgb'=149
'int_to_rgb'=150
'rgb_to_int'=151
'interpolate_linear'=152
'interpolate_area'=153
'text_image'=155
'if'=159
'else'=160
'while'=161
'for'=162
'in'=163
'break'=164
'continue'=165
'return'=166
'sort'=168
'argsort'=169
'argmin'=170
'argmax'=171
'softmax'=172
'softmin'=173
'unique'=174
'flip'=175
'startswith'=176
'endswith'=177
'squeeze'=178
'unsqueeze'=179
'roll'=180
'cov'=181
'entropy'=183
'crop'=184
'gather'=187
'scatter'=188
'scatter_add'=189
'+'=209
'-'=210
'*'=211
'/'=212
'%'=213
'^'=214
'<<'=215
'>>'=216
'>='=217
'>'=218
'<='=219
'<'=220
'=='=221
'='=222
'+='=223
'-='=224
'*='=225
'/='=226
'%='=227
'!='=228
'|'=229
'('=230
')'=231
','=232
';'=233
'->'=234
'['=235
']'=236
'?'=237
':'=238
'{'=239
'}'=240
more_math/Parser/MathExprListener.py
-2523
View File
File diff suppressed because it is too large Load Diff
more_math/Parser/MathExprParser.py
-16684
View File
File diff suppressed because it is too large Load Diff
more_math/Parser/MathExprVisitor.py
-1408
View File
File diff suppressed because it is too large Load Diff
more_math/Parser/UnifiedMathVisitor.py
+50 -38
View File
@@ -10,6 +10,7 @@ import math
import inspect
import weakref
import torch.nn.functional as F
import folder_paths
from comfy.nested_tensor import NestedTensor
from comfy.model_management import throw_exception_if_processing_interrupted
from comfy.k_diffusion import sampling as k_sampling
@@ -156,6 +157,17 @@ _BLEND_MODES = {
}
_FLOAT8_DTYPES = {
dtype for dtype in (
getattr(torch, "float8_e4m3fn", None),
getattr(torch, "float8_e4m3fnuz", None),
getattr(torch, "float8_e5m2", None),
getattr(torch, "float8_e5m2fnuz", None),
)
if dtype is not None
}
def _rec709_luma(r, g, b):
return 0.2126 * r + 0.7152 * g + 0.0722 * b
@@ -190,9 +202,7 @@ def _font_axis_name(axis):
return str(name).lower()
@functools.lru_cache(maxsize=1)
def _system_fonts():
"""Index of installed fonts, scanned once per process (fonts installed later need a restart)."""
def _font_dirs():
if sys.platform == "win32":
search_dirs = [os.path.join(os.environ.get("WINDIR", "C:\\Windows"), "Fonts"), os.path.expandvars(r"%LOCALAPPDATA%\Microsoft\Windows\Fonts")]
elif sys.platform == "darwin":
@@ -205,9 +215,23 @@ def _system_fonts():
os.path.expanduser("~/.local/share/fonts"),
os.path.expanduser("~/.fonts"),
]
return list(dict.fromkeys(search_dirs))
def _is_allowed_font_file(path):
"""Font files given by path must be in a system font folder or in ComfyUI's input folder."""
if not os.path.isfile(path):
return False
path = os.path.normcase(os.path.realpath(path))
roots = _font_dirs() + [folder_paths.get_input_directory()]
return any(path.startswith(os.path.normcase(os.path.realpath(root)) + os.sep) for root in roots)
@functools.lru_cache(maxsize=1)
def _system_fonts():
"""Index of installed fonts, scanned once per process (fonts installed later need a restart)."""
fonts = []
for directory in dict.fromkeys(search_dirs):
for directory in _font_dirs():
for root, _, files in os.walk(directory):
for fname in files:
if not fname.lower().endswith((".ttf", ".otf", ".ttc", ".otc")):
@@ -419,20 +443,8 @@ class UnifiedMathVisitor(MathExprVisitor):
return torch.full(t,val,device=self.device)
return torch.tensor(val, device=self.device)
@staticmethod
def _float8_dtypes():
return {
dtype for dtype in (
getattr(torch, "float8_e4m3fn", None),
getattr(torch, "float8_e4m3fnuz", None),
getattr(torch, "float8_e5m2", None),
getattr(torch, "float8_e5m2fnuz", None),
)
if dtype is not None
}
def _apply_binary_operation(self, a, b, torch_op, scalar_op):
float8_dtypes = self._float8_dtypes()
float8_dtypes = _FLOAT8_DTYPES
original_dtype = None
if self._is_plain_tensor(a) and a.dtype in float8_dtypes:
original_dtype = a.dtype
@@ -449,7 +461,7 @@ class UnifiedMathVisitor(MathExprVisitor):
def _apply_unary_operation(self, value, torch_op, scalar_op):
original_dtype = None
if self._is_plain_tensor(value) and value.dtype in self._float8_dtypes():
if self._is_plain_tensor(value) and value.dtype in _FLOAT8_DTYPES:
original_dtype = value.dtype
value = value.to(torch.float16)
@@ -1342,10 +1354,10 @@ class UnifiedMathVisitor(MathExprVisitor):
dims = self._normalize_dims((yield ctx.expr(2)), val.ndim, ctx, "norm dimension")
dim = tuple(dims)
original_dtype = val.dtype
if original_dtype in self._float8_dtypes():
if original_dtype in _FLOAT8_DTYPES:
val = val.to(torch.float32)
result = torch.linalg.vector_norm(val, ord=p, dim=dim)
if original_dtype in self._float8_dtypes():
if original_dtype in _FLOAT8_DTYPES:
result = result.to(original_dtype)
return result.item() if result.numel() == 1 else result
@@ -1359,13 +1371,13 @@ class UnifiedMathVisitor(MathExprVisitor):
if eps < 0:
raise ValueError(f"{ctx.start.line}:{ctx.start.column}: normalize eps must be non-negative")
original_dtype = val.dtype
if original_dtype in self._float8_dtypes():
if original_dtype in _FLOAT8_DTYPES:
val = val.to(torch.float32)
if val.ndim == 0:
result = val / torch.linalg.vector_norm(val, ord=p).clamp_min(eps)
else:
result = F.normalize(val, p=p, dim=dim, eps=eps)
if original_dtype in self._float8_dtypes():
if original_dtype in _FLOAT8_DTYPES:
result = result.to(original_dtype)
return result.tolist() if return_list else result
