def test_add_license_attr(): p = Package.load_from_json(serve_static_json) assert p.save() is not None v = Version.load_from_json(serve_static_json, package=p) v.save() p.create_version_edge(v) ts1 = v.last_updated assert p.last_updated is not None assert ts1 is not None license_data = serve_static_json["analyses"]["source_licenses"] _, _, licenses = LicenseDetails.load_from_json(license_data) v.add_license_attribute(licenses) assert v.last_updated >= ts1 version_criteria = { 'pecosystem': v.ecosystem_package.ecosystem, 'pname': v.ecosystem_package.name, 'version': v.version } present_version = Version.find_by_criteria('Version', p, version_criteria) assert present_version.last_updated == v.last_updated assert (len(present_version.licenses) == 1) test_set = ['MITNFA'] assert present_version.licenses == test_set Version.delete_by_id(v.id) Package.delete_by_id(p.id)
def test_support_vector_github_detail(): p = Package.load_from_json(serve_static_json) assert p.save() is not None v = Version.load_from_json(serve_static_json, package=p) v.save() p.create_version_edge(v) assert p.last_updated is not None assert v.last_updated is not None github_data = serve_static_json["analyses"]["github_details"] github_result = GithubResult.load_from_json(github_data) assert github_result.last_updated is None v.add_edge_github_details(github_result) ls_before = github_result.last_updated assert (GithubResult.count() == 1) count_before = len(v.get_version_out_edge('has_github_details')) assert count_before == 1 #try adding the edge again v.add_edge_github_details(github_result) count_after = len(v.get_version_out_edge('has_github_details')) ls_after = github_result.last_updated assert count_before == count_after assert ls_after >= ls_before GithubResult.delete_by_id(github_result.id) Version.delete_by_id(v.id) Package.delete_by_id(p.id)
def test_blackduck_graph(): bl_list = [] p = Package.load_from_json(serve_static_json) assert p.save() is not None v = Version.load_from_json(serve_static_json, package=p) v.save() p.create_version_edge(v) assert p.last_updated is not None assert v.last_updated is not None bl_list = bl.load_from_json(input_json['analyses']) assert len(bl_list) == 1 objBlackduck = bl.add_blackduck_issue(bl_list[0]) v.add_blackduck_cve_edge(objBlackduck.id) bl_criteria = {'vulnerability_name': 'CVE-2015-1164'} obj_fetch = SecurityDetails.find_by_criteria('CVE', bl_criteria) assert obj_fetch.last_updated == objBlackduck.last_updated SecurityDetails.delete_by_id(obj_fetch.id) Version.delete_by_id(v.id) Package.delete_by_id(p.id)
def test_person_contributor(): p = Package.load_from_json(serve_static_json) assert p.save() is not None v = Version.load_from_json(serve_static_json, package=p) v.save() p.create_version_edge(v) assert p.last_updated is not None assert v.last_updated is not None contributors_data = serve_static_json["analyses"]["metadata"] contributors_list = Contributor.load_from_json(contributors_data) for c in contributors_list: c.save() assert c.last_updated is not None v.add_edge_contributor(c) assert (Contributor.count() == 0) for c in contributors_list: Contributor.delete_by_id(c.id) Version.delete_by_id(v.id) Package.delete_by_id(p.id)
def test_person_author(): p = Package.load_from_json(serve_static_json) assert p.save() is not None v = Version.load_from_json(serve_static_json, package=p) v.save() p.create_version_edge(v) assert p.last_updated is not None assert v.last_updated is not None authors_data = serve_static_json["analyses"]["metadata"] authors_list = Author.load_from_json(authors_data) ts_list = [] for a in authors_list: a.save() assert a.last_updated is not None ts_list.append(a.last_updated) v.add_edge_author(a) author_before = Author.count() assert (author_before == 1) author_detail = Author(name='Douglas Christopher Wilson', email='*****@*****.**') author_detail.save() assert author_detail.last_updated >= a.last_updated assert (Author.count() == 1) for a in authors_list: Author.delete_by_id(a.id) Author.delete_by_id(author_detail.id) Version.delete_by_id(v.id) Package.delete_by_id(p.id)
