refactor: streamline project and sample creation logic in job_creator and scraper

Author: karlnyr

microSALT/cli.py

@@ -315,14 +315,6 @@ def finish(
         if os.path.isdir(f"{input}/{subfolder}"):
             pool.append(subfolder)
 
-    run_settings = {
-        "input": input,
-        "track": track,
-        "dry": dry,
-        "email": config.regex.mail_recipient,
-        "skip_update": skip_update,
-    }
-
     sampleinfo = review_sampleinfo(sampleinfo_file)
     ext_refs = Referencer(
         log=logger,

microSALT/store/db_manipulator.py

@@ -279,6 +279,18 @@ def delete_sample(self, cg_id: str) -> None:
         self.session.commit()
         self.logger.info(f"Removed sample {cg_id} and its child rows")
 
+    def delete_sample_results(self, cg_id: str) -> None:
+        """Delete only the analysis result rows for a sample (seq_types, resistances, expacs)
+        without removing the Samples row itself."""
+        for obj in self.session.query(Expacs).filter(Expacs.CG_ID_sample == cg_id).all():
+            self.session.delete(obj)
+        for obj in self.session.query(Seq_types).filter(Seq_types.CG_ID_sample == cg_id).all():
+            self.session.delete(obj)
+        for obj in self.session.query(Resistances).filter(Resistances.CG_ID_sample == cg_id).all():
+            self.session.delete(obj)
+        self.session.commit()
+        self.logger.info(f"Cleared analysis results for sample {cg_id}")
+
     def delete_project(self, name: str) -> None:
         """Delete all samples (and their child rows) belonging to a project."""
         for obj in self.session.query(Expacs).filter(Expacs.CG_ID_sample.like(f"{name}%")).all():
@@ -413,6 +425,10 @@ def read_columns(self, tablename: str):
         table = _resolve_orm_table(tablename)
         return dict.fromkeys(table.__table__.columns.keys())
 
+    def get_projects_by_cg_id_project(self, cg_id_project_name: str) -> Projects | None:
+        """Fetch a Projects record by CG_ID_project."""
+        return self.session.query(Projects).filter(Projects.CG_ID_project == cg_id_project_name).scalar()
+
     def read_exists(self, table: str, item: dict[str, str]):
         """Takes a k-v pair and checks for the entrys existence in the given table"""
         orm_table = _resolve_orm_table(table)

microSALT/utils/job_creator.py

@@ -541,8 +541,8 @@ def create_collection(self):
                 self.create_project(lims_project)
                 addedprojs.append(lims_project)
 
-    def create_project(self, name: str):
-        """Creates project in database"""
+    def create_project(self, name: str) -> None:
+        """Creates or updates a project in the database."""
         if not self.sample:
             raise JobCreationError(
                 "No sample information provided. Cannot create project in database."
@@ -552,12 +552,15 @@ def create_project(self, name: str):
             "Customer_ID_project": self.sample["Customer_ID_project"],
             "Customer_ID": self.sample["Customer_ID"],
         }
-        self.db_pusher.add_to_session(self.db_pusher.add_project(**project_data))
-        self.db_pusher.commit_session()
-
-    def create_sample(self):
-        """Creates sample in database"""
+        if self.db_pusher.get_projects_by_cg_id_project(name):
+            update_data = {k: v for k, v in project_data.items() if k != "CG_ID_project"}
+            self.db_pusher.update_project({"CG_ID_project": name}, update_data)
+        else:
+            self.db_pusher.add_to_session(self.db_pusher.add_project(**project_data))
+            self.db_pusher.commit_session()
 
+    def create_sample(self) -> None:
+        """Creates or updates a sample in the database."""
         try:
             if not self.sample:
                 raise JobCreationError(
@@ -581,8 +584,13 @@ def create_sample(self):
                 "method_libprep": self.sample["method_libprep"],
                 "method_sequencing": self.sample["method_sequencing"],
             }
-            self.db_pusher.add_to_session(self.db_pusher.add_sample(**sample_data))
-            self.db_pusher.commit_session()
+            cg_id = self.sample["CG_ID_sample"]
+            if self.db_pusher.read_exists("Samples", {"CG_ID_sample": cg_id}):
+                update_data = {k: v for k, v in sample_data.items() if k != "CG_ID_sample"}
+                self.db_pusher.update_sample({"CG_ID_sample": cg_id}, update_data)
+            else:
+                self.db_pusher.add_to_session(self.db_pusher.add_sample(**sample_data))
+                self.db_pusher.commit_session()
         except JobCreationError as e:
             self.logger.error(f"Unable to add sample {self.name} to database: {e}")
         except KeyError as e:

microSALT/utils/scraper.py

@@ -108,9 +108,7 @@ def scrape_project(self, project=None):
         if project is None:
             project = self.name
         self.db_pusher.delete_project(project)
-        if not self.db_pusher.read_exists("Projects", {"CG_ID_project": project}):
-            self.logger.warning(f"Replacing project {project}")
-            self.job_fallback.create_project(project)
+        self.job_fallback.create_project(project)
 
         # Scrape order matters a lot!
         for item in os.listdir(self.infolder):
@@ -140,17 +138,9 @@ def scrape_sample(self, sample=None):
         """Scrapes a sample folder for information"""
         if sample is None:
             sample = self.name
-        self.db_pusher.delete_sample(sample)
-
-        if not self.db_pusher.read_exists(
-            "Projects", {"CG_ID_project": self.sample.get("CG_ID_project")}
-        ):
-            self.logger.warning(f"Replacing project {self.sample.get('CG_ID_project')}")
-            self.job_fallback.create_project(self.sample.get("CG_ID_project"))
-
-        if not self.db_pusher.read_exists("Samples", {"CG_ID_sample": sample}):
-            self.logger.info(f"Replacing sample {sample}")
-            self.job_fallback.create_sample(sample)
+        self.db_pusher.delete_sample_results(sample)
+        self.job_fallback.create_project(self.sample.get("CG_ID_project"))
+        self.job_fallback.create_sample()
 
         # Scrape order matters a lot!
         self.sampledir = self.infolder