feat: Allow DumpExporter to only export tasks

Author: ranaldmiao

cms/db/util.py

@@ -274,8 +274,8 @@ def get_datasets_to_judge(task):
 
 def enumerate_files(
         session, contest=None,
-        skip_submissions=False, skip_user_tests=False, skip_print_jobs=False,
-        skip_generated=False):
+        skip_submissions=False, skip_user_tests=False, skip_users=False,
+        skip_print_jobs=False, skip_generated=False):
     """Enumerate all the files (by digest) referenced by the
     contest.
 
@@ -302,7 +302,7 @@ def enumerate_files(
     queries.append(dataset_q.join(Dataset.testcases)
                    .with_entities(Testcase.output))
 
-    if not skip_submissions:
+    if not skip_submissions and not skip_users:
         submission_q = task_q.join(Task.submissions)
         queries.append(submission_q.join(Submission.files)
                        .with_entities(File.digest))
@@ -312,7 +312,7 @@ def enumerate_files(
                            .join(SubmissionResult.executables)
                            .with_entities(Executable.digest))
 
-    if not skip_user_tests:
+    if not skip_user_tests and not skip_users:
         user_test_q = task_q.join(Task.user_tests)
         queries.append(user_test_q.with_entities(UserTest.input))
         queries.append(user_test_q.join(UserTest.files)
@@ -327,7 +327,7 @@ def enumerate_files(
             queries.append(user_test_result_q
                            .with_entities(UserTestResult.output))
 
-    if not skip_print_jobs:
+    if not skip_print_jobs and not skip_users:
         queries.append(contest_q.join(Contest.participations)
                        .join(Participation.printjobs)
                        .with_entities(PrintJob.digest))

cmscontrib/DumpExporter.py

@@ -47,6 +47,7 @@
 from cms.db import version as model_version, Codename, Filename, \
     FilenameSchema, FilenameSchemaArray, Digest, SessionGen, Contest, User, \
     Task, Submission, UserTest, SubmissionResult, UserTestResult, PrintJob, \
+    Admin, Message, Question, Announcement, Participation, \
     enumerate_files
 from cms.db.filecacher import FileCacher
 from cmscommon.datetime import make_timestamp
@@ -136,15 +137,18 @@ class DumpExporter:
 
     def __init__(self, contest_ids, export_target,
                  dump_files, dump_model, skip_generated,
-                 skip_submissions, skip_user_tests, skip_print_jobs):
+                 skip_submissions, skip_user_tests, skip_users, skip_print_jobs):
         if contest_ids is None:
             with SessionGen() as session:
                 contests = session.query(Contest).all()
                 self.contests_ids = [contest.id for contest in contests]
-                users = session.query(User).all()
-                self.users_ids = [user.id for user in users]
-                tasks = session.query(Task)\
-                    .filter(Task.contest_id.is_(None)).all()
+                if not skip_users:
+                  users = session.query(User).all()
+                  self.users_ids = [user.id for user in users]
+                  tasks = session.query(Task)\
+                      .filter(Task.contest_id.is_(None)).all()
+                else:
+                  self.user_ids = []
                 self.tasks_ids = [task.id for task in tasks]
         else:
             # FIXME: this is ATM broken, because if you export a contest, you
@@ -158,6 +162,7 @@ def __init__(self, contest_ids, export_target,
         self.skip_generated = skip_generated
         self.skip_submissions = skip_submissions
         self.skip_user_tests = skip_user_tests
+        self.skip_users = skip_users
         self.skip_print_jobs = skip_print_jobs
         self.export_target = export_target
 
@@ -208,6 +213,7 @@ def do_export(self):
                         session, contest,
                         skip_submissions=self.skip_submissions,
                         skip_user_tests=self.skip_user_tests,
+                        skip_users=self.skip_users,
                         skip_print_jobs=self.skip_print_jobs,
                         skip_generated=self.skip_generated)
                     for file_ in files:
@@ -306,6 +312,7 @@ class of the given object), an item for each column property
             val = getattr(obj, prp.key)
             data[prp.key] = encode_value(col.type, val)
 
+        user_related_classes = [User, Admin, UserTest, Submission, PrintJob, Message, Question, Announcement, Participation]
         for prp in cls._rel_props:
             other_cls = prp.mapper.class_
 
@@ -317,6 +324,15 @@ class of the given object), an item for each column property
             if self.skip_user_tests and other_cls is UserTest:
                 continue
 
+            if self.skip_users:
+                skip = False
+                for rel_class in user_related_classes:
+                    if other_cls is rel_class:
+                        skip = True
+                        break                 
+                if skip:
+                  continue
+
             # Skip print jobs if requested
             if self.skip_print_jobs and other_cls is PrintJob:
                 continue
@@ -397,6 +413,8 @@ def main():
                         help="don't export submissions")
     parser.add_argument("-U", "--no-user-tests", action="store_true",
                         help="don't export user tests")
+    parser.add_argument("-X", "--no-users", action="store_true",
+                        help="don't export users")
     parser.add_argument("-P", "--no-print-jobs", action="store_true",
                         help="don't export print jobs")
     parser.add_argument("export_target", action="store",
@@ -412,6 +430,7 @@ def main():
                             skip_generated=args.no_generated,
                             skip_submissions=args.no_submissions,
                             skip_user_tests=args.no_user_tests,
+                            skip_users=args.no_users,
                             skip_print_jobs=args.no_print_jobs)
     success = exporter.do_export()
     return 0 if success is True else 1