Spaces:
Sleeping
Sleeping
File size: 10,361 Bytes
9e61e4e | 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 | """
Gradio EXIF / GPS metadata viewer — multi-image edition.
Features:
- Upload MANY JPEGs at once.
- Extract the ENTIRE metadata set: file info, PIL container info (icc/jfif/dpi),
the base IFD, the Exif sub-IFD (camera/exposure), GPS IFD, and Interop IFD.
- Live progress bar that advances per image, naming the file in progress.
- Per-image NSFW gate + malformed/oversized guards (bad files are reported,
not fatal — the batch keeps going).
- Stateless handler: nothing persisted, nothing shared between users.
Run locally: python app.py
Share with testers: GRADIO_SHARE=1 python app.py
"""
import io
import os
from fractions import Fraction
import gradio as gr
from PIL import Image, ExifTags, UnidentifiedImageError
from PIL.TiffImagePlugin import IFDRational
# ----------------------------------------------------------------------------
# Config
# ----------------------------------------------------------------------------
MAX_BYTES = 15 * 1024 * 1024
ALLOWED_FORMATS = {"JPEG", "MPO"}
MAX_DIMENSION = 12000
NSFW_THRESHOLD = 0.80
SHARE = os.environ.get("GRADIO_SHARE", "0") == "1"
Image.MAX_IMAGE_PIXELS = 100_000_000
# ----------------------------------------------------------------------------
# NSFW classifier (optional: skipped if transformers/torch absent)
# ----------------------------------------------------------------------------
_nsfw_pipe = None
def _get_nsfw_pipe():
global _nsfw_pipe
if _nsfw_pipe is None:
try:
from transformers import pipeline
_nsfw_pipe = pipeline(
"image-classification", model="Falconsai/nsfw_image_detection"
)
except Exception as e: # noqa: BLE001
print(f"[nsfw] classifier unavailable, skipping gate: {e}")
_nsfw_pipe = False
return _nsfw_pipe
def nsfw_score(img: Image.Image) -> float:
pipe = _get_nsfw_pipe()
if not pipe:
return 0.0
for r in pipe(img.convert("RGB")):
if r["label"].lower() == "nsfw":
return float(r["score"])
return 0.0
# ----------------------------------------------------------------------------
# JSON sanitizing — EXIF values are full of rationals, bytes, nested tuples
# ----------------------------------------------------------------------------
def _clean(v, depth=0):
if depth > 6:
return str(v)
if isinstance(v, IFDRational):
try:
return float(v)
except (ZeroDivisionError, ValueError):
return f"{v.numerator}/{v.denominator}"
if isinstance(v, bytes):
if len(v) <= 64:
try:
return v.decode("utf-8", "replace").rstrip("\x00")
except Exception: # noqa: BLE001
return v.hex()
return f"<{len(v)} bytes>"
if isinstance(v, (tuple, list)):
return [_clean(x, depth + 1) for x in v]
if isinstance(v, dict):
return {str(k): _clean(val, depth + 1) for k, val in v.items()}
if isinstance(v, (int, float, str, bool)) or v is None:
return v
return str(v)
# ----------------------------------------------------------------------------
# GPS decoding (needs raw rationals, so done before _clean)
# ----------------------------------------------------------------------------
def _ratio_to_float(x) -> float:
try:
return float(x)
except (TypeError, ValueError):
if isinstance(x, tuple) and len(x) == 2:
return x[0] / x[1]
return float(Fraction(x))
def _dms_to_decimal(dms, ref) -> float:
deg, minute, sec = (_ratio_to_float(x) for x in dms)
dec = deg + minute / 60.0 + sec / 3600.0
return round(-dec if ref in ("S", "W") else dec, 6)
def _decode_gps(gps_ifd):
g = {ExifTags.GPSTAGS.get(k, k): v for k, v in gps_ifd.items()}
if "GPSLatitude" not in g or "GPSLongitude" not in g:
return None
try:
out = {
"latitude": _dms_to_decimal(g["GPSLatitude"], g.get("GPSLatitudeRef", "N")),
"longitude": _dms_to_decimal(g["GPSLongitude"], g.get("GPSLongitudeRef", "E")),
}
if "GPSAltitude" in g:
out["altitude_m"] = round(_ratio_to_float(g["GPSAltitude"]), 1)
return out
except Exception as e: # noqa: BLE001
print(f"[gps] parse error: {e}")
return None
# ----------------------------------------------------------------------------
# Full metadata extraction — every IFD plus PIL container info
# ----------------------------------------------------------------------------
def extract_full_metadata(img: Image.Image, raw_size: int, fname: str):
