Update app.py

app.py

@@ -358,27 +358,20 @@ def extract_pdf_name(text: str, source_pdf: str = None):
                 return name.upper()
 
     return None
-def extract_speed_from_pdf_text_or_filename(full_text: str, source_pdf: str):
-    # 1) Cherche dans le texte PDF
+
+def extract_speed_from_text(full_text: str):
+    if not full_text:
+        return np.nan
+
     patterns = [
         r"VMA\s*(\d+(?:[.,]\d+)?)\s*kmh",
         r"VMA\s*(\d+(?:[.,]\d+)?)\s*km/h",
-        r"Vitesse,\s*km/h\s*(\d+(?:[.,]\d+)?)",
         r"(\d+(?:[.,]\d+)?)\s*kmh",
         r"(\d+(?:[.,]\d+)?)\s*km/h",
     ]
 
-    if full_text:
-        for pattern in patterns:
-            m = re.search(pattern, full_text, flags=re.I)
-            if m:
-                return float(m.group(1).replace(",", "."))
-
-    # 2) Fallback : cherche dans le nom du fichier
-    if source_pdf:
-        m = re.search(r"(\d+(?:[.,]\d+)?)\s*kmh", source_pdf, flags=re.I)
-        if not m:
-            m = re.search(r"(\d+(?:[.,]\d+)?)\s*km/h", source_pdf, flags=re.I)
+    for pattern in patterns:
+        m = re.search(pattern, full_text, flags=re.I)
         if m:
             return float(m.group(1).replace(",", "."))
 
@@ -430,9 +423,8 @@ def parse_zebris_pdf(uploaded_pdf):
             page_texts.append(txt)
 
     full_text = " ".join(page_texts)
-
-    # Extraction de l'allure du PDF (ex: "VMA 14kmh" ou "14,0 km/h")
-    data["speed_kmh"] = extract_speed_from_pdf_text_or_filename(full_text, source_pdf)
+    data["speed_kmh"] = extract_speed_from_text(full_text)
+    st.text(full_text[:2000])
 
     # fallback si nom non récupéré via fichier
     if not data["athlete_name"]: