|
| 1 | +# |
| 2 | +# Copyright (c) nexB Inc. and others. All rights reserved. |
| 3 | +# http://nexb.com and https://github.com/nexB/vulnerablecode/ |
| 4 | +# The VulnTotal software is licensed under the Apache License version 2.0. |
| 5 | +# Data generated with VulnTotal require an acknowledgment. |
| 6 | +# |
| 7 | +# You may not use this software except in compliance with the License. |
| 8 | +# You may obtain a copy of the License at: http://apache.org/licenses/LICENSE-2.0 |
| 9 | +# Unless required by applicable law or agreed to in writing, software distributed |
| 10 | +# under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR |
| 11 | +# CONDITIONS OF ANY KIND, either express or implied. See the License for the |
| 12 | +# specific language governing permissions and limitations under the License. |
| 13 | +# |
| 14 | +# When you publish or redistribute any data created with VulnTotal or any VulnTotal |
| 15 | +# derivative work, you must accompany this data with the following acknowledgment: |
| 16 | +# |
| 17 | +# Generated with VulnTotal and provided on an "AS IS" BASIS, WITHOUT WARRANTIES |
| 18 | +# OR CONDITIONS OF ANY KIND, either express or implied. No content created from |
| 19 | +# VulnTotal should be considered or used as legal advice. Consult an Attorney |
| 20 | +# for any legal advice. |
| 21 | +# VulnTotal is a free software tool from nexB Inc. and others. |
| 22 | +# Visit https://github.com/nexB/vulnerablecode/ for support and download. |
| 23 | + |
| 24 | +import json |
| 25 | +import logging |
| 26 | +from typing import Iterable |
| 27 | +from urllib.parse import quote |
| 28 | + |
| 29 | +import requests |
| 30 | +from bs4 import BeautifulSoup |
| 31 | +from packageurl import PackageURL |
| 32 | + |
| 33 | +from vulntotal.validator import DataSource |
| 34 | +from vulntotal.validator import VendorData |
| 35 | +from vulntotal.vulntotal_utils import snky_constraints_satisfied |
| 36 | + |
| 37 | +logger = logging.getLogger(__name__) |
| 38 | + |
| 39 | + |
| 40 | +class SnykDataSource(DataSource): |
| 41 | + spdx_license_expression = "TODO" |
| 42 | + license_url = "TODO" |
| 43 | + |
| 44 | + def fetch(self, url): |
| 45 | + response = requests.get(url) |
| 46 | + if not response.status_code == 200: |
| 47 | + logger.error(f"Error while fetching {url}") |
| 48 | + return |
| 49 | + if response.headers["content-type"] == "application/json, charset=utf-8": |
| 50 | + return response.json() |
| 51 | + return response.text |
| 52 | + |
| 53 | + def datasource_advisory(self, purl) -> Iterable[VendorData]: |
| 54 | + package_advisory_url = generate_package_advisory_url(purl) |
| 55 | + package_advisories_list = self.fetch(package_advisory_url) |
| 56 | + self._raw_dump.append(package_advisories_list) |
| 57 | + if package_advisories_list: |
| 58 | + advisories = extract_html_json_advisories(package_advisories_list) |
| 59 | + for snyk_id, affected in advisories.items(): |
| 60 | + if "*" in affected or is_purl_in_affected(purl.version, affected): |
| 61 | + advisory_payload = generate_advisory_payload(snyk_id) |
| 62 | + advisory_html = self.fetch(advisory_payload) |
| 63 | + self._raw_dump.append(advisory_html) |
| 64 | + if advisory_html: |
| 65 | + yield parse_html_advisory(advisory_html, snyk_id, affected) |
| 66 | + |
| 67 | + @classmethod |
| 68 | + def supported_ecosystem(cls): |
| 69 | + return { |
| 70 | + "cocoapods": "cocoapods", |
| 71 | + "composer": "composer", |
| 72 | + "golang": "golang", |
| 73 | + "hex": "hex", |
| 74 | + "linux": "linux", |
| 75 | + "maven": "maven", |
| 76 | + "npm": "npm", |
| 77 | + "nuget": "nuget", |
| 78 | + "pypi": "pip", |
| 79 | + "rubygems": "rubygems", |
| 80 | + # any purl.type not in supported_ecosystem shall implicitly be treated as unmanaged type |
| 81 | + "unmanaged": "unmanaged", |
| 82 | + } |
| 83 | + |
| 84 | + |
| 85 | +def generate_package_advisory_url(purl): |
| 86 | + url_package_advisories = "https://security.snyk.io/package/{ecosystem}/{package}" |
