[issue-456] add helper method to remove prefix to support Python 3.7 …

…and Python 3.8

Signed-off-by: Meret Behrens <meret.behrens@tngtech.com>

src/spdx/parser/rdf/checksum_parser.py

@@ -13,7 +13,7 @@
 from spdx.model.checksum import Checksum, ChecksumAlgorithm
 from spdx.parser.logger import Logger
 from spdx.parser.parsing_functions import construct_or_raise_parsing_error, raise_parsing_error_if_logger_has_messages
-from spdx.parser.rdf.graph_parsing_functions import parse_literal
+from spdx.parser.rdf.graph_parsing_functions import parse_literal, remove_prefix
 from spdx.rdfschema.namespace import SPDX_NAMESPACE
 
 
@@ -29,7 +29,7 @@ def parse_checksum(parent_node: URIRef, graph: Graph) -> Checksum:
 
 
 def convert_rdf_to_algorithm(algorithm: str) -> ChecksumAlgorithm:
-    algorithm = algorithm.removeprefix(SPDX_NAMESPACE.checksumAlgorithm_).upper()
+    algorithm = remove_prefix(algorithm, SPDX_NAMESPACE.checksumAlgorithm_).upper()
     if "BLAKE2B" in algorithm:
         algorithm = algorithm.replace("BLAKE2B", "BLAKE2B_")
 

src/spdx/parser/rdf/creation_info_parser.py

@@ -19,7 +19,7 @@
 from spdx.parser.logger import Logger
 from spdx.parser.parsing_functions import construct_or_raise_parsing_error, raise_parsing_error_if_logger_has_messages
 from spdx.parser.rdf.checksum_parser import parse_checksum
-from spdx.parser.rdf.graph_parsing_functions import parse_literal, parse_spdx_id
+from spdx.parser.rdf.graph_parsing_functions import parse_literal, parse_spdx_id, remove_prefix
 from spdx.rdfschema.namespace import SPDX_NAMESPACE, LICENSE_NAMESPACE
 
 from spdx.datetime_conversions import datetime_from_str
@@ -34,7 +34,7 @@ def parse_creation_info(graph: Graph) -> Tuple[CreationInfo, URIRef]:
     namespace, spdx_id, doc_node = parse_namespace_and_spdx_id(graph)
     spec_version = parse_literal(logger, graph, doc_node, SPDX_NAMESPACE.specVersion)
     data_license = parse_literal(logger, graph, doc_node, SPDX_NAMESPACE.dataLicense,
-                                 parsing_method=lambda x: x.removeprefix(LICENSE_NAMESPACE))
+                                 parsing_method=lambda x: remove_prefix(x, LICENSE_NAMESPACE))
     comment = parse_literal(logger, graph, doc_node, RDFS.comment)
     name = parse_literal(logger, graph, doc_node, SPDX_NAMESPACE.name)
 

src/spdx/parser/rdf/file_parser.py

@@ -15,7 +15,7 @@
 from spdx.parser.parsing_functions import construct_or_raise_parsing_error, raise_parsing_error_if_logger_has_messages
 from spdx.parser.rdf.checksum_parser import parse_checksum
 from spdx.parser.rdf.graph_parsing_functions import parse_literal, parse_spdx_id, parse_literal_or_no_assertion_or_none, \
-    get_correct_typed_value
+    get_correct_typed_value, remove_prefix
 from spdx.parser.rdf.license_expression_parser import parse_license_expression
 from spdx.rdfschema.namespace import SPDX_NAMESPACE
 
@@ -67,5 +67,5 @@ def parse_file(file_node: URIRef, graph: Graph, doc_namespace: str) -> File:
 
 
 def convert_uri_ref_to_file_type(file_type_ref: URIRef) -> FileType:
-    file_type = file_type_ref.removeprefix(SPDX_NAMESPACE).replace("fileType_", "").upper()
+    file_type = remove_prefix(file_type_ref, SPDX_NAMESPACE.fileType_).upper()
     return FileType[file_type]

src/spdx/parser/rdf/graph_parsing_functions.py

@@ -41,6 +41,7 @@ def apply_parsing_method_or_log_error(logger: Logger, value: Any, parsing_method
         logger.extend(err.args[0])
     return default
 
+
 def parse_literal_or_no_assertion_or_none(logger: Logger, graph: Graph, subject: Node, predicate: Node,
                                           parsing_method: Callable = str, default: Any = None):
     value = get_unique_value(logger, graph, subject, predicate, default)
@@ -84,7 +85,7 @@ def get_unique_value(logger: Logger, graph: Graph, subject: Node, predicate: Nod
 
