remote, #525: pump fetch-infos instead of GIL-read stderr

+ `handle_process_output()` accepts null-finalizer, to pump completely
stderr before raising any errors.
+ test: Enable `TestGit.test_environment()` on Windows (to checks stderr
consumption).

1 files changed, 6 insertions, 2 deletions

diff --git a/git/cmd.py b/git/cmd.py
index aa711168..ebf2bd75 100644
--- a/git/cmd.py
+++ b/git/cmd.py
@@ -59,7 +59,8 @@ __all__ = ('Git',)
 # Documentation
 ## @{
 
-def handle_process_output(process, stdout_handler, stderr_handler, finalizer, decode_streams=True):
+def handle_process_output(process, stdout_handler, stderr_handler,
+                          finalizer=None, decode_streams=True):
     """Registers for notifications to lean that process output is ready to read, and dispatches lines to
     the respective line handlers.
     This function returns once the finalizer returns
@@ -108,10 +109,13 @@ def handle_process_output(process, stdout_handler, stderr_handler, finalizer, de
         t.start()
         threads.append(t)
 
+    ## FIXME: Why Join??  Will block if `stdin` needs feeding...
+    #
     for t in threads:
         t.join()
 
-    return finalizer(process)
+    if finalizer:
+        return finalizer(process)
 
 
 def dashify(string):