Merge pull request #39 from pycompression/clibuffer

Bigger buffer for the Command Line interface.

Author: rhpvorderman

CHANGELOG.rst

@@ -9,6 +9,9 @@ Changelog
 
 version 0.5.0-dev
 -----------------
++ The command-line interface now reads in blocks of 32K instead of 8K. This
+  improves performance by about 6% when compressing and 11% when decompressing.
+  A hidden ``-b`` flag was added to adjust the buffer size for benchmarks.
 + A ``-c`` or ``--stdout`` flag was added to the CLI interface of isal.igzip.
   This allows it to behave more like the ``gzip`` or ``pigz`` command line
   interfaces.

src/isal/igzip.py

@@ -27,8 +27,6 @@
 import os
 import sys
 
-import _compression
-
 from . import isal_zlib
 
 __all__ = ["IGzipFile", "open", "compress", "decompress", "BadGzipFile"]
@@ -37,8 +35,6 @@
 _COMPRESS_LEVEL_TRADEOFF = isal_zlib.ISAL_DEFAULT_COMPRESSION
 _COMPRESS_LEVEL_BEST = isal_zlib.ISAL_BEST_COMPRESSION
 
-BUFFER_SIZE = _compression.BUFFER_SIZE
-
 try:
     BadGzipFile = gzip.BadGzipFile
 except AttributeError:  # Versions lower than 3.8 do not have BadGzipFile
@@ -278,6 +274,12 @@ def main():
         help="Decompress the file instead of compressing.")
     parser.add_argument("-c", "--stdout", action="store_true",
                         help="write on standard output")
+    # -b flag not taken by either gzip or igzip. Hidden attribute. Above 32K
+    # diminishing returns hit. _compression.BUFFER_SIZE = 8k. But 32K is about
+    # ~6% faster.
+    parser.add_argument("-b", "--buffer-size",
+                        default=32 * 1024, type=int,
+                        help=argparse.SUPPRESS)
     args = parser.parse_args()
 
     compresslevel = args.compresslevel or _COMPRESS_LEVEL_TRADEOFF
@@ -310,7 +312,7 @@ def main():
 
     try:
         while True:
-            block = in_file.read(BUFFER_SIZE)
+            block = in_file.read(args.buffer_size)
             if block == b"":
                 break
             out_file.write(block)