Domi31tls
65c97a39b3
Suppression des 5 hiddenimports optimum* dans les deux specs V6/CLI. Ajout de EXCLUDED_TORCH_STACK + excludes=EXCLUDED_TORCH_STACK dans Analysis() pour éviter que PyInstaller embarque torch (~+2 Go) via optimum à l'analyse statique. Spec GUI V5 legacy inchangée (garde optimum légitimement). Test anti-dérive ajouté (5 cas). Correctif import pytest inutilisé (version.py). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
189 lines
4.8 KiB
Python
189 lines
4.8 KiB
Python
# PyInstaller spec — GUI V6 (build-prep G3-D).
|
|
#
|
|
# Produit `Anonymisation.exe` (V6), source de l'installateur Inno
|
|
# `installer/Anonymisation.iss` qui génère la cible finale `Anonymisation-Setup.exe`.
|
|
#
|
|
# Entrée directe : Pseudonymisation_Gui_V6.py (expose main() + --self-test).
|
|
# Ne construit AUCUN artefact ici : la génération réelle se fait sur Windows.
|
|
|
|
import os
|
|
from pathlib import Path
|
|
|
|
|
|
block_cipher = None
|
|
|
|
project_dir = Path(globals().get("SPECPATH", os.getcwd())).resolve()
|
|
|
|
|
|
def _data_entry(relative_path: str, target_dir: str | None = None):
|
|
src = project_dir / relative_path
|
|
if not src.exists():
|
|
return None
|
|
return (str(src), target_dir or relative_path)
|
|
|
|
|
|
datas = []
|
|
for relative_path, target_dir in [
|
|
("config", "config"),
|
|
("data/bdpm", "data/bdpm"),
|
|
("data/finess", "data/finess"),
|
|
("data/insee", "data/insee"),
|
|
("models/camembert-bio-deid/onnx", "models/camembert-bio-deid/onnx"),
|
|
("detectors", "detectors"),
|
|
("scripts", "scripts"),
|
|
("assets", "assets"),
|
|
]:
|
|
entry = _data_entry(relative_path, target_dir)
|
|
if entry is not None:
|
|
datas.append(entry)
|
|
|
|
for relative_path in [
|
|
"data/stopwords_manuels.txt",
|
|
"data/villes_blacklist.txt",
|
|
"data/dpi_labels_blacklist.txt",
|
|
"data/companion_blacklist.txt",
|
|
]:
|
|
entry = _data_entry(relative_path, "data")
|
|
if entry is not None:
|
|
datas.append(entry)
|
|
|
|
onnxtr_cache_dir = Path(os.environ.get("ONNXTR_CACHE_DIR", Path.home() / ".cache" / "onnxtr"))
|
|
required_onnxtr_weights = [
|
|
"db_resnet50-69ba0015.onnx",
|
|
"crnn_vgg16_bn-743599aa.onnx",
|
|
]
|
|
missing_onnxtr_weights = []
|
|
for filename in required_onnxtr_weights:
|
|
src = onnxtr_cache_dir / "models" / filename
|
|
if src.exists():
|
|
datas.append((str(src), "models/onnxtr/models"))
|
|
else:
|
|
missing_onnxtr_weights.append(str(src))
|
|
if missing_onnxtr_weights:
|
|
raise FileNotFoundError(
|
|
"Poids OCR OnnxTR manquants pour le build frozen : "
|
|
+ ", ".join(missing_onnxtr_weights)
|
|
+ ". Précharger OnnxTR (lancer une OCR une fois) ou définir ONNXTR_CACHE_DIR avant PyInstaller."
