Antoine R. Dumont · Phabricator Migration user · 5e658ed6 · 2f4f2348
--- a/debian/rules
+++ b/debian/rules
 #!/usr/bin/make -f

 export PYBUILD_NAME=swh.loader.tar
-export export PYBUILD_TEST_ARGS=--with-doctest -sv -a !db,!fs
+export PYBUILD_TEST_ARGS=-m 'not db and not fs'
--- a/swh/loader/tar/loader.py
+++ b/swh/loader/tar/loader.py
+    def cleanup(self):
+        """Clean up temporary disk folders used.

        """
-        # Shortcut super() as we use different arguments than the DirLoader.
-        return BufferedLoader.load(self, tar_path=tar_path, origin=origin,
-                                   visit_date=visit_date, revision=revision,
-                                   branch_name=branch_name)
+        if self.debug:
+            self.log.warn('%s Will not clean up temp dir %s' % (
+                DEBUG_MODE, self.temp_directory
+            ))
+            return
+        if os.path.exists(self.temp_directory):
+            self.log.debug('Clean up %s' % self.temp_directory)
+            shutil.rmtree(self.temp_directory)
--- a/swh/loader/tar/tests/test_build.py
+++ b/swh/loader/tar/tests/test_build.py
+            'microseconds': 830834
        })

-        mock_os.lstat.assert_called_once_with('some/path')
-
-    @patch('swh.loader.tar.build.os')
-    def test_time_from_path_with_int(self, mock_os):
-        class MockStat:
-            st_mtime = 1445348286
-
-        mock_os.lstat.return_value = MockStat()
-
-        actual_time = build._time_from_path('some/path')
+    def test_time_from_last_modified_with_int(self):
+        actual_time = build._time_from_last_modified(
+            '2015-10-20T13:38:06+00:00')
--- a/swh/loader/tar/tests/test_build.py
+++ b/swh/loader/tar/tests/test_build.py

-    @patch('swh.loader.tar.build.os')
-    def test_time_from_path_with_float(self, mock_os):
-        class MockStat:
-            st_mtime = 1445348286.8308342
-        mock_os.lstat.return_value = MockStat()
-
-        actual_time = build._time_from_path('some/path')
+    def test_time_from_last_modified_with_float(self):
+        actual_time = build._time_from_last_modified(
+            '2015-10-20T13:38:06.830834+00:00')

        self.assertEqual(actual_time, {
            'seconds': 1445348286,
-            'microseconds': 8308342
+            'microseconds': 830834
--- a/swh/loader/tar/tests/test_loader.py
+++ b/swh/loader/tar/tests/test_loader.py
        """
        # given
        origin = {
-            'url': 'file:///tmp/sample-folder',
-            'type': 'dir'
+            'url': self.repo_url,
+            'type': 'tar'
+        }
--- a/swh/loader/tar/loader.py
+++ b/swh/loader/tar/loader.py
+            ))
+            return
+        if os.path.exists(self.temp_directory):
+            self.log.debug('Clean up %s' % self.temp_directory)
+            shutil.rmtree(self.temp_directory)

    def prepare_origin_visit(self, *, origin, visit_date=None, **kwargs):
+        """Prepare the origin visit information.
+
+        Args:
+            origin (dict): Dict with keys {url, type}
+            visit_date (str): Date representing the date of the
+              visit. None by default will make it the current time
+              during the loading process.
+
+        """
--- a/swh/loader/tar/loader.py
+++ b/swh/loader/tar/loader.py
-    @click.option('--archive-path', required=1, help='Archive path to load')
-    @click.option('--origin-url', required=1, help='Origin url to associate')
-    @click.option('--visit-date', default=None,
-                  help='Visit date time override')
-    def main(archive_path, origin_url, visit_date):
-        """Loading archive tryout."""
-        import datetime
-        origin = {'url': origin_url, 'type': 'tar'}
-        commit_time = int(datetime.datetime.now(
-            tz=datetime.timezone.utc).timestamp())
-        swh_person = {
-            'name': 'Software Heritage',
-            'fullname': 'Software Heritage',
-            'email': 'robot@softwareheritage.org'
+        objects = self.objects
+        self.maybe_load_contents(objects['content'].values())
--- a/swh/loader/tar/loader.py
+++ b/swh/loader/tar/loader.py
+        """
+        self.last_modified = last_modified
+
+    def get_tarball_url_to_retrieve(self):
+        return self.origin['url']
+
+    def build_revision(self, filepath, nature, hashes):
+        """Build the revision with identifier
+
+        We use the `last_modified` date provided by the caller to
+        build the revision.
+
+        """
+        return {
+            **compute_revision(filepath, self.last_modified),
+            'metadata': {
--- a/swh/loader/tar/loader.py
+++ b/swh/loader/tar/loader.py
+        }
+
+    def download(self, url):
+        """Download the remote tarball url locally.
+
+        Args:
+            url (str): Url (file or http*)
+
+        Raises:
+            ValueError in case of failing to query
+
+        Returns:
+            Tuple of local (filepath, hashes of filepath)
+
+        """
+        url_parsed = urlparse(url)