Make shorter Environment hashes

vinisalazar · Apr 24, 2021 · 7961dcb · 7961dcb
1 parent 1a46320
commit 7961dcb
Show file tree

Hide file tree

Showing 4 changed files with 11 additions and 6 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -16,6 +16,7 @@
 * Improve reserved aminoacid characters [x]
 * Add `SeqFile.max_seq` and `.min_seq` properties [x]
 * Patch `Project` deserializer to improve BioProvDocument creation [x]
+* Make shorter Environment hashes [x]
 
 ### v0.1.22
 * Simplify `bp.load_project()` function [x]

diff --git a/bioprov/src/config.py b/bioprov/src/config.py
@@ -270,14 +270,15 @@ def __init__(self):
         Class constructor. All attributes are empty and are initialized with self.update()
         """
         self.env_hash = None
+        self.env_hash_long = None
         self.env_dict = None
         self.user = None
         self.env_namespace = None
         self.update()
         self._actedOnBehalfOf = False
 
     def __repr__(self):
-        return self.env_hash
+        return f"BioProvEnvironment{self.env_hash}"
 
     @property
     def actedOnBehalfOf(self):
@@ -294,9 +295,10 @@ def update(self):
         """
         env_dict = dict(os.environ.items())
         env_hash = dict_to_sha256(env_dict)
-        if env_hash != self.env_hash:
+        if env_hash != self.env_hash_long:
             self.env_dict = env_dict
-            self.env_hash = env_hash
+            self.env_hash = env_hash[:7]
+            self.env_hash_long = env_hash
 
             # this is only to prevent build errors
             try:

diff --git a/bioprov/src/prov.py b/bioprov/src/prov.py
@@ -287,14 +287,14 @@ def _create_program_entities(self, sample, kind="Sample"):
                         ):
                             if self.add_attributes:
                                 self._agents[_env_hash] = _user_bundle.agent(
-                                    f"envs:{_env}",
+                                    f"envs:{_env_hash}",
                                     other_attributes=build_prov_attributes(
                                         _env.env_dict, _env.env_namespace
                                     ),
                                 )
                             else:
                                 self._agents[_env_hash] = _user_bundle.agent(
-                                    f"envs:{_env}"
+                                    f"envs:{_env_hash}"
                                 )
                             if not _env.actedOnBehalfOf:
                                 _user_bundle.actedOnBehalfOf(

diff --git a/bioprov/tests/test_src_prov.py b/bioprov/tests/test_src_prov.py
@@ -31,9 +31,11 @@ def test_EnvProv():
     :return:
     """
     env = Environment()
+    sh = dict_to_sha256(env.env_dict)
     for statement in (
         env.env_dict == dict(environ.items()),
-        env.env_hash == dict_to_sha256(env.env_dict),
+        env.env_hash_long == sh,
+        env.env_hash == sh[:7],
     ):
         assert statement