Reference: #3954

Signed-off-by: Ayan Sinha Mahapatra <[email protected]>

Signed-off-by: Jono Yang <[email protected]>

Signed-off-by: Jono Yang <[email protected]>

addded support to parse labels in dockerfile

Signed-off-by: Varsha U N <[email protected]>

Author: AyanSinhaMahapatra

requirements-linux.txt

@@ -1,4 +1,4 @@
 packagedcode-msitools==0.101.210706
 regipy==3.1.0
 rpm-inspector-rpm==4.16.1.3.210404
-go-inspector==0.3.1
+go-inspector==0.5.0

requirements.txt

@@ -35,7 +35,7 @@ jaraco.functools==4.1.0
 javaproperties==0.8.1
 Jinja2==3.1.3
 jsonstreams==0.6.0
-license-expression==30.3.0
+license-expression==30.4.0
 lxml==5.1.0
 MarkupSafe==2.1.5
 more-itertools==8.13.0

setup-mini.cfg

@@ -149,7 +149,7 @@ packages =
     rpm_inspector_rpm >= 4.16.1.3; platform_system == 'Linux'
     regipy >= 3.1.0; platform_system == 'Linux'
     packagedcode_msitools >= 0.101.210706; platform_system == 'Linux'
-    go-inspector >= 0.3.1; platform_system == 'Linux'
+    go-inspector >= 0.5.0; platform_system == 'Linux'
 
 
 [options.entry_points]

setup.cfg

@@ -84,7 +84,7 @@ install_requires =
     javaproperties >= 0.5
     jinja2 >= 2.7.0
     jsonstreams >= 0.5.0
-    license_expression >= 30.1.1
+    license_expression >= 30.4.0
     lxml >= 4.9.2
     MarkupSafe >= 2.1.2
     packageurl_python >= 0.9.0
@@ -149,7 +149,7 @@ packages =
     rpm_inspector_rpm >= 4.16.1.3; platform_system == 'Linux'
     regipy >= 3.1.0; platform_system == 'Linux'
     packagedcode_msitools >= 0.101.210706; platform_system == 'Linux'
-    go-inspector >= 0.3.1; platform_system == 'Linux'
+    go-inspector >= 0.5.0; platform_system == 'Linux'
 
 
 [options.entry_points]

src/licensedcode/tokenize.py

@@ -336,6 +336,7 @@ def select_ngrams(ngrams, with_pos=False):
     >>> list(select_ngrams(x for x in [(2, 1, 3), (1, 1, 3), (5, 1, 3), (2, 6, 1), (7, 3, 4)]))
     [(2, 1, 3), (1, 1, 3), (5, 1, 3), (2, 6, 1), (7, 3, 4)]
     """
+    ngram = None
     last = None
     for pos, ngram in enumerate(ngrams):
         # FIXME: use a proper hash

src/packagedcode/__init__.py

@@ -20,6 +20,7 @@
 from packagedcode import debian
 from packagedcode import debian_copyright
 from packagedcode import distro
+from packagedcode import dockerfile
 from packagedcode import conda
 from packagedcode import conan
 from packagedcode import cocoapods
@@ -97,6 +98,7 @@
     debian.DebianSourcePackageTarballHandler,
 
     distro.EtcOsReleaseHandler,
+    dockerfile.DockerfileHandler,
 
     freebsd.CompactManifestHandler,
 

src/packagedcode/dockerfile.py

@@ -0,0 +1,59 @@
+#
+# Copyright (c) nexB Inc. and others. All rights reserved.
+# ScanCode is a trademark of nexB Inc.
+# SPDX-License-Identifier: Apache-2.0
+# See http://www.apache.org/licenses/LICENSE-2.0 for the license text.
+# See https://github.com/nexB/scancode-toolkit for support or download.
+# See https://aboutcode.org for more information about nexB OSS projects.
+#
+
+
+
+import io
+from pathlib import Path
+from dockerfile_parse import DockerfileParser  
+from packagedcode import models
+from packagedcode import utils
+import fnmatch
+
+
+class DockerfileHandler(models.DatafileHandler):
+    datasource_id = 'dockerfile_oci_labels'
+
+    @classmethod 
+    def is_datafile(cls, path): 
+        patterns = ['Dockerfile', 'containerfile', '*.dockerfile'] 
+        filename=os.path.basename(path)
+        for pattern in patterns: 
+            if fnmatch.fnmatch(filename, pattern): 
+                return True 
+        return False
+    
+    @classmethod
+    def parse(cls, location, package_only=False):
+        """
+        Parse a Dockerfile and yield one or more PackageData objects with OCI labels and metadata.
+        """
+        labels = cls.extract_oci_labels_from_dockerfile(location)
+        package_data = {
+            'datasource_id': cls.datasource_id,
+            'type': cls.default_package_type,
+            'name': labels.get('name', 'None'),  
+            'version': labels.get('version', 'None'),  
+            'license_expression': labels.get('license', 'None'),
+            'labels': labels,  
+        }
+
+        yield models.PackageData.from_data(package_data, package_only)
+
+    @classmethod
+    def extract_oci_labels_from_dockerfile(cls, dockerfile_path):
+        """
+        Extract OCI labels from the Dockerfile using DockerfileParser.
+        """
+        labels = {}
+        parser = DockerfileParser()
+        with open(dockerfile_path, 'r') as dockerfile:
+          parser.content = dockerfile.read() 
+          labels = parser.labels
+        return labels

src/packagedcode/recognize.py

@@ -113,3 +113,11 @@ def _parse(
 
             if TRACE:
                 raise
+
+        except Exception as e:
+            # We should continue when an Exception has occured when trying to
+            # recognize a package
+            if TRACE:
+                logger_debug(f'_parse: Exception: {str(e)}')
+
+            continue

tests/packagedcode/__pycache__/test_parse_pyproject_toml.cpython-312-pytest-8.3.3.pyc.26520

@@ -0,0 +1,13 @@
+[
+    {
+        "datasource_id": "dockerfile_oci_labels",
+        "type": "default",  
+        "name": "Unknown",  
+        "version": "Unknown",  
+        "license_expression": "GPL-2.0-only AND BSD-2-Clause",
+        "labels": {
+            "source": "https://github.com/kubernetes-sigs/blixt",
+            "licenses": "GPL-2.0-only,BSD-2-Clause"
+        }
+    }
+]