def load_dependencies(ecosystem, dependency_data): dependency_pck_list = [] dependency_ver_list = [] dependency_type = [] dependencies = get_dependencies(dependency_data) for dep_type, dep_list in dependencies.items(): for d in dep_list: splits = d.split(" ") (n, v) = ("", "") if len(splits) >= 1: n = splits[0] if len(splits) >= 2: v = splits[1] pck_criteria_dict = {'ecosystem': ecosystem, 'name': n} pck_dep = Package.find_by_criteria( 'Package', pck_criteria_dict) or Package(ecosystem, n) ver_criteria_dict = { 'pecosystem': ecosystem, 'pname': n, 'version': v } ver_dep = Version.find_by_criteria( 'Version', pck_dep, ver_criteria_dict) or Version(pck_dep, v) dependency_pck_list.append(pck_dep) dependency_ver_list.append(ver_dep) dependency_type.append(dep_type) return dependency_pck_list, dependency_ver_list, dependency_type
def test_empty_github_results(): p = Package.load_from_json(npm_sequence_3) assert p.save() is not None v = Version.load_from_json(npm_sequence_3, package=p) v.save() p.create_version_edge(v) assert p.last_updated is not None assert v.last_updated is not None # input_json = gv.read_from_file('test/data/npm-sequence-3.0.0.json') github_data = npm_sequence_3["analyses"]["github_details"] github_result = GithubResult.load_from_json(github_data) assert (github_result.details is not None) assert github_result.last_updated is None # gid = github_result.save() v.add_edge_github_details(github_result) ls_before = github_result.last_updated assert ls_before is not None assert (github_result.id is not None) v.add_edge_github_details(github_result) ls_after = github_result.last_updated assert ls_after >= ls_before assert GithubResult.count() == 1 GithubResult.delete_by_id(github_result.id) Version.delete_by_id(v.id) Package.delete_by_id(p.id)
def test_support_vector_security(): p = Package.load_from_json(serve_static_json) assert p.save() is not None v = Version.load_from_json(serve_static_json, package=p) v.save() p.create_version_edge(v) assert p.last_updated is not None assert v.last_updated is not None security_data = serve_static_json["analyses"]["security_issues"] security_list, cvss_score, cve_ids = SecurityDetails.load_from_json( security_data) ts_list = [] for s, cvss, cve in zip(security_list, cvss_score, cve_ids): s.save() assert s.last_updated is not None ts_list.append(s.last_updated) v.add_security_edge(s, cvss) security_before = SecurityDetails.count() assert (security_before == 1) present_security = SecurityDetails.find_by_criteria( 'CVE', {'cve_id': 'CVE-2015-1164'}) assert (len(present_security.references) == 5) ref_list = [ "https://github.com/expressjs/serve-static/issues/26", "https://bugzilla.redhat.com/show_bug.cgi?id=1181917", "http://xforce.iss.net/xforce/xfdb/99936", "http://www.securityfocus.com/bid/72064", "http://nodesecurity.io/advisories/serve-static-open-redirect" ] assert (all(r in ref_list for r in present_security.references)) repeat_security_detail = SecurityDetails(cve_id='CVE-2015-1164', cvss=4.3, summary='') repeat_security_detail.issue_has_access('authentication', '') repeat_security_detail.issue_has_access('vector', 'NETWORK') repeat_security_detail.issue_has_access('complexity', 'MEDIUM') repeat_security_detail.issue_has_impact('integrity', 'partial') repeat_security_detail.issue_has_impact('confidentiality', '') repeat_security_detail.issue_has_impact('availability', '') repeat_security_detail.save() assert repeat_security_detail.id == s.id assert repeat_security_detail.last_updated >= ts_list[0] assert (SecurityDetails.count() == 1) for s in security_list: SecurityDetails.delete_by_id(s.id) SecurityDetails.delete_by_id(repeat_security_detail.id) Version.delete_by_id(v.id) Package.delete_by_id(p.id)
