Split old and new summarizer code

    * Remove summarizer2.py

Reference: #2842
Signed-off-by: Jono Yang <jyang@nexb.com>

src/summarycode/legacy_summarizer.py

@@ -47,36 +47,6 @@ def logger_debug(*args):
 """
 
 
-@post_scan_impl
-class ScanSummary(PostScanPlugin):
-    """
-    Summarize a scan at the codebase level.
-    """
-    sort_order = 10
-
-    codebase_attributes = dict(summary=attr.ib(default=attr.Factory(dict)))
-
-    options = [
-        PluggableCommandLineOption(('--summary',),
-            is_flag=True, default=False,
-            help='Summarize license, copyright and other scans at the codebase level.',
-            help_group=POST_SCAN_GROUP,
-            required_options=['classify', 'license_clarity_score']
-        )
-    ]
-
-    def is_enabled(self, summary, **kwargs):
-        return summary
-
-    def process_codebase(self, codebase, summary, **kwargs):
-        if TRACE_LIGHT: logger_debug('ScanSummary:process_codebase')
-        summary = summarize_codebase(codebase, keep_details=False, **kwargs)
-        declared_holders = get_declared_holders(codebase, summary)
-        codebase.attributes.summary['declared_holders'] = declared_holders
-        codebase.attributes.summary['primary_programming_language'] = get_primary_language(summary)
-        remove_declared_from_fields(codebase, summary)
-
-
 class SummaryLegacyPluginDeprecationWarning(DeprecationWarning):
     pass
 