@@ -1380,7 +1392,7 @@ class UnifiedMathVisitor(MathExprVisitor):
raise ValueError(f"{ctx.start.line}:{ctx.start.column}: group_norm groups must be positive and divide the channel count {channels}")
original_dtype = val.dtype
float8_dtypes = self._float8_dtypes()
float8_dtypes = _FLOAT8_DTYPES
if original_dtype in float8_dtypes:
val = val.to(torch.float32)
@@ -1852,7 +1864,7 @@ class UnifiedMathVisitor(MathExprVisitor):
raise ValueError(f"{ctx.start.line}:{ctx.start.column}: slerp expects one vector dimension")
dim = dims[0]
float8_dtypes = self._float8_dtypes()
float8_dtypes = _FLOAT8_DTYPES
input_dtypes = (a.dtype, b.dtype)
output_dtype = next((dtype for dtype in input_dtypes if dtype in float8_dtypes), None)
if output_dtype is None:
@@ -2313,7 +2325,7 @@ class UnifiedMathVisitor(MathExprVisitor):
def visitRfftFunc(self, ctx):
val = self._promote_to_tensor((yield ctx.expr(0)))
dim = self._to_int((yield ctx.expr(1)), ctx, "rfft dimension", strict=True) if len(ctx.expr()) > 1 else -1
if val.dtype in self._float8_dtypes() or not (val.is_floating_point() or val.is_complex()):
if val.dtype in _FLOAT8_DTYPES or not (val.is_floating_point() or val.is_complex()):
val = val.to(torch.float32)
return torch.fft.rfft(val, dim=dim)
@@ -2321,7 +2333,7 @@ class UnifiedMathVisitor(MathExprVisitor):
val = self._promote_to_tensor((yield ctx.expr(0)))
n = self._to_int((yield ctx.expr(1)), ctx, "irfft length", strict=True) if len(ctx.expr()) > 1 else None
dim = self._to_int((yield ctx.expr(2)), ctx, "irfft dimension", strict=True) if len(ctx.expr()) > 2 else -1
if val.dtype in self._float8_dtypes():
if val.dtype in _FLOAT8_DTYPES:
val = val.to(torch.complex64)
return torch.fft.irfft(val, n=n, dim=dim)
@@ -2420,10 +2432,10 @@ class UnifiedMathVisitor(MathExprVisitor):
elif val.ndim > 0:
dims = tuple(range(val.ndim))
original_dtype = val.dtype
if original_dtype in self._float8_dtypes():
if original_dtype in _FLOAT8_DTYPES:
val = val.to(torch.float32)
result = val if dims is None else torch.logsumexp(val, dim=dims)
if original_dtype in self._float8_dtypes():
if original_dtype in _FLOAT8_DTYPES:
result = result.to(original_dtype)
return result.item() if result.numel() == 1 else result
@@ -2672,7 +2684,7 @@ class UnifiedMathVisitor(MathExprVisitor):
def visitOuterFunc(self, ctx):
a = self._promote_to_tensor((yield ctx.expr(0))).flatten()
b = self._promote_to_tensor((yield ctx.expr(1))).flatten()
float8_dtypes = self._float8_dtypes()
float8_dtypes = _FLOAT8_DTYPES
output_dtype = a.dtype if a.dtype in float8_dtypes else b.dtype if b.dtype in float8_dtypes else None
if output_dtype is not None:
a = a.to(torch.float32)
@@ -2685,20 +2697,20 @@ class UnifiedMathVisitor(MathExprVisitor):
if val.ndim < 2 or val.shape[-1] != val.shape[-2]:
raise ValueError(f"{ctx.start.line}:{ctx.start.column}: trace expects square matrices")
original_dtype = val.dtype
if original_dtype in self._float8_dtypes():
if original_dtype in _FLOAT8_DTYPES:
val = val.to(torch.float32)
result = torch.diagonal(val, dim1=-2, dim2=-1).sum(dim=-1)
if original_dtype in self._float8_dtypes():
if original_dtype in _FLOAT8_DTYPES:
result = result.to(original_dtype)
return result.item() if result.numel() == 1 else result
def visitDetFunc(self, ctx):
val = self._promote_to_tensor((yield ctx.expr()))
original_dtype = val.dtype
if original_dtype in self._float8_dtypes():
if original_dtype in _FLOAT8_DTYPES:
val = val.to(torch.float32)
result = torch.linalg.det(val)
if original_dtype in self._float8_dtypes():
if original_dtype in _FLOAT8_DTYPES:
result = result.to(original_dtype)
return result.item() if result.numel() == 1 else result
@@ -2710,10 +2722,10 @@ class UnifiedMathVisitor(MathExprVisitor):
if not pseudo and val.shape[-1] != val.shape[-2]:
raise ValueError(f"{ctx.start.line}:{ctx.start.column}: inv without pseudo requires square matrices")
original_dtype = val.dtype
if original_dtype in self._float8_dtypes():
if original_dtype in _FLOAT8_DTYPES:
val = val.to(torch.float32)
result = torch.linalg.pinv(val) if pseudo else torch.linalg.inv(val)
return result.to(original_dtype) if original_dtype in self._float8_dtypes() else result
return result.to(original_dtype) if original_dtype in _FLOAT8_DTYPES else result
def visitSolveFunc(self, ctx):
a = self._promote_to_tensor((yield ctx.expr(0)))
@@ -2721,7 +2733,7 @@ class UnifiedMathVisitor(MathExprVisitor):
if a.ndim < 2 or a.shape[-1] != a.shape[-2]:
raise ValueError(f"{ctx.start.line}:{ctx.start.column}: solve expects square coefficient matrices")
float8_dtypes = self._float8_dtypes()
float8_dtypes = _FLOAT8_DTYPES
output_dtype = a.dtype if a.dtype in float8_dtypes else b.dtype if b.dtype in float8_dtypes else None
if a.dtype in float8_dtypes or not (a.is_floating_point() or a.is_complex()):
a = a.to(torch.float32)
@@ -2736,7 +2748,7 @@ class UnifiedMathVisitor(MathExprVisitor):
if val.ndim < 2:
raise ValueError(f"{ctx.start.line}:{ctx.start.column}: matrix_rank expects matrices")
original_dtype = val.dtype
if original_dtype in self._float8_dtypes() or not (val.is_floating_point() or val.is_complex()):
if original_dtype in _FLOAT8_DTYPES or not (val.is_floating_point() or val.is_complex()):
val = val.to(torch.float32)