def test_bucket_import_from_folder(): bucket_dir = 'test/data/S3-data' packages = Package.find_all() assert (len(packages) == 0) p = Package() assert (p.id is None) assert (p.label == "Package") assert p.last_updated is None report = import_from_folder(bucket_dir) assert (report.get('status') == 'Success') assert (report.get('count_imported_EPVs') == 3) assert (report.get('last_imported_EPV') == 'npm/sequence/3.0.0.json') assert (report.get('max_finished_at') == '2017-02-24T13:43:11.872916') criteria_dict = {'ecosystem': 'npm', 'name': 'sequence'} p1 = Package.find_by_criteria('Package', criteria_dict) assert p1 is not None assert p1.id is not None assert (p1.ecosystem == 'npm') assert (p1.latest_version == '3.0.0') assert (p1.package_dependents_count == 22) assert (p1.name == 'sequence') criteria_dict = { 'pecosystem': 'npm', 'pname': 'sequence', 'version': '3.0.0' } v1 = Version.find_by_criteria('Version', p1, criteria_dict) assert v1 is not None assert v1.ecosystem_package is not None assert (v1.ecosystem_package.ecosystem == 'npm') packages = Package.find_all() assert (len(packages) == 3) versions = Version.find_all() assert (len(versions) == 4) Package.delete_all() assert (Package.count() == 0) Version.delete_all() assert (Version.count() == 0) GraphMetaData.delete_all() assert (GraphMetaData.count() == 0)
def test_additional_data_as_attr(): add_details = {} pck_obj = Package.load_from_json(serve_static_json) assert pck_obj.last_updated is None assert pck_obj.save() is not None ver_obj = Version.load_from_json(serve_static_json, package=pck_obj) assert ver_obj.last_updated is None ver_obj.save() v_ts1 = ver_obj.last_updated assert pck_obj.last_updated is not None assert v_ts1 is not None pck_obj.create_version_edge(ver_obj) code_metrics = CMR.load_from_json( npm_crumb_data["analyses"]["code_metrics"]) security_data = serve_static_json["analyses"]["security_issues"] security_list, cvss_score, cve_ids = SD.load_from_json(security_data) if len(cvss_score) > 0 and len(cve_ids) > 0: cve_ids = cve_ids cvss = cvss_score add_details["cve_ids"] = cve_ids[0] add_details["cvss"] = cvss[0] ver_obj.add_additional_data_as_attr(code_metrics) ver_obj.add_cve_ids(add_details["cvss"], add_details["cve_ids"]) assert ver_obj.last_updated >= v_ts1 version_criteria = { 'pecosystem': pck_obj.ecosystem, 'pname': pck_obj.name, 'version': ver_obj.version } present_version = Version.find_by_criteria(ver_obj.label, pck_obj, version_criteria) logger.info(present_version.__dict__) assert present_version.cm_loc == 1351 assert present_version.cm_num_files == 16 assert present_version.cm_avg_cyclomatic_complexity == -1 assert present_version.cve_ids == ['CVE-2015-1164:7.5'] assert present_version.last_updated == ver_obj.last_updated Version.delete_by_id(ver_obj.id) Package.delete_by_id(pck_obj.id)
def test_ecosystem_import_from_folder(): ecosystem_dir = 'test/data/S3-data/maven' packages = Package.find_all() assert (len(packages) == 0) p = Package() assert (p.id is None) assert (p.label == "Package") assert p.last_updated is None report = import_from_folder(ecosystem_dir) assert (report.get('status') == 'Success') assert (report.get('count_imported_EPVs') == 2) assert ( report.get('last_imported_EPV') == 'org.slf4j:slf4j-api/1.5.6.json') assert (report.get('max_finished_at') == '2017-02-24T13:43:11.872916') criteria_dict = {'ecosystem': 'maven', 'name': 'junit:junit'} p1 = Package.find_by_criteria('Package', criteria_dict) criteria_dict = { 'pecosystem': 'maven', 'pname': 'junit:junit', 'version': '4.8.2' } v1 = Version.find_by_criteria('Version', p1, criteria_dict) assert v1 is not None assert v1.ecosystem_package is not None assert (v1.ecosystem_package.ecosystem == 'maven') packages = Package.find_all() assert (len(packages) == 2) versions = Version.find_all() assert (len(versions) == 3) Package.delete_all() assert (Package.count() == 0) Version.delete_all() assert (Version.count() == 0) GraphMetaData.delete_all() assert (GraphMetaData.count() == 0)
