From fd990ce8bbf5794abd1aeb6a4f9a021235f1df00 Mon Sep 17 00:00:00 2001
From: Francesco Di Natale <frank.dinatale1988@gmail.com>
Date: Sat, 29 Sep 2018 21:03:53 -0700
Subject: [PATCH] Addition of user enabled workspace hashing (#145)

* Addition of hashing to Study parameterization.

* Addition of the hashws option to argparse.

* Addition of a warning note for users who use labels in steps.
---
 maestrowf/datastructures/core/study.py | 17 ++++++++++++-----
 maestrowf/maestro.py                   |  6 +++++-
 2 files changed, 17 insertions(+), 6 deletions(-)

diff --git a/maestrowf/datastructures/core/study.py b/maestrowf/datastructures/core/study.py
index e70d6e58a..b57bb1300 100644
--- a/maestrowf/datastructures/core/study.py
+++ b/maestrowf/datastructures/core/study.py
@@ -29,6 +29,7 @@
 
 """Class related to the construction of study campaigns."""
 import copy
+from hashlib import md5
 import logging
 import os
 import pickle
@@ -355,7 +356,7 @@ def setup_environment(self):
             self.environment.acquire_environment()
 
     def configure_study(self, submission_attempts=1, restart_limit=1,
-                        throttle=0, use_tmp=False):
+                        throttle=0, use_tmp=False, hash_ws=False):
         """
         Perform initial configuration of a study.
 
@@ -376,6 +377,7 @@ def configure_study(self, submission_attempts=1, restart_limit=1,
         self._restart_limit = restart_limit
         self._submission_throttle = throttle
         self._use_tmp = use_tmp
+        self._hash_ws = hash_ws
 
         logger.info(
             "\n------------------------------------------\n"
@@ -384,9 +386,10 @@ def configure_study(self, submission_attempts=1, restart_limit=1,
             "Submission restart limit =  %d\n"
             "Submission throttle limit = %d\n"
             "Use temporary directory =   %s\n"
+            "Hash workspaces =           %s\n"
             "------------------------------------------",
             self._out_path, submission_attempts, restart_limit, throttle,
-            use_tmp
+            use_tmp, hash_ws
         )
 
     def _stage_parameterized(self, dag):
@@ -579,9 +582,13 @@ def _stage_parameterized(self, dag):
                                 str(combo))
                     # Compute this step's combination name and workspace.
                     combo_str = combo.get_param_string(self.used_params[step])
-                    workspace = \
-                        make_safe_path(self._out_path, step, combo_str)
-                    logger.debug("Workspace: %s", workspace)
+                    if self._hash_ws:
+                        workspace = make_safe_path(
+                            self._out_path, step, md5(combo_str).hexdigest())
+                    else:
+                        workspace = \
+                            make_safe_path(self._out_path, step, combo_str)
+                        logger.debug("Workspace: %s", workspace)
                     combo_str = "{}_{}".format(step, combo_str)
                     self.workspaces[combo_str] = workspace
 
diff --git a/maestrowf/maestro.py b/maestrowf/maestro.py
index 4a4d3c13c..318ec9bdc 100644
--- a/maestrowf/maestro.py
+++ b/maestrowf/maestro.py
@@ -216,7 +216,7 @@ def run_study(args):
     study.setup_environment()
     study.configure_study(
         throttle=args.throttle, submission_attempts=args.attempts,
-        restart_limit=args.rlimit, use_tmp=args.usetmp)
+        restart_limit=args.rlimit, use_tmp=args.usetmp, hash_ws=args.hashws)
 
     # Stage the study.
     path, exec_dag = study.stage()
@@ -319,6 +319,10 @@ def setup_argparser():
     run.add_argument("-fg", action="store_true", default=False,
                      help="Runs the backend conductor in the foreground "
                      "instead of using nohup. [Default: %(default)s]")
+    run.add_argument("--hashws", action="store_true", default=False,
+                     help="Enable hashing of subdirectories in parameterized "
+                     "studies (NOTE: breaks commands that use parameter labels"
+                     " to search directories). [Default: %(default)s]")
 
     prompt_opts = run.add_mutually_exclusive_group()
     prompt_opts.add_argument(