scancode-toolkit/src/summarycode/generated.py at develop · aboutcode-org/scancode-toolkit · GitHub

History

234 lines (180 loc) · 7.09 KB

Raw

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

# ScanCode is a trademark of nexB Inc.

# SPDX-License-Identifier: Apache-2.0

# See http://www.apache.org/licenses/LICENSE-2.0 for the license text.

# See https://github.com/nexB/scancode-toolkit for support or download.

# See https://aboutcode.org for more information about nexB OSS projects.

from itertools import islice

from commoncode.datautils import Boolean

from commoncode.text import toascii

from plugincode.scan import ScanPlugin

from plugincode.scan import scan_impl

from commoncode.cliutils import PluggableCommandLineOption

from commoncode.cliutils import OTHER_SCAN_GROUP

import typecode.contenttype

"""

Tag files as generated code based on conspicuous strings.

"""

# Tracing flag

TRACE = False

def logger_debug(*args):

pass

if TRACE:

import logging

import sys

logger = logging.getLogger(__name__)

logging.basicConfig(stream=sys.stdout)

logger.setLevel(logging.DEBUG)

def logger_debug(*args):

return logger.debug(' '.join(isinstance(a, str) and a or repr(a) for a in args))

@scan_impl

class GeneratedCodeDetector(ScanPlugin):

"""

Tag a file as generated.

"""

resource_attributes = dict(is_generated=Boolean(

help='True if this file is likely an automatically generated file.'))

run_order = 50

sort_order = 50

options = [

PluggableCommandLineOption(('--generated',),

is_flag=True, default=False,

help='Classify automatically generated code files with a flag.',

help_group=OTHER_SCAN_GROUP,

sort_order=50,

)

]

def is_enabled(self, generated, **kwargs):

return generated

def get_scanner(self, **kwargs):

return generated_scanner

def generated_scanner(location, **kwargs):

is_generated = False

for _clue in get_generated_code_hint(location):

# TODO: consider returning the "clue"

is_generated = True

break

return dict(is_generated=is_generated)

GENERATED_KEYWORDS_LOWERED = tuple(g.lower() for g in (

'generated by',

'auto-generated',

'automatically generated',

# Apache Axis

'auto-generated from wsdl',

# jni javahl and others

'do not edit this file',

# jni javahl

'it is machine generated',

'by hibernate tools',

'generated from idl',

# castor generated files

'following schema fragment specifies the',

# Tomcat JSPC

'automatically created by',

# in GNU Classpath

'this file was automatically generated by gnu.localegen from cldr',

'this document is automatically generated by gnu.supplementgen',

# linux kernel/u-boot

'this was automagically generated from',

# angular

'this code is generated',

'this code is generated - do not modify',

# cython

'generated by cython',

# sqlite amalgamation

'this file is an amalgamation of many separate c source files from sqlite',

# various generated or last generated:

'generated on',

'last generated on',

# in freepascal unicode

"this is an automatically created file",

# generated by Postgres ECPG sql to c preprocessor

'processed by ecpg (regression mode)',

'these include files are added by the preprocessor',

'this readme is generated, please do not update',

'this file was automatically generated by',

'any changes will be lost if this file is regenerated',

# yarn lock files

'this is an autogenerated file. do not edit this file directly'

'this file is generated by',

'microsoft visual c++ generated include file',

# OpenJDK:

'generated by the idl-to-java compiler',

'this file was mechanically generated: do not edit!',

'generated by mc.java version 1.0, do not edit by hand!',

'generated from input file',

'generated content - do not edit',

'generators: org.graalvm',

# https://github.com/Microsoft/azure-devops-python-api/blob/0d9537016bd45cf7f2140433c0ec54b44768f726/vsts/vsts/licensing/v4_1/licensing_client.py

'generated file, do not edit',

'changes may cause incorrect behavior and will be lost if the code is regenerated.',

'this file is generated. best not to modify it, as it will likely be overwritten.',

# in Ogg/Vorbis

'function: static codebooks autogenerated by',

# yarn lock

'this is an autogenerated file',

'this is an autogenerated file. do not edit this file directly',

'this file has been automatically created by',

'please do not edit this file, all changes will be lost',

# thrift

'this is an automatically-generated file',

'it could get re-generated if the allow_idl_generation flag is on',

# aws sdk ruby

'warning about generated code',

'this file is generated',

'this file is generated. see the contributing guide for more information',

# seen in Go files:

'code generated file. do not edit',

'code generated by running "go generate". do not edit'

# protoc

'generated by the protocol buffer compiler',

'generated by the protocol buffer compiler. do not edit!',

'code generated by protoc-gen-go. do not edit.',

'code generated by generate-types. do not edit.',

'code generated by generate-protos. do not edit.',

'do not edit -- your changes will be discarded when the file is', # next line: * regenerated.

'do not edit -- generated code',

'generated code -- do not edit!',

'autogenerated by method_dump.sh. do not edit by hand.',

'this file is generated from the .proto files for the well-known', # next line: types. Do not edit!

'this file was generated by upbc (the upb compiler) from the input',

'do not edit -- your changes will be discarded when the file is', # next line: regenerated

# autotools

'makefile.in generated by automake',

'generated automatically by aclocal',

'generated by gnu autoconf',

'this file was automatically generated',

# misc

'this file has been generated by the gui designer. do not modify',

'file is generated by gopy gen. do not edit',

# javacc

'generated by:javacc: do not edit this line',

'generated by:jjtree: do not edit this line',

'generated code (do not edit this line)',

# in some Go projects like buildkit

'code generated by smithy-go-codegen do not edit',

))

def get_generated_code_hint(

location,

max_lines=150,

generated_keywords=GENERATED_KEYWORDS_LOWERED

"""

Return a line of extracted text from a file if that file is likely

generated source code.

for each of the first few lines of a source code file

if generated keywords are found in the line as lowercase

yield the line text as a 'potentially_ generated' annotation

"""

T = typecode.contenttype.get_type(location)

if not T.is_text:

return

with open(location, 'rb') as filein:

for line in islice(filein, max_lines):

text = toascii(line.strip()).lower()

if any(kw in text.lower() for kw in generated_keywords):

# yield only the first 100 chars..

yield text[:100]

scancode-toolkit/src/summarycode/generated.py at develop · aboutcode-org/scancode-toolkit · GitHub

Uh oh!

Uh oh!

scancode-toolkit/src/summarycode/generated.py at develop · aboutcode-org/scancode-toolkit · GitHub

Uh oh!

Uh oh!

Expand file tree

Breadcrumbs

generated.py

Latest commit

History

Breadcrumbs

generated.py

File metadata and controls