meta = {
"file": {
"name": fname,
"size_bytes": raw_size,
"format": img.format,
"format_description": getattr(img, "format_description", None),
"mode": img.mode,
"width": img.width,
"height": img.height,
},
# PIL container info: jfif, dpi, icc_profile, progression, etc.
"info": {k: _clean(v) for k, v in img.info.items()},
}
exif = img.getexif()
meta["exif_base"] = {
ExifTags.TAGS.get(t, str(t)): _clean(v) for t, v in exif.items()
}
# Exif sub-IFD: the photographic metadata (ISO, exposure, lens, dates…)
try:
photo = exif.get_ifd(ExifTags.IFD.Exif)
meta["exif_photo"] = {
ExifTags.TAGS.get(t, str(t)): _clean(v) for t, v in photo.items()
}
except Exception: # noqa: BLE001
meta["exif_photo"] = {}
# GPS IFD: raw (named) + decoded decimal coords
gps_decoded = None
try:
gps_ifd = exif.get_ifd(ExifTags.IFD.GPSInfo)
except Exception: # noqa: BLE001
gps_ifd = {}
if gps_ifd:
meta["gps_raw"] = {
ExifTags.GPSTAGS.get(t, str(t)): _clean(v) for t, v in gps_ifd.items()
}
gps_decoded = _decode_gps(gps_ifd)
if gps_decoded:
meta["gps"] = gps_decoded
# Interop IFD (present on many camera JPEGs)
try:
interop = exif.get_ifd(ExifTags.IFD.Interop)
if interop:
meta["interop"] = {
ExifTags.TAGS.get(t, str(t)): _clean(v) for t, v in interop.items()
}
except Exception: # noqa: BLE001
pass
return meta, gps_decoded
# ----------------------------------------------------------------------------
# Validation
# ----------------------------------------------------------------------------
def load_and_validate(path: str):
size = os.path.getsize(path)
if size == 0:
raise ValueError("Empty file.")
if size > MAX_BYTES:
raise ValueError(f"Too large ({size/1e6:.1f} MB, max {MAX_BYTES/1e6:.0f} MB).")
with open(path, "rb") as f:
raw = f.read()
try:
Image.open(io.BytesIO(raw)).verify()
except (UnidentifiedImageError, OSError, ValueError) as e:
raise ValueError(f"Not a valid image: {e}")
img = Image.open(io.BytesIO(raw))
if img.format not in ALLOWED_FORMATS:
raise ValueError(f"Only JPEG accepted (got {img.format}).")
if max(img.size) > MAX_DIMENSION:
raise ValueError(f"Dimensions too large ({img.size}).")
img.load()
return img, size
# ----------------------------------------------------------------------------
# Handler — multi-image, with a per-image progress bar
# ----------------------------------------------------------------------------
def process(files, progress=gr.Progress()):
if not files:
return [], {}, None, "Upload one or more JPEGs to begin."
gallery, all_meta, coords, rejected = [], {}, [], []
n = len(files)
for i, path in enumerate(files):
name = os.path.basename(path)
progress(i / n, desc=f"Processing {name} ({i + 1}/{n})")
try:
img, raw_size = load_and_validate(path)
except ValueError as e:
rejected.append((name, str(e)))
all_meta[name] = {"error": str(e)}
continue
except Exception as e: # noqa: BLE001
rejected.append((name, f"unreadable: {e}"))
all_meta[name] = {"error": f"unreadable: {e}"}
continue
score = nsfw_score(img)
if score >= NSFW_THRESHOLD:
rejected.append((name, f"content filter ({score:.2f})"))
all_meta[name] = {"error": f"rejected by content filter (score={score:.2f})"}
continue
meta, gps = extract_full_metadata(img, raw_size, name)
all_meta[name] = meta
gallery.append((img.convert("RGB"), name + (" 📍" if gps else "")))
if gps:
coords.append([name, gps["latitude"], gps["longitude"]])
progress(1.0, desc="Done")
summary = f"### Processed {len(gallery)}/{n} image(s)\n"
if coords:
summary += f"- 📍 {len(coords)} with GPS location\n"
if rejected:
summary += f"- ⚠️ {len(rejected)} rejected:\n"
for nm, why in rejected:
summary += f" - **{nm}**: {why}\n"
return gallery, all_meta, (coords or None), summary
# ----------------------------------------------------------------------------
# UI
# ----------------------------------------------------------------------------
with gr.Blocks(title="JPEG Metadata Viewer") as demo:
gr.Markdown(
"# JPEG Metadata & Location Viewer\n"
"Upload one or more JPEGs to inspect their full EXIF metadata. "
"The tool reads embedded GPS location data when present."
)
with gr.Row():
files_in = gr.File(
label="Upload JPEGs",
file_types=[".jpg", ".jpeg"],
file_count="multiple",
type="filepath",
)
status = gr.Markdown()
run_btn = gr.Button("Extract metadata", variant="primary")
gallery = gr.Gallery(label="Images", columns=4, height="auto")
coords_df = gr.Dataframe(headers=["file", "lat", "lon"], label="GPS locations")
meta_out = gr.JSON(label="Full metadata (per image)")
run_btn.click(
process,
inputs=files_in,
outputs=[gallery, meta_out, coords_df, status],
)
demo.queue(default_concurrency_limit=4, max_size=64)
if __name__ == "__main__":
demo.launch(
share=SHARE,
max_file_size=f"{MAX_BYTES // (1024 * 1024)}mb",
show_error=True,
) |