only instantiate get_spdx_licensing() once per module

this takes quite some time and should be done as few times as possible

Signed-off-by: Armin Tänzer <armin.taenzer@tngtech.com>

Author: armintaenzertng

examples/spdx2_document_from_scratch.py

@@ -49,6 +49,7 @@
 # The document currently does not describe anything. Let's create a package that we can add to it.
 # The Package class has quite a few properties (have a look there!),
 # but only name, spdx_id and download_location are mandatory in SPDX v2.3.
+spdx_licensing = get_spdx_licensing()  # this getter takes quite long and should be called as few times as possible
 package = Package(
     name="package name",
     spdx_id="SPDXRef-Package",
@@ -65,9 +66,9 @@
         Checksum(ChecksumAlgorithm.SHA1, "d6a770ba38583ed4bb4525bd96e50461655d2758"),
         Checksum(ChecksumAlgorithm.MD5, "624c1abb3664f4b35547e7c73864ad24"),
     ],
-    license_concluded=get_spdx_licensing().parse("GPL-2.0-only OR MIT"),
-    license_info_from_files=[get_spdx_licensing().parse("GPL-2.0-only"), get_spdx_licensing().parse("MIT")],
-    license_declared=get_spdx_licensing().parse("GPL-2.0-only AND MIT"),
+    license_concluded=spdx_licensing.parse("GPL-2.0-only OR MIT"),
+    license_info_from_files=[spdx_licensing.parse("GPL-2.0-only"), spdx_licensing.parse("MIT")],
+    license_declared=spdx_licensing.parse("GPL-2.0-only AND MIT"),
     license_comment="license comment",
     copyright_text="Copyright 2022 Jane Doe",
     description="package description",
@@ -100,8 +101,8 @@
         Checksum(ChecksumAlgorithm.SHA1, "d6a770ba38583ed4bb4525bd96e50461655d2758"),
         Checksum(ChecksumAlgorithm.MD5, "624c1abb3664f4b35547e7c73864ad24"),
     ],
-    license_concluded=get_spdx_licensing().parse("MIT"),
-    license_info_in_file=[get_spdx_licensing().parse("MIT")],
+    license_concluded=spdx_licensing.parse("MIT"),
+    license_info_in_file=[spdx_licensing.parse("MIT")],
     copyright_text="Copyright 2022 Jane Doe",
 )
 file2 = File(
@@ -110,7 +111,7 @@
     checksums=[
         Checksum(ChecksumAlgorithm.SHA1, "d6a770ba38583ed4bb4525bd96e50461655d2759"),
     ],
-    license_concluded=get_spdx_licensing().parse("GPL-2.0-only"),
+    license_concluded=spdx_licensing.parse("GPL-2.0-only"),
 )
 
 # Assuming the package contains those two files, we create two CONTAINS relationships.

src/spdx_tools/spdx/validation/license_expression_validator.py

@@ -8,6 +8,8 @@
 from spdx_tools.spdx.model import Document, SpdxNoAssertion, SpdxNone
 from spdx_tools.spdx.validation.validation_message import SpdxElementType, ValidationContext, ValidationMessage
 
+spdx_licensing = get_spdx_licensing()
+
 
 def validate_license_expressions(
     license_expressions: List[Union[LicenseExpression, SpdxNoAssertion, SpdxNone]], document: Document, parent_id: str
@@ -40,7 +42,7 @@ def validate_license_expression(
     validation_messages = []
     license_ref_ids: List[str] = [license_ref.license_id for license_ref in document.extracted_licensing_info]
 
