Spaces:
Running
Running
File size: 10,244 Bytes
528c0e7 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 |
#!/usr/bin/env python3
from datetime import datetime
import os
import json
from pathlib import Path
import sys
import shutil
import time
import traceback
import pandas as pd
import streamlit as st
import json
from PIL import Image, ImageDraw
CACHE_TTL = 60 * 60 * 24 * 14
"""
Streamlit app utilities
"""
@st.cache_data(ttl=CACHE_TTL)
def load_json(basedir, name):
if not os.path.exists(f"{basedir}/{name}.json"):
return None
with open(f"{basedir}/{name}.json", "r") as f:
j = json.load(f)
return j
def load_json_no_cache(basedir, name):
if not os.path.exists(f"{basedir}/{name}.json"):
return None
with open(f"{basedir}/{name}.json", "r") as f:
j = json.load(f)
return j
def save_json(basedir, name, data):
with open(f"{basedir}/{name}.json", "w") as f:
json.dump(data, f, indent=4)
@st.cache_data
def load_image(image_file):
img = Image.open(image_file)
return img
@st.cache_resource
def load_page(page_path):
return open(page_path, "rb")
def shorten(s):
# shorten to 100 characters
if len(s) > 100:
s = s[:100] + "..."
return s
@st.cache_data
def parse_arguments(action):
s = []
event_type = action["intent"]
args = action["arguments"]
if event_type == "textInput":
txt = args["text"]
txt = txt.strip()
# escape markdown characters
txt = txt.replace("_", "\\_")
txt = txt.replace("*", "\\*")
txt = txt.replace("`", "\\`")
txt = txt.replace("$", "\\$")
txt = shorten(txt)
s.append(f'"{txt}"')
elif event_type == "change":
s.append(f'{args["value"]}')
elif event_type == "load":
url = args["properties"].get("url") or args.get("url")
short_url = shorten(url)
s.append(f'"[{short_url}]({url})"')
if args["properties"].get("transitionType"):
s.append(f'*{args["properties"]["transitionType"]}*')
s.append(f'*{" ".join(args["properties"]["transitionQualifiers"])}*')
elif event_type == "scroll":
s.append(f'{args["scrollX"]}, {args["scrollY"]}')
elif event_type == "say":
s.append(f'"{args["text"]}"')
elif event_type == "copy":
selected = shorten(args["selected"])
s.append(f'"{selected}"')
elif event_type == "paste":
pasted = shorten(args["pasted"])
s.append(f'"{pasted}"')
elif event_type == "tabcreate":
s.append(f'{args["properties"]["tabId"]}')
elif event_type == "tabremove":
s.append(f'{args["properties"]["tabId"]}')
elif event_type == "tabswitch":
s.append(
f'{args["properties"]["tabIdOrigin"]} -> {args["properties"]["tabId"]}'
)
if args.get("element"):
if event_type == 'click':
x = round(args['metadata']['mouseX'], 1)
y = round(args['metadata']['mouseY'], 1)
uid = args.get('element', {}).get('attributes', {}).get("data-webtasks-id")
s.append(f"*x =* {x}, *y =* {y}, *uid =* {uid}")
else:
top = round(args["element"]["bbox"]["top"], 1)
left = round(args["element"]["bbox"]["left"], 1)
right = round(args["element"]["bbox"]["right"], 1)
bottom = round(args["element"]["bbox"]["bottom"], 1)
s.append(f"*top =* {top}, *left =* {left}, *right =* {right}, *bottom =* {bottom}")
return ", ".join(s)
@st.cache_resource(max_entries=50_000, ttl=CACHE_TTL)
def create_visualization(_img, event_type, bbox, x, y, screenshot_path):
# screenshot_path is not used, but we need it for caching since we can't cache
# PIL images (hence the leading underscore in the variable name to indicate
# that it's not hashed)
_img = _img.convert("RGBA")
draw = ImageDraw.Draw(_img)
# draw a bounding box around the element
color = {
"click": "red",
"hover": "orange",
"textInput": "blue",
"change": "green",
}[event_type]
left = bbox["left"]
top = bbox["top"]
w = bbox["width"]
h = bbox["height"]
draw.rectangle((left, top, left + w, top + h), outline=color, width=2)
if event_type in ["click", "hover"]:
r = 15
for i in range(1, 5):
rx = r * i
draw.ellipse((x - rx, y - rx, x + rx, y + rx), outline=color, width=3)
draw.ellipse((x - r, y - r, x + r, y + r), fill=color)
return _img
@st.cache_data(max_entries=50_000, ttl=CACHE_TTL)
def get_screenshot_minimal(screenshot_path, event_type, bbox, x, y, new_width=None):
img = load_image(screenshot_path)
# vis = None
if event_type in ["click", "textInput", "change", "hover"]:
img = create_visualization(img, event_type, bbox, x, y, screenshot_path)
if new_width is not None:
# Resize to 800px wide
w, h = img.size
