updated Dockerfile. Replaced is_valid_dir to is_valid_dir_or_url in cli

Dockerfile

@@ -1,12 +1,23 @@
-FROM python:3.8-buster
+FROM ubuntu:20.04
 
 MAINTAINER Stefano Dalla Palma
 
-RUN python3.8 -m pip install --upgrade pip
+# Install python
+RUN apt-get update \
+  && apt-get install -y python3-pip python3-dev \
+  && cd /usr/local/bin \
+  && ln -s /usr/bin/python3 python \
+  && pip3 install --upgrade pip
+
+# Install git
+RUN apt-get install git -y
 
 COPY . /app
 WORKDIR /app
 
-RUN pip install repository-miner==0.7.0
+# Install application (latest)
+RUN pip install -r requirements.txt
+RUN pip install repository-miner
 
-CMD repo-miner -h
+# Environment variable for temporary repositories
+ENV TMP_REPOSITORIES_DIR=/tmp/

README.md

@@ -27,11 +27,37 @@ pip install .
 
 # How to test
 
-```
+```text
 pip install pytest
 unzip test_data.zip -d .
-pytest tests/
+pytest
 ```
 
+
+# How to build Docker container
+
+`docker build --tag repo-miner:latest .`
+
+# How to run Docker container
+
+First create or define a directory to mount inside the Docker container to access the results once generated.
+For the sake of the example let's use `/tmp/`.
+
+## Mine
+
+Using the `github` argument:
+
+`docker run -v /tmp:/app  -e GITHUB_ACCESS_TOKEN=$GITHUB_ACCESS_TOKEN repo-miner:test repo-miner mine github ansible adriagalin/ansible.motd . --verbose`
+
+Using the `github` argument:
+
+`docker run -v /tmp:/app  -e GITLAB_ACCESS_TOKEN=$GITHUB_ACCESS_TOKEN repo-miner:test repo-miner mine github ansible adriagalin/ansible.motd . --verbose`
+
+
+## Extract metrics
+
+`docker run -v /tmp:/app  repo-miner:test repo-miner extract-metrics https://github.com/<owner>/<repository>.git /tmp/failure_prone_files.json ansible all release . --verbose`
+
+
 ## CHANGELOG
 See the [CHANGELOG](CHANGELOG.md) for information about the release history.

docs/index.md

@@ -17,7 +17,7 @@ pip install .
 
 To avoid affecting the current environment, it is strongly recommended to create and activate a virtual environment:
 
-```
+```text
 sudo apt install python3-venv
 python3 -m venv repo-miner-env
 source repo-miner-env/bin/activate

repominer/cli.py

@@ -14,7 +14,19 @@
 from repominer.mining.tosca import ToscaMiner
 from repominer.report import create_report
 
-VERSION = '0.7.0'
+VERSION = '0.7.1'
+
+
+def valid_dir_or_url(x: str) -> str:
+    """
+    Check if x is a directory and exists, or a remote url
+    :param x: a path
+    :return: the path if exists or is a remote url; raise an ArgumentTypeError otherwise
+    """
+    if not (os.path.isdir(x) or x.startswith("git@") or x.startswith("https://")):
+        raise ArgumentTypeError('Insert a valid path or url')
+
+    return x
 
 
 def valid_dir(x: str) -> str:
@@ -84,7 +96,7 @@ def set_extract_metrics_parser(subparsers):
 
     parser.add_argument(action='store',
                         dest='path_to_repo',
-                        type=valid_dir,
+                        type=valid_dir_or_url,
                         help='the absolute path to a cloned repository or the url to a remote repository')
 
     parser.add_argument(action='store',
@@ -204,7 +216,8 @@ def extract_metrics(args: Namespace):
     global extractor
 
     if args.verbose:
-        print(f'Extracting metrics from {args.path_to_repo} using report {args.src} [started at: {datetime.now().hour}:{datetime.now().minute}]')
+        print(
+            f'Extracting metrics from {args.path_to_repo} using report {args.src} [started at: {datetime.now().hour}:{datetime.now().minute}]')
 
     with open(args.src, 'r') as f:
         labeled_files = json.load(f, cls=FailureProneFileDecoder)

repominer/metrics/base.py

@@ -45,8 +45,13 @@ def __init__(self, path_to_repo: str, at: str = 'release'):
 
         self.path_to_repo = path_to_repo
 
-        # If path_to_repo is a remote url, then clone it in os.getenv('TMP_REPOSITORIES_DIR')
-        if is_remote(path_to_repo):
+        if os.path.isdir(path_to_repo):
+            self.repo_miner = RepositoryMining(path_to_repo=path_to_repo,
+                                               only_releases=True if at == 'release' else False,
+                                               order='date-order')
+
+        elif is_remote(path_to_repo):
+            # If path_to_repo is a remote url, then clone it to os.getenv('TMP_REPOSITORIES_DIR')
             self.repo_miner = RepositoryMining(path_to_repo=path_to_repo,
                                                clone_repo_to=os.getenv('TMP_REPOSITORIES_DIR'),
                                                only_releases=True if at == 'release' else False,
@@ -57,9 +62,7 @@ def __init__(self, path_to_repo: str, at: str = 'release'):
             self.path_to_repo = os.path.join(os.getenv('TMP_REPOSITORIES_DIR'), repo_name)
 
         else:
-            self.repo_miner = RepositoryMining(path_to_repo=path_to_repo,
-                                               only_releases=True if at == 'release' else False,
-                                               order='date-order')
+            raise ValueError(f'{path_to_repo} does not seem a path or url to a Git repository.')
 
         self.dataset = pd.DataFrame()
 

setup.py

@@ -8,7 +8,7 @@
 with open("README.md", "r") as fh:
     long_description = fh.read()
 
-VERSION = '0.7.0'
+VERSION = '0.7.1'
 
 setup(name='repository-miner',
       version=VERSION,

tests/test_cli_extract_metrics.py

@@ -11,19 +11,29 @@ class CLIExtractMetricsTestCase(unittest.TestCase):
     @classmethod
     def setUpClass(cls):
         cls.path_to_repo = os.path.join(os.getcwd(), 'test_data', 'repositories', 'ansible.motd')
+        cls.url_to_repo = 'https://github.com/adriagalin/ansible.motd.git'
         cls.path_to_report = os.path.join(os.getcwd(), 'test_data', 'ansible_report.json')
 
         cls.path_to_tmp_dir = os.path.join(os.getcwd(), 'test_data', 'tmp')
         os.mkdir(cls.path_to_tmp_dir)
+        os.environ["TMP_REPOSITORIES_DIR"] = cls.path_to_tmp_dir
 
     @classmethod
     def tearDownClass(cls):
         shutil.rmtree(cls.path_to_tmp_dir)
+        del os.environ["TMP_REPOSITORIES_DIR"]
 
-    def test_mine(self):
+    def test_remote_url(self):
+        result = os.system('repo-miner extract-metrics {0} {1} ansible all release {2}'.format(self.url_to_repo,
+                                                                                               self.path_to_report,
+                                                                                               self.path_to_tmp_dir))
+        assert result == 0
+        assert 'metrics.csv' in os.listdir(self.path_to_tmp_dir)
+
+    def test_local_path(self):
         result = os.system('repo-miner extract-metrics {0} {1} ansible all release {2}'.format(self.path_to_repo,
-                                                                                                self.path_to_report,
-                                                                                                self.path_to_tmp_dir))
+                                                                                               self.path_to_report,
+                                                                                               self.path_to_tmp_dir))
         assert result == 0
         assert 'metrics.csv' in os.listdir(self.path_to_tmp_dir)