Files
SnapOtter/packages/ai/python/ocr.py
T
SnapOtterandGitHub 6e3a14ec6b fix: remove automatic third-party egress of user data + optional strict offline mode (OSM tiles, Scalar fonts, editor fonts, AI model downloads) (#422)
* fix: remove all automatic third-party egress (OSM tiles, Scalar fonts, editor Google Fonts, AI model download fallbacks)

Phone-home audit follow-up. The product no longer makes any automatic
third-party request; user-initiated click-outs stay, and production now
fails closed on missing AI models.

1. GPS leak via OSM tiles: the strip-metadata panel auto-loaded
   tile.openstreetmap.org tiles encoding the photo's GPS position. The
   Leaflet mini-map is gone; coordinates render as text plus an explicit
   View on map link (openstreetmap.org, opens on click only). Removed
   tile.openstreetmap.org from the CSP img-src, dropped the leaflet
   dependency, added the viewOnMap i18n key to all 21 locales.

2. Scalar docs fonts: /api/docs loaded Inter and JetBrains Mono from
   fonts.scalar.com. Scalar now renders with withDefaultFonts: false and
   both --scalar-font and --scalar-font-code pinned to system stacks;
   fonts.scalar.com removed from the docs CSP font-src. Verified by
   injecting GET /api/docs/: config carries withDefaultFonts false and
   the served page has no fonts.scalar.com reference.

3. Editor Google Fonts: the editor font picker built
   fonts.googleapis.com stylesheet URLs for 25 web fonts the served CSP
   already blocked. The remote loading path is deleted; the picker now
   offers system fonts only, with a SELF_HOSTED_FONTS seam (FontFace API,
   same origin) for bundling fonts later. Unknown families saved in old
   documents fall back to the browser default.

4. Python sidecar fails closed on model downloads: new
   packages/ai/python/offline_guard.py gates every runtime download
   fallback (inpaint, outpaint, restore, noise_removal, detect_faces,
   enhance_faces, face_landmarks, red_eye_removal, remove_bg, ocr,
   transcribe, upscale) behind SNAPOTTER_ALLOW_MODEL_DOWNLOAD=1 with an
   actionable error. Bundled models keep working untouched.

5. OCR and transcription library-internal downloads: unbundled PaddleOCR
   language and detection fallbacks now raise the guard error naming the
   language instead of resolving models over the network; faster-whisper
   gets local_files_only when downloads are off.

6. GFPGAN and CodeFormer cwd-relative weights: facexlib and
   codeformer-pip resolve helper weights relative to the process cwd and
   fetch them from GitHub when absent. They are now symlinked from the
   installed bundle files under MODELS_PATH/gfpgan/facelib before the
   libraries load, failing closed when unresolvable.

Defense in depth: HF_HUB_OFFLINE=1 and TRANSFORMERS_OFFLINE=1 are set in
the runtime image and in the sidecar spawn env; install_feature.py lifts
them for user-initiated bundle installs and restores them afterwards
(it can run in-process inside the dispatcher). SNAPOTTER_ALLOW_MODEL_DOWNLOAD
is documented in .env.example, default off.

Validation: typecheck 9/9 workspaces, Biome clean on touched files,
5178 unit tests pass, py_compile on all touched scripts, guard behavior
exercised in both dispatcher exec and per-request import modes, zero
remaining runtime references to the three hosts. Docker build and live
AI inference need post-merge verification on the GPU host.

Claude-Session: https://claude.ai/code/session_01XGB4pGvTvb7sUX4JN745U7

* fix: allow AI model downloads by default, make strict offline mode opt-in

Product call: ease of use first. The download gating from the previous
commit inverts its default: runtime model fetches (public model weights
only, never user data) are allowed out of the box so AI tools self-heal,
and SNAPOTTER_ALLOW_MODEL_DOWNLOAD=0 becomes the explicit strict offline
mode for airgapped deployments, where every fallback raises the
actionable error instead of fetching.