tol = (yield ctx.expr(1)) if len(ctx.expr()) > 1 else None
if self._is_tensor(tol):
@@ -3536,7 +3548,7 @@ class UnifiedMathVisitor(MathExprVisitor):
tier = _FONT_WEIGHT_TIERS[bisect.bisect_left((150, 250, 350, 450, 550, 650, 750, 850), weight_val)]
synthetic_italic = synthetic_bold = False
variation_weight, variation_italic = weight_val, is_italic
if os.path.isfile(font_name):
if _is_allowed_font_file(font_name):
font_path = font_name
else:
font_path = _find_font(font_name, tier, weight_val, is_italic)
more_math/Parser/inbuilt_symbols.py
+1 -1
View File
@@ -264,7 +264,7 @@ INBUILT_FUNCTION_META = {
'interpolate_area': {'min_args': 2, 'max_args': 2, 'signature': 'interpolate_area(x, scale)', 'forms': [['x', 'scale']], 'description': 'resizes x [N, C, ...] with area interpolation. A number scales every spatial dimension; a list is the target spatial size', 'details': '', 'params': {'x': 'tensor shaped [N, C, ...] with 1-3 spatial dimensions', 'scale': 'a number scales every spatial dimension; a list gives the target spatial size'}, 'aliases': []},
'interpolate_nearest': {'min_args': 2, 'max_args': 2, 'signature': 'interpolate_nearest(x, scale)', 'forms': [['x', 'scale']], 'description': 'resizes x [N, C, ...] with nearest-neighbour interpolation. A number scales every spatial dimension; a list is the target spatial size', 'details': '', 'params': {'x': 'tensor shaped [N, C, ...] with 1-3 spatial dimensions', 'scale': 'a number scales every spatial dimension; a list gives the target spatial size'}, 'aliases': ['interpolate_nearest_exact']},
'interpolate_nearest_exact': {'min_args': 2, 'max_args': 2, 'signature': 'interpolate_nearest_exact(x, scale)', 'forms': [['x', 'scale']], 'description': 'resizes x [N, C, ...] with nearest-neighbour interpolation. A number scales every spatial dimension; a list is the target spatial size', 'details': '', 'params': {'x': 'tensor shaped [N, C, ...] with 1-3 spatial dimensions', 'scale': 'a number scales every spatial dimension; a list gives the target spatial size'}, 'aliases': ['interpolate_nearest']},
'text_image': {'min_args': 3, 'max_args': 9, 'signature': 'text_image(text, font, size, [max_width], [weight], [rotation_angle], [line_spacing], [italic], [underline])', 'forms': [['text', 'font', 'size', '[max_width]', '[weight]', '[rotation_angle]', '[line_spacing]', '[italic]', '[underline]']], 'description': 'renders text to a 2D tensor with values 0-1', 'details': '', 'params': {'text': 'text to render; line breaks start new lines', 'font': 'font name to look up in the system fonts, or a path to a font file', 'size': 'font size in pixels', 'max_width': 'wrap lines wider than this many pixels (default: 0, no wrapping)', 'weight': 'font weight 100-900 as in CSS (default: 400)', 'rotation_angle': 'rotation in degrees, counter-clockwise (default: 0)', 'line_spacing': 'line distance as a multiple of the line height (default: 1)', 'italic': 'truthy for italic text (default: false)', 'underline': 'truthy to underline the text (default: false)'}, 'aliases': []},
'text_image': {'min_args': 3, 'max_args': 9, 'signature': 'text_image(text, font, size, [max_width], [weight], [rotation_angle], [line_spacing], [italic], [underline])', 'forms': [['text', 'font', 'size', '[max_width]', '[weight]', '[rotation_angle]', '[line_spacing]', '[italic]', '[underline]']], 'description': 'renders text to a 2D tensor with values 0-1', 'details': '', 'params': {'text': 'text to render; line breaks start new lines', 'font': 'font name to look up in the system fonts, or a path to a font file in a system font folder or the ComfyUI input folder', 'size': 'font size in pixels', 'max_width': 'wrap lines wider than this many pixels (default: 0, no wrapping)', 'weight': 'font weight 100-900 as in CSS (default: 400)', 'rotation_angle': 'rotation in degrees, counter-clockwise (default: 0)', 'line_spacing': 'line distance as a multiple of the line height (default: 1)', 'italic': 'truthy for italic text (default: false)', 'underline': 'truthy to underline the text (default: false)'}, 'aliases': []},
'as_nested_tensor': {'min_args': 1, 'max_args': 1, 'signature': 'as_nested_tensor(values)', 'forms': [['values']], 'description': 'converts a list to a nested tensor (special object containing tensors of different shapes behaving like a tensor)', 'details': '', 'params': {'values': 'list of tensors; their shapes may differ'}, 'aliases': ['as_nested']},
'as_nested': {'min_args': 1, 'max_args': 1, 'signature': 'as_nested(values)', 'forms': [['values']], 'description': 'converts a list to a nested tensor (special object containing tensors of different shapes behaving like a tensor)', 'details': '', 'params': {'values': 'list of tensors; their shapes may differ'}, 'aliases': ['as_nested_tensor']},
'svd': {'min_args': 1, 'max_args': 2, 'signature': 'svd(x, [full_matrices])', 'forms': [['x', '[full_matrices]']], 'description': 'computes the singular value decomposition of x. Returns a list of 3 tensors U, S, Vh such that x = matmul(matmul(U, diagonal_matrix(S, shape(U))), Vh)', 'details': '', 'params': {'x': 'matrix, or a batch of matrices in the last two dimensions', 'full_matrices': 'truthy to return full-size U and Vh (default: false)'}, 'aliases': ['singular_value_decomposition']},
