__init__.py 文件源码

def extract_phenocode_from_filepath(phenolist, regex):
    print("NOTE: working with {} phenos".format(len(phenolist)))
    if not isinstance(regex, re._pattern_type):
        regex = re.compile(regex)
    for pheno in phenolist:
        if 'assoc_files' not in pheno:
            raise PheWebError("ERROR: At least one phenotype doesn't have the key 'assoc_files'.")
        if not pheno['assoc_files']:
            raise PheWebError("ERROR: At least one phenotype has an empty 'assoc_files' list.")
        phenocodes = []
        for assoc_filepath in pheno['assoc_files']:
            match = re.search(regex, assoc_filepath)
            if match is None:
                raise PheWebError("ERROR: The regex {!r} doesn't match the filepath {!r}".format(regex.pattern, assoc_filepath))
            groups = match.groups()
            if len(groups) != 1:
                raise PheWebError("ERROR: The regex {!r} doesn't capture any groups on the filepath {!r}!  You're using parentheses without backslashes, right?".format(regex.pattern, assoc_filepath))
            phenocodes.append(groups[0])
        if len(set(phenocodes)) != 1:
            raise PheWebError("ERROR: At least one phenotype gets multiple different phenocodes from its several association filepaths.  Here they are: {!r}".format(list(set(phenocodes))))
        if 'phenocode' in pheno:
            if pheno['phenocode'] != phenocodes[0]:
                raise PheWebError("""\
ERROR: The regex {!r} matched the filepaths {!r} to produce the phenocode {!r}.  But that phenotype already had a phenocode, {!r}.
""".format(regex.pattern, pheno['assoc_files'], phenocodes[0], pheno['phenocode']))
        pheno['phenocode'] = phenocodes[0]
    return phenolist