def saveResults(audioMetadata, metadata, children):
for key in audioMetadata.keys():
#cherrypy.log("key:", key)
if not in_blacklist(key, audioMetadata[key]):
#cherrypy.log("TYPE:", type(audioMetadata[key]))
value = audioMetadata[key]
if hasattr(value, 'data'):
#an embedded binary
name = "cover"
if hasattr(value, 'desc'):
name = value.desc
childpath = tempfile.mkstemp(dir=insiderer.TMP_DIR)[1]
try:
tmp_handle = open(childpath, "wb")
tmp_handle.write(value.data)
tmp_handle.close()
child = insiderer.get_metadata(childpath, name)
children.append(child)
finally:
insiderer.safedelete(childpath)
continue
if not hasattr(value, '__len__') and hasattr(value, 'text'):
value = value.text
if hasattr(value, '__len__') and len(value) == 1:
value = value[0]
metadata[key] = value
def saveResults(audioMetadata, metadata, children):
for key in audioMetadata.keys():
#cherrypy.log("key:", key)
if not in_blacklist(key, audioMetadata[key]):
#cherrypy.log("TYPE:", type(audioMetadata[key]))
value = audioMetadata[key]
if hasattr(value, 'data'):
#an embedded binary
name = "cover"
if hasattr(value, 'desc'):
name = value.desc
childpath = tempfile.mkstemp(dir=insiderer.TMP_DIR)[1]
try:
with open(childpath, "wb") as tmp_handle:
tmp_handle.write(value.data)
tmp_handle.close()
child = insiderer.get_metadata(childpath, name)
children.append(child)
finally:
insiderer.safedelete(childpath)
continue
if not hasattr(value, '__len__') and hasattr(value, 'text'):
value = value.text
if hasattr(value, '__len__') and len(value) == 1:
value = value[0]
metadata[key] = value
def process_a_jpeg(jpeg_data):
try:
jpeg_path = tempfile.mkstemp(dir=insiderer.TMP_DIR)[1]
jpeg_handle = open(jpeg_path, "wb")
jpeg_handle.write(jpeg_data)
jpeg_handle.close()
import mimes.image
jpeg_metadata = dict()
mimes.image.image(jpeg_path, jpeg_metadata, None)
except Exception as e:
cherrypy.log("PDF JPEG exception", e)
finally:
insiderer.safedelete(jpeg_path)
return jpeg_metadata
def image_svg_xml(path, metadata, children):
svg = etree.parse(path)
fields = {
'export-filename': '//@inkscape:export-filename',
'docname': '//@sodipodi:docname',
'description': '//svg:desc',
'title': '//svg:title',
'dca': '//@*[namespace-uri()="%s"][not(ancestor::dc:*)]' % namespaces[
"dc"], #because selecting by namespace attributes doesn't seem to work
'rdf': '//rdf:*[not(ancestor::rdf:*)]'
}
for key in fields.keys():
results = svg.xpath(fields[key], namespaces=namespaces)
for result in results:
key_name = key
if '*' in fields[key] and len(results) > 0:
if hasattr(result, "attrname"):
key_name = re.sub(r'\{.*?\}', '', result.attrname)
else:
key_name = result.xpath('name()')
key_name = insiderer.de_dup(key_name, metadata)
if isinstance(result, etree._ElementUnicodeResult):
metadata[key_name] = str(result)
else:
metadata[key_name] = xmltodict.parse(
etree.tostring(result, encoding='utf8',
method='xml').decode('utf-8'))
base64images = svg.xpath('//svg:image', namespaces=namespaces)
for image in base64images:
key = image.xpath('local-name()')
href = image.attrib.get('{%s}href' % namespaces["xlink"])
name = image.attrib.get('id') or "unknown"
if href.startswith("data:") and base64_prefix in href[0:100]:
image_data = base64.standard_b64decode(
href[href.find(base64_prefix) + len(base64_prefix):])
try:
image_path = tempfile.mkstemp(dir=insiderer.TMP_DIR)[1]
image_handle = open(image_path, "wb")
image_handle.write(image_data)
image_handle.close()
child = insiderer.get_metadata(image_path, name)
children.append(child)
finally:
insiderer.safedelete(image_path)
def application_vnd_oasis_opendocument_text(path, metadata, children, from_doc=False):
odtzip = zipfile.ZipFile(path)
try:
tmp_path = None
tmp_path = tempfile.mkdtemp(dir=insiderer.TMP_DIR)
odtzip.extractall(tmp_path)
for name in odtzip.namelist():
childpath = tmp_path + "/" + name
try:
#cherrypy.log(childpath, name)
if name == "meta.xml":
metaxml = etree.parse(childpath)
metadatas = metaxml.xpath('/*/*/*')
for item in metadatas:
key = item.xpath('local-name()')
if from_doc is True and key == "generator":
pass
else:
metadata[key] = item.text
elif name == "content.xml":
metaxml = etree.parse(childpath)
trackchanges = metaxml.xpath('//*[local-name() = "change-info"]')
if len(trackchanges) > 0:
metadata['track_changes'] = []
for trackchange in trackchanges:
trackchange_dict = {}
for item in trackchange.xpath('*'):
trackchange_dict[item.xpath('local-name()')] = item.text