more_math/Parser/optical_flow_utils.py
+17 -13
View File
@@ -6,25 +6,29 @@ from torchvision.utils import flow_to_image as torchvision_flow_to_image
import numpy as np
import folder_paths
import comfy.utils
import comfy.model_management
import comfy.model_patcher
_raft_model = None
folder_paths.folder_names_and_paths.setdefault("optical_flow", ([os.path.join(folder_paths.models_dir, "optical_flow")], folder_paths.supported_pt_extensions))
def get_raft_model(device):
global _raft_model
if _raft_model is None:
# Small (~20MB) and offloaded by model management like any other model, so it is loaded once per process
_raft_patcher = None
def get_raft_model():
global _raft_patcher
if _raft_patcher is None:
names = folder_paths.get_filename_list("optical_flow")
model = raft_large(weights=None, progress=False)
model.load_state_dict(comfy.utils.load_torch_file(folder_paths.get_full_path_or_raise("optical_flow", names[0]), safe_load=True))
_raft_patcher = comfy.model_patcher.ModelPatcher(model.eval(), load_device=comfy.model_management.get_torch_device(), offload_device=comfy.model_management.unet_offload_device())
comfy.model_management.load_model_gpu(_raft_patcher)
return _raft_patcher.model
if names:
model = raft_large(weights=None, progress=False)
model.load_state_dict(comfy.utils.load_torch_file(folder_paths.get_full_path_or_raise("optical_flow", names[0]), safe_load=True))
else:
# No local checkpoint: download torchvision's RAFT weights once and keep a copy in models/optical_flow
weights = Raft_Large_Weights.DEFAULT
model = raft_large(weights=weights, progress=True)
save_dir = folder_paths.get_folder_paths("optical_flow")[0]
os.makedirs(save_dir, exist_ok=True)
torch.save(model.state_dict(), os.path.join(save_dir, os.path.basename(weights.url)))
_raft_model = model.to(device).eval()
return _raft_model
def warp(x, flow, padding_mode='reflection'):
"""
@@ -90,8 +94,8 @@ def get_optical_flow(img1, img2, tiling_size=0, iterations=12, multi_scale=False
img1, img2: [H, W, C] or [B, H, W, C]
Returns: [B, H, W, 2] flow vectors
"""
device = img1.device
model = get_raft_model(device)
model = get_raft_model()
device = comfy.model_management.get_torch_device()
img1_proc = preprocess_image(img1, device)
img2_proc = preprocess_image(img2, device)
@@ -176,7 +180,7 @@ def get_optical_flow(img1, img2, tiling_size=0, iterations=12, multi_scale=False
flow[:, 1] *= float(h) / h_new
# [B, 2, H, W] -> [B, H, W, 2]
return flow.movedim(1, -1)
return flow.movedim(1, -1).to(img1.device)
def calculate_occlusion_mask(flow):
"""
more_math/ScriptTextWindow.py
+1 -1
View File
@@ -22,4 +22,4 @@ class ScriptTextInput(io.ComfyNode):
@classmethod
def execute(cls, script):
return (remove_dead_code(parse_expr(script)),)
return (remove_dead_code(parse_expr.__wrapped__(script)),)
more_math/StringMathNode.py
+1 -3
View File
@@ -69,7 +69,7 @@ class StringMathNode(io.ComfyNode):
"z": F.get("F3", 0.0) if F.get("F3") is not None else 0.0
}
v_stacked, v_cnt = get_v_variable(V, length_mismatch="do nothing")
v_stacked, v_cnt = get_v_variable(V)
if v_stacked is not None:
variables["V"] = v_stacked
variables["Vcnt"] = float(v_cnt)
@@ -105,9 +105,7 @@ class StringMathNode(io.ComfyNode):
for start in range(0, len(s), n):
yield s[start:start+n]
stack = stack if remember_stack else copy.deepcopy(stack)
return (chunks(result,batching), stack)
else:
stack = stack if remember_stack else copy.deepcopy(stack)
return ([result], stack)
more_math/VideoMathNode.py
+4 -4
View File
@@ -169,10 +169,10 @@ class VideoMathNode(io.ComfyNode):
variables[v_name].is_lazy_var = True
def lazy_v_stack():
return get_v_variable(V_norm, length_mismatch=length_mismatch)[0]
return get_v_variable(V_norm)[0]
lazy_v_stack.is_lazy_var = True
def lazy_v_count():
return float(get_v_variable(V_norm, length_mismatch=length_mismatch)[1])
return float(get_v_variable(V_norm)[1])
lazy_v_count.is_lazy_var = True
variables["V"] = lazy_v_stack
variables["Vcnt"] = lazy_v_count
@@ -251,10 +251,10 @@ class VideoMathNode(io.ComfyNode):
variables[v_name].is_lazy_var = True
def lazy_v_stack():
return get_v_variable(V_norm_waveforms, length_mismatch=length_mismatch)[0]
return get_v_variable(V_norm_waveforms)[0]
lazy_v_stack.is_lazy_var = True
def lazy_v_count():
return float(get_v_variable(V_norm_waveforms, length_mismatch=length_mismatch)[1])
return float(get_v_variable(V_norm_waveforms)[1])
lazy_v_count.is_lazy_var = True
variables["V"] = lazy_v_stack
variables["Vcnt"] = lazy_v_count
more_math/deprecated/NoiseMathNode.py
+1 -2
View File
@@ -100,8 +100,7 @@ class NoiseExecutor:
self.z = z
self.expr = Noise
self.tree = parse_expr(Noise)
seed = -1
self.seed = next((n.seed for n in (a, b, c, d) if n is not None), 0)
def generate_noise(self, input_latent: torch.Tensor) -> torch.Tensor:
samples = input_latent["samples"]
more_math/helper_functions.py
+6 -2
View File
@@ -1,3 +1,5 @@
import functools
from antlr4.error.ErrorListener import ErrorListener
from antlr4 import InputStream, CommonTokenStream
from antlr4.atn.PredictionMode import PredictionMode
@@ -66,8 +68,10 @@ def as_tensor(value, shape):
return torch.cat(value)