 def parse_enum_value(enum_str: str, enum_class: Type[Enum], prefix: str) -> Enum:
     try:
-        enum_without_rdf_prefix = enum_str.removeprefix(prefix)
+        enum_without_rdf_prefix = remove_prefix(enum_str, prefix)
         value = camel_case_to_snake_case(enum_without_rdf_prefix).upper()
         return enum_class[value]
     except KeyError:
@@ -100,3 +101,11 @@ def parse_spdx_id(resource: URIRef, doc_namespace: str, graph: Graph) -> Optiona
         namespace_manager = NamespaceManager(graph)
         return namespace_manager.normalizeUri(resource)
     return resource.toPython() or None
+
+
+# Python 3.9 introduced the method removeprefix() for strings, but as we are also supporting Python 3.7 and 3.8 we need
+# to write our own helper method to delete prefixes.
+def remove_prefix(string: str, prefix: str) -> str:
+    if string.startswith(prefix):
+        return string[len(prefix):]
+    return string

src/spdx/parser/rdf/license_expression_parser.py

@@ -14,15 +14,17 @@
 from license_expression import LicenseExpression, get_spdx_licensing
 from rdflib.term import Identifier, URIRef, BNode, Node
 from spdx.parser.error import SPDXParsingError
+from spdx.parser.rdf.graph_parsing_functions import remove_prefix
 
 from spdx.rdfschema.namespace import SPDX_NAMESPACE, LICENSE_NAMESPACE
 
 
-def parse_license_expression(license_expression_node: Union[URIRef, BNode, Node], graph: Graph, doc_namespace: str) -> LicenseExpression:
+def parse_license_expression(license_expression_node: Union[URIRef, BNode, Node], graph: Graph,
+                             doc_namespace: str) -> LicenseExpression:
     spdx_licensing = get_spdx_licensing()
     expression = ""
     if license_expression_node.startswith(LICENSE_NAMESPACE):
-        expression = license_expression_node.removeprefix(LICENSE_NAMESPACE)
+        expression = remove_prefix(license_expression_node, LICENSE_NAMESPACE)
         return spdx_licensing.parse(expression)
     if license_expression_node.startswith(doc_namespace):
         expression = license_expression_node.fragment
@@ -60,7 +62,7 @@ def parse_license_expression(license_expression_node: Union[URIRef, BNode, Node]
 
 def parse_license_exception(exception_node: Identifier, graph: Graph) -> str:
     if exception_node.startswith(LICENSE_NAMESPACE):
-        exception = exception_node.removeprefix(LICENSE_NAMESPACE)
+        exception = remove_prefix(exception_node, LICENSE_NAMESPACE)
     else:
         exception = graph.value(exception_node, SPDX_NAMESPACE.licenseExceptionId).toPython()
     return exception

src/spdx/parser/rdf/package_parser.py

@@ -20,7 +20,7 @@
 from spdx.parser.parsing_functions import raise_parsing_error_if_logger_has_messages, construct_or_raise_parsing_error
 from spdx.parser.rdf.checksum_parser import parse_checksum
 from spdx.parser.rdf.graph_parsing_functions import parse_spdx_id, parse_literal, parse_enum_value, \
-    parse_literal_or_no_assertion_or_none, get_correct_typed_value, parse_literal_or_no_assertion
+    parse_literal_or_no_assertion_or_none, get_correct_typed_value, parse_literal_or_no_assertion, remove_prefix
 from spdx.parser.rdf.license_expression_parser import parse_license_expression
 from spdx.rdfschema.namespace import SPDX_NAMESPACE, REFERENCE_NAMESPACE
 
@@ -128,7 +128,7 @@ def parse_external_package_ref(external_package_ref_node: URIRef, graph: Graph)
         logger, graph, external_package_ref_node, SPDX_NAMESPACE.referenceCategory,
         parsing_method=lambda x: parse_enum_value(x, ExternalPackageRefCategory, SPDX_NAMESPACE.referenceCategory_, ))
     ref_type = parse_literal(logger, graph, external_package_ref_node, SPDX_NAMESPACE.referenceType,
-                             parsing_method=lambda x: x.removeprefix(REFERENCE_NAMESPACE))
+                             parsing_method=lambda x: remove_prefix(x, REFERENCE_NAMESPACE))
     comment = parse_literal(logger, graph, external_package_ref_node, RDFS.comment)
 
     raise_parsing_error_if_logger_has_messages(logger)

tests/spdx/parser/rdf/test_graph_parsing_function.py

@@ -11,7 +11,7 @@
 import pytest
 from rdflib import URIRef, Graph, Namespace
 
-from spdx.parser.rdf.graph_parsing_functions import parse_spdx_id
+from spdx.parser.rdf.graph_parsing_functions import parse_spdx_id, remove_prefix
 
 
 @pytest.mark.parametrize("resource,doc_namespace,ext_namespace_mapping,expected",
@@ -27,3 +27,11 @@ def test_parse_spdx_id(resource, doc_namespace, ext_namespace_mapping, expected)
     spdx_id = parse_spdx_id(resource, doc_namespace, graph)
 
     assert spdx_id == expected
+
+
+@pytest.mark.parametrize("string,prefix,expected", [("prefixString", "prefix", "String"),
+                                                    ("prefixString", "refix", "prefixString")])
+def test_remove_prefix(string, prefix, expected):
+    shorten_string = remove_prefix(string, prefix)
+
+    assert expected == shorten_string