Changes: offline_guard blocks only on an explicit 0/false; the
unconditional HF_HUB_OFFLINE/TRANSFORMERS_OFFLINE image ENV is removed
and bridge.ts sets those flags for the sidecar only in strict mode;
.env.example documents the new default; install_feature's lift/restore
stays. All bundled-path preferences, pre-existence checks, and symlink
pre-placement remain, so installed bundles never trigger a download.
The OSM, Scalar font, and editor font fixes are unchanged.

Validation rerun: typecheck 9/9, Biome clean on touched files, 5178
unit tests pass, py_compile on touched scripts, guard behavior verified
for unset/1 (allowed) and 0/false (blocked with the new message).

Claude-Session: https://claude.ai/code/session_01XGB4pGvTvb7sUX4JN745U7
2026-07-04 05:46:52 +00:00

436 lines
16 KiB
Python

"""Text extraction from images using Tesseract, PaddleOCR PP-OCRv5, or PaddleOCR-VL 1.5."""
import sys
import json
import os
# Prevent PaddlePaddle C++ runtime from probing for CUDA on CPU-only systems.
# Without these, paddlepaddle-gpu can segfault during import on machines without
# a GPU, because the C++ layer attempts GPU initialization before Python-level
# device routing takes effect. Must run before any PaddleOCR import.
from gpu import gpu_available
if not gpu_available():
if not os.environ.get("FLAGS_use_cuda"):
os.environ["FLAGS_use_cuda"] = "0"
if not os.environ.get("FLAGS_use_cudnn"):
os.environ["FLAGS_use_cudnn"] = "0"
# Lazy-loaded VLM instance (stays resident in dispatcher process)
_paddleocr_vl_instance = None
def emit_progress(percent, stage):
"""Emit structured progress to stderr for bridge.ts to capture."""
print(json.dumps({"progress": percent, "stage": stage}), file=sys.stderr, flush=True)
# OCR quality tiers backed by PaddleOCR. PaddleOCR ships as the GPU build
# (paddlepaddle-gpu) in the amd64 bundle; its native libs dlopen libcuda.so.1 at
# import and segfault on a CPU-only host, so these tiers need a usable GPU.
PADDLE_QUALITY_TIERS = ("balanced", "best")
def effective_quality(requested):
"""Return the OCR quality tier that can actually run on this host.
On a CPU-only host the PaddleOCR tiers (balanced/best) cannot load, so they
transparently fall back to "fast" (Tesseract), which runs on CPU. This keeps a
GPU-less host from importing paddlepaddle-gpu, whose import segfaults and wedges
the shared AI dispatcher.
"""
if requested in PADDLE_QUALITY_TIERS and not gpu_available():
return "fast"
return requested
TESSERACT_LANG_MAP = {
"en": "eng", "de": "deu", "fr": "fra", "es": "spa",
"zh": "chi_sim", "ja": "jpn", "ko": "kor",
}
PADDLE_LANG_MAP = {
"en": "en", "de": "latin", "fr": "latin", "es": "latin",
"zh": "ch", "ja": "japan", "ko": "korean",
}
# Bundled PaddleOCR models (shipped by the OCR feature bundle into MODELS_PATH).
# Pinning the constructor at these dirs keeps OCR fully offline / air-gapped and
# skips slow HuggingFace model resolution on first use. PP-OCRv5 server rec
# covers Chinese+English; latin covers en/de/fr/es; plus a dedicated Korean rec.
PADDLE_DET_MODEL = "PP-OCRv5_server_det"
PADDLE_TEXTLINE_MODEL = "PP-LCNet_x1_0_textline_ori"
PADDLE_REC_MODEL = {
"ch": "PP-OCRv5_server_rec",
"en": "latin_PP-OCRv5_mobile_rec",
"latin": "latin_PP-OCRv5_mobile_rec",
"korean": "korean_PP-OCRv5_mobile_rec",
}
def _bundled_paddle_kwargs(paddle_lang):
"""Build PaddleOCR kwargs that use the bundled models in MODELS_PATH.
Only pins a component when its model is actually present on disk, so a