# check_lazy_status and execute both parse the same expression; trees are only read, so they can be shared.
@functools.lru_cache(maxsize=64)
def parse_expr(expr: str):
"""Parse a math expression and return the parse tree."""
"""Parse a math expression and return the parse tree. The tree is shared, use parse_expr.__wrapped__ for one you modify."""
input_stream = InputStream(expr)
lexer = MathExprLexer(input_stream)
stream = CommonTokenStream(lexer)
@@ -299,7 +303,7 @@ def normalize_to_common_shape(*tensors, mode="pad"):
result.append(t.contiguous())
return tuple(result)
def get_v_variable(v_norm_dict, length_mismatch="error"):
def get_v_variable(v_norm_dict):
"""
Collects V0, V1, ... from the dict (ignoring NestedTensor component keys
like V0_0, V1_1), and returns them as a list and count.
more_math/nodes.py
+2 -2
View File
@@ -100,7 +100,7 @@ class BoolToIntNode(io.ComfyNode):
display_name="Bool -> Int",
category="More math/converters",
inputs=[
io.Boolean.Input(id="value", default=0),
io.Boolean.Input(id="value", default=False),
],
outputs=[
io.Int.Output(),
@@ -125,7 +125,7 @@ class IntToBoolNode(io.ComfyNode):
category="More math/converters",
display_name="Int -> bool",
inputs=[
io.Int.Input(id="value", default=0.0),
io.Int.Input(id="value", default=0),
],
outputs=[
io.Boolean.Output(),
requirements.txt
+3 -1
View File
@@ -1,3 +1,5 @@
antlr4-python3-runtime
torch
pillow
pillow
numpy
scipy
test_bitwise_fix.py
-133
View File
@@ -1,133 +0,0 @@
#!/usr/bin/env python3
"""Quick test to verify bitwise operations work with tensors"""
import sys
sys.path.insert(0, r'D:\stability\Data\Packages\ComfyUI')
import torch
from custom_nodes.more_math.more_math.Parser.UnifiedMathVisitor import UnifiedMathVisitor
def test_bitwise_xor_float():
"""Test XOR with float tensors (the reported bug)"""
print("Testing bitwise XOR with float tensors...")
# Create float tensors (this was causing the error)
a = torch.tensor([1.0, 2.0, 3.0], dtype=torch.float32)
b = torch.tensor([4.0, 5.0, 6.0], dtype=torch.float32)
visitor = UnifiedMathVisitor({"a": a, "b": b})
try:
# This should convert to int64, perform XOR, then convert back
result = visitor._bitwise_op(a, b, torch.bitwise_xor, lambda x, y: x ^ y)
print("✓ XOR succeeded!")
print(f" Input a (float32): {a}")
print(f" Input b (float32): {b}")
print(f" Result: {result}")
print(f" Result dtype: {result.dtype}")
return True
except Exception as e:
print(f"✗ XOR failed: {e}")
return False
def test_bitwise_and_float():
"""Test AND with float tensors"""
print("\nTesting bitwise AND with float tensors...")
a = torch.tensor([15.0, 14.0, 13.0], dtype=torch.float32)
b = torch.tensor([7.0, 3.0, 1.0], dtype=torch.float32)
visitor = UnifiedMathVisitor({"a": a, "b": b})
try:
result = visitor._bitwise_op(a, b, torch.bitwise_and, lambda x, y: x & y)
print("✓ AND succeeded!")
print(f" Input a (float32): {a}")
print(f" Input b (float32): {b}")
print(f" Result: {result}")
print(f" Result dtype: {result.dtype}")
return True
except Exception as e:
print(f"✗ AND failed: {e}")
return False
def test_bitwise_or_float():
"""Test OR with float tensors"""
print("\nTesting bitwise OR with float tensors...")
a = torch.tensor([15.0, 14.0, 13.0], dtype=torch.float32)
b = torch.tensor([7.0, 3.0, 1.0], dtype=torch.float32)
visitor = UnifiedMathVisitor({"a": a, "b": b})
try:
result = visitor._bitwise_op(a, b, torch.bitwise_or, lambda x, y: x | y)
print("✓ OR succeeded!")
print(f" Input a (float32): {a}")
print(f" Input b (float32): {b}")
print(f" Result: {result}")
print(f" Result dtype: {result.dtype}")
return True
except Exception as e:
print(f"✗ OR failed: {e}")
return False
def test_bitwise_not_float():
"""Test NOT with float tensors"""
print("\nTesting bitwise NOT with float tensors...")
