Bug fix/Patch: firstnames with no spaces

Author: Lucaterre

.gitignore

@@ -179,4 +179,5 @@ cython_debug/
 #index_dil/
 *.seg
 *.toc
-*WRITELOCK
+*WRITELOCK
+index_dil/*

api/api_utils.py

@@ -0,0 +1,34 @@
+"""api_utils.py
+
+Utility functions for the API.
+"""
+import re
+
+_COMMA_FIX = re.compile(r'\s*,+\s*')
+_MULTI_SPACE = re.compile(r'\s{2,}')
+
+def normalize_firstnames(v: str | None) -> str | None:
+    """Normalize firstnames by removing extra spaces and fixing commas.
+
+    Examples:
+    >>> normalize_firstnames("  Auguste,Titus  ")
+    'Auguste, Titus'
+    >>> normalize_firstnames("  Auguste ,  Titus  ")
+    'Auguste, Titus'
+    >>> normalize_firstnames("  Auguste  , Titus, ")
+    'Auguste, Titus'
+    >>> normalize_firstnames(None)
+    None
+
+    :param v: The input string to normalize.
+    :type v: str | None
+    :return: The normalized string or None if input is None.
+    """
+    if not v:
+        return v
+    s1 = v.strip()
+    s = _COMMA_FIX.sub(', ', s1)   # "Auguste,Titus" -> "Auguste, Titus" ; "Auguste ,  Titus" -> "Auguste, Titus"
+    s = re.sub(r',\s*$', '', s)   # remove final comma
+    s = _MULTI_SPACE.sub(' ', s)  # remove multiple spaces
+    #print(s1," > ", s, " : ", s == s1)
+    return s

api/crud.py

@@ -19,6 +19,7 @@
                             Address,
                             Image)
 from .models.constants import type_patent_relations
+from api.api_utils import normalize_firstnames
 
 MARKUP_HTML_FIELDS = {'personal_information', 'professional_information'}
 inverted_type_relations = {v: k for k, v in type_patent_relations.items()}
@@ -67,7 +68,7 @@ def enhance_patent_response(db: Session,
                 'patent_relations': [{
                     '_id_dil': str(patent_relation.person_related._id_dil) if patent_relation.person_related._id_dil else None,
                     'lastname': patent_relation.person_related.lastname,
-                    'firstnames': patent_relation.person_related.firstnames,
+                    'firstnames': normalize_firstnames(patent_relation.person_related.firstnames),
                     'type': inverted_type_relations.get(patent_relation.type)
 
                 } for patent_relation in patent.patent_relations if len(patent.patent_relations) > 0]
@@ -88,7 +89,7 @@ def enhance_printer_response(db: Session,
     return {
         "_id_dil": str(printer._id_dil) if printer._id_dil else None,
         "lastname": printer.lastname,
-        "firstnames": printer.firstnames,
+        "firstnames": normalize_firstnames(printer.firstnames),
         "birth_date": printer.birth_date,
         "birth_city_label": printer.birth_city_label,
         "birth_city_id": str(get_city(db, {
@@ -135,7 +136,7 @@ def get_printers(db: Session, args: dict, enhance: bool = False):
                 {
                     "_id_dil": str(printer._id_dil) if printer._id_dil else None,
                     "lastname": printer.lastname,
-                    "firstnames": printer.firstnames,
+                    "firstnames": normalize_firstnames(printer.firstnames),
                     "total_patents": len(printer.patents),
                 }
                 for printer in printers

api/routes.py

@@ -48,6 +48,7 @@
                                City,
                                Address)
 from api.index_fts.search_utils import search_whoosh
+from api.api_utils import normalize_firstnames
 
 api_router = APIRouter()
 
@@ -170,7 +171,7 @@ def get_cities_with_printers(
             if person_key not in city_map[city_id]["persons"]:
                 city_map[city_id]["persons"][person_key] = {
                     "id": person_key,
-                    "firstnames": row.firstnames,
+                    "firstnames": normalize_firstnames(row.firstnames),
                     "lastname": row.lastname,
                     "city_patent": row.patent_city_label
                 }
@@ -311,6 +312,7 @@ def make_cache_key(name: str, content: str):
     raw = f"{name or ''}|{content or ''}"
     return hashlib.sha256(raw.encode()).hexdigest()
 
+
 @cached(cache=cache, key=lambda name, content: make_cache_key(name, content))
 def cached_search(name, content):
     return search_whoosh(query_firstnames_lastname=name, query_content=content)
@@ -407,12 +409,13 @@ def read_printers(
         transformed_items = []
         for p in paginated.items:
             highlight = whoosh_hits.get(str(p.id_dil), {}).get("highlight")
+            firstnames = normalize_firstnames(p.firstnames)
 
             transformed_items.append(
                 PrinterMinimalResponseOut(
                     _id_dil=str(p.id_dil),
                     lastname=p.lastname,
-                    firstnames=p.firstnames,
+                    firstnames=firstnames,
                     total_patents=p.total_patents,
                     highlight=highlight
                 )
@@ -427,6 +430,7 @@ def read_printers(
         )
 
     except Exception as e:
+        print(e)
         import traceback
         return JSONResponse(status_code=500, content={"message": f"Erreur serveur: {e}"})
 

api/schemas.py

@@ -8,6 +8,7 @@
 from typing import Union, List, Optional
 from pydantic import BaseModel, Field
 
+
 class BaseMeta(BaseModel):
     """An abstract base class for meta schemas."""
     id_dil: str = Field(..., alias="_id_dil")