cloudcat 0.3.6__tar.gz → 0.3.8__tar.gz

{cloudcat-0.3.6 → cloudcat-0.3.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cloudcat
-Version: 0.3.6
+Version: 0.3.8
 Summary: Preview and analyze data files in Google Cloud Storage, AWS S3, and Azure Data Lake Storage Gen2 from your terminal
 Home-page: https://github.com/jonathansudhakar1/cloudcat
 Author: Jonathan Sudhakar

{cloudcat-0.3.6 → cloudcat-0.3.8}/cloudcat/__init__.py

@@ -2,4 +2,4 @@
 #!/usr/bin/env python
 """cloudcat - A CLI utility to read and display files from cloud storage."""
 
-__version__ = "0.3.6"
+__version__ = "0.3.8"

{cloudcat-0.3.6 → cloudcat-0.3.8}/cloudcat/cli.py

@@ -470,7 +470,12 @@ def read_data_streaming(
     rows_to_read = (offset + num_rows) if num_rows > 0 else 0
 
     # For columnar formats without external compression, try native PyArrow filesystem
-    if input_format in ('parquet', 'orc') and compression is None and supports_pyarrow_fs():
+    use_native_fs = input_format in ('parquet', 'orc') and compression is None
+    if use_native_fs and not supports_pyarrow_fs():
+        click.echo(Fore.YELLOW + "Note: pyarrow.fs not available, downloading full file instead of streaming" + Style.RESET_ALL)
+        use_native_fs = False
+
+    if use_native_fs:
         try:
             pyarrow_fs, _ = get_pyarrow_filesystem(
                 service,
@@ -894,13 +899,13 @@ def main(path, output_format, input_format, columns, num_rows, offset, where, sc
 
                 file_list = get_files_for_multiread(service, bucket, object_path, input_format, max_size_mb, quiet=True)
 
-                # For a single file with columnar format, use streaming read for efficiency
-                if len(file_list) == 1 and input_format in ('parquet', 'orc'):
+                # For a single file, use streaming read for efficiency
+                if len(file_list) == 1:
                     single_file_path = file_list[0][0]
                     file_name = single_file_path.split('/')[-1]
                     update_progress(f"Reading {file_name}...")
 
-                    # Use streaming read which supports PyArrow native filesystem
+                    # Use streaming read for all formats
                     df, full_schema, streaming_stats = read_data_streaming(
                         service, bucket, single_file_path, input_format, num_rows, columns, delimiter, offset
                     )
@@ -909,7 +914,7 @@ def main(path, output_format, input_format, columns, num_rows, offset, where, sc
                     click.echo(Fore.BLUE + f"Inferred input format: {input_format}" + Style.RESET_ALL)
                     multi_file_list = file_list
                     total_record_count = None
-                else:
+                elif len(file_list) > 1:
                     # Read data from multiple files with progress updates
                     update_progress(f"Reading {len(file_list)} files...")
                     df, full_schema, rows_in_files = read_data_from_multiple_files(

{cloudcat-0.3.6 → cloudcat-0.3.8}/cloudcat/compression.py

@@ -46,8 +46,9 @@ def detect_compression(path: str) -> Optional[str]:
     path_lower = path.lower()
 
     # For Parquet and ORC, snappy is an internal codec - PyArrow handles it
-    # Files like .parquet.snappy or .orc.snappy don't need external decompression
-    if path_lower.endswith('.parquet.snappy') or path_lower.endswith('.orc.snappy'):
+    # Files like .snappy.parquet (Spark convention) or .parquet.snappy don't need external decompression
+    if (path_lower.endswith('.snappy.parquet') or path_lower.endswith('.parquet.snappy') or
+        path_lower.endswith('.snappy.orc') or path_lower.endswith('.orc.snappy')):
         return None
 
     if path_lower.endswith('.gz') or path_lower.endswith('.gzip'):
@@ -122,8 +123,11 @@ def strip_compression_extension(path: str) -> str:
     """
     path_lower = path.lower()
 
-    # For .parquet.snappy and .orc.snappy, strip .snappy to get the format
-    # (snappy is internal codec, not external compression)
+    # For Parquet/ORC with snappy naming conventions, snappy is internal codec
+    # .snappy.parquet (Spark) - already ends in .parquet, return as-is
+    # .parquet.snappy - strip .snappy to get .parquet
+    if path_lower.endswith('.snappy.parquet') or path_lower.endswith('.snappy.orc'):
+        return path  # Already ends with format extension
     if path_lower.endswith('.parquet.snappy'):
         return path[:-len('.snappy')]
     if path_lower.endswith('.orc.snappy'):

{cloudcat-0.3.6 → cloudcat-0.3.8}/cloudcat.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cloudcat
-Version: 0.3.6
+Version: 0.3.8
 Summary: Preview and analyze data files in Google Cloud Storage, AWS S3, and Azure Data Lake Storage Gen2 from your terminal
 Home-page: https://github.com/jonathansudhakar1/cloudcat
 Author: Jonathan Sudhakar