a = torch.tensor([1.0, 2.0, 3.0], dtype=torch.float32)
visitor = UnifiedMathVisitor({"a": a})
try:
result = visitor._bitwise_not(a)
print("✓ NOT succeeded!")
print(f" Input a (float32): {a}")
print(f" Result: {result}")
print(f" Result dtype: {result.dtype}")
return True
except Exception as e:
print(f"✗ NOT failed: {e}")
return False
def test_int_tensors_preserved():
"""Ensure int tensors still work as before"""
print("\nTesting that int tensor dtypes are preserved...")
a = torch.tensor([15, 14, 13], dtype=torch.int16)
b = torch.tensor([7, 3, 1], dtype=torch.int16)
visitor = UnifiedMathVisitor({"a": a, "b": b})
try:
result = visitor._bitwise_op(a, b, torch.bitwise_and, lambda x, y: x & y)
assert result.dtype == torch.int16, f"Expected int16, got {result.dtype}"
print("✓ Int16 dtype preserved!")
print(f" Input a (int16): {a}")
print(f" Input b (int16): {b}")
print(f" Result (int16): {result}")
return True
except Exception as e:
print(f"✗ Int16 test failed: {e}")
return False
if __name__ == "__main__":
print("=" * 70)
print("Bitwise Operations Fix Verification")
print("=" * 70)
results = [
test_bitwise_xor_float(),
test_bitwise_and_float(),
test_bitwise_or_float(),
test_bitwise_not_float(),
test_int_tensors_preserved(),
]
print("\n" + "=" * 70)
if all(results):
print("✓ All tests passed!")
else:
print("✗ Some tests failed")
sys.exit(1)
print("=" * 70)
test_bitwise_shifts.py
-170
View File
@@ -1,170 +0,0 @@
#!/usr/bin/env python3
"""
Test bitwise shift operators and bit count function
"""
import torch
import sys
sys.path.insert(0, 'custom_nodes/more_math')
from more_math.Parser.MathExprParser import MathExprParser
from more_math.Parser.UnifiedMathVisitor import UnifiedMathVisitor
from antlr4 import InputStream, CommonTokenFactory
def parse_and_evaluate(expression, variables=None):
"""Parse and evaluate a math expression"""
if variables is None:
variables = {}
input_stream = InputStream(expression)
lexer = MathExprParser(input_stream).lexer
stream = CommonTokenFactory()
parser = MathExprParser(input_stream)
tree = parser.start()
visitor = UnifiedMathVisitor(variables, device='cpu')
result = visitor.visit(tree)
return result
def test_bit_shifts():
"""Test bitwise shift operators"""
print("=" * 60)
print("Testing Bitwise Shift Operators")
print("=" * 60)
test_cases = [
# Left shift: 5 << 2 = 20 (0101 << 2 = 10100)
("5 << 2", {}, 20),
# Right shift: 20 >> 2 = 5 (10100 >> 2 = 0101)
("20 >> 2", {}, 5),
# Left shift with variable
("x << 3", {"x": 4}, 32), # 4 << 3 = 32
# Right shift with variable
("x >> 2", {"x": 16}, 4), # 16 >> 2 = 4
# Chained shifts
("(8 << 2) >> 3", {}, 4), # (32) >> 3 = 4
]
for expr, vars, expected in test_cases:
try:
result = parse_and_evaluate(expr, vars)
status = "✓" if result == expected else "✗"
print(f"{status} {expr:30} = {result:10} (expected {expected})")
except Exception as e:
print(f"✗ {expr:30} ERROR: {e}")
def test_bit_count():
"""Test bit count function"""
print("\n" + "=" * 60)
print("Testing Bitwise Bit Count Function")
print("=" * 60)
test_cases = [
# bitcount(5) = 2 (0101 has 2 set bits)
("bitcount(5)", {}, 2),
# bitcount(15) = 4 (1111 has 4 set bits)
("bitcount(15)", {}, 4),
# bitcount(7) = 3 (111 has 3 set bits)
("bitcount(7)", {}, 3),
# bitcount(255) = 8 (11111111 has 8 set bits)
("bitcount(255)", {}, 8),
# bitcount(0) = 0
("bitcount(0)", {}, 0),
# With variable
("bitcount(x)", {"x": 31}, 5), # 31 = 11111 = 5 bits set
]
for expr, vars, expected in test_cases:
try:
result = parse_and_evaluate(expr, vars)
status = "✓" if result == expected else "✗"
print(f"{status} {expr:30} = {result:10} (expected {expected})")
except Exception as e:
print(f"✗ {expr:30} ERROR: {e}")
def test_bit_shifts_with_tensors():
"""Test bitwise shifts with tensors"""
print("\n" + "=" * 60)
print("Testing Bitwise Shifts with Tensors")
print("=" * 60)
vars = {
"a": torch.tensor([1, 2, 4, 8], dtype=torch.int32),
"shift": 2,
}
try:
result = parse_and_evaluate("a << shift", vars)
expected = torch.tensor([4, 8, 16, 32], dtype=torch.int32)
match = torch.equal(result, expected)
status = "✓" if match else "✗"
print(f"{status} tensor_shift_left: [1,2,4,8] << 2 = {result.tolist()}")
except Exception as e:
print(f"✗ tensor_shift_left ERROR: {e}")
try:
result = parse_and_evaluate("a >> shift", vars)
expected = torch.tensor([0, 0, 1, 2], dtype=torch.int32)
match = torch.equal(result, expected)
status = "✓" if match else "✗"
print(f"{status} tensor_shift_right: [1,2,4,8] >> 2 = {result.tolist()}")
except Exception as e:
print(f"✗ tensor_shift_right ERROR: {e}")
def test_bit_count_with_tensors():
"""Test bit count with tensors"""
print("\n" + "=" * 60)
print("Testing Bit Count with Tensors")
print("=" * 60)
vars = {
"nums": torch.tensor([5, 15, 7, 255], dtype=torch.int32),
}
try:
result = parse_and_evaluate("bitcount(nums)", vars)
# Expected: [2, 4, 3, 8] set bits
print(f"✓ bitcount([5,15,7,255]): {result}")
except Exception as e:
print(f"✗ bitcount_tensor ERROR: {e}")
def test_combinations():
"""Test combinations of shift and bit count"""
print("\n" + "=" * 60)
print("Testing Combinations")
print("=" * 60)
test_cases = [
# Shift then count bits
("bitcount(5 << 2)", {}, 2), # 5 << 2 = 20 (10100) = 2 bits
# Combined with other operators
("(5 << 2) | (3 << 4)", {}, 0xCC), # 20 | 48 = 0xCC = 204
]
for expr, vars, expected in test_cases:
try:
result = parse_and_evaluate(expr, vars)
status = "✓" if result == expected else "✗"
print(f"{status} {expr:35} = {result:10} (expected {expected})")
except Exception as e:
print(f"✗ {expr:35} ERROR: {e}")
if __name__ == "__main__":
test_bit_shifts()
test_bit_count()
test_bit_shifts_with_tensors()
test_bit_count_with_tensors()
test_combinations()
print("\n" + "=" * 60)
print("All tests completed!")
