Merge pull request #971 from Hendriela/kwargs_in_populate

Keyword arguments for populate()

Author: guzman-raphael

CHANGELOG.md

@@ -16,6 +16,7 @@
 * Bugfix - Fix count for left-joined `QueryExpressions` (#951) PR #966
 * Bugfix - Fix assertion error when performing a union into a join (#930) PR #967
 * Update `~jobs.error_stack` from blob to mediumblob to allow error stacks >64kB in jobs (#984) PR #986
+* Add - Allow optional keyword arguments for `make()` in `populate()` PR #971
 
 ### 0.13.2 -- May 7, 2021
 * Update `setuptools_certificate` dependency to new name `otumat`

datajoint/autopopulate.py

@@ -128,10 +128,11 @@ def _jobs_to_do(self, restrictions):
 
     def populate(self, *restrictions, suppress_errors=False, return_exception_objects=False,
                  reserve_jobs=False, order="original", limit=None, max_calls=None,
-                 display_progress=False, processes=1):
+                 display_progress=False, processes=1, make_kwargs=None):
         """
-        table.populate() calls table.make(key) for every primary key in self.key_source
-        for which there is not already a tuple in table.
+        ``table.populate()`` calls ``table.make(key)`` for every primary key in
+        ``self.key_source`` for which there is not already a tuple in table.
+
         :param restrictions: a list of restrictions each restrict
             (table.key_source - target.proj())
         :param suppress_errors: if True, do not terminate execution.
@@ -143,6 +144,10 @@ def populate(self, *restrictions, suppress_errors=False, return_exception_object
         :param display_progress: if True, report progress_bar
         :param processes: number of processes to use. When set to a large number, then
             uses as many as CPU cores
+        :param make_kwargs: Keyword arguments which do not affect the result of computation
+            to be passed down to each ``make()`` call. Computation arguments should be
+            specified within the pipeline e.g. using a `dj.Lookup` table.
+        :type make_kwargs: dict, optional
         """
         if self.connection.in_transaction:
             raise DataJointError('Populate cannot be called during a transaction.')
@@ -176,7 +181,8 @@ def handler(signum, frame):
         error_list = []
         populate_kwargs = dict(
             suppress_errors=suppress_errors,
-            return_exception_objects=return_exception_objects)
+            return_exception_objects=return_exception_objects,
+            make_kwargs=make_kwargs)
 
         if processes == 1:
             for key in tqdm(keys, desc=self.__class__.__name__) if display_progress else keys:
@@ -207,7 +213,7 @@ def handler(signum, frame):
         if suppress_errors:
             return error_list
 
-    def _populate1(self, key, jobs, suppress_errors, return_exception_objects):
+    def _populate1(self, key, jobs, suppress_errors, return_exception_objects, make_kwargs=None):
         """
         populates table for one source key, calling self.make inside a transaction.
         :param jobs: the jobs table or None if not reserve_jobs
@@ -228,7 +234,7 @@ def _populate1(self, key, jobs, suppress_errors, return_exception_objects):
                 logger.info('Populating: ' + str(key))
                 self.__class__._allow_insert = True
                 try:
-                    make(dict(key))
+                    make(dict(key), **(make_kwargs or {}))
                 except (KeyboardInterrupt, SystemExit, Exception) as error:
                     try:
                         self.connection.cancel_transaction()

docs-parts/intro/Releases_lang1.rst

@@ -15,6 +15,7 @@
 * Bugfix - Fix assertion error when performing a union into a join (#930) PR #967
 * Bugfix - Fix regression issue with `DISTINCT` clause and `GROUP_BY` (#914) PR #963
 * Update `~jobs.error_stack` from blob to mediumblob to allow error stacks >64kB in jobs (#984) PR #986
+* Add - Allow optional keyword arguments for `make()` in `populate()` PR #971
 
 0.13.2 -- May 7, 2021
 ----------------------