fix video downloading in check_install.

Added a more robust downloading function in `utils.py` that handles errors more gracefully.
The download is skipped if the file already exists.

In previous implementation, the benchmarking video was downloaded to the home directory,
which in windows could result in permission issues when reading the file. This is now changed to
the check_install directory.

OpenCV was silently failing to read the video file, resulting in a frame count of zero.
Now a ValueError is raised when the video cannot be read.

Author: deruyter92

dlclive/check_install/check_install.py

@@ -9,28 +9,22 @@
 import urllib.request
 import argparse
 import shutil
-import sys
+
 import urllib.request
 import warnings
 from pathlib import Path
 
+from dlclive.utils import download_file
 from dlclibrary.dlcmodelzoo.modelzoo_download import download_huggingface_model
 
+import dlclive
 from dlclive.benchmark import benchmark_videos
 from dlclive.engine import Engine
 
 MODEL_NAME = "superanimal_quadruped"
 SNAPSHOT_NAME = "snapshot-700000.pb"
 
 
-def urllib_pbar(count, blockSize, totalSize):
-    percent = int(count * blockSize * 100 / totalSize)
-    outstr = f"{round(percent)}%"
-    sys.stdout.write(outstr)
-    sys.stdout.write("\b" * len(outstr))
-    sys.stdout.flush()
-
-
 def main():
     parser = argparse.ArgumentParser(
         description="Test DLC-Live installation by downloading and evaluating a demo DLC project!"
@@ -46,22 +40,25 @@ def main():
     if not display:
         print("Running without displaying video")
 
-    # make temporary directory in $HOME
-    # TODO: why create this temp directory in $HOME?
+    # make temporary directory
     print("\nCreating temporary directory...\n")
-    tmp_dir = Path().home() / "dlc-live-tmp"
+    tmp_dir = Path(dlclive.__file__).parent / "check_install" / "dlc-live-tmp"
     tmp_dir.mkdir(mode=0o775, exist_ok=True)
 
     video_file = str(tmp_dir / "dog_clip.avi")
     model_dir = tmp_dir / "DLC_Dog_resnet_50_iteration-0_shuffle-0"
 
     # download dog test video from github:
-    if not os.path.exists(video_file):
+    # Use raw.githubusercontent.com for direct file access
+    if not Path(video_file).exists():
         print(f"Downloading Video to {video_file}")
-        url_link = "https://github.com/DeepLabCut/DeepLabCut-live/blob/main/check_install/dog_clip.avi?raw=True"
-        urllib.request.urlretrieve(url_link, video_file, reporthook=urllib_pbar)
+        url_link = "https://raw.githubusercontent.com/DeepLabCut/DeepLabCut-live/master/check_install/dog_clip.avi"
+        try:
+            download_file(url_link, video_file)
+        except (urllib.error.URLError, IOError) as e:
+            raise RuntimeError(f"Failed to download video file: {e}") from e
     else:
-        print(f"Video already exists at {video_file}")
+        print(f"Video file already exists at {video_file}, skipping download.")
 
     # download model from the DeepLabCut Model Zoo
     if Path(model_dir / SNAPSHOT_NAME).exists():

dlclive/utils.py

@@ -6,8 +6,10 @@
 """
 
 import warnings
-
+from pathlib import Path
 import numpy as np
+import urllib.request
+import urllib.error
 
 from dlclive.exceptions import DLCLiveWarning
 
@@ -31,6 +33,13 @@
         DLCLiveWarning,
     )
 
+try:
+    from tqdm import tqdm
+
+    has_tqdm = True
+except ImportError:
+    has_tqdm = False
+
 
 def convert_to_ubyte(frame: np.ndarray) -> np.ndarray:
     """Converts an image to unsigned 8-bit integer numpy array.
@@ -203,3 +212,68 @@ def decode_fourcc(cc):
         decoded = ""
 
     return decoded
+
+
+def download_file(url: str, filepath: str, chunk_size: int = 8192) -> None:
+    """
+    Download a file from a URL with progress bar and error handling.
+    
+    Args:
+        url: URL to download from
+        filepath: Local path to save the file
+        chunk_size: Size of chunks to read (default: 8192 bytes)
+    
+    Raises:
+        urllib.error.URLError: If the download fails
+        IOError: If the file cannot be written
+    """
+    filepath = Path(filepath)
+    
+    # Check if file already exists
+    if filepath.exists():
+        print(f"File already exists at {filepath}, skipping download.")
+        return
+    
+    # Ensure parent directory exists
+    filepath.parent.mkdir(parents=True, exist_ok=True)
+    
+    try:
+        # Open the URL
+        with urllib.request.urlopen(url) as response:
+            # Get file size if available
+            total_size = int(response.headers.get('Content-Length', 0))
+            
+            # Create progress bar if tqdm is available
+            if has_tqdm and total_size > 0:
+                pbar = tqdm(total=total_size, unit='B', unit_scale=True, desc="Downloading")
+            else:
+                pbar = None
+                print("Downloading...")
+            
+            # Download and write file
+            downloaded = 0
+            with open(filepath, 'wb') as f:
+                while True:
+                    chunk = response.read(chunk_size)
+                    if not chunk:
+                        break
+                    f.write(chunk)
+                    downloaded += len(chunk)
+                    if pbar:
+                        pbar.update(len(chunk))
+            
+            if pbar:
+                pbar.close()
+            
+            # Verify file was written
+            if not filepath.exists() or filepath.stat().st_size == 0:
+                raise IOError(f"Downloaded file is empty or was not written to {filepath}")
+            
+            print(f"Successfully downloaded to {filepath}")
+            
+    except urllib.error.HTTPError as e:
+        raise urllib.error.URLError(f"HTTP error {e.code}: {e.reason} when downloading from {url}")
+    except urllib.error.URLError as e:
+        raise urllib.error.URLError(f"Failed to download from {url}: {e.reason}")
+    except IOError as e:
+        raise IOError(f"Failed to write file to {filepath}: {e}")