def test_create_package_entity(): packages = Package.find_all() assert (len(packages) == 0) list_epv_1 = [{ 'ecosystem': 'maven', 'name': 'org.slf4j:slf4j-api', 'version': '1.5.6' }] import_epv_from_folder('test/data/S3-data', list_epv=list_epv_1) criteria_dict = {'ecosystem': 'maven', 'name': 'org.slf4j:slf4j-api'} p = Package.find_by_criteria('Package', criteria_dict) assert p.latest_version == '1.7.22' p.save() # must be an update assert (Package.count() == 2) p.create() # duplicate should not create new node assert (Package.count() == 2) criteria_dict = {'ecosystem': 'maven', 'name': 'junit:junit'} p2 = Package.find_by_criteria('Package', criteria_dict) assert p2.latest_version == '-1' list_epv_2 = [{ 'ecosystem': 'maven', 'name': 'junit:junit', 'version': '4.8.2' }] import_epv_from_folder('test/data/S3-data', list_epv=list_epv_2) criteria_dict = {'ecosystem': 'maven', 'name': 'junit:junit'} p3 = Package.find_by_criteria('Package', criteria_dict) assert p3.latest_version == '4.12' p.save() # must be an update assert (Package.count() == 2) Package.delete_all() assert (Package.count() == 0) Version.delete_all() assert (Version.count() == 0)
def test_version_dependencies(): p = Package.load_from_json(serve_static_json) assert p.save() is not None v = Version.load_from_json(serve_static_json, package=p) v.save() p.create_version_edge(v) dependency_data = serve_static_json["analyses"]["dependency_snapshot"] dependency_pck_list, dependency_ver_list, dependency_type = \ vdv.load_dependencies(v.ecosystem_package.ecosystem, dependency_data) for d_pck, d_ver, d_type in zip(dependency_pck_list, dependency_ver_list, dependency_type): assert d_pck.save() is not None d_ver.save() d_pck.create_version_edge(d_ver) v.add_edge_dependency(d_ver, d_type) assert (Version.count_dependency(v.id) == 4) assert (Version.count() + Package.count() == 10) for pd, vd in zip(dependency_pck_list, dependency_ver_list): Version.delete_by_id(vd.id) Package.delete_by_id(pd.id) Version.delete_by_id(v.id) Package.delete_by_id(p.id)
def test_add_code_metrics_non_empty(): p = Package.load_from_json(npm_crumb_data) assert p.save() is not None v = Version.load_from_json(npm_crumb_data, package=p) v.save() p.create_version_edge(v) assert p.last_updated is not None assert v.last_updated is not None code_metrics_data = npm_crumb_data["analyses"]["code_metrics"] code_metrics = CodeMetricsResult.load_from_json(code_metrics_data) assert code_metrics.last_updated is None assert code_metrics.id is None v.add_code_metrics_edge(code_metrics) assert code_metrics.id >= 0 assert code_metrics.last_updated is not None assert code_metrics.last_updated > v.last_updated assert (CodeMetricsResult.count() == 1) assert (CodeMetricsLanguage.count() == 6) count_before = len(v.get_version_out_edge('has_code_metrics')) assert count_before == 1 #try adding the edge again, should not let this happen results2 = v.add_code_metrics_edge(code_metrics) count_after = len(v.get_version_out_edge('has_code_metrics')) assert count_after == count_before assert CodeMetricsLanguage.count() == 6 lang_nodes = CodeMetricsLanguage.find_all() for lang_node in lang_nodes: CodeMetricsLanguage.delete_by_id(lang_node.id) CodeMetricsLanguage.delete_by_id(code_metrics.id) Version.delete_by_id(v.id) Package.delete_by_id(p.id)