metadata['track_changes'].append(trackchange_dict)
elif os.path.isdir(childpath):
pass
elif name == "mimetype" or name == "manifest.rdf" or name == "META-INF/manifest.xml" or name == "current.xml" or name == "styles.xml" or name == "settings.xml" or name == "" or name == "Configurations2/accelerator/current.xml" or name == "layout-cache":
pass
elif from_doc is True and name == "Thumbnails/thumbnail.png":
pass
else:
child = insiderer.get_metadata(childpath, name)
children.append(child)
finally:
if not os.path.isdir(childpath):
insiderer.safedelete(childpath)
finally:
if tmp_path:
shutil.rmtree(tmp_path)
def application_pdf(path, metadata, children):
try:
uncompressed_pdf = tempfile.mkstemp(dir=insiderer.TMP_DIR)[1]
stdout = subprocess.check_output(
["pdftk", path, "output", uncompressed_pdf, "uncompress"])
with open(uncompressed_pdf, 'rb') as pdfhandle:
pdfdata = pdfhandle.read()
children.extend(extract_jpegs(pdfdata))
pdf_document = PyPDF2.PdfFileReader(pdfhandle)
if pdf_document.isEncrypted:
pdf_document.decrypt("")
metadata["info"] = pdf_document.getDocumentInfo()
metadata["xmp"] = dict()
xmp = pdf_document.getXmpMetadata()
if xmp:
for name in dir(xmp):
try:
xmp_data = getattr(xmp, name)
str_xmp_data = ""
if isinstance(xmp_data, datetime.datetime):
metadata["xmp"][name] = str(xmp_data.now())
else:
str_xmp_data = str(xmp_data)
try:
metadata["xmp"][name] = json.loads(
str_xmp_data)
except Exception as e:
if str_xmp_data is None:
pass
elif str_xmp_data.startswith("<"):
try:
metadata["xmp"][
name] = xmltodict.parse(
xmp_data.toxml())
except Exception as e:
pass
else:
metadata["xmp"][name] = str_xmp_data
except Exception as e:
cherrypy.log("Can't serialize %s. %s", name, e)
except Exception as e:
cherrypy.log("PDF exception", e)
finally:
insiderer.safedelete(uncompressed_pdf)
def image_svg_xml(path, metadata, children):
svg = etree.parse(path)
fields = {
'export-filename': '//@inkscape:export-filename',
'docname': '//@sodipodi:docname',
'description': '//svg:desc',
'title': '//svg:title',
'dca': '//@*[namespace-uri()="%s"][not(ancestor::dc:*)]' % namespaces["dc"], #because selecting by namespace attributes doesn't seem to work
'rdf': '//rdf:*[not(ancestor::rdf:*)]'
}
for key in fields.keys():
results = svg.xpath(fields[key], namespaces=namespaces)
for result in results:
key_name = key
if '*' in fields[key] and len(results) > 0:
if hasattr(result, "attrname"):
key_name = re.sub(r'\{.*?\}', '', result.attrname)
else:
key_name = result.xpath('name()')
key_name = insiderer.de_dup(key_name, metadata)
if isinstance(result, etree._ElementUnicodeResult):
metadata[key_name] = str(result)
else:
metadata[key_name] = xmltodict.parse(etree.tostring(result, encoding='utf8', method='xml').decode('utf-8'))
base64images = svg.xpath('//svg:image', namespaces=namespaces)
for image in base64images:
key = image.xpath('local-name()')
href = image.attrib.get('{%s}href' % namespaces["xlink"])
name = image.attrib.get('id') or "unknown"
if href.startswith("data:") and base64_prefix in href[0:100]:
image_data = base64.standard_b64decode( href[href.find(base64_prefix) + len(base64_prefix):] )
try:
image_path = tempfile.mkstemp(dir=insiderer.TMP_DIR)[1]
image_handle = open(image_path, "wb")
image_handle.write(image_data)
image_handle.close()
child = insiderer.get_metadata(image_path, name)
children.append(child)
finally:
insiderer.safedelete(image_path)
def application_zip(path, metadata, children, from_doc=False):
aZipFile = zipfile.ZipFile(path)
try:
tmp_path = None
tmp_path = tempfile.mkdtemp(dir=insiderer.TMP_DIR)
aZipFile.extractall(tmp_path)
for name in aZipFile.namelist():
childpath = os.path.join(tmp_path, name)
try:
if os.path.isdir(childpath):
pass
else:
child = insiderer.get_metadata(childpath, name)
children.append(child)
finally:
if not os.path.isdir(childpath):
insiderer.safedelete(childpath)
finally:
if tmp_path:
shutil.rmtree(tmp_path)
def application_zip(path, metadata, children, from_doc=False):
aZipFile = zipfile.ZipFile(path)
try:
tmp_path = None
tmp_path = tempfile.mkdtemp(dir=insiderer.TMP_DIR)
aZipFile.extractall(tmp_path)
for name in aZipFile.namelist():
childpath = os.path.join(tmp_path, name)
try:
if os.path.isdir(childpath):
pass
else:
child = insiderer.get_metadata(childpath, name)
children.append(child)
finally:
if not os.path.isdir(childpath):
insiderer.safedelete(childpath)
finally:
if tmp_path:
shutil.rmtree(tmp_path)