fix: Generalize tar decompression (#88)

Change tarfile open mode from "r:gz" assuming that file is gzip
compressed (which is not true for .tar files) to more general "r:*".

---------

Co-authored-by: Filip Knefel <[email protected]>

Author: ds-filipknefel

CHANGELOG.md

@@ -1,3 +1,12 @@
+## 0.0.10
+
+### Enhancements
+
+* "Fix tar extraction" - tar extraction function assumed archive was gzip compressed which isn't true for supported `.tar` archives. Updated to work for both compressed and uncompressed tar archives.
+
+### Fixes
+
+
 ## 0.0.9
 
 ### Enhancements
@@ -10,7 +19,7 @@
 ### Fixes
 
 **Fix uncompress logic** Use of the uncompress process wasn't being leveraged in the pipeline correctly. Updated to use the new loca download path for where the partitioned looks for the new file.  
->>>>>>> d7a2cab (Add entry to changelog)
+
 
 ## 0.0.8
 

unstructured_ingest/__version__.py

@@ -1 +1 @@
-__version__ = "0.0.9"  # pragma: no cover
+__version__ = "0.0.10"  # pragma: no cover

unstructured_ingest/utils/compression.py

@@ -63,7 +63,8 @@ def uncompress_tar_file(tar_filename: str, path: Optional[str] = None) -> str:
 
     path = path if path else os.path.join(head, f"{tail}-tar-uncompressed")
     logger.info(f"extracting tar {tar_filename} -> {path}")
-    with tarfile.open(tar_filename, "r:gz") as tfile:
+    # NOTE: "r:*" mode opens both compressed (e.g ".tar.gz") and uncompressed ".tar" archives
+    with tarfile.open(tar_filename, "r:*") as tfile:
         # NOTE(robinson: Mitigate against malicious content being extracted from the tar file.
         # This was added in Python 3.12
         # Ref: https://docs.python.org/3/library/tarfile.html#extraction-filters