Spaces:
Running
on
L40S
Running
on
L40S
File size: 8,772 Bytes
4450790 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 |
from PIL import Image
from ..log import log
from ..utils import comfy_dir, font_path, pil2tensor
# class MtbExamples:
# """MTB Example Images"""
# def __init__(self):
# pass
# @classmethod
# @lru_cache(maxsize=1)
# def get_root(cls):
# return here / "examples" / "samples"
# @classmethod
# def INPUT_TYPES(cls):
# input_dir = cls.get_root()
# files = [f.name for f in input_dir.iterdir() if f.is_file()]
# return {
# "required": {"image": (sorted(files),)},
# }
# RETURN_TYPES = ("IMAGE", "MASK")
# FUNCTION = "do_mtb_examples"
# CATEGORY = "fun"
# def do_mtb_examples(self, image, index):
# image_path = (self.get_root() / image).as_posix()
# i = Image.open(image_path)
# i = ImageOps.exif_transpose(i)
# image = i.convert("RGB")
# image = np.array(image).astype(np.float32) / 255.0
# image = torch.from_numpy(image)[None,]
# if "A" in i.getbands():
# mask = np.array(i.getchannel("A")).astype(np.float32) / 255.0
# mask = 1.0 - torch.from_numpy(mask)
# else:
# mask = torch.zeros((64, 64), dtype=torch.float32, device="cpu")
# return (image, mask)
# @classmethod
# def IS_CHANGED(cls, image):
# image_path = (cls.get_root() / image).as_posix()
# m = hashlib.sha256()
# with open(image_path, "rb") as f:
# m.update(f.read())
# return m.digest().hex()
class MTB_UnsplashImage:
"""Unsplash Image given a keyword and a size"""
@classmethod
def INPUT_TYPES(cls):
return {
"required": {
"width": (
"INT",
{"default": 512, "max": 8096, "min": 0, "step": 1},
),
"height": (
"INT",
{"default": 512, "max": 8096, "min": 0, "step": 1},
),
"random_seed": (
"INT",
{"default": 0, "max": 1e5, "min": 0, "step": 1},
),
},
"optional": {
"keyword": ("STRING", {"default": "nature"}),
},
}
RETURN_TYPES = ("IMAGE",)
FUNCTION = "do_unsplash_image"
CATEGORY = "mtb/generate"
def do_unsplash_image(self, width, height, random_seed, keyword=None):
import io
import requests
base_url = "https://source.unsplash.com/random/"
if width and height:
base_url += f"/{width}x{height}"
if keyword:
keyword = keyword.replace(" ", "%20")
base_url += f"?{keyword}&{random_seed}"
else:
base_url += f"?&{random_seed}"
try:
log.debug(f"Getting unsplash image from {base_url}")
response = requests.get(base_url)
response.raise_for_status()
image = Image.open(io.BytesIO(response.content))
return (
pil2tensor(
image,
),
)
except requests.exceptions.RequestException as e:
print("Error retrieving image:", e)
return (None,)
def bbox_dim(bbox):
left, upper, right, lower = bbox
width = right - left
height = lower - upper
return width, height
# TODO: Auto install the base font to ComfyUI/fonts
class MTB_TextToImage:
"""Utils to convert text to image using a font.
The tool looks for any .ttf file in the Comfy folder hierarchy.
"""
fonts = {}
DESCRIPTION = """# Text to Image
This node look for any font files in comfy_dir/fonts.
by default it fallsback to a default font.
"""
def __init__(self):
# - This is executed when the graph is executed,
# - we could conditionaly reload fonts there
pass
@classmethod
def CACHE_FONTS(cls):
font_extensions = ["*.ttf", "*.otf", "*.woff", "*.woff2", "*.eot"]
fonts = [font_path]
for extension in font_extensions:
try:
if comfy_dir.exists():
fonts.extend(comfy_dir.glob(f"fonts/**/{extension}"))
else:
log.warn(f"Directory {comfy_dir} does not exist.")
except Exception as e:
log.error(f"Error during font caching: {e}")
for font in fonts:
log.debug(f"Adding font {font}")
MTB_TextToImage.fonts[font.stem] = font.as_posix()
@classmethod
def INPUT_TYPES(cls):
if not cls.fonts:
cls.CACHE_FONTS()
else:
log.debug(f"Using cached fonts (count: {len(cls.fonts)})")
return {
"required": {
"text": (
"STRING",
{"default": "Hello world!"},
),
"font": ((sorted(cls.fonts.keys())),),
"wrap": ("BOOLEAN", {"default": True}),
"trim": ("BOOLEAN", {"default": True}),
"line_height": (
"FLOAT",
{"default": 1.0, "min": 0, "step": 0.1},
),
"font_size": (
"INT",
{"default": 32, "min": 1, "max": 2500, "step": 1},
),
"width": (
"INT",
{"default": 512, "min": 1, "max": 8096, "step": 1},
),
"height": (
"INT",
{"default": 512, "min": 1, "max": 8096, "step": 1},
),
"color": (
"COLOR",
{"default": "black"},
),
"background": (
"COLOR",
{"default": "white"},
),
"h_align": (("left", "center", "right"), {"default": "left"}),
"v_align": (("top", "center", "bottom"), {"default": "top"}),
"h_offset": (
"INT",
{"default": 0, "min": 0, "max": 8096, "step": 1},
),
"v_offset": (
"INT",
{"default": 0, "min": 0, "max": 8096, "step": 1},
),
"h_coverage": (
"INT",
{"default": 100, "min": 1, "max": 100, "step": 1},
),
}
}
RETURN_TYPES = ("IMAGE",)
RETURN_NAMES = ("image",)
FUNCTION = "text_to_image"
CATEGORY = "mtb/generate"
def text_to_image(
self,
text: str,
font,
wrap,
trim,
line_height,
font_size,
width,
height,
color,
background,
h_align="left",
v_align="top",
h_offset=0,
v_offset=0,
h_coverage=100,
):
import textwrap
from PIL import Image, ImageDraw, ImageFont
font_path = self.fonts[font]
text = (
text.encode("ascii", "ignore").decode().strip() if trim else text
)
# Handle word wrapping
if wrap:
wrap_width = (((width / 100) * h_coverage) / font_size) * 2
lines = textwrap.wrap(text, width=wrap_width)
else:
lines = [text]
font = ImageFont.truetype(font_path, size=font_size)
log.debug(f"Lines: {lines}")
img = Image.new("RGBA", (width, height), background)
draw = ImageDraw.Draw(img)
line_height_px = line_height * font_size
# Vertical alignment
if v_align == "top":
y_text = v_offset
elif v_align == "center":
y_text = ((height - (line_height_px * len(lines))) // 2) + v_offset
else: # bottom
y_text = (height - (line_height_px * len(lines))) - v_offset
def get_width(line):
if hasattr(font, "getsize"):
return font.getsize(line)[0]
else:
return font.getlength(line)
# Draw each line of text
for line in lines:
line_width = get_width(line)
# Horizontal alignment
if h_align == "left":
x_text = h_offset
elif h_align == "center":
x_text = ((width - line_width) // 2) + h_offset
else: # right
x_text = (width - line_width) - h_offset
draw.text((x_text, y_text), line, fill=color, font=font)
y_text += line_height_px
return (pil2tensor(img),)
__nodes__ = [
MTB_UnsplashImage,
MTB_TextToImage,
# MtbExamples,
]
|