choosealicense in badges. Fixes #533

src/somef/regular_expressions.py

@@ -181,6 +181,31 @@ def extract_support_channels(readme_text, repository_metadata: Result, readme_so
 
     return repository_metadata
 
+def extract_license_badges(readme_text, repository_metadata, readme_source):
+    """
+    Extracts license information from choosealicense.com URLs found in README badges.
+    """
+    matches = re.finditer(constants.REGEXP_CHOOSE_LICENSE, readme_text)
+    for match in matches:
+        license_url = match.group(0)  # full URL
+        license_info = detect_license_spdx(license_url, 'HEADER')
+        if license_info:
+            result = {
+                constants.PROP_VALUE: license_info['spdx_id'],
+                constants.PROP_TYPE: constants.PROP_LICENSE,
+                constants.PROP_NAME: license_info['name'],
+                constants.PROP_SPDX_ID: license_info['spdx_id'],
+                constants.PROP_URL: license_info['url'],
+                constants.PROP_IDENTIFIER: license_info['url'],
+            }
+            repository_metadata.add_result(
+                constants.CAT_LICENSE,
+                result,
+                1,
+                constants.TECHNIQUE_REGULAR_EXPRESSION,
+                readme_source
+            )
+    return repository_metadata
 
 def extract_repo_status(unfiltered_text, repository_metadata: Result, readme_source) -> Result:
     """
@@ -995,7 +1020,29 @@ def detect_license_spdx(license_text, type):
     -------
     A JSON dictionary with name and spdx id
     """
-
+    match = re.search(constants.REGEXP_CHOOSE_LICENSE, license_text)
+    if match:
+        slug = match.group(1).lower().rstrip('/')
+        for license_name, license_info in constants.LICENSES_DICT.items():
+            if license_info['spdx_id'].lower() == slug:
+                spdx_id = license_info['spdx_id']
+                spdx_url = f"https://spdx.org/licenses/{spdx_id}"
+                if type == 'JSON':
+                    return {
+                        "name": license_name,
+                        "spdx_id": spdx_id,
+                        "@id": spdx_url,
+                        "url": spdx_url,
+                        "identifier": spdx_url
+                    }
+                else:
+                    return {
+                        "name": license_name,
+                        "spdx_id": spdx_id,
+                        "identifier": spdx_url,
+                        "url": spdx_url
+                    }
+
     for license_name, license_info in constants.LICENSES_DICT.items():
         if re.search(license_info["regex"], license_text, re.IGNORECASE):
             spdx_id = license_info['spdx_id']

src/somef/somef_cli.py

@@ -230,6 +230,8 @@ def cli_get_data(threshold, ignore_classifiers, repo_url=None, doc_src=None, loc
                                                                      repository_metadata, readme_source, def_branch)
             repository_metadata = regular_expressions.extract_arxiv_links(readme_unfiltered_text, repository_metadata,
                                                                           readme_source)
+            repository_metadata = regular_expressions.extract_license_badges(readme_unfiltered_text, repository_metadata, readme_source)
+
             logging.info("Completed extracting regular expressions")
 
         return repository_metadata

src/somef/test/test_JSON_export.py

@@ -1004,3 +1004,38 @@ def test_issue_980_reconciliation_requirements(self):
         os.remove(output_path)
 
 
+    def test_issue_533_choosealicense_badge(self):
+        """
+        Checks that a license badge with a choosealicense.com URL is detected and resolved to SPDX.
+        """
+        output_path = test_data_path + "test_issue_533_choosealicense_badge.json"
+
+        somef_cli.run_cli(threshold=0.8,
+                            ignore_classifiers=False,
+                            repo_url=None,
+                            local_repo=None,
+                            doc_src=test_data_path + "README-manim.md",
+                            in_file=None,
+                            output=output_path,
+                            graph_out=None,
+                            graph_format="turtle",
+                            codemeta_out=None,
+                            pretty=True,
+                            missing=False,
+                            readme_only=False)
+
+        with open(output_path, "r") as f:
+            json_content = json.loads(f.read())
+
+        licenses = json_content.get(constants.CAT_LICENSE, [])
+
+        mit_license = next(
+            (l for l in licenses if l[constants.PROP_RESULT].get(constants.PROP_SPDX_ID) == "MIT"),
+            None
+        )
+        assert mit_license is not None, f"Expected a MIT license resolved from choosealicense.com badge, got: {licenses}"
+
+        techniques = mit_license.get("technique", [])
+        assert constants.TECHNIQUE_REGULAR_EXPRESSION in techniques, f"Expected 'regular_expressions' in techniques, got: {techniques}"
+
+        os.remove(output_path)

src/somef/utils/constants.py

@@ -77,6 +77,9 @@
 REGEXP_MPL2 = r'(?i)mozilla\s+public\s+license\s*,?\s*version\s*2\.0'
 REGEXP_UNLICENSE = r'(?i)the\s+unlicense'
 
+# detect choosealicense in badges
+REGEXP_CHOOSE_LICENSE = r'choosealicense\.com/licenses/([^/\s]+)'
+
 # Detect organization in authors.md
 # REGEXP_LTD_INC = r'\b(inc|ltd|llc|corporation)([.,]|\b)'
 REGEXP_LTD_INC = r'\b(inc|ltd|llc|corporation|foundation|community|project|team|group|society|institute|association|consortium|organization|organisation)([.,]|\b)'