Integrate Euclid galaxy images and dating descriptions from HuggingFace datasets

requirements.txt

@@ -1,4 +1,7 @@
 dash==2.14.1
 dash-bootstrap-components==1.5.0
 gunicorn==21.2.0
+huggingface-hub==0.20.1
 python-dotenv==1.1.1
+datasets==2.16.1
+pillow==10.1.0

src/components.py

@@ -279,7 +279,7 @@ def _create_star_field(n=80):
 
 
 def create_galaxy_card(galaxy_id, side="left", is_champion=False):
-    """Build a single galaxy profile card -- image + name only."""
+    """Build a single galaxy profile card -- image + name + description."""
     profile = GALAXY_PROFILES[galaxy_id]
     btn_id = f"{side}-card-btn"
     
@@ -297,6 +297,9 @@ def create_galaxy_card(galaxy_id, side="left", is_champion=False):
         card_style["borderRadius"] = "12px"
         card_style["animation"] = "galaxyWin 1.5s ease-in-out"
     
+    # Get description text
+    description = profile.get('description', profile.get('bio', ''))
+    
     card_contents = [
         html.Img(
             src=f"/galaxy-images/{galaxy_id}.jpg",
@@ -316,7 +319,22 @@ def create_galaxy_card(galaxy_id, side="left", is_champion=False):
             ],
             className="galaxy-card-name",
         ),
+        # Add description below the name
+        html.Div(
+            description,
+            className="galaxy-card-description",
+            style={
+                "fontSize": "0.8rem",
+                "color": "rgba(255,255,255,0.7)", 
+                "padding": "8px 16px 16px",
+                "lineHeight": "1.4",
+                "fontFamily": "'Outfit', sans-serif",
+            }
+        ) if description else None,
     ]
+    
+    # Filter out None elements
+    card_contents = [item for item in card_contents if item is not None]
 