tests/summarycode/test_legacy_summarizer.py

@@ -0,0 +1,141 @@
+#
+# Copyright (c) nexB Inc. and others. All rights reserved.
+# ScanCode is a trademark of nexB Inc.
+# SPDX-License-Identifier: Apache-2.0
+# See http://www.apache.org/licenses/LICENSE-2.0 for the license text.
+# See https://github.com/nexB/scancode-toolkit for support or download.
+# See https://aboutcode.org for more information about nexB OSS projects.
+#
+
+
+from os import path
+
+import pytest
+
+from commoncode.testcase import FileDrivenTesting
+
+from scancode.cli_test_utils import check_json_scan
+from scancode.cli_test_utils import check_jsonlines_scan
+from scancode.cli_test_utils import run_scan_click
+from scancode_config import REGEN_TEST_FIXTURES
+
+
+pytestmark = pytest.mark.scanslow
+
+
+class TestScanSummaryLegacy(FileDrivenTesting):
+
+    test_data_dir = path.join(path.dirname(__file__), 'data')
+
+    def test_copyright_summary_base(self):
+        test_dir = self.get_test_loc('summary_legacy/copyright_summary/scan')
+        result_file = self.get_temp_file('json')
+        expected_file = self.get_test_loc('summary_legacy/copyright_summary/summary.expected.json')
+        run_scan_click(['-c', '--summary-legacy', '--json-pp', result_file, test_dir])
+        check_json_scan(expected_file, result_file, remove_file_date=True, regen=REGEN_TEST_FIXTURES)
+
+    def test_copyright_summary_with_details(self):
+        test_dir = self.get_test_loc('summary_legacy/copyright_summary/scan')
+        result_file = self.get_temp_file('json')
+        expected_file = self.get_test_loc('summary_legacy/copyright_summary/summary_details.expected.json')
+        run_scan_click(['-c', '--summary-with-details', '--json-pp', result_file, test_dir])
+        check_json_scan(expected_file, result_file, remove_file_date=True, regen=REGEN_TEST_FIXTURES)
+
+    def test_copyright_summary_with_details_plain_json(self):
+        test_dir = self.get_test_loc('summary_legacy/copyright_summary/scan')
+        result_file = self.get_temp_file('json')
+        expected_file = self.get_test_loc('summary_legacy/copyright_summary/summary_details.expected2.json')
+        run_scan_click(['-c', '--summary-with-details', '--json', result_file, test_dir])
+        check_json_scan(expected_file, result_file, remove_file_date=True, regen=REGEN_TEST_FIXTURES)
+
+    def test_copyright_summary_does_not_crash(self):
+        test_dir = self.get_test_loc('summary_legacy/copyright_summary/scan2')
+        result_file = self.get_temp_file('json')
+        expected_file = self.get_test_loc('summary_legacy/copyright_summary/summary2.expected.json')
+        run_scan_click(['-c', '--summary-legacy', '--json-pp', result_file, test_dir])
+        check_json_scan(expected_file, result_file, remove_file_date=True, regen=REGEN_TEST_FIXTURES)
+
+    def test_full_summary_base(self):
+        test_dir = self.get_test_loc('summary_legacy/full_summary/scan')
+        result_file = self.get_temp_file('json')
+        expected_file = self.get_test_loc('summary_legacy/full_summary/summary.expected.json')
+        run_scan_click(['-clip', '--summary-legacy', '--json-pp', result_file, test_dir])
+        check_json_scan(expected_file, result_file, remove_instance_uuid=True, remove_file_date=True, regen=REGEN_TEST_FIXTURES)
+
+    def test_full_summary_with_details(self):
+        test_dir = self.get_test_loc('summary_legacy/full_summary/scan')
+        result_file = self.get_temp_file('json')
+        expected_file = self.get_test_loc('summary_legacy/full_summary/summary_details.expected.json')
+        run_scan_click(['-clip', '--summary-with-details', '--json-pp', result_file, test_dir])
+        check_json_scan(expected_file, result_file, remove_instance_uuid=True, remove_file_date=True, regen=REGEN_TEST_FIXTURES)
+
+    def test_copyright_summary_key_files(self):
+        test_dir = self.get_test_loc('summary_legacy/copyright_summary/scan')
+        result_file = self.get_temp_file('json')
+        expected_file = self.get_test_loc('summary_legacy/copyright_summary/summary_key_files.expected.json')
+        run_scan_click(
+            ['-c', '-i', '--classify', '--summary-legacy', '--summary-key-files',
+             '--json-pp', result_file, test_dir])
+
+        check_json_scan(expected_file, result_file, remove_file_date=True, regen=REGEN_TEST_FIXTURES)
+
+    def test_full_summary_key_files(self):
+        test_dir = self.get_test_loc('summary_legacy/full_summary/scan')
+        result_file = self.get_temp_file('json')
+        expected_file = self.get_test_loc('summary_legacy/full_summary/summary_key_files.expected.json')
+        run_scan_click(
+            ['-cli', '--classify', '--summary-legacy', '--summary-key-files',
+             '--json-pp', result_file, test_dir])
+        check_json_scan(expected_file, result_file, remove_file_date=True, regen=REGEN_TEST_FIXTURES)
+
+    def test_full_summary_key_files_json_lines(self):
+        test_dir = self.get_test_loc('summary_legacy/full_summary/scan')
+        result_file = self.get_temp_file('json')
+        expected_file = self.get_test_loc('summary_legacy/full_summary/summary_key_files-details.expected.json-lines')
+        run_scan_click(
+            ['-cli', '--classify', '--summary-legacy', '--summary-key-files',
+             '--json-lines', result_file, test_dir])
+        check_jsonlines_scan(expected_file, result_file, remove_file_date=True, regen=REGEN_TEST_FIXTURES)
+
+    def test_full_summary_by_facet(self):
+        test_dir = self.get_test_loc('summary_legacy/full_summary/scan')
+        result_file = self.get_temp_file('json')
+        expected_file = self.get_test_loc('summary_legacy/full_summary/summary_by_facet.expected.json')
+        run_scan_click([
+            '-clpieu',
+            '--facet', 'dev=*.java',
+            '--facet', 'dev=*.cs',
+            '--facet', 'dev=*ada*',
+            '--facet', 'data=*.S',
+            '--facet', 'tests=*infback9*',
+            '--facet', 'docs=*README',
+            '--summary-legacy',
+            '--summary-by-facet',
+            '--json-pp', result_file, test_dir
+        ])
+        check_json_scan(expected_file, result_file, remove_instance_uuid=True, remove_file_date=True, regen=REGEN_TEST_FIXTURES)
+
+    def test_end2end_summary_and_classify_works_with_empty_dir_and_empty_values(self):
+        test_dir = self.extract_test_tar('summary_legacy/end-2-end/bug-1141.tar.gz')
+        result_file = self.get_temp_file('json')
+        expected_file = self.get_test_loc('summary_legacy/end-2-end/bug-1141.expected.json')
+        run_scan_click([
+            '-clip',
+            '--classify',
+            '--facet', 'dev=*.java',
+            '--summary-legacy',
+            '--summary-key-files',
+            '--json-pp', result_file, test_dir
+        ])
+        check_json_scan(expected_file, result_file, remove_instance_uuid=True, remove_file_date=True, regen=REGEN_TEST_FIXTURES)
+
+    def test_summary_with_packages_reports_packages_with_files(self):
+        test_dir = self.get_test_loc('summary_legacy/packages/scan')
+        result_file = self.get_temp_file('json')
+        expected_file = self.get_test_loc('summary_legacy/packages/expected.json')
+        run_scan_click([
+            '--package',
+            '--summary-legacy',
+            '--json-pp', result_file, test_dir
+        ])
+        check_json_scan(expected_file, result_file, remove_instance_uuid=True, remove_file_date=True, regen=REGEN_TEST_FIXTURES)