-    for non_spdx_token in get_spdx_licensing().validate(license_expression).invalid_symbols:
+    for non_spdx_token in spdx_licensing.validate(license_expression).invalid_symbols:
         if non_spdx_token not in license_ref_ids:
             validation_messages.append(
                 ValidationMessage(
@@ -51,14 +53,14 @@ def validate_license_expression(
             )
 
     try:
-        get_spdx_licensing().parse(str(license_expression), validate=True, strict=True)
+        spdx_licensing.parse(str(license_expression), validate=True, strict=True)
     except ExpressionParseError as err:
         # This error is raised when an exception symbol is used as a license symbol and vice versa.
         # So far, it only catches the first such error in the provided string.
         validation_messages.append(ValidationMessage(f"{err}. for license_expression: {license_expression}", context))
     except ExpressionError:
         # This error is raised for invalid symbols within the license_expression, but it provides only a string of
-        # these. On the other hand, get_spdx_licensing().validate() gives an actual list of invalid symbols, so this is
+        # these. On the other hand, spdx_licensing.validate() gives an actual list of invalid symbols, so this is
         # handled above.
         pass
 

src/spdx_tools/spdx/writer/rdf/license_expression_writer.py

@@ -18,6 +18,8 @@
 from spdx_tools.spdx.model import SpdxNoAssertion, SpdxNone
 from spdx_tools.spdx.rdfschema.namespace import LICENSE_NAMESPACE, SPDX_NAMESPACE
 
+spdx_licensing = get_spdx_licensing()
+
 
 def add_license_expression_or_none_or_no_assertion(
     value: Union[
@@ -75,7 +77,7 @@ def add_license_expression_to_graph(
 
 
 def license_or_exception_is_on_spdx_licensing_list(license_symbol: LicenseSymbol) -> bool:
-    symbol_info: ExpressionInfo = get_spdx_licensing().validate(license_symbol)
+    symbol_info: ExpressionInfo = spdx_licensing.validate(license_symbol)
     return not symbol_info.errors
 
 

src/spdx_tools/spdx3/bump_from_spdx2/license_expression.py

@@ -28,6 +28,8 @@
 )
 from spdx_tools.spdx.model import ExtractedLicensingInfo, SpdxNoAssertion, SpdxNone
 
+spdx_licensing = get_spdx_licensing()
+
 
 def bump_license_expression_or_none_or_no_assertion(
     element: Union[LicenseExpression, SpdxNoAssertion, SpdxNone],
@@ -61,7 +63,7 @@ def bump_license_expression(
             subject_addition=bump_license_exception(license_expression.exception_symbol, extracted_licensing_info),
         )
     if isinstance(license_expression, LicenseSymbol):
-        if not get_spdx_licensing().validate(license_expression).invalid_symbols:
+        if not spdx_licensing.validate(license_expression).invalid_symbols:
             return ListedLicense(license_expression.key, license_expression.obj, "blank")
         else:
             for licensing_info in extracted_licensing_info:
@@ -80,7 +82,7 @@ def bump_license_expression(
 def bump_license_exception(
     license_exception: LicenseSymbol, extracted_licensing_info: List[ExtractedLicensingInfo]
 ) -> LicenseAddition:
-    if not get_spdx_licensing().validate(license_exception).invalid_symbols:
+    if not spdx_licensing.validate(license_exception).invalid_symbols:
         return ListedLicenseException(license_exception.key, "", "")
     else:
         for licensing_info in extracted_licensing_info:

