Merge remote-tracking branch 'origin/master' into development

CODEOWNER

@@ -0,0 +1,3 @@
+* @JessyBarrette
+* @n-a-t-e
+* @JorinHakai

README.md

@@ -28,7 +28,7 @@ If you just want to see how a dataset is harvested by CDE:
 - Rename .env.sample from the root directory to .env and change any settings if needed. If you are running on your local machine these settings don't need to change
 
 - Start a local database using `docker`:
-  `docker-compose up db -d`
+  `docker-compose up -d db`
 - Setup Python virtual env and install Python modules:
 
   ```sh

harvester/cde_harvester/__main__.py

@@ -24,6 +24,8 @@
 load_dotenv()
 
 logging.getLogger("urllib3").setLevel(logging.WARNING)
+logger = logging.getLogger()
+
 sentry_sdk.init(
     dsn=os.environ.get("SENTRY_DSN"),
     integrations=[
@@ -38,15 +40,17 @@
 
 def setup_logging(log_time, log_level):
     # setup logging
-    logging.basicConfig(
-        level=logging.getLevelName(log_level.upper()),
-        format=(
-            "%(asctime)s - %(name)s : %(message)s"
-            if log_time
-            else "%(name)s : %(message)s"
-        ),
+    logger.setLevel(logging.getLevelName(log_level.upper()))
+    logger.handlers.clear()
+    # Define the stream log format and level
+    c_handler = logging.StreamHandler()
+    c_handler.setLevel(logging.getLevelName(log_level.upper()))
+    c_format = logging.Formatter(
+        ("%(asctime)s - " if log_time else "")
+        + "%(levelname)-8s - %(name)s : %(message)s"
     )
-    logger = logging.getLogger()
+    c_handler.setFormatter(c_format)
+    logger.addHandler(c_handler)
     return logger
 
 

harvester/cde_harvester/dataset.py

@@ -253,7 +253,7 @@ def get_eovs(self):
         return eovs
 
     def get_platform_code(self):
-        platform = self.globals["platform"]
+        platform = self.globals.get("platform")
         platform_vocabulary = self.globals.get("platform_vocabulary")
 
         if not (platform and platform_vocabulary):
@@ -278,13 +278,20 @@ def get_platform_code(self):
                 subset=["l06_label"]
             )
 
-            if platform in list(platforms_nerc_ioos["l06_label"]):
+            if platform.lower() in list(platforms_nerc_ioos["l06_label"]):
                 return platforms_nerc_ioos_no_duplicates.query(
-                    f"l06_label=='{platform}'"
+                    f"l06_label=='{platform.lower()}'"
                 )["category"].item()
             else:
                 self.logger.error("Found unsupported L06 platform: %s", platform)
 
+        self.logger.warning(
+            "Found platform='%s' without known vocabulary='%s', setting to 'unknown'",
+            platform,
+            platform_vocabulary,
+        )
+        return "unknown"
+
     def get_metadata(self):
         "get all the global and variable metadata for a dataset"
 
@@ -354,8 +361,7 @@ def get_metadata(self):
             filter(None, set([globals_dict.get(x) for x in organization_fields]))
         )
 
-        if self.globals.get("platform"):
-            self.platform = self.get_platform_code()
+        self.platform = self.get_platform_code()
 
     def get_logger(self):
         logger = logging.getLogger(f"{self.erddap_server.domain} - {self.id}")