partial bundle (or an unbundled language such as Japanese) falls back to
PaddleOCR's default resolution for that component. The doc-orientation and
doc-unwarping models are not bundled and not needed for plain OCR, so they
are disabled to avoid a runtime HuggingFace download.
"""
models_dir = os.environ.get("MODELS_PATH", "/data/ai/models")
def model_dir(name):
if not name:
return None
path = os.path.join(models_dir, name)
return path if os.path.isdir(path) else None
kwargs = {"use_doc_orientation_classify": False, "use_doc_unwarping": False}
det = model_dir(PADDLE_DET_MODEL)
if det:
kwargs["text_detection_model_name"] = PADDLE_DET_MODEL
kwargs["text_detection_model_dir"] = det
rec_name = PADDLE_REC_MODEL.get(paddle_lang)
rec = model_dir(rec_name)
if rec:
kwargs["text_recognition_model_name"] = rec_name
kwargs["text_recognition_model_dir"] = rec
textline = model_dir(PADDLE_TEXTLINE_MODEL)
if textline:
kwargs["textline_orientation_model_name"] = PADDLE_TEXTLINE_MODEL
kwargs["textline_orientation_model_dir"] = textline
kwargs["use_textline_orientation"] = True
else:
kwargs["use_textline_orientation"] = False
return kwargs
def auto_detect_language(input_path):
"""Detect the predominant script in the image using Tesseract multi-lang.
Runs a quick Tesseract pass with all installed language packs,
then analyzes the Unicode character ranges in the output to
determine which PaddleOCR language model to use.
"""
import subprocess
try:
result = subprocess.run(
["tesseract", input_path, "stdout", "-l", "eng+kor+chi_sim+jpn"],
capture_output=True, text=True, timeout=30,
)
text = result.stdout.strip()
if not text:
return "en"
hangul = sum(1 for c in text if "\uAC00" <= c <= "\uD7AF" or "\u1100" <= c <= "\u11FF")
cjk = sum(1 for c in text if "\u4E00" <= c <= "\u9FFF")
hiragana = sum(1 for c in text if "\u3040" <= c <= "\u309F")
katakana = sum(1 for c in text if "\u30A0" <= c <= "\u30FF")
latin = sum(1 for c in text if c.isascii() and c.isalpha())
total = hangul + cjk + hiragana + katakana + latin
if total == 0:
return "en"
if hangul / total > 0.3:
return "ko"
if (hiragana + katakana) / total > 0.2:
return "ja"
if cjk / total > 0.3:
return "zh"
return "en"
except Exception:
return "en"
def run_tesseract(input_path, language, is_auto=False):
"""Run Tesseract OCR (Fast tier)."""
import subprocess
# When auto-detected, use all installed language packs for best coverage
if is_auto:
tess_lang = "eng+kor+chi_sim+jpn+deu+fra+spa"
else:
tess_lang = TESSERACT_LANG_MAP.get(language, "eng")
emit_progress(30, "Scanning")
result = subprocess.run(
["tesseract", input_path, "stdout", "-l", tess_lang],
capture_output=True,
text=True,
timeout=120,
)
emit_progress(70, "Extracting text")
text = result.stdout.strip()
if result.returncode != 0 and not text:
raise RuntimeError(result.stderr.strip() or "Tesseract failed")
return text
def _extract_ocr_texts(results):
"""Extract text from PaddleOCR 3.x result objects.
Handles multiple result formats across PaddleOCR versions:
- 3.4.x: OCRResult with .json["res"]["rec_texts"]
- Earlier: result objects with .res dict containing "text" list
"""
text_parts = []
for res in results:
# PaddleOCR 3.4.x format: OCRResult with .json dict
if hasattr(res, "json") and isinstance(res.json, dict):
inner = res.json.get("res", {})
rec_texts = inner.get("rec_texts", [])
if rec_texts:
text_parts.extend(rec_texts)
continue
# Older format: .res dict with "text" list
if hasattr(res, "res") and isinstance(res.res, dict):
