summaryrefslogtreecommitdiff
path: root/lib/parser.py
blob: 58dd7fadfcde81cb1f7ac968ed670e526693dfd1 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
'''
    Parent class of all parser
'''

import hachoir_core
import hachoir_editor

import os

import mat

NOMETA = ('.bmp', '.rdf', '.txt', '.xml', '.rels')
#bmp : image
#rdf : text
#txt : plain text
#xml : formated text
#rels : openxml foramted text


class GenericParser(object):
    '''
        Parent class of all parsers
    '''
    def __init__(self, filename, parser, mime, backup, add2archive):
        self.filename = ''
        self.parser = parser
        self.mime = mime
        self.backup = backup
        self.editor = hachoir_editor.createEditor(parser)
        self.realname = filename
        try:
            self.filename = hachoir_core.cmd_line.unicodeFilename(filename)
        except TypeError:  # get rid of "decoding Unicode is not supported"
            self.filename = filename
        basename, ext = os.path.splitext(filename)
        self.output = basename + '.cleaned' + ext
        self.basename = os.path.basename(filename)  # only filename

    def is_clean(self):
        '''
            Check if the file is clean from harmful metadatas
        '''
        for field in self.editor:
            if self._should_remove(field):
                return False
        return True

    def remove_all(self):
        '''
            Remove all the files that are compromizing
        '''
        for field in self.editor:
            if self._should_remove(field):
                self._remove(field.name)
        hachoir_core.field.writeIntoFile(self.editor, self.output)
        self.do_backup()

    def remove_all_ugly(self):
        '''
            If the remove_all() is not efficient enough,
            this method is implemented :
            It is efficient, but destructive.
            In a perfect world, with nice fileformat,
            this method would not exist.
        '''
        self.remove_all()

    def _remove(self, field):
        '''
            Delete the given field
        '''
        del self.editor[field]

    def get_meta(self):
        '''
            Return a dict with all the meta of the file
        '''
        metadata = {}
        for field in self.editor:
            if self._should_remove(field):
                try:
                    metadata[field.name] = field.value
                except:
                    metadata[field.name] = 'harmful content'
        return metadata

    def _should_remove(self, key):
        '''
            return True if the field is compromizing
            abstract method
        '''
        raise NotImplementedError

    def do_backup(self):
        '''
            Do a backup of the file if asked,
            and change his creation/access date
        '''
        if self.backup is True:
            os.utime(self.output, (0, 0))
        else:
            mat.secure_remove(self.filename)
            os.rename(self.output, self.filename)
            os.utime(self.filename, (0, 0))