Derive variant display names from upstream data

tools/device_mapper.py

@@ -463,6 +463,40 @@ def extract_codes(text: str) -> List[str]:
     return [code.strip() for code in BACKTICK_RE.findall(text) if code.strip()]
 
 
+def infer_base_variant_name(variant_name: str, entry_title: str) -> Optional[str]:
+    base = re.split(r"\s+/\s+", variant_name.strip(), maxsplit=1)[0].strip()
+    if not base:
+        return None
+
+    base = re.sub(r"\s*(?:国行版|国内版|中国版|印度版|欧洲版|国际版|北美版|日本版|韩国版|港版|台版|海外版)\s*$", "", base)
+    base = re.sub(
+        r"\s+(?:China|Chinese|India|Europe|European|Global|International|North America|North American|Japan|Korea|Hong Kong|Taiwan|US|USA|T-Mobile|Verizon|AT&T|SIM Free|SoftBank)\s*$",
+        "",
+        base,
+        flags=re.IGNORECASE,
+    )
+    base = " ".join(base.split())
+    if not base or normalize_text(base) not in normalize_text(entry_title):
+        return None
+    return base
+
+
+def split_variant_groups(entry_title: str, title_codes: Iterable[str], variants: list[tuple[list[str], str]]) -> list[tuple[str, Set[str]]]:
+    groups: dict[str, Set[str]] = {}
+    for variant_codes, variant_name in variants:
+        base_name = infer_base_variant_name(variant_name, entry_title)
+        if not base_name:
+            return []
+        aliases = groups.setdefault(base_name, set(title_codes))
+        aliases.add(base_name)
+        aliases.add(variant_name)
+        aliases.update(variant_codes)
+
+    if len(groups) < 2:
+        return []
+    return list(groups.items())
+
+
 def parse_brand_file(path: Path) -> List[DeviceRecord]:
     file_stem = path.stem
     brand = canonical_brand(file_stem)
@@ -473,43 +507,50 @@ def parse_brand_file(path: Path) -> List[DeviceRecord]:
 
     section = ""
     current_title = ""
+    current_title_codes: List[str] = []
     current_aliases: Set[str] = set()
+    current_variants: list[tuple[list[str], str]] = []
 
     def flush_current() -> None:
-        nonlocal current_title, current_aliases
+        nonlocal current_title, current_title_codes, current_aliases, current_variants
         if not current_title:
             return
 
-        aliases = sorted({alias.strip() for alias in current_aliases if alias.strip()})
-        record_id = f"{file_stem}:{len(records) + 1}"
-        device_type = infer_device_type(
-            device_name=current_title,
-            section=section,
-            source_file=path.name,
-            aliases=aliases,
-            default_type=default_type,
-        )
-        records.append(
-            DeviceRecord(
-                id=record_id,
-                device_name=current_title,
-                brand=brand,
-                manufacturer_brand=brand,
-                parent_brand=resolve_parent_brand(brand),
-                market_brand=infer_market_brand(
-                    manufacturer_brand=brand,
-                    device_name=current_title,
-                    section=section,
-                    aliases=aliases,
-                ),
-                device_type=device_type,
-                aliases=aliases,
-                source_file=f"brands/{path.name}",
+        split_groups = split_variant_groups(current_title, current_title_codes, current_variants)
+        record_groups = split_groups or [(current_title, current_aliases)]
+        for device_name, raw_aliases in record_groups:
+            aliases = sorted({alias.strip() for alias in raw_aliases if alias.strip()})
+            record_id = f"{file_stem}:{len(records) + 1}"
+            device_type = infer_device_type(
+                device_name=device_name,
                 section=section,
+                source_file=path.name,
+                aliases=aliases,
+                default_type=default_type,
+            )
+            records.append(
+                DeviceRecord(
+                    id=record_id,
+                    device_name=device_name,
+                    brand=brand,
+                    manufacturer_brand=brand,
+                    parent_brand=resolve_parent_brand(brand),
+                    market_brand=infer_market_brand(
+                        manufacturer_brand=brand,
+                        device_name=device_name,
+                        section=section,
+                        aliases=aliases,
+                    ),
+                    device_type=device_type,
+                    aliases=aliases,
+                    source_file=f"brands/{path.name}",
+                    section=section,
+                )
             )
-        )
         current_title = ""
+        current_title_codes = []
         current_aliases = set()
+        current_variants = []
 
     for raw in lines:
         line = raw.strip()
@@ -527,8 +568,10 @@ def parse_brand_file(path: Path) -> List[DeviceRecord]:
             flush_current()
             raw_title = entry_match.group(1).strip()
             current_title = clean_entry_title(raw_title)
-            current_aliases = set(extract_codes(raw_title))
+            current_title_codes = extract_codes(raw_title)
+            current_aliases = set(current_title_codes)
             current_aliases.add(current_title)
+            current_variants = []
             continue
 
         if not current_title:
@@ -538,6 +581,7 @@ def parse_brand_file(path: Path) -> List[DeviceRecord]:
         if variant_match:
             variant_codes = extract_codes(variant_match.group(1))
             variant_name = variant_match.group(2).strip()
+            current_variants.append((variant_codes, variant_name))
             current_aliases.update(variant_codes)
             current_aliases.add(variant_name)