diff options
| author | Richard Purdie <richard.purdie@linuxfoundation.org> | 2020-07-07 14:05:26 +0100 |
|---|---|---|
| committer | Richard Purdie <richard.purdie@linuxfoundation.org> | 2020-07-08 10:37:12 +0100 |
| commit | 0dae63359e8c4298b218dd4dbc238511f637115c (patch) | |
| tree | 0593337df1942259ec1b0dc9b55d0f8b430db32c | |
| parent | 5c2381f9ebcfb9b7991cc40429f286ae3557fd1e (diff) | |
| download | poky-0dae63359e8c4298b218dd4dbc238511f637115c.tar.gz | |
spdx: Remove the class as its obsolete
The project is a strong supporter of SPDX but this class is old code using
a dated approach which now misleads people. Remove it.
The meta-sdpxscanner layer is a much more modern and active approach to
handling this and we should be pointing people there.
(From OE-Core rev: 871599bf76b3fb635efd416340803f7bf3341f7e)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
| -rw-r--r-- | meta/classes/spdx.bbclass | 360 |
1 files changed, 0 insertions, 360 deletions
diff --git a/meta/classes/spdx.bbclass b/meta/classes/spdx.bbclass deleted file mode 100644 index fb78e274a8..0000000000 --- a/meta/classes/spdx.bbclass +++ /dev/null | |||
| @@ -1,360 +0,0 @@ | |||
| 1 | # This class integrates real-time license scanning, generation of SPDX standard | ||
| 2 | # output and verifiying license info during the building process. | ||
| 3 | # It is a combination of efforts from the OE-Core, SPDX and Fossology projects. | ||
| 4 | # | ||
| 5 | # For more information on FOSSology: | ||
| 6 | # http://www.fossology.org | ||
| 7 | # | ||
| 8 | # For more information on FOSSologySPDX commandline: | ||
| 9 | # https://github.com/spdx-tools/fossology-spdx/wiki/Fossology-SPDX-Web-API | ||
| 10 | # | ||
| 11 | # For more information on SPDX: | ||
| 12 | # http://www.spdx.org | ||
| 13 | # | ||
| 14 | |||
| 15 | # SPDX file will be output to the path which is defined as[SPDX_MANIFEST_DIR] | ||
| 16 | # in ./meta/conf/licenses.conf. | ||
| 17 | |||
| 18 | SPDXSSTATEDIR = "${WORKDIR}/spdx_sstate_dir" | ||
| 19 | |||
| 20 | # If ${S} isn't actually the top-level source directory, set SPDX_S to point at | ||
| 21 | # the real top-level directory. | ||
| 22 | SPDX_S ?= "${S}" | ||
| 23 | |||
| 24 | python do_spdx () { | ||
| 25 | import os, sys | ||
| 26 | import json, shutil | ||
| 27 | |||
| 28 | info = {} | ||
| 29 | info['workdir'] = d.getVar('WORKDIR') | ||
| 30 | info['sourcedir'] = d.getVar('SPDX_S') | ||
| 31 | info['pn'] = d.getVar('PN') | ||
| 32 | info['pv'] = d.getVar('PV') | ||
| 33 | info['spdx_version'] = d.getVar('SPDX_VERSION') | ||
| 34 | info['data_license'] = d.getVar('DATA_LICENSE') | ||
| 35 | |||
| 36 | sstatedir = d.getVar('SPDXSSTATEDIR') | ||
| 37 | sstatefile = os.path.join(sstatedir, info['pn'] + info['pv'] + ".spdx") | ||
| 38 | |||
| 39 | manifest_dir = d.getVar('SPDX_MANIFEST_DIR') | ||
| 40 | info['outfile'] = os.path.join(manifest_dir, info['pn'] + ".spdx" ) | ||
| 41 | |||
| 42 | info['spdx_temp_dir'] = d.getVar('SPDX_TEMP_DIR') | ||
| 43 | info['tar_file'] = os.path.join(info['workdir'], info['pn'] + ".tar.gz" ) | ||
| 44 | |||
| 45 | # Make sure important dirs exist | ||
| 46 | try: | ||
| 47 | bb.utils.mkdirhier(manifest_dir) | ||
| 48 | bb.utils.mkdirhier(sstatedir) | ||
| 49 | bb.utils.mkdirhier(info['spdx_temp_dir']) | ||
| 50 | except OSError as e: | ||
| 51 | bb.error("SPDX: Could not set up required directories: " + str(e)) | ||
| 52 | return | ||
| 53 | |||
