Mercurial > repos > iuc > table_compute

comparison scripts/safety.py @ 1:dddadbbac949 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
"planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/table_compute commit 6820ec9431a22576f3716c40feeb27f0b8cf5e83"
author iuc
date Fri, 30 Aug 2019 05:28:18 -0400
parents 1b0f96ed73f2
children 02c3e335a695
comparison
equal deleted inserted replaced
0:1b0f96ed73f2 1:dddadbbac949
9 9
10 __allowed_tokens = ( 10 __allowed_tokens = (
11 '(', ')', 'if', 'else', 'or', 'and', 'not', 'in', 11 '(', ')', 'if', 'else', 'or', 'and', 'not', 'in',
12 '+', '-', '*', '/', '%', ',', '!=', '==', '>', '>=', '<', '<=', 12 '+', '-', '*', '/', '%', ',', '!=', '==', '>', '>=', '<', '<=',
13 'min', 'max', 'sum', 13 'min', 'max', 'sum',
14 'str', 'int', 'float'
14 ) 15 )
15 __allowed_ref_types = { 16 __allowed_ref_types = {
16 'pd.DataFrame': { 17 'pd.DataFrame': {
17 'abs', 'add', 'agg', 'aggregate', 'align', 'all', 'any', 'append', 18 'abs', 'add', 'agg', 'aggregate', 'align', 'all', 'any', 'append',
18 'apply', 'applymap', 'as_matrix', 'asfreq', 'at', 'axes', 'bool', 19 'apply', 'applymap', 'as_matrix', 'asfreq', 'at', 'axes', 'bool',
161 and their methods 162 and their methods
162 """ 163 """
163 164
164 safe = True 165 safe = True
165 # examples of user-expressions 166 # examples of user-expressions
166 # '-math.log(1 - elem/4096) * 4096 if elem != bn else elem - 0.5' 167 # '-math.log(1 - elem/4096) * 4096 if elem != 1 else elem - 0.5'
167 # 'vec.median() + vec.sum()' 168 # 'vec.median() + vec.sum()'
168 169
169 # 1. Break expressions into tokens 170 # 1. Break expressions into tokens
170 # e.g., 171 # e.g.,
171 # [ 172 # [
172 # '-', 'math.log', '(', '1', '-', 'elem', '/', '4096', ')', '*', 173 # '-', 'math.log', '(', '1', '-', 'elem', '/', '4096', ')', '*',
173 # '4096', 'if', 'elem', '!=', 'bn', 'else', 'elem', '-', '0.5' 174 # '4096', 'if', 'elem', '!=', '1', 'else', 'elem', '-', '0.5'
174 # ] 175 # ]
175 # or 176 # or
176 # ['vec.median', '(', ')', '+', 'vec.sum', '(', ')'] 177 # ['vec.median', '(', ')', '+', 'vec.sum', '(', ')']
177 tokens = [ 178 tokens = [
178 e for e in re.split( 179 e for e in re.split(
179 r'([a-zA-Z0-9_.]+|[^a-zA-Z0-9_.() ]+|[()])', self.expr 180 r'("[a-zA-Z%0-9_.]+"|[a-zA-Z0-9_.]+|[^a-zA-Z0-9_.() ]+|[()])', self.expr
180 ) if e.strip() 181 ) if e.strip()
181 ] 182 ]
182 183
183 # 2. Subtract allowed standard tokens 184 # 2. Subtract allowed standard tokens
184 rem = [e for e in tokens if e not in self.__allowed_tokens] 185 rem = [e for e in tokens if e not in self.__allowed_tokens]
185
186 # 3. Subtract allowed qualified objects from allowed modules 186 # 3. Subtract allowed qualified objects from allowed modules
187 # and whitelisted references and their attributes 187 # and whitelisted references and their attributes
188 rem2 = [] 188 rem2 = []
189 for e in rem: 189 for e in rem:
190 parts = e.split('.') 190 parts = e.split('.')
192 if parts[0] in self.these: 192 if parts[0] in self.these:
193 continue 193 continue
194 if len(parts) == 2: 194 if len(parts) == 2:
195 if parts[0] in self.these: 195 if parts[0] in self.these:
196 parts[0] = '_this' 196 parts[0] = '_this'
197 elif parts[0] == "":
198 # e.g. '.T' gives ['','.T']
199 # Here we assume that the blank part[0] refers to the
200 # self.ref_type (e.g. "pd.DataFrame"), and that
201 # the second part is a function of that type.
202 if parts[1] in self.allowed_qualified['_this']:
203 continue
204
197 if parts[0] in self.allowed_qualified: 205 if parts[0] in self.allowed_qualified:
198 if parts[1] in self.allowed_qualified[parts[0]]: 206 if parts[1] in self.allowed_qualified[parts[0]]:
199 continue 207 continue
208
200 rem2.append(e) 209 rem2.append(e)
201 210
202 # 4. Assert that rest are real numbers 211 # Debug
212 # for x in (tokens, rem, rem2):print(x)
213
214 # 4. Assert that rest are real numbers or strings
203 e = '' 215 e = ''
204 for e in rem2: 216 for e in rem2:
205 try: 217 try:
206 _ = float(e) 218 _ = float(e)
207 except ValueError: 219 except ValueError:
208 safe = False 220 # e.g. '"TEXT"' is okay.
209 break 221 if not(e[0] == '"' and e[-1] == '"'):
222 safe = False
223 break
210 224
211 return safe, e 225 return safe, e