Add timestamp to scan summary and json output file #925

Signed-off-by: Tushar Mittal <[email protected]>

src/formattedcode/output_json.py

@@ -64,6 +64,7 @@ def process_codebase(self, codebase, output_json, files_count,
                    files_count=files_count,
                    scancode_version=scancode_version,
                    scancode_notice=scancode_notice,
+                   scan_start = codebase.scan_start,
                    pretty_options=pretty_options,
                    pretty=False)
 
@@ -92,18 +93,21 @@ def process_codebase(self, codebase, output_json_pp, files_count,
                    files_count=files_count,
                    scancode_version=scancode_version,
                    scancode_notice=scancode_notice,
+                   scan_start = codebase.scan_start,
                    pretty_options=pretty_options,
                    pretty=True)
 
 
 def write_json(results, output_file, files_count,
                scancode_version, scancode_notice,
-               pretty_options, pretty=False):
+               scan_start, pretty_options, 
+               pretty=False):
 
     scan = OrderedDict([
         ('scancode_notice', scancode_notice),
         ('scancode_version', scancode_version),
         ('scancode_options', pretty_options),
+        ('scan_start', scan_start),
         ('files_count', files_count),
         ('files', results),
     ])

src/formattedcode/output_jsonlines.py

@@ -61,6 +61,7 @@ def process_codebase(self, codebase, output_json_lines, files_count,
             ('scancode_notice', scancode_notice),
             ('scancode_version', scancode_version),
             ('scancode_options', pretty_options),
+            ('scan_start', codebase.scan_start),
             ('files_count', files_count)
         ]))
 

src/scancode/cli.py

@@ -724,7 +724,7 @@ def scancode(ctx, input,  # NOQA
         # TODO: this is weird: may be the timings should NOt be stored on the
         # codebase, since they exist in abstract of it??
         codebase.timings.update(setup_timings)
-
+        codebase.scan_start = scan_start
         codebase.timings['inventory'] = time() - inventory_start
         files_count, dirs_count, size_count = codebase.compute_counts()
         codebase.summary['initial:files_count'] = files_count
@@ -1202,6 +1202,10 @@ def display_summary(codebase, scan_names, processes, verbose):
                                 '%(final_size_count)s' % locals())
 
     echo_stderr('Timings:')
+
+    timestamp = codebase.scan_start
+    echo_stderr('  scan_start: %(timestamp)s'% locals())
+
     for name, value, in codebase.timings.items():
         if value > 0.1:
             echo_stderr('  %(name)s: %(value).2fs' % locals())

src/scancode/cli_test_utils.py

@@ -115,7 +115,7 @@ def check_json_scan(expected_file, result_file, regen=False,
     convenient for updating tests expectations. But use with caution.
     """
     scan_results = load_json_result(result_file, strip_dates, clean_errs)
-
+    scan_results.pop('scan_start',None)
     if regen:
         with open(expected_file, 'wb') as reg:
             json.dump(scan_results, reg, indent=2, separators=(',', ': '))

src/scancode/resource.py

@@ -216,6 +216,9 @@ def __init__(self, location, resource_class=None,
         # This is populated automatically.
         self.timings = OrderedDict()
 
+        # stores the timestamp when the scan started as string 
+        self.scan_start = None
+
         # list of errors from collecting the codebase details (such as
         # unreadable file, etc).
         self.errors = []

tests/formattedcode/test_output_json.py

@@ -32,6 +32,7 @@
 from commoncode.testcase import FileDrivenTesting
 from scancode.cli_test_utils import check_json_scan
 from scancode.cli_test_utils import run_scan_click
+from scancode.cli_test_utils import load_json_result
 
 test_env = FileDrivenTesting()
 test_env.test_data_dir = os.path.join(os.path.dirname(__file__), 'data')
@@ -71,3 +72,10 @@ def test_scan_output_does_not_truncate_copyright_with_json_to_stdout():
     run_scan_click(args)
     expected = test_env.get_test_loc('json/tree/expected.json')
     check_json_scan(test_env.get_test_loc(expected), result_file, strip_dates=True)
+
+def test_scan_output_for_timestamp():
+    test_dir = test_env.get_test_loc('json/simple')
+    result_file = test_env.get_temp_file('json')
+    run_scan_click(['-clip', test_dir, '--json', result_file])
+    result_json = load_json_result(result_file)
+    assert "scan_start" in result_json