Merge branch 'develop' into feat/custom-config-imports

Author: Theofilos Manitaras

docs/config_reference.rst

@@ -2118,6 +2118,14 @@ General Configuration
    - ``pretty``: (default) Generate a pretty table
 
 
+.. py:attribute:: general.table_format_delim
+
+   :required: No
+   :default: ``","``
+
+   Delimiter to use when emitting tables in CSV format.
+
+
 .. py:attribute:: general.timestamp_dirs
 
    :required: No

docs/manpage.rst

@@ -1204,6 +1204,14 @@ Miscellaneous options
 
    .. versionadded:: 4.7
 
+.. option:: --table-format-delim[=DELIM]
+
+   Delimiter to use when emitting tables in CSV format using the :option:`--table-format=csv` option.
+
+   The default delimiter is ``,``.
+
+   .. versionadded:: 4.9
+
 .. option:: --upgrade-config-file=OLD[:NEW]
 
    Convert the old-style configuration file ``OLD``, place it into the new file ``NEW`` and exit.
@@ -2316,6 +2324,21 @@ Whenever an environment variable is associated with a configuration option, its
 
    .. versionadded:: 4.7
 
+.. envvar:: RFM_TABLE_FORMAT_DELIM
+
+   Delimiter for CSV tables.
+
+
+   .. table::
+      :align: left
+
+      ================================== ==================
+      Associated command line option     :option:`--table-format-delim`
+      Associated configuration parameter :attr:`~config.general.table_format_delim`
+      ================================== ==================
+
+   .. versionadded:: 4.9
+
 
 .. envvar:: RFM_TIMESTAMP_DIRS
 

reframe/frontend/cli.py

@@ -697,6 +697,11 @@ def main():
         help='Table formatting',
         envvar='RFM_TABLE_FORMAT', configvar='general/table_format'
     )
+    misc_options.add_argument(
+        '--table-format-delim', action='store',
+        help='The delimiter to use when using `--table-format=csv`',
+        envvar='RFM_TABLE_FORMAT_DELIM', configvar='general/table_format_delim'
+    )
     misc_options.add_argument(
         '-v', '--verbose', action='count',
         help='Increase verbosity level of output',
@@ -1083,9 +1088,7 @@ def restrict_logging():
                                   lambda htype: htype != 'stream')
         with exit_gracefully_on_error('failed to retrieve session data',
                                       printer):
-            printer.info(jsonext.dumps(reporting.session_info(
-                options.describe_stored_sessions
-            ), indent=2))
+            printer.info(reporting.session_info(options.describe_stored_sessions))
             sys.exit(0)
 
     if options.describe_stored_testcases:

reframe/frontend/printer.py

@@ -277,8 +277,9 @@ def performance_report(self, data, **kwargs):
         self.info('')
 
     def _table_as_csv(self, data):
+        delim = rt.runtime().get_option('general/0/table_format_delim')
         for line in data:
-            self.info(','.join(str(x) for x in line))
+            self.info(delim.join(str(x) for x in line))
 
     def table(self, data, **kwargs):
         '''Print tabular data'''

reframe/frontend/reporting/__init__.py

@@ -860,8 +860,10 @@ def testcase_data(spec, namepatt=None, test_filter=None):
 @time_function
 def session_info(query):
     '''Retrieve session details as JSON'''
-
-    return StorageBackend.default().fetch_sessions(parse_query_spec(query))
+    sessions = StorageBackend.default().fetch_sessions(
+        parse_query_spec(query), False
+    )
+    return rf'[{",".join(sessions)}]'
 
 
 @time_function

reframe/frontend/reporting/storage.py

@@ -54,12 +54,14 @@ def fetch_testcases(self, selector: QuerySelector, name_patt=None,
         '''
 
     @abc.abstractmethod
-    def fetch_sessions(self, selector: QuerySelector):
+    def fetch_sessions(self, selector: QuerySelector, decode=True):
         '''Fetch sessions based on the specified query selector.
 
