648 max_chunk_size can be now set to control output_stream behavior

bpiotr · Byron · commit 7be3486dc7f9 · 2018-05-18T13:59:50.000+02:00
diff --git a/git/cmd.py b/git/cmd.py
@@ -44,10 +44,10 @@
 )
 
 
-execute_kwargs = {'istream', 'with_extended_output', 'with_exceptions',
-                  'as_process', 'stdout_as_string', 'output_stream',
-                  'with_stdout', 'kill_after_timeout', 'universal_newlines',
-                  'shell', 'env'}
+execute_kwargs = {'istream', 'with_extended_output',
+                  'with_exceptions', 'as_process', 'stdout_as_string',
+                  'output_stream', 'with_stdout', 'kill_after_timeout',
+                  'universal_newlines', 'shell', 'env', 'max_chunk_size'}
 
 log = logging.getLogger(__name__)
 log.addHandler(logging.NullHandler())
@@ -174,8 +174,6 @@ def __setstate__(self, d):
         dict_to_slots_and__excluded_are_none(self, d, excluded=self._excluded_)
 
     # CONFIGURATION
-    # The size in bytes read from stdout when copying git's output to another stream
-    max_chunk_size = io.DEFAULT_BUFFER_SIZE
 
     git_exec_name = "git"           # default that should work on linux and windows
 
@@ -597,6 +595,7 @@ def execute(self, command,
                 universal_newlines=False,
                 shell=None,
                 env=None,
+                max_chunk_size=io.DEFAULT_BUFFER_SIZE,
                 **subprocess_kwargs
                 ):
         """Handles executing the command on the shell and consumes and returns
@@ -642,6 +641,11 @@ def execute(self, command,
 
         :param env:
             A dictionary of environment variables to be passed to `subprocess.Popen`.
+            
+        :param max_chunk_size:
+            Maximum number of bytes in one chunk of data passed to the output_stream in
+            one invocation of write() method. If the given number is not positive then
+            the default value is used.
 
         :param subprocess_kwargs:
             Keyword arguments to be passed to subprocess.Popen. Please note that
@@ -788,7 +792,8 @@ def _kill_process(pid):
                     stderr_value = stderr_value[:-1]
                 status = proc.returncode
             else:
-                stream_copy(proc.stdout, output_stream, self.max_chunk_size)
+                max_chunk_size = max_chunk_size if max_chunk_size and max_chunk_size > 0 else io.DEFAULT_BUFFER_SIZE
+                stream_copy(proc.stdout, output_stream, max_chunk_size)
                 stdout_value = output_stream
                 stderr_value = proc.stderr.read()
                 # strip trailing "\n"
diff --git a/git/test/test_repo.py b/git/test/test_repo.py
@@ -5,6 +5,7 @@
 # This module is part of GitPython and is released under
 # the BSD License: http://www.opensource.org/licenses/bsd-license.php
 import glob
+import io
 from io import BytesIO
 import itertools
 import os
@@ -220,6 +221,22 @@ def test_clone_from_pathlib(self, rw_dir):
 
         Repo.clone_from(original_repo.git_dir, pathlib.Path(rw_dir) / "clone_pathlib")
 
+    @with_rw_repo('HEAD')
+    def test_max_chunk_size(self, repo):
+        class TestOutputStream(object):
+            def __init__(self, max_chunk_size):
+                self.max_chunk_size = max_chunk_size
+
+            def write(self, b):
+                assert_true(len(b) <= self.max_chunk_size)
+
+        for chunk_size in [16, 128, 1024]:
+            repo.git.status(output_stream=TestOutputStream(chunk_size), max_chunk_size=chunk_size)
+
+        repo.git.log(n=100, output_stream=TestOutputStream(io.DEFAULT_BUFFER_SIZE), max_chunk_size=None)
+        repo.git.log(n=100, output_stream=TestOutputStream(io.DEFAULT_BUFFER_SIZE), max_chunk_size=-10)
+        repo.git.log(n=100, output_stream=TestOutputStream(io.DEFAULT_BUFFER_SIZE))
+
     def test_init(self):
         prev_cwd = os.getcwd()
         os.chdir(tempfile.gettempdir())