| 87 | + |
| 88 | + # Pseudo API, unfortunately gives only 30 vulnerability per package, but this is the best we have for unmanaged packages |
| 89 | + url_unmanaged_package_advisories = ( |
| 90 | + "https://security.snyk.io/api/listing?search={package}&type=unmanaged" |
| 91 | + ) |
| 92 | + supported_ecosystem = SnykDataSource.supported_ecosystem() |
| 93 | + |
| 94 | + if purl.type == "unmanaged" or purl.type not in supported_ecosystem: |
| 95 | + return url_unmanaged_package_advisories.format( |
| 96 | + package=purl.name if not purl.namespace else f"{purl.namespace}/{purl.name}", |
| 97 | + ) |
| 98 | + |
| 99 | + purl_name = purl.name |
| 100 | + if purl.type == "maven": |
| 101 | + if not purl.namespace: |
| 102 | + logger.error(f"Invalid Maven PURL {str(purl)}") |
| 103 | + return |
| 104 | + purl_name = quote(f"{purl.namespace}:{purl.name}", safe="") |
| 105 | + |
| 106 | + elif purl.type in ("golang", "composer"): |
| 107 | + if purl.namespace: |
| 108 | + purl_name = quote(f"{purl.namespace}/{purl.name}", safe="") |
| 109 | + |
| 110 | + elif purl.type == "linux": |
| 111 | + distro = purl.qualifiers["distro"] |
| 112 | + purl_name = f"{distro}/{purl.name}" |
| 113 | + |
| 114 | + return url_package_advisories.format( |
| 115 | + ecosystem=supported_ecosystem[purl.type], |
| 116 | + package=purl_name, |
| 117 | + ) |
| 118 | + |
| 119 | + |
| 120 | +def extract_html_json_advisories(package_advisories): |
| 121 | + vulnerablity = {} |
| 122 | + |
| 123 | + # If advisories are json and is obtained through pseudo API |
| 124 | + if isinstance(package_advisories, dict): |
| 125 | + if package_advisories["status"] == "ok": |
| 126 | + for vuln in package_advisories["vulnerabilities"]: |
| 127 | + vulnerablity[vuln["id"]] = vuln["semver"]["vulnerable"] |
| 128 | + else: |
| 129 | + soup = BeautifulSoup(package_advisories, "html.parser") |
| 130 | + vulns_table = soup.find("tbody", class_="vue--table__tbody") |
| 131 | + if vulns_table: |
| 132 | + vulns_rows = vulns_table.find_all("tr", class_="vue--table__row") |
| 133 | + for row in vulns_rows: |
| 134 | + anchor = row.find(class_="vue--anchor") |
| 135 | + ranges = row.find_all( |
| 136 | + "span", class_="vue--chip vulnerable-versions__chip vue--chip--default" |
| 137 | + ) |
| 138 | + affected_versions = [vers.text.strip() for vers in ranges] |
| 139 | + vulnerablity[anchor["href"].rsplit("/", 1)[-1]] = affected_versions |
| 140 | + return vulnerablity |
| 141 | + |
| 142 | + |
| 143 | +def parse_html_advisory(advisory_html, snyk_id, affected) -> VendorData: |
| 144 | + aliases = [] |
| 145 | + fixed_versions = [] |
| 146 | + |
| 147 | + advisory_soup = BeautifulSoup(advisory_html, "html.parser") |
| 148 | + cve_span = advisory_soup.find("span", class_="cve") |
| 149 | + if cve_span: |
| 150 | + cve_anchor = cve_span.find("a", class_="vue--anchor") |
| 151 | + aliases.append(cve_anchor["id"]) |
| 152 | + |
| 153 | + how_to_fix = advisory_soup.find( |
| 154 | + "div", class_="vue--block vuln-page__instruction-block vue--block--instruction" |
| 155 | + ) |
| 156 | + if how_to_fix: |
| 157 | + fixed = how_to_fix.find("p").text.split(" ") |
| 158 | + if "Upgrade" in fixed: |
| 159 | + lower = fixed.index("version") if "version" in fixed else fixed.index("versions") |
| 160 | + upper = fixed.index("or") |
| 161 | + fixed_versions = "".join(fixed[lower + 1 : upper]).split(",") |
| 162 | + aliases.append(snyk_id) |
| 163 | + return VendorData( |
| 164 | + aliases=aliases, |
| 165 | + affected_versions=affected, |
| 166 | + fixed_versions=fixed_versions, |
| 167 | + ) |
| 168 | + |
| 169 | + |
| 170 | +def is_purl_in_affected(version, affected): |
| 171 | + for affected_range in affected: |
| 172 | + if snky_constraints_satisfied(affected_range, version): |
| 173 | + return True |
| 174 | + return False |
| 175 | + |
| 176 | + |
| 177 | +def generate_advisory_payload(snyk_id): |
| 178 | + return f"https://security.snyk.io/vuln/{snyk_id}" |
0 commit comments