make id generator consistent

GraphScope · Jan 20, 2025 · 4f4a35c · 4f4a35c
1 parent 0665ee2
commit 4f4a35c
Show file tree

Hide file tree

Showing 3 changed files with 3 additions and 2 deletions.
diff --git a/python/graphy/utils/cryptography.py b/python/graphy/utils/cryptography.py
@@ -7,6 +7,7 @@
 
 
 def id_generator(value: str, length: int = 16) -> str:
+    value = value.lower()
     return hashlib.sha256(value.encode()).hexdigest()[:length]
 
 

diff --git a/python/graphy/utils/data_extractor.py b/python/graphy/utils/data_extractor.py
@@ -159,7 +159,7 @@ def extract_data(self, dimension_node_names=[]):
                 if not paper_data.get("id", ""):
                     # this is a bug of inconsistent data
                     if "title" in paper_data:
-                        paper_data["id"] = id_generator(paper_data["title"].lower())
+                        paper_data["id"] = id_generator(paper_data["title"])
                     else:
                         continue
                 # some hacking messy stuff

diff --git a/python/graphy/utils/paper_struct.py b/python/graphy/utils/paper_struct.py
@@ -123,7 +123,7 @@ def parse_dict(meta: dict):
 
         paper_id = parsed_meta.get("id", "")
         if paper_id == "":
-            parsed_meta["id"] = id_generator(parsed_meta.get("title", "").lower())
+            parsed_meta["id"] = id_generator(parsed_meta.get("title", ""))
 
         if not parsed_meta.get("published", ""):
             try:
Original file line number	Diff line number	Diff line change
Expand Up		@@ -7,6 +7,7 @@


		def id_generator(value: str, length: int = 16) -> str:
		value = value.lower()
		return hashlib.sha256(value.encode()).hexdigest()[:length]


Expand Down