tests/spdx/examples/test_spdx2_document_from_scratch.py

@@ -51,6 +51,7 @@ def test_spdx2_document_from_scratch():
     # The document currently does not describe anything. Let's create a package that we can add to it.
     # The Package class has quite a few properties (have a look there!),
     # but only name, spdx_id and download_location are mandatory in SPDX v2.3.
+    spdx_licensing = get_spdx_licensing()  # this getter takes quite long and should be called as few times as possible
     package = Package(
         name="package name",
         spdx_id="SPDXRef-Package",
@@ -67,9 +68,9 @@ def test_spdx2_document_from_scratch():
             Checksum(ChecksumAlgorithm.SHA1, "d6a770ba38583ed4bb4525bd96e50461655d2758"),
             Checksum(ChecksumAlgorithm.MD5, "624c1abb3664f4b35547e7c73864ad24"),
         ],
-        license_concluded=get_spdx_licensing().parse("GPL-2.0-only OR MIT"),
-        license_info_from_files=[get_spdx_licensing().parse("GPL-2.0-only"), get_spdx_licensing().parse("MIT")],
-        license_declared=get_spdx_licensing().parse("GPL-2.0-only AND MIT"),
+        license_concluded=spdx_licensing.parse("GPL-2.0-only OR MIT"),
+        license_info_from_files=[spdx_licensing.parse("GPL-2.0-only"), spdx_licensing.parse("MIT")],
+        license_declared=spdx_licensing.parse("GPL-2.0-only AND MIT"),
         license_comment="license comment",
         copyright_text="Copyright 2022 Jane Doe",
         description="package description",
@@ -102,8 +103,8 @@ def test_spdx2_document_from_scratch():
             Checksum(ChecksumAlgorithm.SHA1, "d6a770ba38583ed4bb4525bd96e50461655d2758"),
             Checksum(ChecksumAlgorithm.MD5, "624c1abb3664f4b35547e7c73864ad24"),
         ],
-        license_concluded=get_spdx_licensing().parse("MIT"),
-        license_info_in_file=[get_spdx_licensing().parse("MIT")],
+        license_concluded=spdx_licensing.parse("MIT"),
+        license_info_in_file=[spdx_licensing.parse("MIT")],
         copyright_text="Copyright 2022 Jane Doe",
     )
     file2 = File(
@@ -112,7 +113,7 @@ def test_spdx2_document_from_scratch():
         checksums=[
             Checksum(ChecksumAlgorithm.SHA1, "d6a770ba38583ed4bb4525bd96e50461655d2759"),
         ],
-        license_concluded=get_spdx_licensing().parse("GPL-2.0-only"),
+        license_concluded=spdx_licensing.parse("GPL-2.0-only"),
     )
 
     # Assuming the package contains those two files, we create two CONTAINS relationships.

tests/spdx/fixtures.py

@@ -32,6 +32,9 @@
     Version,
 )
 
+spdx_licensing = get_spdx_licensing()
+
+
 # Utility methods to create data model instances. All properties have valid defaults, so they don't need to be
 # specified unless relevant for the test.
 
@@ -88,7 +91,7 @@ def file_fixture(
     spdx_id="SPDXRef-File",
     checksums=None,
     file_types=None,
-    license_concluded=get_spdx_licensing().parse("MIT and GPL-2.0"),
+    license_concluded=spdx_licensing.parse("MIT and GPL-2.0"),
     license_info_in_file=None,
     license_comment="licenseComment",
     copyright_text="copyrightText",
@@ -100,7 +103,7 @@ def file_fixture(
     checksums = [checksum_fixture()] if checksums is None else checksums
     file_types = [FileType.TEXT] if file_types is None else file_types
     license_info_in_file = (
-        [get_spdx_licensing().parse("MIT"), get_spdx_licensing().parse("GPL-2.0"), SpdxNoAssertion()]
+        [spdx_licensing.parse("MIT"), spdx_licensing.parse("GPL-2.0"), SpdxNoAssertion()]
         if license_info_in_file is None
         else license_info_in_file
     )
@@ -135,9 +138,9 @@ def package_fixture(
     checksums=None,
     homepage="https://homepage.com",
     source_info="sourceInfo",
-    license_concluded=get_spdx_licensing().parse("MIT and GPL-2.0"),
+    license_concluded=spdx_licensing.parse("MIT and GPL-2.0"),
     license_info_from_files=None,
-    license_declared=get_spdx_licensing().parse("MIT and GPL-2.0"),
+    license_declared=spdx_licensing.parse("MIT and GPL-2.0"),
     license_comment="packageLicenseComment",
     copyright_text="packageCopyrightText",
     summary="packageSummary",
@@ -152,7 +155,7 @@ def package_fixture(
 ) -> Package:
     checksums = [checksum_fixture()] if checksums is None else checksums
     license_info_from_files = (
-        [get_spdx_licensing().parse("MIT"), get_spdx_licensing().parse("GPL-2.0"), SpdxNoAssertion()]
+        [spdx_licensing.parse("MIT"), spdx_licensing.parse("GPL-2.0"), SpdxNoAssertion()]
         if license_info_from_files is None
         else license_info_from_files
     )
@@ -208,7 +211,7 @@ def snippet_fixture(
     file_spdx_id="SPDXRef-File",
     byte_range=(1, 2),
     line_range=(3, 4),
-    license_concluded=get_spdx_licensing().parse("MIT and GPL-2.0"),
+    license_concluded=spdx_licensing.parse("MIT and GPL-2.0"),
     license_info_in_snippet=None,
     license_comment="snippetLicenseComment",
     copyright_text="licenseCopyrightText",
@@ -217,7 +220,7 @@ def snippet_fixture(
     attribution_texts=None,
 ) -> Snippet:
     license_info_in_snippet = (
-        [get_spdx_licensing().parse("MIT"), get_spdx_licensing().parse("GPL-2.0"), SpdxNone()]
+        [spdx_licensing.parse("MIT"), spdx_licensing.parse("GPL-2.0"), SpdxNone()]
         if license_info_in_snippet is None
         else license_info_in_snippet
     )