print("=" * 60)
tests/test_dead_code.py
+2 -2
View File
@@ -9,7 +9,7 @@ from more_math.helper_functions import parse_expr, checkLazyNew
def run(script, variables=None, optimize=True, stack=None):
tree = parse_expr(script)
tree = parse_expr.__wrapped__(script)
if optimize:
remove_dead_code(tree)
out = io.StringIO()
@@ -109,5 +109,5 @@ def test_last_statement_of_a_block_stays():
def test_unused_inputs_are_no_longer_requested():
tree = remove_dead_code(parse_expr("u = V1 * 2; V0"))
tree = remove_dead_code(parse_expr.__wrapped__("u = V1 * 2; V0"))
assert checkLazyNew(tree, {"V0": None, "V1": None}, {}) == {"V0"}
tests/test_guider_math.py
+10
View File
@@ -102,6 +102,16 @@ class TestMathGuider(unittest.TestCase):
# At step 2 with 2 steps: 2 / 2 = 1.0
self.assertTrue(torch.allclose(res2, torch.tensor(1.0)))
def test_math_guider_batched_sigma_and_substeps(self):
# Samplers pass sigma with one entry per batch item; heun-like samplers also call the model between sigmas
math_guider = MathGuider({"V0": MockGuider(1.0)}, {}, "current_step + sigma * 0", "V0", {})
math_guider.sigmas = torch.tensor([10.0, 5.0, 0.0])
math_guider.steps = 2
x = torch.zeros((2, 1, 1, 1))
self.assertTrue(torch.allclose(math_guider(x, torch.tensor([5.0, 5.0])), torch.tensor(1.0)))
self.assertTrue(torch.allclose(math_guider(x, torch.tensor([7.0, 7.0])), torch.tensor(0.0)))
self.assertTrue(torch.allclose(math_guider(x, torch.tensor([10.0, 10.0])), torch.tensor(0.0)))
def test_math_guider_crop_function(self):
"""Test crop function within guider context"""
g0 = MockGuider(1.0)
tests/test_more_math.py
+2
View File
@@ -360,6 +360,7 @@ def test_noise_math_5d():
class MockNoise:
def __init__(self, tensor):
self.tensor = tensor
self.seed = 0
def generate_noise(self, input_latent):
return self.tensor
@@ -382,6 +383,7 @@ def test_noise_math_autogrow():
class MockNoise:
def __init__(self, tensor):
self.tensor = tensor
self.seed = 0
def generate_noise(self, input_latent):
return self.tensor
tests/test_unified_visitor_regressions.py
+35 -6
View File
@@ -9,6 +9,9 @@ from PIL import ImageFont
from more_math.Parser.UnifiedMathVisitor import ChangeTrackingDict, MathDict, UnifiedMathVisitor
from more_math.helper_functions import checkLazyNew, parse_expr
from more_math.FloatMathNode import FloatMathNode
from more_math.ImageMathNode import ImageMathNode
from more_math.LatentMathNode import LatentMathNode
from more_math.MaskMathNode import MaskMathNode
def _visit(expression, variables):
@@ -80,17 +83,12 @@ def test_lambda_calls_require_exact_argument_count(call):
def test_text_image_size_expression_is_evaluated_once(monkeypatch):
visitor_module = importlib.import_module("more_math.Parser.UnifiedMathVisitor")
default_font = ImageFont.load_default()
original_isfile = os.path.isfile
monkeypatch.setattr(
visitor_module.ImageFont,
"truetype",
lambda *args, **kwargs: default_font,
)
monkeypatch.setattr(
visitor_module.os.path,
"isfile",
lambda path: path == "mock-font" or original_isfile(path),
)
monkeypatch.setattr(visitor_module, "_is_allowed_font_file", lambda path: path == "mock-font")
values = MathDict()
_visit(
@@ -101,6 +99,15 @@ def test_text_image_size_expression_is_evaluated_once(monkeypatch):
assert len(values) == 1
def test_text_image_rejects_font_path_outside_font_folders(monkeypatch):
visitor_module = importlib.import_module("more_math.Parser.UnifiedMathVisitor")
monkeypatch.setattr(visitor_module, "_find_font", lambda *args: None)
path = os.path.abspath(__file__).replace("\\", "/")
with pytest.raises(FileNotFoundError):
_visit(f'text_image("x", "{path}", 12)', {})
def test_stack_list_of_tensors_with_negative_dimension():
first = torch.tensor([1.0, 2.0])
second = torch.tensor([3.0, 4.0])
@@ -280,6 +287,28 @@ def test_unconnected_stack_is_not_shared_between_runs():
assert stack == {0: [1.0]}
@pytest.mark.parametrize("node, value, extra", [
(ImageMathNode, torch.zeros(2, 4, 4, 3), {"use_compute_device": False}),
(MaskMathNode, torch.zeros(2, 4, 4), {}),
(LatentMathNode, {"samples": torch.zeros(2, 4, 4, 4)}, {"use_compute_device": False}),
])
@pytest.mark.parametrize("batching", [0, 1])
def test_stack_across_list_runs(node, value, extra, batching):
# ComfyUI runs the node once per list item and passes the same stack object every time
def run(stack, remember):
return node.execute(V={"V0": value}, F={}, Expression="stack_push(0, 1); V0", batching=batching, remember_stack=remember, stack=stack, **extra)[-1]
shared = {}
run(shared, True)
assert run(shared, True) == {0: [1.0, 1.0]}
fresh = {}
first, second = run(fresh, False), run(fresh, False)
assert fresh == {}
assert first == second == {0: [1.0]}
assert first is not second
def test_top_level_return_stops_script():
