BUG: Avoid multiprocessing trouble on Windos (#6)

* BUG: Avoid BrokenProcessPool error on Windos

* REF: Avoid any multiprocessing exceptions in `Backtest.optimize()`

... and instead switch to non-parallel computation.

* REF: Support parallelism only if start method is 'fork'

* REF: In case of non-'fork' multiprocessing, use a simple loop

* MNT: quiet codecov

* TST: Remove redundant test

The code path is covered by any other Backtest.optimize() test.

* fix flake8 warning

.codecov.yml

@@ -1,10 +1,12 @@
+comment: off
 coverage:
   range: 75..95
   precision: 0
   status:
     patch:
       default:
-        target: '95'
+        target: 90
     project:
       default:
         target: auto
+        threshold: 5

backtesting/backtesting.py

@@ -5,6 +5,7 @@ module directly, e.g.
 
     from backtesting import Backtest, Strategy
 """
+import multiprocessing as mp
 import os
 import re
 import sys
@@ -812,11 +813,22 @@ class Backtest:
             for i in range(0, len(seq), n):
                 yield seq[i:i + n]
 
-        with ProcessPoolExecutor() as executor:
-            futures = [executor.submit(self._mp_task, params)
-                       for params in _batch(param_combos)]
-            for future in _tqdm(as_completed(futures), total=len(futures)):
-                for params, stats in future.result():
+        # If multiprocessing start method is 'fork' (i.e. on POSIX), use
+        # a pool of processes to compute results in parallel.
+        # Otherwise (i.e. on Windos), sequential computation will be "faster".
+        if mp.get_start_method(allow_none=False) == 'fork':
+            with ProcessPoolExecutor() as executor:
+                futures = [executor.submit(self._mp_task, params)
+                           for params in _batch(param_combos)]
+                for future in _tqdm(as_completed(futures), total=len(futures)):
+                    for params, stats in future.result():
+                        heatmap[tuple(params.values())] = maximize(stats)
+        else:
+            if os.name == 'posix':
+                warnings.warn("For multiprocessing support in `Backtest.optimize()` "
+                              "set multiprocessing start method to 'fork'.")
+            for params in _tqdm(param_combos):
+                for _, stats in self._mp_task([params]):
                     heatmap[tuple(params.values())] = maximize(stats)
 
         best_params = heatmap.idxmax()