tests/spdx/parser/jsonlikedict/test_license_expression_parser.py

@@ -10,12 +10,14 @@
 from spdx_tools.spdx.parser.error import SPDXParsingError
 from spdx_tools.spdx.parser.jsonlikedict.license_expression_parser import LicenseExpressionParser
 
+spdx_licensing = get_spdx_licensing()
+
 
 @pytest.mark.parametrize(
     "license_expression_str, expected_license",
     [
-        ("First License", get_spdx_licensing().parse("First License")),
-        ("Second License", get_spdx_licensing().parse("Second License")),
+        ("First License", spdx_licensing.parse("First License")),
+        ("Second License", spdx_licensing.parse("Second License")),
         ("NOASSERTION", SpdxNoAssertion()),
         ("NONE", SpdxNone()),
     ],

tests/spdx/parser/rdf/test_file_parser.py

@@ -13,6 +13,8 @@
 from spdx_tools.spdx.parser.rdf.file_parser import parse_file
 from spdx_tools.spdx.rdfschema.namespace import SPDX_NAMESPACE
 
+spdx_licensing = get_spdx_licensing()
+
 
 def test_parse_file():
     graph = Graph().parse(os.path.join(os.path.dirname(__file__), "data/file_to_test_rdf_parser.rdf.xml"))
@@ -29,10 +31,10 @@ def test_parse_file():
     assert file.comment == "fileComment"
     assert file.copyright_text == "copyrightText"
     assert file.contributors == ["fileContributor"]
-    assert file.license_concluded == get_spdx_licensing().parse("MIT AND GPL-2.0")
+    assert file.license_concluded == spdx_licensing.parse("MIT AND GPL-2.0")
     TestCase().assertCountEqual(
         file.license_info_in_file,
-        [get_spdx_licensing().parse("MIT"), get_spdx_licensing().parse("GPL-2.0"), SpdxNoAssertion()],
+        [spdx_licensing.parse("MIT"), spdx_licensing.parse("GPL-2.0"), SpdxNoAssertion()],
     )
     assert file.license_comment == "licenseComment"
     assert file.notice == "fileNotice"

tests/spdx/parser/rdf/test_license_expression_parser.py

@@ -11,6 +11,8 @@
 from spdx_tools.spdx.parser.rdf.license_expression_parser import parse_license_expression
 from spdx_tools.spdx.rdfschema.namespace import SPDX_NAMESPACE
 