         :arg selector: an instance of :class:`QuerySelector` that will specify
             the actual type of query requested.
-        :returns: A list of matching sessions.
+        :arg decode: If set to :obj:`False`, do not decode the returned
+            sessions and leave them JSON-encoded.
+        :returns: A list of matching sessions, either decoded or not.
         '''
 
     @abc.abstractmethod
@@ -101,6 +103,7 @@ def _db_file(self):
 
             self._db_create()
 
+        self._db_create_indexes()
         self._db_schema_check()
         return self.__db_file
 
@@ -161,12 +164,20 @@ def _db_create(self):
                          'uuid TEXT, '
                          'FOREIGN KEY(session_uuid) '
                          'REFERENCES sessions(uuid) ON DELETE CASCADE)')
+
+        # Update DB file mode
+        os.chmod(self.__db_file, self.__db_file_mode)
+
+    def _db_create_indexes(self):
+        clsname = type(self).__name__
+        getlogger().debug(f'{clsname}: creating database indexes if needed')
+        with self._db_connect(self.__db_file) as conn:
             conn.execute('CREATE INDEX IF NOT EXISTS index_testcases_time '
                          'on testcases(job_completion_time_unix)')
             conn.execute('CREATE TABLE IF NOT EXISTS metadata('
                          'schema_version TEXT)')
-        # Update DB file mode
-        os.chmod(self.__db_file, self.__db_file_mode)
+            conn.execute('CREATE INDEX IF NOT EXISTS index_sessions_time '
+                         'on sessions(session_start_unix)')
 
     def _db_schema_check(self):
         with self._db_read(self.__db_file) as conn:
@@ -232,10 +243,16 @@ def store(self, report, report_file=None):
             return self._db_store_report(conn, report, report_file)
 
     @time_function
-    def _decode_sessions(self, results, sess_filter):
-        '''Decode sessions from the raw DB results.
+    def _mass_json_decode(self, *json_objs):
+        data = rf'[{",".join(json_objs)}]'
+        getlogger().debug(f'decoding JSON raw data of length {len(data)}')
+        return json.loads(data)
 
-        Return a map of session uuids to decoded session data
+    @time_function
+    def _fetch_sessions(self, results, sess_filter):
+        '''Fetch JSON-encoded sessions from the DB by applying a filter.
+
+        :returns: A list of the JSON-encoded valid sessions.
         '''
         sess_info_patt = re.compile(
             r'\"session_info\":\s+(?P<sess_info>\{.*?\})'
@@ -245,34 +262,31 @@ def _decode_sessions(self, results, sess_filter):
         def _extract_sess_info(s):
             return sess_info_patt.search(s).group('sess_info')
 
-        @time_function
-        def _mass_json_decode(json_objs):
-            data = '[' + ','.join(json_objs) + ']'
-            getlogger().debug(
-                f'decoding JSON raw data of length {len(data)}'
-            )
-            return json.loads(data)
-
         session_infos = {}
         sessions = {}
         for uuid, json_blob in results:
             sessions.setdefault(uuid, json_blob)
             session_infos.setdefault(uuid, _extract_sess_info(json_blob))
 
-        # Find the UUIDs to decode fully by inspecting only the session info
+        # Find the relevant sessions by inspecting only the session info
         uuids = []
-        for sess_info in _mass_json_decode(session_infos.values()):
+        infos = self._mass_json_decode(*session_infos.values())
+        for sess_info in infos:
             try:
                 if self._db_filter_json(sess_filter, sess_info):
                     uuids.append(sess_info['uuid'])
             except Exception:
                 continue
 
-        # Decode selected sessions
-        reports = _mass_json_decode(sessions[uuid] for uuid in uuids)
+        return [sessions[uuid] for uuid in uuids]
 
-        # Return only the selected sessions
-        return {rpt['session_info']['uuid']: rpt for rpt in reports}
+    def _decode_and_index_sessions(self, json_blobs):
+        '''Decode the sessions and index them by their uuid.
+
+        :returns: A dictionary with uuids as keys and the sessions as values.
+        '''
+        return {sess['session_info']['uuid']: sess
+                for sess in self._mass_json_decode(*json_blobs)}
 
