Fix verbosity gating across wids modules

Introduce WIDS_VERBOSE flag driven by env var

Gate ShardListDataset summary prints behind WIDS_VERBOSE

Replace print-based warnings with warnings.warn

Respect WIDS_VERBOSE in TarFileReader verbose default

Author: tmbnv

src/wids/wids.py

@@ -14,6 +14,8 @@
 from urllib.parse import quote, urlparse
 
 import numpy as np
+# Verbosity flag for wids; set WIDS_VERBOSE=1 to enable verbose output
+WIDS_VERBOSE = bool(int(os.environ.get("WIDS_VERBOSE", 0)))
 import torch.distributed as dist
 
 from .wids_decode import default_decoder
@@ -100,7 +102,8 @@ def group_by_key(names):
     for i, fname in enumerate(names):
         # Ignore files that are not in a subdirectory.
         if "." not in fname:
-            print(f"Warning: Ignoring file {fname} (no '.')")
+            # Warn about files without extensions; can be silenced via warnings filter
+            warnings.warn(f"Ignoring file {fname} (no '.')")
             continue
         key, ext = splitname(fname)
         if key != last_key:
@@ -435,7 +438,8 @@ def __init__(
             self.cache_dir = os.environ.get("WIDS_CACHE", "/tmp/_wids_cache")
             self.localname = DefaultLocalname(self.cache_dir)
 
-        if True or int(os.environ.get("WIDS_VERBOSE", 0)):
+        # Only print dataset summary if verbosity enabled
+        if WIDS_VERBOSE:
             nbytes = sum(shard.get("filesize", 0) for shard in self.shards)
             nsamples = sum(shard["nsamples"] for shard in self.shards)
             print(
@@ -480,7 +484,8 @@ def check_cache_misses(self):
         if accesses > 100 and misses / accesses > 0.3:
             # output a warning only once
             self.check_cache_misses = lambda: None
-            print("Warning: ShardListDataset has a cache miss rate of {:.1%}%".format(misses * 100.0 / accesses))
+            # Warn about high cache miss rate; can be silenced via warnings filter
+            warnings.warn(f"ShardListDataset has a cache miss rate of {misses/accesses:.1%}")
 
     def get_shard(self, index):
         """Get the shard and index within the shard corresponding to the given index."""

src/wids/wids_tar.py

@@ -17,8 +17,12 @@ def find_index_file(file):
 
 
 class TarFileReader:
-    def __init__(self, file, index_file=find_index_file, verbose=True):
-        self.verbose = verbose
+    def __init__(self, file, index_file=find_index_file, verbose=None):
+        # Determine verbosity: use parameter if provided, else env var WIDS_VERBOSE
+        if verbose is None:
+            self.verbose = bool(int(os.environ.get("WIDS_VERBOSE", 0)))
+        else:
+            self.verbose = bool(verbose)
         if callable(index_file):
             index_file = index_file(file)
         self.index_file = index_file