| 54 | ## get everything from cache. use it to decide if | ||
| 55 | ## something needs to be rerun | ||
| 56 | cur_ver_code = get_ver_code(info['sourcedir']) | ||
| 57 | cache_cur = False | ||
| 58 | if os.path.exists(sstatefile): | ||
| 59 | ## cache for this package exists. read it in | ||
| 60 | cached_spdx = get_cached_spdx(sstatefile) | ||
| 61 | |||
| 62 | if cached_spdx['PackageVerificationCode'] == cur_ver_code: | ||
| 63 | bb.warn("SPDX: Verification code for " + info['pn'] | ||
| 64 | + "is same as cache's. do nothing") | ||
| 65 | cache_cur = True | ||
| 66 | else: | ||
| 67 | local_file_info = setup_foss_scan(info, True, cached_spdx['Files']) | ||
| 68 | else: | ||
| 69 | local_file_info = setup_foss_scan(info, False, None) | ||
| 70 | |||
| 71 | if cache_cur: | ||
| 72 | spdx_file_info = cached_spdx['Files'] | ||
| 73 | foss_package_info = cached_spdx['Package'] | ||
| 74 | foss_license_info = cached_spdx['Licenses'] | ||
| 75 | else: | ||
| 76 | ## setup fossology command | ||
| 77 | foss_server = d.getVar('FOSS_SERVER') | ||
| 78 | foss_flags = d.getVar('FOSS_WGET_FLAGS') | ||
| 79 | foss_full_spdx = d.getVar('FOSS_FULL_SPDX') == "true" or False | ||
| 80 | foss_command = "wget %s --post-file=%s %s"\ | ||
| 81 | % (foss_flags, info['tar_file'], foss_server) | ||
| 82 | |||
| 83 | foss_result = run_fossology(foss_command, foss_full_spdx) | ||
| 84 | if foss_result is not None: | ||
| 85 | (foss_package_info, foss_file_info, foss_license_info) = foss_result | ||
| 86 | spdx_file_info = create_spdx_doc(local_file_info, foss_file_info) | ||
| 87 | ## write to cache | ||
| 88 | write_cached_spdx(sstatefile, cur_ver_code, foss_package_info, | ||
| 89 | spdx_file_info, foss_license_info) | ||
| 90 | else: | ||
| 91 | bb.error("SPDX: Could not communicate with FOSSology server. Command was: " + foss_command) | ||
| 92 | return | ||
| 93 | |||
| 94 | ## Get document and package level information | ||
| 95 | spdx_header_info = get_header_info(info, cur_ver_code, foss_package_info) | ||
| 96 | |||
| 97 | ## CREATE MANIFEST | ||
| 98 | create_manifest(info, spdx_header_info, spdx_file_info, foss_license_info) | ||
| 99 | |||
| 100 | ## clean up the temp stuff | ||
| 101 | shutil.rmtree(info['spdx_temp_dir'], ignore_errors=True) | ||
| 102 | if os.path.exists(info['tar_file']): | ||
| 103 | remove_file(info['tar_file']) | ||
| 104 | } | ||
| 105 | addtask spdx after do_patch before do_configure | ||
| 106 | |||
| 107 | def create_manifest(info, header, files, licenses): | ||
| 108 | import codecs | ||
| 109 | with codecs.open(info['outfile'], mode='w', encoding='utf-8') as f: | ||
| 110 | # Write header | ||
| 111 | f.write(header + '\n') | ||
| 112 | |||
| 113 | # Write file data | ||
| 114 | for chksum, block in files.iteritems(): | ||
| 115 | f.write("FileName: " + block['FileName'] + '\n') | ||
| 116 | for key, value in block.iteritems(): | ||
| 117 | if not key == 'FileName': | ||
| 118 | f.write(key + ": " + value + '\n') | ||
| 119 | f.write('\n') | ||
| 120 | |||
| 121 | # Write license data | ||
| 122 | for id, block in licenses.iteritems(): | ||
| 123 | f.write("LicenseID: " + id + '\n') | ||
| 124 | for key, value in block.iteritems(): | ||
| 125 | f.write(key + ": " + value + '\n') | ||
| 126 | f.write('\n') | ||
| 127 | |||
| 128 | def get_cached_spdx(sstatefile): | ||
| 129 | import json | ||
| 130 | import codecs | ||
| 131 | cached_spdx_info = {} | ||