     return html.Button(
         card_contents,

src/galaxy_data_loader.py

@@ -0,0 +1,202 @@
+"""Load galaxy data from HuggingFace datasets and update profiles."""
+
+import os
+import logging
+from typing import Dict, Any
+import requests
+from PIL import Image
+import io
+
+try:
+    from datasets import load_dataset
+    DATASETS_AVAILABLE = True
+except ImportError:
+    DATASETS_AVAILABLE = False
+    logging.warning("datasets library not available, using fallback data")
+
+logger = logging.getLogger(__name__)
+
+def load_galaxy_data() -> Dict[str, Dict[str, Any]]:
+    """Load galaxy data from HuggingFace datasets and return structured profiles."""
+    
+    if not DATASETS_AVAILABLE:
+        logger.warning("Using fallback galaxy data")
+        return get_fallback_galaxy_data()
+    
+    try:
+        # Load both datasets
+        euclid_dataset = load_dataset("mwalmsley/gz_euclid", "tiny", split="test")
+        descriptions_dataset = load_dataset("Smith42/dating_pool_but_galaxies", split="train")
+        
+        # Convert to dictionaries for easier access
+        euclid_data = {row['id_str']: row for row in euclid_dataset}
+        descriptions_data = {row['id_str']: row for row in descriptions_dataset}
+        
+        galaxy_profiles = {}
+        
+        # Match on id_str and create profiles with numbered galaxy IDs
+        matched_galaxies = []
+        for galaxy_id_str in euclid_data.keys():
+            if galaxy_id_str in descriptions_data:
+                matched_galaxies.append(galaxy_id_str)
+        
+        # Sort for consistent ordering
+        matched_galaxies.sort()
+        
+        # Create numbered galaxy profiles (galaxy_01, galaxy_02, etc.)
+        for i, galaxy_id_str in enumerate(matched_galaxies, 1):
+            numbered_id = f"galaxy_{i:02d}"
+            euclid_row = euclid_data[galaxy_id_str]
+            desc_row = descriptions_data[galaxy_id_str]
+            
+            # Extract name and description from caption
+            caption = desc_row.get('caption', '')
+            bio = ""
+            description = ""
+            name = f"Galaxy {i}"
+            
+            # Parse the caption to extract bio and name if available
+            if caption:
+                lines = caption.split('\\n')
+                if lines and lines[0].startswith('Bio:'):
+                    bio_part = lines[0].replace('Bio:', '').strip()
+                    bio = bio_part[:100] + "..." if len(bio_part) > 100 else bio_part
+                    description = bio_part
+                else:
+                    description = caption.strip()
+                    bio = description[:100] + "..." if len(description) > 100 else description
+            
+            galaxy_profiles[numbered_id] = {
+                "name": name,
+                "bio": bio,
+                "description": description,  
+                "tags": ["Cosmic", "Mysterious"],
+                "color": generate_color_from_id(galaxy_id_str),
+                "id_str": galaxy_id_str,
+                "image_data": euclid_row.get('image'),  # PIL Image object
+                "euclid_features": {k: v for k, v in euclid_row.items() if k not in ['image', 'id_str']}
+            }
+        
+        logger.info(f"Loaded {len(galaxy_profiles)} galaxy profiles from HuggingFace datasets")
+        return galaxy_profiles
+        
+    except Exception as e:
+        logger.error(f"Error loading from HuggingFace datasets: {e}")
+        return get_fallback_galaxy_data()
+
+def generate_color_from_id(id_str: str) -> str:
+    """Generate a consistent color from galaxy ID string."""
+    colors = [
+        "#A688C9", "#D68B8B", "#E8A0D0", "#7BC9A0", "#E87D5A",
+        "#6B8DD6", "#D4A76A", "#C9A688", "#B8A0C9", "#A0D4E8",
+        "#E8D4A0", "#A0E8B8", "#E8A0A0", "#A0A0E8", "#E8E8A0",
+        "#D0A0E8", "#A0E8D0", "#E8C4A0", "#C4A0E8", "#A0C4E8",
+        "#E8A0C4", "#B4E8A0", "#A0E8E8"
+    ]
+    # Use hash to get consistent color index
+    hash_val = sum(ord(c) for c in id_str)
+    return colors[hash_val % len(colors)]
+
+def get_fallback_galaxy_data() -> Dict[str, Dict[str, Any]]:
+    """Fallback galaxy data using original profiles with numbered IDs."""
+    fallback_profiles = {
+        "galaxy_01": {
+            "name": "Velvet Vortex",
+            "bio": "Smooth operator with a soft spiral glow.",
+            "description": "I keep my arms tight and my luminosity low-key. Looking for someone who appreciates subtlety over spectacle.",
+            "tags": ["Smooth Spiral", "Low-Key", "Gentle Glow"],
+            "color": "#A688C9",
+            "image_url": None,
+            "image_data": None
+        },
+        "galaxy_02": {
+            "name": "Crimson Drift", 
+            "bio": "Redshifted and proud.",
+            "description": "I have been expanding away from everyone for billions of years and I am NOT slowing down. Commitment-phobic? Maybe. Mysterious? Definitely.",
+            "tags": ["High Redshift", "Distant", "Loner Vibes"],
+            "color": "#D68B8B",
+            "image_url": None,
+            "image_data": None
+        }
+        # Add more as needed...
+    }
+    
+    # Generate numbered galaxy profiles if we have fewer than 23
+    for i in range(1, 24):
+        galaxy_id = f"galaxy_{i:02d}"
+        if galaxy_id not in fallback_profiles:
+            fallback_profiles[galaxy_id] = {
+                "name": f"Galaxy {i}",
+                "bio": "A mysterious galaxy in the cosmic dating scene.",
+                "description": "This galaxy is looking for its perfect cosmic match.",
+                "tags": ["Mysterious", "Cosmic"],
+                "color": "#A688C9",
+                "image_url": None,
+                "image_data": None
+            }
+    
+    return fallback_profiles
+
+def download_galaxy_image(image_data, galaxy_id: str, images_dir: str = "images") -> bool:
+    """Download and save a galaxy image from HuggingFace dataset."""
+    if not image_data:
+        return False
+    
+    try:
+        # Ensure images directory exists
+        os.makedirs(images_dir, exist_ok=True)
+        
+        # If image_data is a PIL Image (from datasets)
+        if hasattr(image_data, 'save') and hasattr(image_data, 'mode'):
+            image_path = os.path.join(images_dir, f"{galaxy_id}.jpg")
+            # Convert to RGB if needed (in case it's RGBA or other format)
+            if image_data.mode != 'RGB':
+                image_data = image_data.convert('RGB')
+            image_data.save(image_path, 'JPEG', quality=85)
+            logger.info(f"Saved image for {galaxy_id}")
+            return True
+            
+        # If image_data has a URL (fallback)
+        elif isinstance(image_data, dict) and 'url' in image_data:
+            response = requests.get(image_data['url'])
+            if response.status_code == 200:
+                image = Image.open(io.BytesIO(response.content))
+                if image.mode != 'RGB':
+                    image = image.convert('RGB')
+                image_path = os.path.join(images_dir, f"{galaxy_id}.jpg")
+                image.save(image_path, 'JPEG', quality=85)
+                logger.info(f"Downloaded and saved image for {galaxy_id}")
+                return True
+                
+        return False
+        
+    except Exception as e:
+        logger.error(f"Error downloading image for {galaxy_id}: {e}")
+        return False
+
+def update_galaxy_images():
+    """Download all galaxy images from the dataset."""
+    galaxy_data = load_galaxy_data()
+    
+    for galaxy_id, profile in galaxy_data.items():
+        if profile.get('image_data'):
+            download_galaxy_image(profile['image_data'], galaxy_id)
+        elif profile.get('image_url'):
+            # Try to download from URL
+            try:
+                response = requests.get(profile['image_url'])
+                if response.status_code == 200:
+                    image = Image.open(io.BytesIO(response.content))
+                    download_galaxy_image(image, galaxy_id)
+            except Exception as e:
+                logger.error(f"Error downloading from URL for {galaxy_id}: {e}")
+
+if __name__ == "__main__":
+    # Test the data loading
+    data = load_galaxy_data()
+    print(f"Loaded {len(data)} galaxies")
+    for galaxy_id, profile in list(data.items())[:3]:
+        print(f"{galaxy_id}: {profile['name']} - {profile['bio'][:50]}...")
+    
+    # Update images
+    update_galaxy_images()

