Dom/rpa_vision_v3
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: aligner capture monitors[0] partout (cause de la régression)
Some checks failed
security-audit / Bandit (scan statique) (push) Successful in 12s
Details
security-audit / pip-audit (CVE dépendances) (push) Successful in 12s
Details
security-audit / Scan secrets (grep) (push) Successful in 8s
Details
tests / Lint (ruff + black) (push) Successful in 14s
Details
tests / Tests unitaires (sans GPU) (push) Failing after 15s
Details
tests / Tests sécurité (critique) (push) Has been skipped
Details

Browse Source 
La capture VWB utilisait monitors[0] (composite) mais l'exécution
utilisait monitors[1] (premier écran). Images incompatibles → CLIP
retournait 0.00 sur un écran identique.

Tous les fichiers alignés sur monitors[0].

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
Dom
2026-04-21 10:52:13 +02:00
parent 2fbb305f65
commit c50adab3a1
2 changed files with 5 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
core/execution/input_handler.py
View File

@@ -93,7 +93,7 @@ def check_screen_for_patterns() -> Optional[Dict[str, Any]]:
lib = UIPatternLibrary() lib = UIPatternLibrary()
with mss.mss() as sct: with mss.mss() as sct:
monitor = sct.monitors[1] monitor = sct.monitors[0]
screenshot = sct.grab(monitor) screenshot = sct.grab(monitor)
screen = Image.frombytes('RGB', screenshot.size, screenshot.bgra, 'raw', 'BGRX') screen = Image.frombytes('RGB', screenshot.size, screenshot.bgra, 'raw', 'BGRX')
@@ -160,7 +160,7 @@ def handle_detected_pattern(pattern: Dict[str, Any]) -> bool:
return extractor.extract_words_from_image(img) return extractor.extract_words_from_image(img)
with mss.mss() as sct: with mss.mss() as sct:
monitor = sct.monitors[1] monitor = sct.monitors[0]
screenshot = sct.grab(monitor) screenshot = sct.grab(monitor)
screen = Image.frombytes('RGB', screenshot.size, screenshot.bgra, 'raw', 'BGRX') screen = Image.frombytes('RGB', screenshot.size, screenshot.bgra, 'raw', 'BGRX')
@@ -262,7 +262,7 @@ def vlm_reason_about_screen(objective: str = "", context: str = "") -> Optional[
from PIL import Image from PIL import Image
with mss.mss() as sct: with mss.mss() as sct:
monitor = sct.monitors[1] monitor = sct.monitors[0]
screenshot = sct.grab(monitor) screenshot = sct.grab(monitor)
screen = Image.frombytes('RGB', screenshot.size, screenshot.bgra, 'raw', 'BGRX') screen = Image.frombytes('RGB', screenshot.size, screenshot.bgra, 'raw', 'BGRX')
@@ -432,7 +432,7 @@ def _capture_screen():
from PIL import Image as PILImage from PIL import Image as PILImage
with mss.mss() as sct: with mss.mss() as sct:
monitor = sct.monitors[1] monitor = sct.monitors[0]
screenshot = sct.grab(monitor) screenshot = sct.grab(monitor)
screen = PILImage.frombytes('RGB', screenshot.size, screenshot.bgra, 'raw', 'BGRX') screen = PILImage.frombytes('RGB', screenshot.size, screenshot.bgra, 'raw', 'BGRX')
return screen, monitor['width'], monitor['height'] return screen, monitor['width'], monitor['height']

2
visual_workflow_builder/backend/services/intelligent_executor.py
View File

@@ -700,7 +700,7 @@ def find_and_click(
import mss import mss
with mss.mss() as sct: with mss.mss() as sct:
monitor = sct.monitors[1] # Premier écran monitor = sct.monitors[0] # Écran composite (identique à la capture VWB)
screenshot = sct.grab(monitor) screenshot = sct.grab(monitor)
screen_image = Image.frombytes('RGB', screenshot.size, screenshot.bgra, 'raw', 'BGRX') screen_image = Image.frombytes('RGB', screenshot.size, screenshot.bgra, 'raw', 'BGRX')