Refactor external licenses code

Refactor according to the feedback.

Signed-off-by: Ayan Sinha Mahapatra <ayansmahapatra@gmail.com>

azure-pipelines.yml

@@ -32,7 +32,9 @@ jobs:
                   --ignore=tests/licensedcode/test_detection_datadriven1.py \
                   --ignore=tests/licensedcode/test_detection_datadriven2.py \
                   --ignore=tests/licensedcode/test_detection_datadriven3.py \
-                  --ignore=tests/licensedcode/test_detection_datadriven4.py
+                  --ignore=tests/licensedcode/test_detection_datadriven4.py \
+                  --ignore=tests/licensedcode/test_additional_license.py \
+                  tests/licensedcode
 
               license_datadriven1_2: |
                 venv/bin/pytest -n 3 -vvs --test-suite=all \

src/licensedcode/cache.py

@@ -72,9 +72,9 @@ def load_or_build(
         On the side, we load cached or build license db, SPDX symbols and other
         license-related data structures.
 
-        - If the cache exists, it is returned unless corrupted, ``force`` is True, or if we pass in additional
-          directories containing licenses that are not present in the existing cache.
-        - If the cache does not exist, a new index is built and cached.
+        - If the cache exists, it is returned unless corrupted.
+        - If ``force`` is True, or if the cache does not exist a new index is built
+          and cached.
         - If ``index_all_languages`` is True, include texts in all languages when
           building the license index. Otherwise, only include the English license
           texts and rules (the default)

src/licensedcode/models.py

@@ -444,6 +444,8 @@ def write(location, byte_string):
         if text:
             write(self.text_file(licenses_data_dir=licenses_data_dir), text.encode('utf-8'))
 
+        return self
+
     def load(self, data_file, text_file):
         """
         Populate license data from a YAML file stored in ``data_file`` and  ``text_file``.

src/licensedcode/reindex.py

@@ -24,7 +24,7 @@
     '--only-builtin',
     is_flag=True,
     help='Rebuild the license index excluding any additional '
-         'license directory or additional license plugins which'
+         'license directory or additional license plugins which '
          'were added previously, i.e. with only builtin scancode '
          'license and rules.',
     conflicting_options=['additional_directory'],

tests/licensedcode/data/plugin_license/license_url.expected.json → tests/licensedcode/data/plugin_license/license_url/license_url.expected.json

@@ -1,15 +1,15 @@
 {
   "files": [
     {
-      "path": "license_url",
+      "path": "scan",
       "type": "directory",
       "licenses": [],
       "license_expressions": [],
       "percentage_of_license_text": 0,
       "scan_errors": []
     },
     {
-      "path": "license_url/apache-1.0.txt",
+      "path": "scan/apache-1.0.txt",
       "type": "file",
       "licenses": [
         {

tests/licensedcode/test_plugin_license.py

@@ -245,7 +245,7 @@ def test_reindex_licenses_works():
 
 @pytest.mark.scanslow
 def test_scan_license_with_url_template():
-    test_dir = test_env.get_test_loc('plugin_license/license_url', copy=True)
+    test_dir = test_env.get_test_loc('plugin_license/license_url/scan/', copy=True)
     result_file = test_env.get_temp_file('json')
     args = [
         '--license',
@@ -254,7 +254,7 @@ def test_scan_license_with_url_template():
         '--json-pp', result_file,
         test_dir,
     ]
-    test_loc = test_env.get_test_loc('plugin_license/license_url.expected.json')
+    test_loc = test_env.get_test_loc('plugin_license/license_url/license_url.expected.json')
     run_scan_click(args)
     check_json_scan(test_loc, result_file, regen=REGEN_TEST_FIXTURES)