Fixed since_commit and tests (#219)

* Fixed since_commit and tests

* Updated testing comment

Co-authored-by: Dylan Ayrey <dxa4481@rit.edu>

.gitignore

@@ -3,3 +3,4 @@
 /truffleHog.egg-info/
 **/__pycache__/
 **/*.pyc
+env/

test_all.py

@@ -30,12 +30,12 @@ def test_unicode_expection(self):
             self.fail("Unicode print error")
 
     def test_return_correct_commit_hash(self):
-        # Start at commit d15627104d07846ac2914a976e8e347a663bbd9b, which 
+        # Start at commit 202564cf776b402800a4aab8bb14fa4624888475, which 
         # is immediately followed by a secret inserting commit:
-        # https://github.com/dxa4481/truffleHog/commit/9ed54617547cfca783e0f81f8dc5c927e3d1e345
-        since_commit = 'd15627104d07846ac2914a976e8e347a663bbd9b'
-        commit_w_secret = '9ed54617547cfca783e0f81f8dc5c927e3d1e345'
-        cross_valdiating_commit_w_secret_comment = 'OH no a secret'
+        # https://github.com/dxa4481/truffleHog/commit/d15627104d07846ac2914a976e8e347a663bbd9b
+        since_commit = '202564cf776b402800a4aab8bb14fa4624888475'
+        commit_w_secret = 'd15627104d07846ac2914a976e8e347a663bbd9b'
+        cross_valdiating_commit_w_secret_comment = 'Oh no a secret file'
 
         json_result = ''
         if sys.version_info >= (3,):
@@ -54,7 +54,8 @@ def test_return_correct_commit_hash(self):
 
         json_result_list = tmp_stdout.getvalue().split('\n')
         results = [json.loads(r) for r in json_result_list if bool(r.strip())]
-        filtered_results = list(filter(lambda r: r['commitHash'] == commit_w_secret, results))
+        filtered_results = list(filter(lambda r: r['commitHash'] == commit_w_secret and r['branch'] == 'origin/master', results))
+
         self.assertEqual(1, len(filtered_results))
         self.assertEqual(commit_w_secret, filtered_results[0]['commitHash'])
         # Additionally, we cross-validate the commit comment matches the expected comment

truffleHog/truffleHog.py

@@ -344,9 +344,7 @@ def find_strings(git_url, since_commit=None, max_depth=1000000, printJson=False,
             commitHash = curr_commit.hexsha
             if commitHash == since_commit:
                 since_commit_reached = True
-            if since_commit and since_commit_reached:
-                prev_commit = curr_commit
-                continue
+                break
             # if not prev_commit, then curr_commit is the newest commit. And we have nothing to diff with.
             # But we will diff the first commit with NULL_TREE here to check the oldest code.
             # In this way, no commit will be missed.
@@ -364,8 +362,16 @@ def find_strings(git_url, since_commit=None, max_depth=1000000, printJson=False,
             foundIssues = diff_worker(diff, curr_commit, prev_commit, branch_name, commitHash, custom_regexes, do_entropy, do_regex, printJson, surpress_output, path_inclusions, path_exclusions, allow)
             output = handle_results(output, output_dir, foundIssues)
             prev_commit = curr_commit
-        # Handling the first commit
-        diff = curr_commit.diff(NULL_TREE, create_patch=True)
+
+        # Check if since_commit was used to check which diff should be grabbed
+        if since_commit_reached:
+            # Handle when there's no prev_commit (used since_commit on the most recent commit)
+            if prev_commit is None:
+                continue
+            diff = prev_commit.diff(curr_commit, create_patch=True)
+        else:
+            diff = curr_commit.diff(NULL_TREE, create_patch=True)
+
         foundIssues = diff_worker(diff, curr_commit, prev_commit, branch_name, commitHash, custom_regexes, do_entropy, do_regex, printJson, surpress_output, path_inclusions, path_exclusions, allow)
         output = handle_results(output, output_dir, foundIssues)
     output["project_path"] = project_path