Issue 14 check local branch

.gitignore

@@ -2,3 +2,4 @@
 /dist/
 /truffleHog.egg-info/
 */__pycache__/
+files.txt

README.md

@@ -13,11 +13,44 @@ truffleHog file:///user/dxa4481/codeprojects/truffleHog/
 
 ![Example](https://i.imgur.com/YAXndLD.png)
 
+## Usage Details
+```
+usage: truffleHog [-h] [--json] git_repo
+
+Find secrets hidden in the depths of git.
+
+positional arguments:
+  git_repo    Git Repository: Remote URL or local filesystem for secret searching
+
+optional arguments:
+  -h, --help  show this help message and exit
+  --json      Output in JSON
+```
+
 ## Install
+Automatically install via the [Python Package Indexer, PIP](https://pypi.python.org/pypi/pip)
 ```
 pip install truffleHog
 ```
 
+or manually install by via clone/download the repository and install with setup.py
+
+```
+git clone https://github.com/dxa4481/truffleHog.git
+cd truffleHog.git
+python setup.py install --record files.txt
+```
+
+## Uninstall
+```
+pip uninstall truffleHog
+```
+
+or manually uninstall if installed locally via repo
+```
+cat files.txt | xargs rm -rf
+```
+
 ## How it works
 This module will go through the entire commit history of each branch, and check each diff from each commit, and evaluate the shannon entropy for both the base64 char set and hexidecimal char set for every blob of text greater than 20 characters comprised of those character sets in each diff. If at any point a high entropy string >20 characters is detected, it will print to the screen.
 

truffleHog/truffleHog.py

@@ -8,22 +8,68 @@
 import argparse
 import tempfile
 import os
+import subprocess
 import json
 import stat
+import urllib
 from git import Repo
 
+# Global Vars
+BASE64_CHARS = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/="
+HEX_CHARS = "1234567890abcdefABCDEF"
+
 def main():
     parser = argparse.ArgumentParser(description='Find secrets hidden in the depths of git.')
     parser.add_argument('--json', dest="output_json", action="store_true", help="Output in JSON")
-    parser.add_argument('git_url', type=str, help='URL for secret searching')
+    parser.add_argument('git_repo', type=str, help='Git Repository: Remote URL or local filesystem for secret searching')
     args = parser.parse_args()
-    output = find_strings(args.git_url, args.output_json)
+    output = find_strings(args.git_repo, args.output_json)
     project_path = output["project_path"]
     shutil.rmtree(project_path, onerror=del_rw)
 
+def is_abs_url(url=None):
+    """Checks if string is absolute URL or not and returns boolean denoting True or False if valid URL.  
 
-BASE64_CHARS = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/="
-HEX_CHARS = "1234567890abcdefABCDEF"
+    :param url: A string, absolute URL.
+    """
+    if url is None or url.strip() == '':
+        return False
+    if not url.strip().lower()[:4] == 'http': 
+        return False
+    try:
+        rc = urllib.urlopen(url).getcode()
+        if rc == 200:
+            return True
+        else:
+            sys.stderr.write("Unable to reach remote Git repo over HTTP due to response code : " + str(rc))
+            return False
+    except Exception as e:
+        sys.stderr.write("Invalid URL " + e.message)
+        return False
+
+def is_git_dir(filepath=None):
+    """Checks if filepath is directory and if directory itself is a git repository.
+    Returns boolean denoting True or False if valid directory and git repository.  
+
+    :param filepath: A string, filepath of the directory.
+    """
+    if filepath is None or filepath.strip() == '':
+        return False
+
+    if filepath[-4:] != '.git':
+        filepath += '/.git'
+
+    # verify actual git directory
+    if not os.path.isdir(filepath) and os.path.exists(filepath):
+        try:
+            rc = subprocess.call(['git', 'rev-parse'], cwd=filepath)
+            return (rc == 0)
+        except OSError as e:
+            sys.stderr.write("Unable to invoke cmd 'git rev-parse' due to following: " + e.message)
+            return False  
+    else:
+        sys.stderr.write("Invalid filepath - directory does not exist " + filepath)
+        return False
 
 def del_rw(action, name, exc):
     os.chmod(name, stat.S_IWRITE)
@@ -42,7 +88,6 @@ def shannon_entropy(data, iterator):
             entropy += - p_x*math.log(p_x, 2)
     return entropy
 
-
 def get_strings_of_set(word, char_set, threshold=20):
     count = 0
     letters = ""
@@ -70,9 +115,13 @@ class bcolors:
     BOLD = '\033[1m'
     UNDERLINE = '\033[4m'
 
-def find_strings(git_url, printJson=False):
-    project_path = tempfile.mkdtemp()
-    Repo.clone_from(git_url, project_path)
+def find_strings(git_repo, printJson=False):
+    if is_abs_url(git_repo) or is_git_dir(git_repo):
+        project_path = tempfile.mkdtemp()
+        Repo.clone_from(git_repo, project_path)
+    else:
+        sys.exit("\nUnable to scan for secrets due to invalid Git repository " + git_repo)
+
     output = {"entropicDiffs": []}
     repo = Repo(project_path)
     already_searched = set()