fix: Do not use nbsp in strings (#2140)

* Use normal spaces in ingest

*In preprocessing replace nbsps with normal spaces.

ingest/scripts/prepare_metadata.py

@@ -42,7 +42,7 @@ def split_authors(authors: str) -> str:
 
     for i in range(0, len(single_split), 2):
         if i + 1 < len(single_split):
-            result.append(single_split[i + 1].strip() + "\u00a0" + single_split[i].strip())
+            result.append(single_split[i + 1].strip() + " " + single_split[i].strip())
         else:
             result.append(single_split[i].strip())
 

preprocessing/nextclade/src/loculus_preprocessing/config.py

@@ -93,7 +93,7 @@ def get_config() -> Config:
     # Overwrite config with config in config_file
     if args.config_file:
         config = load_config_from_yaml(args.config_file, config)
-    if config.backend_host is None:  # Check if backend_host wasn't set during initialization
+    if not config.backend_host:  # Check if backend_host wasn't set during initialization
         config.backend_host = f"http://127.0.0.1:8079/{config.organism}"
 
     # Use environment variables if available

preprocessing/nextclade/src/loculus_preprocessing/prepro.py

@@ -52,7 +52,9 @@ def parse_ndjson(ndjson_data: str) -> Sequence[UnprocessedEntry]:
     for json_str in ndjson_data.split("\n"):
         if len(json_str) == 0:
             continue
-        json_object = json.loads(json_str)
+        # Loculus currently cannot handle non-breaking spaces.
+        json_str_processed = json_str.replace("\N{NO-BREAK SPACE}", " ")
+        json_object = json.loads(json_str_processed)
         unprocessed_data = UnprocessedData(
             metadata=json_object["data"]["metadata"],
             unalignedNucleotideSequences=json_object["data"]["unalignedNucleotideSequences"],