remove

score/app.py

@@ -52,7 +52,7 @@ def pypi(package_name):
 def pypi_score(package_name, source_url: Optional[str] = None):
     package_data = get_pypi_package_data_cached(package_name)
 
-    if source_url is None:
+    if not source_url:
         source_url = package_data.get("source_url")
     source_data = None
     if source_url:
@@ -81,7 +81,7 @@ def npm(package_name):
 def npm_score(package_name, source_url: Optional[str] = None):
     package_data = get_npm_package_data_cached(package_name)
 
-    if source_url is None:
+    if not source_url:
         source_url = package_data.get("source_url")
     source_data = None
     if source_url:
@@ -114,7 +114,7 @@ def conda(channel, package_name):
 def conda_score(channel, package_name, source_url: Optional[str] = None):
     package_data = get_conda_package_data_cached(channel, package_name)
 
-    if source_url is None:
+    if not source_url:
         source_url = package_data.get("source_url")
     source_data = None
     if source_url:

score/git_vcs/license_detection.py

@@ -1,4 +1,5 @@
 import pandas as pd
+from typing import Union
 from pathlib import Path
 from functools import lru_cache
 from strsimpy import SorensenDice
@@ -116,13 +117,19 @@
 
 def identify_license(license_content: str) -> dict:
 
+    license_content_without_copyright = "".join(
+        [line for line in license_content.splitlines() if not copyright_line(line)]
+    )
+
     sd = SorensenDice()
     similarities = []
     for license_name, ref_license in get_all_licenses().items():
         similarities.append(
             {
                 "name": license_name,
-                "similarity": sd.similarity(license_content.strip(), ref_license),
+                "similarity": sd.similarity(
+                    license_content_without_copyright.strip(), ref_license
+                ),
             }
         )
     similarities = pd.DataFrame(similarities).set_index("name")
@@ -152,12 +159,22 @@ def identify_license(license_content: str) -> dict:
 license_dir = Path(__file__).parent / "licenses"
 
 
+def copyright_line(line: Union[bytes | str]):
+    copyright = b"copyright" if isinstance(line, bytes) else "copyright"
+    if line.strip().lower().startswith(copyright):
+        return True
+    return False
+
+
 @lru_cache
 def get_all_licenses():
     licenses = {}
     for license_file in license_dir.glob("*"):
         with open(license_file, "rb") as f:
-            data = f.read()
+            lines = f.readlines()
+            lines = [line for line in lines if not copyright_line(line)]
+            data = b"".join(lines)
+
         data = data.decode(errors="ignore")
         licenses[license_file.name] = data.strip()
     return licenses

score/score/app_score.py

@@ -1,5 +1,5 @@
 from datetime import timedelta
-
+import re
 from .score import safe_date_diff
 from .maturity import build_maturity_score
 from .legal import build_legal_score
@@ -10,13 +10,20 @@
 from ..notes import Note
 
 
+def pypi_normalize(name):
+    if not name:
+        return None
+
+    return re.sub(r"[-_.]+", "-", name).lower()
+
+
 def score_python(package_data: dict, source_data: dict, score: Score):
 
     if not package_data:
         return
 
-    expected_name = source_data.get("py_package")
-    actual_name = package_data.get("name")
+    expected_name = pypi_normalize(source_data.get("py_package"))
+    actual_name = pypi_normalize(package_data.get("name"))
 
     if not expected_name:
         score.add_note(Note.NO_PROJECT_NAME)