| 132 | with codecs.open(sstatefile, mode='r', encoding='utf-8') as f: | ||
| 133 | try: | ||
| 134 | cached_spdx_info = json.load(f) | ||
| 135 | except ValueError as e: | ||
| 136 | cached_spdx_info = None | ||
| 137 | return cached_spdx_info | ||
| 138 | |||
| 139 | def write_cached_spdx(sstatefile, ver_code, package_info, files, license_info): | ||
| 140 | import json | ||
| 141 | import codecs | ||
| 142 | spdx_doc = {} | ||
| 143 | spdx_doc['PackageVerificationCode'] = ver_code | ||
| 144 | spdx_doc['Files'] = {} | ||
| 145 | spdx_doc['Files'] = files | ||
| 146 | spdx_doc['Package'] = {} | ||
| 147 | spdx_doc['Package'] = package_info | ||
| 148 | spdx_doc['Licenses'] = {} | ||
| 149 | spdx_doc['Licenses'] = license_info | ||
| 150 | with codecs.open(sstatefile, mode='w', encoding='utf-8') as f: | ||
| 151 | f.write(json.dumps(spdx_doc)) | ||
| 152 | |||
| 153 | def setup_foss_scan(info, cache, cached_files): | ||
| 154 | import errno, shutil | ||
| 155 | import tarfile | ||
| 156 | file_info = {} | ||
| 157 | cache_dict = {} | ||
| 158 | |||
| 159 | for f_dir, f in list_files(info['sourcedir']): | ||
| 160 | full_path = os.path.join(f_dir, f) | ||
| 161 | abs_path = os.path.join(info['sourcedir'], full_path) | ||
| 162 | dest_dir = os.path.join(info['spdx_temp_dir'], f_dir) | ||
| 163 | dest_path = os.path.join(info['spdx_temp_dir'], full_path) | ||
| 164 | |||
| 165 | checksum = hash_file(abs_path) | ||
| 166 | if not checksum is None: | ||
| 167 | file_info[checksum] = {} | ||
| 168 | ## retain cache information if it exists | ||
| 169 | if cache and checksum in cached_files: | ||
| 170 | file_info[checksum] = cached_files[checksum] | ||
| 171 | ## have the file included in what's sent to the FOSSology server | ||
| 172 | else: | ||
| 173 | file_info[checksum]['FileName'] = full_path | ||
| 174 | try: | ||
| 175 | bb.utils.mkdirhier(dest_dir) | ||
| 176 | shutil.copyfile(abs_path, dest_path) | ||
| 177 | except OSError as e: | ||
| 178 | bb.warn("SPDX: mkdirhier failed: " + str(e)) | ||
| 179 | except shutil.Error as e: | ||
| 180 | bb.warn("SPDX: copyfile failed: " + str(e)) | ||
| 181 | except IOError as e: | ||
| 182 | bb.warn("SPDX: copyfile failed: " + str(e)) | ||
| 183 | else: | ||
| 184 | bb.warn("SPDX: Could not get checksum for file: " + f) | ||
| 185 | |||
| 186 | with tarfile.open(info['tar_file'], "w:gz") as tar: | ||
| 187 | tar.add(info['spdx_temp_dir'], arcname=os.path.basename(info['spdx_temp_dir'])) | ||
| 188 | |||
| 189 | return file_info | ||
| 190 | |||
| 191 | def remove_file(file_name): | ||
| 192 | try: | ||
| 193 | os.remove(file_name) | ||
| 194 | except OSError as e: | ||
| 195 | pass | ||
| 196 | |||
| 197 | def list_files(dir): | ||
| 198 | for root, subFolders, files in os.walk(dir): | ||
| 199 | for f in files: | ||
| 200 | rel_root = os.path.relpath(root, dir) | ||
| 201 | yield rel_root, f | ||
| 202 | return | ||
| 203 | |||
| 204 | def hash_file(file_name): | ||
| 205 | from bb.utils import sha1_file | ||
| 206 | return sha1_file(file_name) | ||
| 207 | |||
| 208 | def hash_string(data): | ||
| 209 | import hashlib | ||
| 210 | sha1 = hashlib.sha1() | ||
| 211 | sha1.update(data.encode('utf-8')) | ||
| 212 | return sha1.hexdigest() | ||
| 213 | |||
| 214 | def run_fossology(foss_command, full_spdx): | ||
| 215 | import string, re | ||
| 216 | import subprocess | ||
| 217 | |||
| 218 | try: | ||
| 219 | foss_output = subprocess.check_output(foss_command.split(), | ||