new_w = new_width
new_h = int(new_w * h / w)
img = img.resize((new_w, new_h))
print(f"Resized '{screenshot_path}' to", new_w, new_h)
return img
def get_event_info(d):
event_type = d["action"]["intent"]
try:
bbox = d["action"]["arguments"]["element"]["bbox"]
except KeyError:
bbox = None
try:
x = d["action"]["arguments"]["properties"]["x"]
y = d["action"]["arguments"]["properties"]["y"]
except KeyError:
x = None
y = None
return event_type, bbox, x, y
def get_screenshot(d, basedir, new_width=None):
screenshot_filename = d["state"]["screenshot"]
if not screenshot_filename:
return None
event_type, bbox, x, y = get_event_info(d)
screenshot_path = f"{basedir}/screenshots/{screenshot_filename}"
return get_screenshot_minimal(
screenshot_path, event_type, bbox, x, y, new_width=new_width
)
def text_bubble(text, color):
text = text.replace("\n", "<br>").replace("\t", " " * 8)
return f'<div style="background-color:{color}; padding: 8px; margin: 6px; border-radius:10px; display:inline-block;">{text}</div>'
def gather_chat_history(data, example_index):
chat = []
for i, d in enumerate(data):
if d["type"] == "chat":
if i >= example_index:
break
chat.append(d)
# # leave out just 5 last messages
# if len(chat) > 5:
# chat = chat[-5:]
return reversed(chat)
def format_chat_message(d):
if d["speaker"] == "instructor":
return text_bubble("🧑 " + d["utterance"], "rgba(63, 111, 255, 0.35)")
else:
return text_bubble("🤖 " + d["utterance"], "rgba(185,185,185,0.35)")
def find_screenshot(data, example_index, basedir):
# keep looking at previous screenshots until we find one
# if there is none, return None
for i in range(example_index, -1, -1):
d = data[i]
if d["type"] == "chat":
continue
screenshot = get_screenshot(d, basedir)
if screenshot:
return screenshot
return None
def create_visualization_2(_img, bbox, color, width, x, y):
_img = _img.convert("RGBA")
draw = ImageDraw.Draw(_img)
if bbox:
left = bbox["left"]
top = bbox["top"]
w = bbox["width"]
h = bbox["height"]
draw.rectangle((left, top, left + w, top + h), outline=color, width=width)
if x and y:
r = 8
for i in range(1, 4):
rx = r * i
draw.ellipse((x - rx, y - rx, x + rx, y + rx), outline=color, width=2)
draw.ellipse((x - r, y - r, x + r, y + r), fill=color)
return _img
def rescale_bbox(bbox, scaling_factor):
return {
k: bbox[k] * scaling_factor
for k in ["top", "left", "width", "height", "right", "bottom"]
if k in bbox
}
def show_overlay(
_img,
pred,
ref,
turn_args,
turn_metadata,
scale_pred=True,
show=("pred_coords", "ref", "pred_elem"),
):
scaling_factor = turn_metadata.get("zoomLevel", 1.0)
if "pred_elem" in show:
# First, draw red box around predicted element
if pred.get("element") and pred["element"].get("bbox"):
# rescale the bbox by scaling_factor
bbox = rescale_bbox(pred["element"]["bbox"], scaling_factor)
_img = create_visualization_2(
_img, bbox, color="red", width=9, x=None, y=None
)
if "ref" in show:
# Finally, draw a blue box around the reference element (if it exists)
if ref.get("element") and ref["element"].get("bbox"):
# rescale the bbox
bbox = rescale_bbox(ref["element"]["bbox"], scaling_factor)
x = turn_args.get("properties", {}).get("x")
y = turn_args.get("properties", {}).get("y")
_img = create_visualization_2(_img, bbox, color="blue", width=6, x=x, y=y)
if "pred_coords" in show:
# Second draw a green box and x/y coordinate based on predicted coordinates
# The predicted coordinates are the raw output of the model,
# Whereas the predicted element is the inferred element from the predicted coordinates
if pred["args"].get("x") and pred["args"].get("y"):
x = pred["args"]["x"]
y = pred["args"]["y"]
if scale_pred:
x = x * scaling_factor
y = y * scaling_factor
else:
x = None
y = None
# If the predicted element is a bounding box, draw a green box around it
if all(c in pred["args"] for c in ["top", "left", "right", "bottom"]):
bbox = {
"top": pred["args"]["top"],
"left": pred["args"]["left"],
"width": (pred["args"]["right"] - pred["args"]["left"]),
"height": (pred["args"]["bottom"] - pred["args"]["top"]),
"right": pred["args"]["right"],
"bottom": pred["args"]["bottom"],
}
if scale_pred:
bbox = rescale_bbox(bbox, scaling_factor)
else:
# Otherwise, do nothing
bbox = None
_img = create_visualization_2(_img, bbox=bbox, color="green", width=3, x=x, y=y)
return _img
|