     @time_function
     def _fetch_testcases_raw(self, condition):
@@ -289,7 +303,11 @@ def _fetch_testcases_raw(self, condition):
             results = conn.execute(query).fetchall()
 
         getprofiler().exit_region()
-        sessions = self._decode_sessions(results, None)
+
+        # Fetch, decode and index the sessions by their uuid
+        sessions = self._decode_and_index_sessions(
+            self._fetch_sessions(results, None)
+        )
 
         # Extract the test case data by extracting their UUIDs
         getprofiler().enter_region('sqlite testcase query')
@@ -319,8 +337,8 @@ def _fetch_testcases_raw(self, condition):
         return testcases
 
     @time_function
-    def _fetch_testcases_from_session(self, selector,
-                                      name_patt=None, test_filter=None):
+    def _fetch_testcases_from_session(self, selector, name_patt=None,
+                                      test_filter=None):
         query = 'SELECT uuid, json_blob from sessions'
         if selector.by_session_uuid():
             query += f' WHERE uuid == "{selector.uuid}"'
@@ -338,9 +356,11 @@ def _fetch_testcases_from_session(self, selector,
         if not results:
             return []
 
-        sessions = self._decode_sessions(
-            results,
-            selector.sess_filter if selector.by_session_filter() else None
+        sessions = self._decode_and_index_sessions(
+            self._fetch_sessions(
+                results,
+                selector.sess_filter if selector.by_session_filter() else None
+            )
         )
         return [tc for sess in sessions.values()
                 for run in sess['runs'] for tc in run['testcases']
@@ -366,15 +386,15 @@ def fetch_testcases(self, selector: QuerySelector,
                         name_patt=None, test_filter=None):
         if selector.by_session():
             return self._fetch_testcases_from_session(
-                selector, name_patt, test_filter
+                selector, name_patt, test_filter,
             )
         else:
             return self._fetch_testcases_time_period(
                 *selector.time_period, name_patt, test_filter
             )
 
     @time_function
-    def fetch_sessions(self, selector: QuerySelector):
+    def fetch_sessions(self, selector: QuerySelector, decode=True):
         query = 'SELECT uuid, json_blob FROM sessions'
         if selector.by_time_period():
             ts_start, ts_end = selector.time_period
@@ -389,11 +409,14 @@ def fetch_sessions(self, selector: QuerySelector):
             results = conn.execute(query).fetchall()
 
         getprofiler().exit_region()
-        session = self._decode_sessions(
+        raw_sessions = self._fetch_sessions(
             results,
             selector.sess_filter if selector.by_session_filter() else None
         )
-        return [*session.values()]
+        if decode:
+            return [*self._decode_and_index_sessions(raw_sessions).values()]
+        else:
+            return raw_sessions
 
     def _do_remove(self, conn, uuids):
         '''Remove sessions'''

reframe/frontend/reporting/utility.py

@@ -200,7 +200,9 @@ def by_session_filter(self):
 
     def __repr__(self):
         clsname = type(self).__name__
-        return f'{clsname}(value={self.__value}, kind={self.__kind})'
+        return (f'{clsname}(uuid={self.__uuid!r}, '
+                f'time_period={self.__time_period!r}, '
+                f'sess_filter={self.__sess_filter!r})')
 
 
 def parse_time_period(s):

reframe/schemas/config.json

@@ -538,6 +538,7 @@
                     "save_log_files": {"type": "boolean"},
                     "target_systems": {"$ref": "#/defs/system_ref"},
                     "table_format": {"enum": ["csv", "plain", "pretty"]},
+                    "table_format_delim": {"type": "string"},
                     "timestamp_dirs": {"type": "string"},
                     "topology_prefix": {"type": "string"},
                     "trap_job_errors": {"type": "boolean"},
@@ -609,6 +610,7 @@
         "general/resolve_module_conflicts": true,
         "general/save_log_files": false,
         "general/table_format": "pretty",
+        "general/table_format_delim": ",",
         "general/target_systems": ["*"],
         "general/topology_prefix": "${HOME}/.reframe/topology",
         "general/timestamp_dirs": "%Y%m%dT%H%M%S%z",