def test_package_import_from_folder(): package_dir = 'test/data/S3-data/maven/junit:junit' packages = Package.find_all() assert (len(packages) == 0) versions = Version.find_all() assert (len(versions) == 0) p = Package() assert (p.id is None) assert (p.label == "Package") assert p.last_updated is None report = import_from_folder(package_dir) assert (report.get('status') == 'Success') assert (report.get('count_imported_EPVs') == 1) assert (report.get('last_imported_EPV') == '4.8.2.json') assert (report.get('max_finished_at') == '2017-02-24T13:42:29.665786') criteria_dict = {'ecosystem': 'maven', 'name': 'junit:junit'} p1 = Package.find_by_criteria('Package', criteria_dict) assert p1 is not None assert p1.id is not None assert (p1.ecosystem == 'maven') assert (p1.latest_version == '4.12') assert (p1.package_dependents_count == -1) assert (p1.name == 'junit:junit') criteria_dict = { 'pecosystem': 'maven', 'pname': 'junit:junit', 'version': '4.8.2' } v1 = Version.find_by_criteria('Version', p1, criteria_dict) assert v1 is not None assert v1.ecosystem_package is not None assert (v1.ecosystem_package.ecosystem == 'maven') packages = Package.find_all() assert (len(packages) == 1) versions = Version.find_all() assert (len(versions) == 1) Package.delete_by_id(p1.id) assert (Package.count() == 0) Version.delete_by_id(v1.id) assert (Version.count() == 0) GraphMetaData.delete_all() assert (GraphMetaData.count() == 0)
def test_support_vector_license(): p = Package.load_from_json(serve_static_json) assert p.save() is not None v = Version.load_from_json(serve_static_json, package=p) v.save() p.create_version_edge(v) assert p.last_updated is not None assert v.last_updated is not None license_data = serve_static_json["analyses"]["source_licenses"] license_details_list, license_counts_list, _ = LicenseDetails.load_from_json( license_data) ts_list = [] for license_detail, license_count in zip(license_details_list, license_counts_list): license_detail.save() assert license_detail.last_updated is not None ts_list.append(license_detail.last_updated) v.add_license_edge(license_detail, license_count) assert (LicenseDetails.count() == 1) new_license_detail = LicenseDetails(name='MITNFA') new_license_detail.save() assert new_license_detail.last_updated >= ts_list[0] # Duplicate license should not be inserted assert (LicenseDetails.count() == 1) for l_id in license_details_list: LicenseDetails.delete_by_id(l_id.id) LicenseDetails.delete_by_id(new_license_detail.id) Version.delete_by_id(v.id) Package.delete_by_id(p.id)
def load_package_version_values(filename): input_json = gv.read_from_file(filename) objpackage = Package.load_from_json(input_json) objversion = Version.load_from_json(input_json, objpackage) github_data = input_json["analyses"]["github_details"] github_result = GithubResult.load_from_json(github_data) objversion.github_details = github_result authors_data = input_json["analyses"]["metadata"] authors_list = Author.load_from_json(authors_data) for objauthor in authors_list: objversion.version_authored_by(objauthor) contributors_data = input_json["analyses"]["metadata"] contributor_list = Contributor.load_from_json(contributors_data) for objcontributor in contributor_list: objversion.version_contributed_by(objcontributor) license_data = input_json["analyses"]["source_licenses"] license_details_list, license_counts_list, license_names = LicenseDetails.load_from_json( license_data) for objlicense, license_count in zip(license_details_list, license_counts_list): objversion.version_covered_under(objlicense, license_count) objversion.licenses = license_names dependency_data = input_json["analyses"]["dependency_snapshot"] _, dependency_ver_list, dependency_type = \ vdv.load_dependencies( objversion.ecosystem_package.ecosystem, dependency_data) for d_ver, d_type in zip(dependency_ver_list, dependency_type): objversion.version_depends_on(d_ver, d_type) security_data = input_json["analyses"]["security_issues"] security_list, __, _ = SecurityDetails.load_from_json(security_data) for objsecurity in security_list: objversion.version_has_nvd_issues(objsecurity) return objversion
