Rename var. Add str as supported type on site id.

src/vorta/scheduler.py

@@ -20,7 +20,7 @@
 from vorta.notifications import VortaNotifications
 
 logger = logging.getLogger(__name__)
-DEBUG = True
+DEBUG = False
 
 
 # TODO: refactor to use QtCore.QTimer directly
@@ -317,16 +317,16 @@ def __init__(self, site_id, nb_workers_running: QtCore.QSemaphore, threadpool):
         self.current_job = None
         self.nb_workers_running = nb_workers_running
         self.threadpool = threadpool
-        self.start_site = QtCore.QMutex()
+        self.mut_start_site = QtCore.QMutex()
 
     def add_job(self, task: Job):
         self.__p_queue.put(task)
         # If the site is dead, run the site again
-        self.start_site.lock()
+        self.mut_start_site.lock()
         if self.worker_is_running is False:
             self.threadpool.start(self)
             self.worker_is_running = True
-        self.start_site.unlock()
+        self.mut_start_site.unlock()
         # TODO This function must add the job to the database
         # self.add_to_db(job)
 
@@ -337,7 +337,7 @@ def cancel_all_jobs(self):
         end_job = Job()
         end_job.set_status(JobStatus.CANCEL)
 
-        self.start_site.lock()
+        self.mut_start_site.lock()
         # Stop the loop
         self.worker_is_running = False
         # if the site is waiting for a Job, send a cancel Job
@@ -346,7 +346,7 @@ def cancel_all_jobs(self):
         if self.current_job is not None:
             self.current_job.cancel()
 
-        self.start_site.unlock()
+        self.mut_start_site.unlock()
 
     def cancel_job(self, job: Job):
         # Dequeue the job
@@ -411,18 +411,18 @@ def __init__(self):
         self.load_from_db()
         # use a threadpool -> This could be changed in the future
         self.threadpool = QThreadPool()
-        self.lock_queues = QtCore.QMutex()
+        self.mut_queues = QtCore.QMutex()
 
     def test_get_site(self, site):
         return self.__queues.get(site)
 
     def get_site(self, site):
-        self.lock_queues.lock()
+        self.mut_queues.lock()
         if site not in self.__queues:
             if DEBUG:
                 print("Create a site ", site)
             self.__queues[site] = _Queue(site, JobsManager.nb_workers_running, self.threadpool)
-        self.lock_queues.unlock()
+        self.mut_queues.unlock()
         return self.__queues.get(site)
 
     def load_from_db(self):
@@ -434,23 +434,28 @@ def add_job(self, job: Job):
         if DEBUG:
             print("Add Job on site ", job.get_site_id(), type(job.get_site_id()))
 
-        if type(job.get_site_id()) is not int:
+        if type(job.get_site_id()) is not (int or str):
             print("get_site_id must return an integer. A ", type(job.get_site_id()), " has be returned.")
             return 1
         self.get_site(job.get_site_id()).add_job(job)
 
-    # Ask to all queues to cancel all jobs. This is what the user expects when he presses the cancel button.
+    """
+    Ask to all queues to cancel all jobs.
+    There is no guarantee that all tasks will be removed. If some jobs are added before lock from another thread,
+    it will not be cancelled. That's why, it's strongly advised to add jobs from the main UI loop.
+    """
     def cancel_all_jobs(self):
         # Lock dict to avoid someone else adding a new site during cancel operation
-        self.lock_queues.lock()
+        self.mut_queues.lock()
         for id_site, site in self.__queues.items():
+            # don't use get_site since mut_queue is already locked
             site.cancel_all_jobs()
         self.__queues.clear()
         # reset the semaphore
         JobsManager.reset_nb_workers()
         if DEBUG:
             print("End Cancel")
-        self.lock_queues.unlock()
+        self.mut_queues.unlock()
 
     def cancel_job(self, job: Job):
         # call cancel job of the site queue