assert _visit("x = 1; return 5; x = 2; x", {}) == 5
web/inbuilt_symbols.js
+1 -1
View File
@@ -267,7 +267,7 @@ export const FUNCTION_META = {
"interpolate_area": { minArgs: 2, maxArgs: 2, signature: "interpolate_area(x, scale)", forms: [["x", "scale"]], description: "resizes x [N, C, ...] with area interpolation. A number scales every spatial dimension; a list is the target spatial size", details: "", params: {"x": "tensor shaped [N, C, ...] with 1-3 spatial dimensions", "scale": "a number scales every spatial dimension; a list gives the target spatial size"}, aliases: [] },
"interpolate_nearest": { minArgs: 2, maxArgs: 2, signature: "interpolate_nearest(x, scale)", forms: [["x", "scale"]], description: "resizes x [N, C, ...] with nearest-neighbour interpolation. A number scales every spatial dimension; a list is the target spatial size", details: "", params: {"x": "tensor shaped [N, C, ...] with 1-3 spatial dimensions", "scale": "a number scales every spatial dimension; a list gives the target spatial size"}, aliases: ["interpolate_nearest_exact"] },
"interpolate_nearest_exact": { minArgs: 2, maxArgs: 2, signature: "interpolate_nearest_exact(x, scale)", forms: [["x", "scale"]], description: "resizes x [N, C, ...] with nearest-neighbour interpolation. A number scales every spatial dimension; a list is the target spatial size", details: "", params: {"x": "tensor shaped [N, C, ...] with 1-3 spatial dimensions", "scale": "a number scales every spatial dimension; a list gives the target spatial size"}, aliases: ["interpolate_nearest"] },
"text_image": { minArgs: 3, maxArgs: 9, signature: "text_image(text, font, size, [max_width], [weight], [rotation_angle], [line_spacing], [italic], [underline])", forms: [["text", "font", "size", "[max_width]", "[weight]", "[rotation_angle]", "[line_spacing]", "[italic]", "[underline]"]], description: "renders text to a 2D tensor with values 0-1", details: "", params: {"text": "text to render; line breaks start new lines", "font": "font name to look up in the system fonts, or a path to a font file", "size": "font size in pixels", "max_width": "wrap lines wider than this many pixels (default: 0, no wrapping)", "weight": "font weight 100-900 as in CSS (default: 400)", "rotation_angle": "rotation in degrees, counter-clockwise (default: 0)", "line_spacing": "line distance as a multiple of the line height (default: 1)", "italic": "truthy for italic text (default: false)", "underline": "truthy to underline the text (default: false)"}, aliases: [] },
"text_image": { minArgs: 3, maxArgs: 9, signature: "text_image(text, font, size, [max_width], [weight], [rotation_angle], [line_spacing], [italic], [underline])", forms: [["text", "font", "size", "[max_width]", "[weight]", "[rotation_angle]", "[line_spacing]", "[italic]", "[underline]"]], description: "renders text to a 2D tensor with values 0-1", details: "", params: {"text": "text to render; line breaks start new lines", "font": "font name to look up in the system fonts, or a path to a font file in a system font folder or the ComfyUI input folder", "size": "font size in pixels", "max_width": "wrap lines wider than this many pixels (default: 0, no wrapping)", "weight": "font weight 100-900 as in CSS (default: 400)", "rotation_angle": "rotation in degrees, counter-clockwise (default: 0)", "line_spacing": "line distance as a multiple of the line height (default: 1)", "italic": "truthy for italic text (default: false)", "underline": "truthy to underline the text (default: false)"}, aliases: [] },
"as_nested_tensor": { minArgs: 1, maxArgs: 1, signature: "as_nested_tensor(values)", forms: [["values"]], description: "converts a list to a nested tensor (special object containing tensors of different shapes behaving like a tensor)", details: "", params: {"values": "list of tensors; their shapes may differ"}, aliases: ["as_nested"] },
"as_nested": { minArgs: 1, maxArgs: 1, signature: "as_nested(values)", forms: [["values"]], description: "converts a list to a nested tensor (special object containing tensors of different shapes behaving like a tensor)", details: "", params: {"values": "list of tensors; their shapes may differ"}, aliases: ["as_nested_tensor"] },
"svd": { minArgs: 1, maxArgs: 2, signature: "svd(x, [full_matrices])", forms: [["x", "[full_matrices]"]], description: "computes the singular value decomposition of x. Returns a list of 3 tensors U, S, Vh such that x = matmul(matmul(U, diagonal_matrix(S, shape(U))), Vh)", details: "", params: {"x": "matrix, or a batch of matrices in the last two dimensions", "full_matrices": "truthy to return full-size U and Vh (default: false)"}, aliases: ["singular_value_decomposition"] },