comparison scripts/safety.py @ 2:02c3e335a695 draft

"planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/table_compute commit d00a518202228b990aeeea2ec3f842501fd2ec09"

1:dddadbbac949

    __allowed_tokens = (
        '(', ')', 'if', 'else', 'or', 'and', 'not', 'in',
        '+', '-', '*', '/', '%', ',', '!=', '==', '>', '>=', '<', '<=',
        'min', 'max', 'sum',
        'str', 'int', 'float'
    )
    __allowed_ref_types = {
        'pd.DataFrame': {
            'abs', 'add', 'agg', 'aggregate', 'align', 'all', 'any', 'append',
            'apply', 'applymap', 'as_matrix', 'asfreq', 'at', 'axes', 'bool',

        # ]
        # or
        # ['vec.median', '(', ')', '+', 'vec.sum', '(', ')']
        tokens = [
            e for e in re.split(
                r'("[a-zA-Z%0-9_.]+"|[a-zA-Z0-9_.]+|[^a-zA-Z0-9_.() ]+|[()])', self.expr
            ) if e.strip()
        ]

        # 2. Subtract allowed standard tokens
        rem = [e for e in tokens if e not in self.__allowed_tokens]
        # 3. Subtract allowed qualified objects from allowed modules
        #    and whitelisted references and their attributes
        rem2 = []
        for e in rem:
            parts = e.split('.')

                if parts[0] in self.these:
                    continue
            if len(parts) == 2:
                if parts[0] in self.these:
                    parts[0] = '_this'
                elif parts[0] == "":
                    # e.g. '.T' gives ['','.T']
                    # Here we assume that the blank part[0] refers to the
                    # self.ref_type (e.g. "pd.DataFrame"), and that
                    # the second part is a function of that type.
                    if parts[1] in self.allowed_qualified['_this']:
                        continue

                if parts[0] in self.allowed_qualified:
                    if parts[1] in self.allowed_qualified[parts[0]]:
                        continue

            rem2.append(e)

        # Debug
        # for x in (tokens, rem, rem2):print(x)

        # 4. Assert that rest are real numbers or strings
        e = ''
        for e in rem2:
            try:
                _ = float(e)
            except ValueError:
                # e.g. '"TEXT"' is okay.
                if not(e[0] == '"' and e[-1] == '"'):
                    safe = False
                    break

        return safe, e

2:02c3e335a695

    __allowed_tokens = (
        '(', ')', 'if', 'else', 'or', 'and', 'not', 'in',
        '+', '-', '*', '/', '%', ',', '!=', '==', '>', '>=', '<', '<=',
        'min', 'max', 'sum',
    )
    __allowed_ref_types = {
        'pd.DataFrame': {
            'abs', 'add', 'agg', 'aggregate', 'align', 'all', 'any', 'append',
            'apply', 'applymap', 'as_matrix', 'asfreq', 'at', 'axes', 'bool',

        # ]
        # or
        # ['vec.median', '(', ')', '+', 'vec.sum', '(', ')']
        tokens = [
            e for e in re.split(
                r'([a-zA-Z0-9_.]+|[^a-zA-Z0-9_.() ]+|[()])', self.expr
            ) if e.strip()
        ]

        # 2. Subtract allowed standard tokens
        rem = [e for e in tokens if e not in self.__allowed_tokens]

        # 3. Subtract allowed qualified objects from allowed modules
        #    and whitelisted references and their attributes
        rem2 = []
        for e in rem:
            parts = e.split('.')

                if parts[0] in self.these:
                    continue
            if len(parts) == 2:
                if parts[0] in self.these:
                    parts[0] = '_this'
                if parts[0] in self.allowed_qualified:
                    if parts[1] in self.allowed_qualified[parts[0]]:
                        continue
            rem2.append(e)

        # 4. Assert that rest are real numbers or strings
        e = ''
        for e in rem2:
            try:
                _ = float(e)
            except ValueError:
                safe = False
                break

        return safe, e