+spdx_licensing = get_spdx_licensing()
+
 
 def test_license_expression_parser():
     graph = Graph().parse(os.path.join(os.path.dirname(__file__), "data/file_to_test_rdf_parser.rdf.xml"))
@@ -19,7 +21,7 @@ def test_license_expression_parser():
 
     license_expression = parse_license_expression(license_expression_node, graph, "https://some.namespace#")
 
-    assert license_expression == get_spdx_licensing().parse("GPL-2.0 AND MIT")
+    assert license_expression == spdx_licensing.parse("GPL-2.0 AND MIT")
 
 
 def test_license_expression_parser_with_coupled_licenses():
@@ -30,19 +32,19 @@ def test_license_expression_parser_with_coupled_licenses():
     packages_by_spdx_id = {package.spdx_id: package for package in doc.packages}
     files_by_spdx_id = {file.spdx_id: file for file in doc.files}
 
-    assert packages_by_spdx_id["SPDXRef-Package"].license_declared == get_spdx_licensing().parse(
+    assert packages_by_spdx_id["SPDXRef-Package"].license_declared == spdx_licensing.parse(
         "LGPL-2.0-only AND LicenseRef-3"
     )
-    assert packages_by_spdx_id["SPDXRef-Package"].license_concluded == get_spdx_licensing().parse(
+    assert packages_by_spdx_id["SPDXRef-Package"].license_concluded == spdx_licensing.parse(
         "LGPL-2.0-only OR LicenseRef-3"
     )
     TestCase().assertCountEqual(
         packages_by_spdx_id["SPDXRef-Package"].license_info_from_files,
         [
-            get_spdx_licensing().parse("GPL-2.0"),
-            get_spdx_licensing().parse("LicenseRef-1"),
-            get_spdx_licensing().parse("LicenseRef-2"),
+            spdx_licensing.parse("GPL-2.0"),
+            spdx_licensing.parse("LicenseRef-1"),
+            spdx_licensing.parse("LicenseRef-2"),
         ],
     )
 
-    assert files_by_spdx_id["SPDXRef-JenaLib"].license_concluded == get_spdx_licensing().parse("LicenseRef-1")
+    assert files_by_spdx_id["SPDXRef-JenaLib"].license_concluded == spdx_licensing.parse("LicenseRef-1")

tests/spdx/parser/rdf/test_package_parser.py

@@ -22,6 +22,8 @@
 from spdx_tools.spdx.parser.rdf.package_parser import parse_external_package_ref, parse_package
 from spdx_tools.spdx.rdfschema.namespace import SPDX_NAMESPACE
 
+spdx_licensing = get_spdx_licensing()
+
 
 def test_package_parser():
     graph = Graph().parse(os.path.join(os.path.dirname(__file__), "data/file_to_test_rdf_parser.rdf.xml"))
@@ -41,11 +43,11 @@ def test_package_parser():
     assert package.files_analyzed is True
     assert package.checksums == [Checksum(ChecksumAlgorithm.SHA1, "71c4025dd9897b364f3ebbb42c484ff43d00791c")]
     assert package.source_info == "sourceInfo"
-    assert package.license_concluded == get_spdx_licensing().parse("MIT AND GPL-2.0")
-    assert package.license_declared == get_spdx_licensing().parse("MIT AND GPL-2.0")
+    assert package.license_concluded == spdx_licensing.parse("MIT AND GPL-2.0")
+    assert package.license_declared == spdx_licensing.parse("MIT AND GPL-2.0")
     TestCase().assertCountEqual(
         package.license_info_from_files,
-        [get_spdx_licensing().parse("MIT"), get_spdx_licensing().parse("GPL-2.0"), SpdxNoAssertion()],
+        [spdx_licensing.parse("MIT"), spdx_licensing.parse("GPL-2.0"), SpdxNoAssertion()],
     )
     assert package.license_comment == "packageLicenseComment"
     assert package.copyright_text == "packageCopyrightText"