text_parts.extend(res.res.get("text", []))
return "\n".join(text_parts)
def run_paddleocr_v5(input_path, language):
"""Run PaddleOCR PP-OCRv5 server models (Balanced tier)."""
# GPU-only: paddlepaddle-gpu segfaults at import on a CPU-only host (libcuda
# absent). Refuse before importing so the caller falls back to Tesseract.
if not gpu_available():
raise ImportError(
"PaddleOCR (paddlepaddle-gpu) requires a GPU; the amd64 bundle ships the "
"GPU build, which cannot load on a CPU-only host. Use quality=fast (Tesseract)."
)
os.environ["PADDLE_PDX_DISABLE_MODEL_SOURCE_CHECK"] = "True"
stdout_fd = os.dup(1)
os.dup2(2, 1)
try:
import logging
from paddleocr import PaddleOCR
# Suppress PaddleOCR internal logging (replaces removed show_log param)
for name in ("ppocr", "paddleocr", "paddle"):
logging.getLogger(name).setLevel(logging.ERROR)
paddle_lang = PADDLE_LANG_MAP.get(language, "en")
device = "gpu:0" if gpu_available() else "cpu"
emit_progress(20, "Loading")
mk = _bundled_paddle_kwargs(paddle_lang)
# When a bundled recognizer is pinned, the model selects the script, so
# we omit lang (this is the proven fully-offline path). Lang-based
# resolution for a language without a bundled recognizer (e.g. ja) and
# a missing detection model both make PaddleOCR fetch models over the
# network, which strict offline mode blocks with a clear error.
if "text_recognition_model_dir" not in mk:
from offline_guard import ensure_download_allowed
ensure_download_allowed(f"PaddleOCR recognition model for language '{language}'")
mk["lang"] = paddle_lang
if "text_detection_model_dir" not in mk:
from offline_guard import ensure_download_allowed
ensure_download_allowed(f"PaddleOCR text detection model ({PADDLE_DET_MODEL})")
ocr = PaddleOCR(
device=device,
ocr_version="PP-OCRv5",
enable_mkldnn=False,
**mk,
)
emit_progress(30, "Scanning")
results = ocr.predict(input=input_path)
emit_progress(70, "Extracting text")
text = _extract_ocr_texts(results)
finally:
os.dup2(stdout_fd, 1)
os.close(stdout_fd)
return text
def run_paddleocr_vl(input_path):
"""Run PaddleOCR-VL 1.5 vision-language model (Best tier).
The VLM is lazy-loaded on first call and stays resident in the
dispatcher process for subsequent requests.
Requires PaddlePaddle >= 3.2 for fused_rms_norm_ext.
"""
global _paddleocr_vl_instance
# GPU-only: see run_paddleocr_v5. Refuse before importing paddle on CPU.
if not gpu_available():
raise ImportError(
"PaddleOCR-VL (paddlepaddle-gpu) requires a GPU; the amd64 bundle ships the "
"GPU build, which cannot load on a CPU-only host. Use quality=balanced or fast."
)
os.environ["PADDLE_PDX_DISABLE_MODEL_SOURCE_CHECK"] = "True"
stdout_fd = os.dup(1)
os.dup2(2, 1)
try:
if _paddleocr_vl_instance is None:
emit_progress(15, "Loading model")
from paddleocr import PaddleOCRVL
device = "gpu" if gpu_available() else "cpu"
_paddleocr_vl_instance = PaddleOCRVL(device=device)
emit_progress(30, "Scanning")
output = _paddleocr_vl_instance.predict(input_path)
emit_progress(70, "Extracting text")
text_parts = []
for res in output:
# PaddleOCR-VL 1.5+: markdown_texts holds the extracted text
if hasattr(res, "markdown") and isinstance(res.markdown, dict):
md_text = res.markdown.get("markdown_texts", "")
if md_text:
text_parts.append(md_text)
continue
if hasattr(res, "parsing_res_list"):
for block in res.parsing_res_list:
content = block.get("block_content", "")
if content:
text_parts.append(content)
