def match(cls, file_path, entity):
mime_types = [normalize_mimetype(m, default=None) for m in cls.MIME_TYPES]
mime_types = [m for m in mime_types if m is not None]
for mime_type in entity.get("mimeType"):
if mime_type in mime_types:
return cls.SCORE
extensions = [normalize_extension(e) for e in cls.EXTENSIONS]
for file_name in entity.get("fileName"):
extension = normalize_extension(file_name)
if extension is not None and extension in extensions:
return cls.SCORE
return -1
def load_mime_extensions():
media_types = {}
for xcd_file in FILES:
doc = etree.parse(xcd_file)
path = './*[@oor:package="org.openoffice.TypeDetection"]/node/node'
for tnode in doc.xpath(path, namespaces=NS):
node = {}
for prop in tnode.findall('./prop'):
name = prop.get(NAME)
for value in prop.findall('./value'):
node[name] = value.text
media_type = normalize_mimetype(node.get('MediaType'),
default=None)
if media_type is None:
continue
extensions = node.get('Extensions')
if extensions is None:
continue
extension = normalize_extension(extensions.split(' ')[0])
if extension is not None:
media_types[media_type] = extension
return media_types
def match(cls, file_path, result=None):
mime_types = [
normalize_mimetype(m, default=None) for m in cls.MIME_TYPES
] # noqa
mime_types = [m for m in mime_types if m is not None]
mime_type = normalize_mimetype(result.mime_type, default=None)
if mime_type in mime_types:
return cls.SCORE
extensions = [normalize_extension(e) for e in cls.EXTENSIONS]
extensions = [e for e in extensions if e is not None]
extension = normalize_extension(result.file_name)
if extension in extensions:
return cls.SCORE
return -1
def parse_extensions(self, extensions):
if extensions is not None:
for ext in extensions.split(' '):
if ext == '*':
continue
ext = normalize_extension(ext)
if ext is not None:
yield ext
async def convert(request):
data = await request.post()
upload = data['file']
extension = normalize_extension(upload.filename)
mime_type = normalize_mimetype(upload.content_type, default=None)
log.info('PDF convert: %s [%s]', upload.filename, mime_type)
fd, upload_file = mkstemp()
os.close(fd)
fd, out_file = mkstemp(suffix='.pdf')
os.close(fd)
log.info('Source: %s, target: %s', upload_file, out_file)
try:
with open(upload_file, 'wb') as fh:
shutil.copyfileobj(upload.file, fh, BUFFER_SIZE)
filters = list(FORMATS.get_filters(extension, mime_type))
timeout = int(request.query.get('timeout', 300))
await asyncio.sleep(0)
converter.convert_file(upload_file, out_file, filters,
timeout=timeout)
out_size = os.path.getsize(out_file)
if out_size == 0:
raise ConversionFailure("No PDF version was generated.")
await asyncio.sleep(0)
response = web.StreamResponse()
response.content_length = out_size
response.content_type = 'application/pdf'
await response.prepare(request)
with open(out_file, 'rb') as f:
while True:
chunk = f.read(BUFFER_SIZE)
if not chunk:
break
await response.write(chunk)
return response
except ConversionFailure as fail:
log.info("Failed to convert: %s", fail)
return web.Response(text=str(fail), status=400)
except Exception as exc:
log.exception('System error: %s.', exc)
converter.terminate()
finally:
os.remove(upload_file)
os.remove(out_file)
def convert():
acquired = lock.acquire(timeout=2)
if not acquired:
return ("BUSY", 503)
try:
if os.path.exists(OUT_PATH):
os.unlink(OUT_PATH)
upload = request.files['file']
extension = normalize_extension(upload.filename)
mime_type = normalize_mimetype(upload.mimetype, default=None)
if extension is None:
extension = extensions.get(mime_type)
log.info('PDF convert: %s [%s]', upload.filename, mime_type)
fd, upload_file = mkstemp(suffix='.%s' % extension)
fh = os.fdopen(fd, mode='wb')
upload.save(fh)
fh.close()
if listener.poll() is not None:
log.error("Listener has terminated.")
app.is_dead = True
return ("DEAD", 503)
args = [
'unoconv', '-f', 'pdf', '-o', OUT_PATH, '-i',
'MacroExecutionMode=0', '-i', 'ReadOnly=1', '-e',
'SelectPdfVersion=1', '-e', 'MaxImageResolution=300',
'--no-launch', upload_file
]
err = subprocess.call(args, timeout=TIMEOUT)
if err != 0 or not os.path.exists(OUT_PATH):
return ('The document could not be converted to PDF.', 400)
return send_file(OUT_PATH)
except subprocess.TimeoutExpired:
log.error("Timeout exceeded: %s", upload.filename)
app.is_dead = True
return ('Processing the document timed out.', 400)
finally:
lock.release()