|
|
)
|
|
|
|
|
|
hiddenimports = [
|
|
# Entrée + package GUI V6
|
|
"Pseudonymisation_Gui_V6",
|
|
"gui_v6",
|
|
"gui_v6.app",
|
|
"gui_v6.theme",
|
|
"gui_v6.license_client",
|
|
"gui_v6.license_store",
|
|
"gui_v6.machine_id",
|
|
"gui_v6.engine_bridge",
|
|
"gui_v6.config_state",
|
|
"gui_v6.version",
|
|
"gui_v6._build_version",
|
|
"gui_v6.processing_runner",
|
|
"gui_v6.tabs",
|
|
"gui_v6.tabs.tab_about",
|
|
"gui_v6.tabs.tab_config",
|
|
"gui_v6.tabs.tab_usage",
|
|
# UI customtkinter
|
|
"customtkinter",
|
|
"darkdetect",
|
|
# Réseau licence
|
|
"requests",
|
|
# Moteur + modules support (inchangés vs V5)
|
|
"anonymizer_core_refactored_onnx",
|
|
"admin_mode",
|
|
"admin_rules",
|
|
"config_defaults",
|
|
"profile_defaults",
|
|
"gui_batch_paths",
|
|
"manual_masking",
|
|
"pdf_mask_designer",
|
|
"format_converter",
|
|
"ner_manager_onnx",
|
|
"camembert_ner_manager",
|
|
"eds_pseudo_manager",
|
|
"gliner_manager",
|
|
"vlm_manager",
|
|
"build_info",
|
|
# OCR OnnxTR (ONNX Runtime, remplace docTR — sans torch ni doctr)
|
|
"onnxtr",
|
|
"onnxtr.io",
|
|
"onnxtr.models",
|
|
"onnxtr.models.detection",
|
|
"onnxtr.models.recognition",
|
|
"onnxtr.utils",
|
|
"onnxtr.utils.data",
|
|
# Dépendances transitives OnnxTR (hiddenimports défensifs vs omission PyInstaller)
|
|
"pyclipper",
|
|
"scipy.cluster.hierarchy",
|
|
"scipy.special",
|
|
"cv2",
|
|
"edsnlp",
|
|
"edsnlp.pipes",
|
|
"edsnlp.pipes.ner",
|
|
"edsnlp.pipes.ner.pseudo",
|
|
"spacy",
|
|
"spacy.lang.fr",
|
|
"gliner",
|
|
"onnxruntime",
|
|
"transformers",
|
|
"tokenizers",
|
|
"pdfplumber",
|
|
"fitz",
|
|
"PIL",
|
|
"yaml",
|
|
"loguru",
|
|
"regex",
|
|
]
|
|
|
|
# P0-3 (Plan 3) : exclusion dure de la pile torch. Le core fait un
|
|
# `import torch` lazy (try/except no-op) dans _configure_torch_threads que
|
|
# l'analyse statique suivrait ; en frozen l'ImportError est attendue et gérée.
|
|
EXCLUDED_TORCH_STACK = [
|
|
"torch",
|
|
"torchvision",
|
|
"optimum",
|
|
"doctr",
|
|
]
|
|
|
|
|
|
a = Analysis(
|
|
[str(project_dir / "Pseudonymisation_Gui_V6.py")],
|
|
pathex=[str(project_dir)],
|
|
datas=datas,
|
|
hiddenimports=hiddenimports,
|
|
excludes=EXCLUDED_TORCH_STACK,
|
|
cipher=block_cipher,
|
|
noarchive=False,
|
|
)
|
|
pyz = PYZ(a.pure, a.zipped_data, cipher=block_cipher)
|
|
|
|
splash = Splash(
|
|
str(project_dir / "assets" / "splash.png"),
|
|
binaries=a.binaries,
|
|
datas=a.datas,
|
|
text_pos=(60, 195),
|
|
text_size=10,
|
|
text_color="white",
|
|
minify_script=True,
|
|
always_on_top=False,
|
|
)
|
|
|
|
exe = EXE(
|
|
pyz,
|
|
a.scripts,
|
|
splash,
|
|
splash.binaries,
|
|
a.binaries,
|
|
a.zipfiles,
|
|
a.datas,
|
|
[],
|
|
name="Anonymisation",
|
|
debug=False,
|
|
strip=False,
|
|
upx=False,
|
|
console=False,
|
|
icon=str(project_dir / "assets" / "icons" / "app.ico"),
|
|
)
|