elif hasattr(res, "rec_text"):
text_parts.append(res.rec_text)
# Also try the json-based extraction as fallback
elif hasattr(res, "json") and isinstance(res.json, dict):
inner = res.json.get("res", {})
rec_texts = inner.get("rec_texts", [])
text_parts.extend(rec_texts)
text = "\n".join(text_parts)
finally:
os.dup2(stdout_fd, 1)
os.close(stdout_fd)
return text
def main():
input_path = sys.argv[1]
settings = json.loads(sys.argv[2]) if len(sys.argv) > 2 else {}
quality = settings.get("quality", None)
language = settings.get("language", "auto")
enhance = settings.get("enhance", True)
# Backward compat: old "engine" param maps to quality
if quality is None:
engine = settings.get("engine", "tesseract")
quality = "fast" if engine == "tesseract" else "balanced"
# On a CPU-only host, downgrade GPU-only tiers (PaddleOCR) to Tesseract so we
# never import paddlepaddle-gpu, whose import segfaults and wedges the dispatcher.
downgraded = effective_quality(quality)
if downgraded != quality:
print(
json.dumps({"info": f"{quality} OCR needs a GPU; using {downgraded} (Tesseract) on CPU"}),
file=sys.stderr,
flush=True,
)
quality = downgraded
preprocessed_path = None
try:
emit_progress(5, "Preparing")
# Preprocessing (if enabled)
if enhance:
emit_progress(8, "Enhancing image")
try:
from ocr_preprocess import preprocess
preprocessed_path = input_path + "_enhanced.png"
preprocess(input_path, preprocessed_path)
input_path = preprocessed_path
except Exception as e:
print(json.dumps({"warning": f"Enhancement skipped: {e}"}), file=sys.stderr, flush=True)
preprocessed_path = None
# Language auto-detection
was_auto = language == "auto"
if was_auto:
emit_progress(10, "Detecting language")
language = auto_detect_language(input_path)
engine_used = quality
# Route to engine based on quality tier
if quality == "fast":
try:
text = run_tesseract(input_path, language, is_auto=was_auto)
engine_used = "tesseract"
except FileNotFoundError:
print(json.dumps({"success": False, "error": "Tesseract is not installed"}))
sys.exit(1)
elif quality == "balanced":
try:
text = run_paddleocr_v5(input_path, language)
engine_used = "paddleocr-v5"
except ImportError as e:
print(json.dumps({
"success": False,
"error": (
f"PaddleOCR is not installed: {e}. "
"Install the OCR feature or use quality=fast for Tesseract."
),
}))
sys.exit(1)
except Exception as e:
print(json.dumps({
"success": False,
"error": (
f"PaddleOCR PP-OCRv5 failed: {type(e).__name__}: {e}. "
"Install the OCR feature or use quality=fast for Tesseract."
),
}))
sys.exit(1)
elif quality == "best":
try:
text = run_paddleocr_vl(input_path)
engine_used = "paddleocr-vl"
except ImportError as e:
print(json.dumps({
"success": False,
"error": (
f"PaddleOCR-VL is not available: {e}. "
"Install the OCR feature or use quality=balanced for PP-OCRv5."
),
}))
sys.exit(1)
except Exception as e:
print(json.dumps({
"success": False,
"error": (
f"PaddleOCR-VL failed: {type(e).__name__}: {e}. "
"Install the OCR feature or use quality=balanced for PP-OCRv5."
),
}))
sys.exit(1)
else:
print(json.dumps({"success": False, "error": f"Unknown quality: {quality}"}))
sys.exit(1)
emit_progress(95, "Done")
print(json.dumps({"success": True, "text": text, "engine": engine_used}))
except Exception as e:
print(json.dumps({"success": False, "error": str(e)}))
sys.exit(1)
finally:
# Clean up preprocessed temp file
if preprocessed_path:
try:
os.remove(preprocessed_path)
except OSError:
pass
if __name__ == "__main__":
main()