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
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
|
''' Care about office's formats
'''
import logging
import os
import shutil
import tempfile
import xml.dom.minidom as minidom
import zipfile
try:
import cairo
from gi.repository import Poppler
except ImportError:
logging.info('office.py loaded without PDF support')
pass
import parser
import archive
class OpenDocumentStripper(archive.TerminalZipStripper):
''' An open document file is a zip, with xml file into.
The one that interest us is meta.xml
'''
def get_meta(self):
''' Return a dict with all the meta of the file by
trying to read the meta.xml file.
'''
metadata = super(OpenDocumentStripper, self).get_meta()
zipin = zipfile.ZipFile(self.filename, 'r')
try:
content = zipin.read('meta.xml')
dom1 = minidom.parseString(content)
elements = dom1.getElementsByTagName('office:meta')
for i in elements[0].childNodes:
if i.tagName != 'meta:document-statistic':
nodename = ''.join(i.nodeName.split(':')[1:])
metadata[nodename] = ''.join([j.data for j in i.childNodes])
else:
# thank you w3c for not providing a nice
# method to get all attributes of a node
pass
except KeyError: # no meta.xml file found
logging.debug('%s has no opendocument metadata' % self.filename)
zipin.close()
return metadata
def remove_all(self):
''' Removes metadata
'''
return super(OpenDocumentStripper, self).remove_all(ending_blacklist=['meta.xml'])
def is_clean(self):
''' Check if the file is clean from harmful metadatas
'''
clean_super = super(OpenDocumentStripper, self).is_clean()
if clean_super is False:
return False
zipin = zipfile.ZipFile(self.filename, 'r')
try:
zipin.getinfo('meta.xml')
except KeyError: # no meta.xml in the file
return True
zipin.close()
return False
class OpenXmlStripper(archive.TerminalZipStripper):
''' Represent an office openxml document, which is like
an opendocument format, with some tricky stuff added.
It contains mostly xml, but can have media blobs, crap, ...
(I don't like this format.)
'''
def remove_all(self):
return super(OpenXmlStripper, self).remove_all(
beginning_blacklist=('docProps/'), whitelist=('.rels'))
def is_clean(self):
''' Check if the file is clean from harmful metadatas.
This implementation is faster than something like
"return this.get_meta() == {}".
'''
clean_super = super(OpenXmlStripper, self).is_clean()
if clean_super is False:
return False
zipin = zipfile.ZipFile(self.filename, 'r')
for item in zipin.namelist():
if item.startswith('docProps/'):
return False
zipin.close()
return True
def get_meta(self):
''' Return a dict with all the meta of the file
'''
metadata = super(OpenXmlStripper, self).get_meta()
zipin = zipfile.ZipFile(self.filename, 'r')
for item in zipin.namelist():
if item.startswith('docProps/'):
metadata[item] = 'harmful content'
zipin.close()
return metadata
class PdfStripper(parser.GenericParser):
''' Represent a PDF file
'''
def __init__(self, filename, parser, mime, backup, is_writable, **kwargs):
super(PdfStripper, self).__init__(filename, parser, mime, backup, is_writable, **kwargs)
self.uri = 'file://' + os.path.abspath(self.filename)
self.password = None
try:
self.pdf_quality = kwargs['low_pdf_quality']
except KeyError:
self.pdf_quality = False
self.meta_list = frozenset(['title', 'author', 'subject',
'keywords', 'creator', 'producer', 'metadata'])
def is_clean(self):
''' Check if the file is clean from harmful metadatas
'''
document = Poppler.Document.new_from_file(self.uri, self.password)
for key in self.meta_list:
if document.get_property(key):
return False
return True
def remove_all(self):
''' Opening the PDF with poppler, then doing a render
on a cairo pdfsurface for each pages.
http://cairographics.org/documentation/pycairo/2/
The use of an intermediate tempfile is necessary because
python-cairo segfaults on unicode.
See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=699457
'''
document = Poppler.Document.new_from_file(self.uri, self.password)
try:
output = tempfile.mkstemp()[1]
page = document.get_page(0)
# assume that every pages are the same size
page_width, page_height = page.get_size()
surface = cairo.PDFSurface(output, page_width, page_height)
context = cairo.Context(surface) # context draws on the surface
logging.debug('PDF rendering of %s' % self.filename)
for pagenum in range(document.get_n_pages()):
page = document.get_page(pagenum)
context.translate(0, 0)
if self.pdf_quality:
page.render(context) # render the page on context
else:
page.render_for_printing(context) # render the page on context
context.show_page() # draw context on surface
surface.finish()
shutil.move(output, self.output)
except:
logging.error('Something went wrong when cleaning %s.' % self.filename)
return False
try:
import pdfrw # For now, poppler cannot write meta, so we must use pdfrw
logging.debug('Removing %s\'s superficial metadata' % self.filename)
trailer = pdfrw.PdfReader(self.output)
trailer.Info.Producer = None
trailer.Info.Creator = None
writer = pdfrw.PdfWriter()
writer.trailer = trailer
writer.write(self.output)
self.do_backup()
except:
logging.error('Unable to remove all metadata from %s, please install pdfrw' % self.output)
return False
return True
def get_meta(self):
''' Return a dict with all the meta of the file
'''
document = Poppler.Document.new_from_file(self.uri, self.password)
metadata = {}
for key in self.meta_list:
if document.get_property(key):
metadata[key] = document.get_property(key)
return metadata
|