def test_version_entity(): p = Package.load_from_json(serve_static_json) assert p.save() is not None v = Version.load_from_json(serve_static_json, package=p) v.save() assert (Version.count() == 1) ls_before = v.last_updated assert ls_before is not None criteria_dict = { 'pecosystem': 'npm', 'pname': 'serve-static', 'version': '1.7.1' } v2 = Version.find_by_criteria('Version', p, criteria_dict) assert v2.last_updated == v.last_updated v.save() ls_after = v.last_updated assert (Version.count() == 1) assert ls_after >= ls_before assert v.last_updated >= v2.last_updated test_packaged_in = ['nodejs-serve-static'] test_published_in = [] assert (all(pck in test_packaged_in for pck in v.is_packaged_in)) assert (all(pub in test_published_in for pck in v.is_published_in)) # now create an edge edge_count_before = Package.edge_count() p.create_version_edge(v) edge_count_after = Package.edge_count() assert (edge_count_after == edge_count_before + 1) # now try to create an edge again edge_count_before = Package.edge_count() p.create_version_edge(v) edge_count_after = Package.edge_count() assert (edge_count_after == edge_count_before) # this should return all versions associated with this package p_versions = p.get_versions() assert (len(p_versions) == 1) Version.delete_by_id(v.id) Package.delete_by_id(p.id)
def populate_from_json(cls, input_json): # NPM packages with dependencies, versions i.e. Package version # insertion logger.info("Instantiating package ...") package = Package.load_from_json(input_json) logger.info("Saving package ...") pkg_id = package.save() logger.info(" Package node ID: %s" % pkg_id) if pkg_id is None: return version = Version.load_from_json(input_json, package=package) ver_id = version.save() logger.info(" Version node ID: %s" % ver_id) package.create_version_edge(version) analyses = input_json["analyses"] if "dependency_snapshot" in analyses: dependency_snapshot = analyses["dependency_snapshot"] dependency_pck_list, dependency_ver_list, dependency_type = vdv.load_dependencies( version.ecosystem_package.ecosystem, dependency_snapshot) for d_pck, d_ver, d_type in zip(dependency_pck_list, dependency_ver_list, dependency_type): d_pck.save() if d_pck is None: continue d_ver.save() d_pck.create_version_edge(d_ver) version.add_edge_dependency(d_ver, d_type) if "metadata" in analyses: meta_data = analyses["metadata"] print(" Adding authors_list") authors_list = Author.load_from_json(meta_data) for author in authors_list: a_id = author.save() print(" author ID: %s" % a_id) version.add_edge_author(author) print(" Adding contributor_list") contributor_list = Contributor.load_from_json(meta_data) for contributor in contributor_list: c_id = contributor.save() print(" contributor ID: %s" % c_id) version.add_edge_author(contributor) # License Information if "source_licenses" in analyses: print(" Adding source_licenses") licenses = set() license_data = analyses["source_licenses"] license_details_list, license_counts_list, licenses = LicenseDetails.load_from_json( license_data) for used_license, license_count in zip(license_details_list, license_counts_list): lic_id = used_license.save() print(" license_data ID: %s" % lic_id) version.add_license_edge(used_license, license_count) version.add_license_attribute(licenses) # NVD Security Information if "security_issues" in analyses: print(" Adding security_issues") security_data = analyses["security_issues"] security_list, cvss_score, cve_ids = SecurityDetails.load_from_json( security_data) for s, cvss, cve_id in zip(security_list, cvss_score, cve_ids): ss_id = s.save() print(" security_data ID: %s" % ss_id) version.add_security_edge(s, cvss) version.add_cve_ids(cvss, cve_id) # GitHub Details if "github_details" in analyses: print(" Adding github_details") github_data = analyses["github_details"] github_result = GithubResult.load_from_json(github_data) package.add_github_details_as_attr(github_result) version.add_edge_github_details(github_result) # Code Metrics if "code_metrics" in analyses: print(" Adding code_metrics") code_metrics_data = analyses["code_metrics"] code_metrics = CodeMetricsResult.load_from_json(code_metrics_data) version.add_code_metrics_edge(code_metrics) version.add_additional_data_as_attr(code_metrics) if "blackduck" in analyses: print("Adding extra security info via blackduck") blackduck_cve = analyses["blackduck"] issue_list = bl.load_from_json(blackduck_cve) for issue in issue_list: bl_obj = bl.add_blackduck_issue(issue) version.add_blackduck_cve_edge(bl_obj.id)
def test_full_import_and_incr_update(): data_dir = 'test/data' # Let us make sure that target graph has no metadata graph_meta = GraphPopulator.get_metadata() assert (graph_meta is None) # Full import: insert all the EPVs from the given data source src_dir = os.path.join(data_dir, 'full_import') report = import_bulk(data_source=LocalFileSystem(src_dir=src_dir), book_keeper=None) assert (report.get('status') == 'Success') assert (report.get('count_imported_EPVs') == 1) assert (report.get('last_imported_EPV') == 'npm/serve-static/1.7.1.json') assert (report.get('max_finished_at') == '2017-02-08T12:26:51.962609') graph_meta = GraphPopulator.get_metadata() assert (graph_meta is not None) assert (graph_meta.last_incr_update_ts == '2017-02-08T12:26:51.962609') # Incremental update 1: # Let us mimic a scenario where a new EPV was inserted recently: npm/send/0.10.1 src_dir = os.path.join(data_dir, 'incr_update1') book_keeping_json = os.path.join(data_dir, 'book_keeping1.json') report = import_bulk( data_source=LocalFileSystem(src_dir=src_dir), book_keeper=JsonBookKeeper(json_file_name=book_keeping_json)) assert (report.get('status') == 'Success') assert (report.get('count_imported_EPVs') == 1) assert (report.get('last_imported_EPV') == 'npm/send/0.10.1.json') assert (report.get('max_finished_at') == '2017-02-22T15:34:59.469864') graph_meta = GraphPopulator.get_metadata() assert (graph_meta is not None) assert (graph_meta.last_incr_update_ts == '2017-02-22T15:34:59.469864') # Incremental update 2: # Let us mimic a scenario where a new EPV was inserted recently: npm/parseurl/1.3.1 # and also an already existing EPV was updated recently: npm/serve-static/1.7.1 src_dir = os.path.join(data_dir, 'incr_update2') book_keeping_json = os.path.join(data_dir, 'book_keeping2.json') report = import_bulk( data_source=LocalFileSystem(src_dir=src_dir), book_keeper=JsonBookKeeper(json_file_name=book_keeping_json)) assert (report.get('status') == 'Success') assert (report.get('count_imported_EPVs') == 2) assert (report.get('last_imported_EPV') == 'npm/serve-static/1.7.1.json') assert (report.get('max_finished_at') == '2017-02-22T15:35:51.962609') graph_meta = GraphPopulator.get_metadata() assert (graph_meta is not None) assert (graph_meta.last_incr_update_ts == '2017-02-22T15:35:51.962609') # Cleanup GraphMetaData.delete_all() assert (GraphMetaData.count() == 0) LicenseDetails.delete_all() assert (LicenseDetails.count() == 0) Author.delete_all() assert (Author.count() == 0) CodeMetricsResult.delete_all() assert (CodeMetricsResult.count() == 0) CodeMetricsLanguage.delete_all() assert (CodeMetricsLanguage.count() == 0) GithubResult.delete_all() assert (GithubResult.count() == 0) Contributor.delete_all() assert (Contributor.count() == 0) Package.delete_all() assert (Package.count() == 0) Version.delete_all() assert (Version.count() == 0)