| 220 | stderr=subprocess.STDOUT).decode('utf-8') | ||
| 221 | except subprocess.CalledProcessError as e: | ||
| 222 | return None | ||
| 223 | |||
| 224 | foss_output = foss_output.replace('\r', '') | ||
| 225 | |||
| 226 | # Package info | ||
| 227 | package_info = {} | ||
| 228 | if full_spdx: | ||
| 229 | # All mandatory, only one occurrence | ||
| 230 | package_info['PackageCopyrightText'] = re.findall('PackageCopyrightText: (.*?</text>)', foss_output, re.S)[0] | ||
| 231 | package_info['PackageLicenseDeclared'] = re.findall('PackageLicenseDeclared: (.*)', foss_output)[0] | ||
| 232 | package_info['PackageLicenseConcluded'] = re.findall('PackageLicenseConcluded: (.*)', foss_output)[0] | ||
| 233 | # These may be more than one | ||
| 234 | package_info['PackageLicenseInfoFromFiles'] = re.findall('PackageLicenseInfoFromFiles: (.*)', foss_output) | ||
| 235 | else: | ||
| 236 | DEFAULT = "NOASSERTION" | ||
| 237 | package_info['PackageCopyrightText'] = "<text>" + DEFAULT + "</text>" | ||
| 238 | package_info['PackageLicenseDeclared'] = DEFAULT | ||
| 239 | package_info['PackageLicenseConcluded'] = DEFAULT | ||
| 240 | package_info['PackageLicenseInfoFromFiles'] = [] | ||
| 241 | |||
| 242 | # File info | ||
| 243 | file_info = {} | ||
| 244 | records = [] | ||
| 245 | # FileName is also in PackageFileName, so we match on FileType as well. | ||
| 246 | records = re.findall('FileName:.*?FileType:.*?</text>', foss_output, re.S) | ||
| 247 | for rec in records: | ||
| 248 | chksum = re.findall('FileChecksum: SHA1: (.*)\n', rec)[0] | ||
| 249 | file_info[chksum] = {} | ||
| 250 | file_info[chksum]['FileCopyrightText'] = re.findall('FileCopyrightText: ' | ||
| 251 | + '(.*?</text>)', rec, re.S )[0] | ||
| 252 | fields = ['FileName', 'FileType', 'LicenseConcluded', 'LicenseInfoInFile'] | ||
| 253 | for field in fields: | ||
| 254 | file_info[chksum][field] = re.findall(field + ': (.*)', rec)[0] | ||
| 255 | |||
| 256 | # Licenses | ||
| 257 | license_info = {} | ||
| 258 | licenses = [] | ||
| 259 | licenses = re.findall('LicenseID:.*?LicenseName:.*?\n', foss_output, re.S) | ||
| 260 | for lic in licenses: | ||
| 261 | license_id = re.findall('LicenseID: (.*)\n', lic)[0] | ||
| 262 | license_info[license_id] = {} | ||
| 263 | license_info[license_id]['ExtractedText'] = re.findall('ExtractedText: (.*?</text>)', lic, re.S)[0] | ||
| 264 | license_info[license_id]['LicenseName'] = re.findall('LicenseName: (.*)', lic)[0] | ||
| 265 | |||
| 266 | return (package_info, file_info, license_info) | ||
| 267 | |||
| 268 | def create_spdx_doc(file_info, scanned_files): | ||
| 269 | import json | ||
| 270 | ## push foss changes back into cache | ||
| 271 | for chksum, lic_info in scanned_files.iteritems(): | ||
| 272 | if chksum in file_info: | ||
| 273 | file_info[chksum]['FileType'] = lic_info['FileType'] | ||
| 274 | file_info[chksum]['FileChecksum: SHA1'] = chksum | ||
| 275 | file_info[chksum]['LicenseInfoInFile'] = lic_info['LicenseInfoInFile'] | ||
| 276 | file_info[chksum]['LicenseConcluded'] = lic_info['LicenseConcluded'] | ||
| 277 | file_info[chksum]['FileCopyrightText'] = lic_info['FileCopyrightText'] | ||
| 278 | else: | ||
| 279 | bb.warn("SPDX: " + lic_info['FileName'] + " : " + chksum | ||
| 280 | + " : is not in the local file info: " | ||
| 281 | + json.dumps(lic_info, indent=1)) | ||
| 282 | return file_info | ||
| 283 | |||
| 284 | def get_ver_code(dirname): | ||