src/galaxy_profiles.py

@@ -1,146 +1,17 @@
-"""Hardcoded galaxy dating profiles for the 23 tournament galaxies."""
+"""Galaxy profiles loaded from HuggingFace datasets."""
 
-GALAXY_PROFILES = {
-    "galaxy_01": {
-        "name": "Velvet Vortex",
-        "bio": "Smooth operator with a soft spiral glow. I keep my arms tight and my luminosity low-key. Looking for someone who appreciates subtlety over spectacle.",
-        "tags": ["Smooth Spiral", "Low-Key", "Gentle Glow"],
-        "color": "#A688C9",
-    },
-    "galaxy_02": {
-        "name": "Crimson Drift",
-        "bio": "Redshifted and proud. I have been expanding away from everyone for billions of years and I am NOT slowing down. Commitment-phobic? Maybe. Mysterious? Definitely.",
-        "tags": ["High Redshift", "Distant", "Loner Vibes"],
-        "color": "#D68B8B",
-    },
-    "galaxy_03": {
-        "name": "Stardust Siren",
-        "bio": "Bright core, hypnotic arms, and enough HII regions to light up a whole cluster. Warning: I will outshine your ex.",
-        "tags": ["Star Forming", "Bright Core", "Show-Off"],
-        "color": "#E8A0D0",
-    },
-    "galaxy_04": {
-        "name": "Phantom Edge",
-        "bio": "Edge-on and enigmatic. You can see my dust lane but you will never see my full face. Some call it mysterious, I call it good angles.",
-        "tags": ["Edge-On", "Dust Lane", "Mysterious"],
-        "color": "#7BC9A0",
-    },
-    "galaxy_05": {
-        "name": "Nova Charm",
-        "bio": "Compact, intense, and full of surprises. My surface brightness might be modest but my personality is SUPERNOVA. Size is just a number.",
-        "tags": ["Compact", "Intense", "Underdog"],
-        "color": "#E87D5A",
-    },
-    "galaxy_06": {
-        "name": "Cosmic Flirt",
-        "bio": "Slightly tilted, slightly chaotic, totally charming. I have got asymmetric arms and I am not afraid to use them. Swipe right if you like a little imperfection.",
-        "tags": ["Asymmetric", "Charming", "Tilted"],
-        "color": "#6B8DD6",
-    },
-    "galaxy_07": {
-        "name": "Nebula Crush",
-        "bio": "Face-on perfection. My spiral arms are textbook and my bar is prominent. I know what I bring to the gravitational table.",
-        "tags": ["Face-On", "Barred Spiral", "Confident"],
-        "color": "#D4A76A",
-    },
-    "galaxy_08": {
-        "name": "Twilight Ember",
-        "bio": "Fading gracefully but still got it. My stellar population is old and wise. Looking for someone who values maturity over flashy star formation.",
-        "tags": ["Evolved", "Mature", "Elegant"],
-        "color": "#C9A688",
-    },
-    "galaxy_09": {
-        "name": "Galactic Riot",
-        "bio": "Currently mid-merger and THRIVING. Tidal tails everywhere, stars flying, gas colliding. If you cannot handle me at my most disturbed you do not deserve me at my relaxed.",
-        "tags": ["Merger", "Chaotic", "Tidal Tails"],
-        "color": "#E85A7D",
-    },
-    "galaxy_10": {
-        "name": "Sapphire Haze",
-        "bio": "Diffuse, dreamy, and delightfully low surface brightness. You will need dark skies and patience to appreciate me. Worth it though.",
-        "tags": ["Low Surface Brightness", "Dreamy", "Patient"],
-        "color": "#88A6D6",
-    },
-    "galaxy_11": {
-        "name": "Blaze Runner",
-        "bio": "Starburst energy that will not quit. Forming stars at a reckless pace and producing galactic superwinds. Not here for a long time, here for a BRIGHT time.",
-        "tags": ["Starburst", "Superwind", "Intense"],
-        "color": "#E8A05A",
-    },
-    "galaxy_12": {
-        "name": "Silk Spiral",
-        "bio": "Grand design spiral with arms so well-defined they could be on a textbook cover. Photogenic from every angle. Yes, I am always this pretty.",
-        "tags": ["Grand Design", "Photogenic", "Classic"],
-        "color": "#A78BFA",
-    },