| 285 | chksums = [] | ||
| 286 | for f_dir, f in list_files(dirname): | ||
| 287 | path = os.path.join(dirname, f_dir, f) | ||
| 288 | hash = hash_file(path) | ||
| 289 | if not hash is None: | ||
| 290 | chksums.append(hash) | ||
| 291 | else: | ||
| 292 | bb.warn("SPDX: Could not hash file: " + path) | ||
| 293 | ver_code_string = ''.join(chksums).lower() | ||
| 294 | ver_code = hash_string(ver_code_string) | ||
| 295 | return ver_code | ||
| 296 | |||
| 297 | def get_header_info(info, spdx_verification_code, package_info): | ||
| 298 | """ | ||
| 299 | Put together the header SPDX information. | ||
| 300 | Eventually this needs to become a lot less | ||
| 301 | of a hardcoded thing. | ||
| 302 | """ | ||
| 303 | from datetime import datetime | ||
| 304 | import os | ||
| 305 | head = [] | ||
| 306 | DEFAULT = "NOASSERTION" | ||
| 307 | |||
| 308 | package_checksum = hash_file(info['tar_file']) | ||
| 309 | if package_checksum is None: | ||
| 310 | package_checksum = DEFAULT | ||
| 311 | |||
| 312 | ## document level information | ||
| 313 | head.append("## SPDX Document Information") | ||
| 314 | head.append("SPDXVersion: " + info['spdx_version']) | ||
| 315 | head.append("DataLicense: " + info['data_license']) | ||
| 316 | head.append("DocumentComment: <text>SPDX for " | ||
| 317 | + info['pn'] + " version " + info['pv'] + "</text>") | ||
| 318 | head.append("") | ||
| 319 | |||
| 320 | ## Creator information | ||
| 321 | ## Note that this does not give time in UTC. | ||
| 322 | now = datetime.now().strftime('%Y-%m-%dT%H:%M:%SZ') | ||
| 323 | head.append("## Creation Information") | ||
| 324 | ## Tools are supposed to have a version, but FOSSology+SPDX provides none. | ||
| 325 | head.append("Creator: Tool: FOSSology+SPDX") | ||
| 326 | head.append("Created: " + now) | ||
| 327 | head.append("CreatorComment: <text>UNO</text>") | ||
| 328 | head.append("") | ||
| 329 | |||
| 330 | ## package level information | ||
| 331 | head.append("## Package Information") | ||
| 332 | head.append("PackageName: " + info['pn']) | ||
| 333 | head.append("PackageVersion: " + info['pv']) | ||
| 334 | head.append("PackageFileName: " + os.path.basename(info['tar_file'])) | ||
| 335 | head.append("PackageSupplier: Person:" + DEFAULT) | ||
| 336 | head.append("PackageDownloadLocation: " + DEFAULT) | ||
| 337 | head.append("PackageSummary: <text></text>") | ||
| 338 | head.append("PackageOriginator: Person:" + DEFAULT) | ||
| 339 | head.append("PackageChecksum: SHA1: " + package_checksum) | ||
| 340 | head.append("PackageVerificationCode: " + spdx_verification_code) | ||
| 341 | head.append("PackageDescription: <text>" + info['pn'] | ||
| 342 | + " version " + info['pv'] + "</text>") | ||
| 343 | head.append("") | ||
| 344 | head.append("PackageCopyrightText: " | ||
| 345 | + package_info['PackageCopyrightText']) | ||
| 346 | head.append("") | ||
| 347 | head.append("PackageLicenseDeclared: " | ||
| 348 | + package_info['PackageLicenseDeclared']) | ||
| 349 | head.append("PackageLicenseConcluded: " | ||
| 350 | + package_info['PackageLicenseConcluded']) | ||
| 351 | |||
| 352 | for licref in package_info['PackageLicenseInfoFromFiles']: | ||
| 353 | head.append("PackageLicenseInfoFromFiles: " + licref) | ||
| 354 | head.append("") | ||
| 355 | |||
| 356 | ## header for file level | ||
| 357 | head.append("## File Information") | ||
| 358 | head.append("") | ||
| 359 | |||
| 360 | return '\n'.join(head) | ||