-    "galaxy_13": {
-        "name": "Orbit Tease",
-        "bio": "I have got a companion galaxy and we are in a complicated situationship. Tidal interactions keep things spicy. Open to new gravitational bonds.",
-        "tags": ["Interacting", "Situationship", "Spicy"],
-        "color": "#F472B6",
-    },
-    "galaxy_14": {
-        "name": "Celestial Sage",
-        "bio": "Elliptical and proud. No flashy spiral arms, just a smooth, uniform glow built over billions of years of mergers. I contain multitudes (of old stars).",
-        "tags": ["Elliptical", "Old Soul", "Smooth"],
-        "color": "#C9B688",
-    },
-    "galaxy_15": {
-        "name": "Pixel Punk",
-        "bio": "Small, faint, but with ATTITUDE. I may be barely resolved but my irregular morphology says I play by my own gravitational rules.",
-        "tags": ["Irregular", "Rebellious", "Compact"],
-        "color": "#5AE8C8",
-    },
-    "galaxy_16": {
-        "name": "Halo Queen",
-        "bio": "Extended halo, dominant bulge, and a dark matter envelope that goes on for megaparsecs. I take up space and I am not apologizing for it.",
-        "tags": ["Massive Halo", "Dominant", "Expansive"],
-        "color": "#D6A66B",
-    },
-    "galaxy_17": {
-        "name": "Aurora Wisp",
-        "bio": "Delicate and ethereal. My faint tidal features hint at a turbulent past but I have learned to glow softly. Healing era galaxy.",
-        "tags": ["Tidal Features", "Ethereal", "Healing Era"],
-        "color": "#88D6C9",
-    },
-    "galaxy_18": {
-        "name": "Midnight Mirage",
-        "bio": "Is that a ring? A shell? A tidal stream? Nobody can quite figure me out and I prefer it that way. Morphologically ambiguous and thriving.",
-        "tags": ["Peculiar", "Ambiguous", "Enigmatic"],
-        "color": "#8B6BD6",
-    },
-    "galaxy_19": {
-        "name": "Solar Flare",
-        "bio": "Blazing nucleus with jets that could cross a cluster. My AGN is active and so is my social life. Warning: I radiate in ALL wavelengths.",
-        "tags": ["Active Nucleus", "Jets", "Radiant"],
-        "color": "#E8D45A",
-    },
-    "galaxy_20": {
-        "name": "Glimmer Ghost",
-        "bio": "Ultra-diffuse and loving it. You could look right through me and miss the magic. Only the most dedicated observers get to know the real me.",
-        "tags": ["Ultra-Diffuse", "Elusive", "Hidden Gem"],
-        "color": "#A0C9E8",
-    },
-    "galaxy_21": {
-        "name": "Quasar Babe",
-        "bio": "Bright, bold, and unapologetically luminous. My central engine outshines entire galaxies. If you need someone low-maintenance, keep swiping.",
-        "tags": ["Luminous", "Bold", "High Maintenance"],
-        "color": "#D65A8B",
-    },
-    "galaxy_22": {
-        "name": "Lenticular Lux",
-        "bio": "Not quite spiral, not quite elliptical. I exist in the elegant in-between. My dust disk is subtle and my bulge is refined. Classy, never trashy.",
-        "tags": ["Lenticular", "Refined", "Classy"],
-        "color": "#C9A0E8",
-    },
-    "galaxy_23": {
-        "name": "Cluster Crush",
-        "bio": "Living my best life in a dense galaxy cluster. Surrounded by neighbors, feeding off the intracluster medium, and thriving in the chaos. Extrovert energy.",
-        "tags": ["Cluster Member", "Social", "Extrovert"],
-        "color": "#5AE87D",
-    },
-}
+import logging
+from typing import Dict, List
+from .galaxy_data_loader import load_galaxy_data
 
+logger = logging.getLogger(__name__)
+
+# Load galaxy profiles from HuggingFace datasets
+_galaxy_data = load_galaxy_data()
+GALAXY_PROFILES = _galaxy_data
 GALAXY_IDS = list(GALAXY_PROFILES.keys())
 NUM_GALAXIES = len(GALAXY_IDS)
 TOTAL_PAIRS = NUM_GALAXIES * (NUM_GALAXIES - 1) // 2
+
+logger.info(f"Loaded {NUM_GALAXIES} galaxies from datasets")
+logger.info(f"Galaxy IDs: {GALAXY_IDS[:5]}{'...' if len(GALAXY_IDS) > 5 else ''}")