def _update(self, data, mime_type=None):
new_digest = hashlib.md5(data).hexdigest()
if new_digest == self.digest:
return
self.digest = new_digest
self.data = data
self.size = len(data)
if mime_type:
self.mime_type = mime_type
# TODO Else: use a sniffer
# TODO: This should be asynchronous
if self.mime_type != "application/pdf":
try:
self.pdf = converter.to_pdf(self.digest, self.data, self.mime_type)
except ConversionError:
traceback.print_exc()
else:
self.pdf = self.data
try:
self.text = converter.to_text(self.digest, self.data, self.mime_type)
except ConversionError:
self.text = u""
traceback.print_exc()
try:
self.extra_metadata = converter.get_metadata(self.digest, self.data, self.mime_type)
except ConversionError:
self.extra_metadata = {}
traceback.print_exc()
if self.text:
self.language = guessLanguageName(self.text)
self.page_num = self.extra_metadata.get("PDF:Pages", 1)
def XXXtest_excel_to_text(self):
blob = self.read_file("test.xls")
text = converter.to_text("", blob, "application/excel")
def XXXtest_wordx_to_text(self):
blob = self.read_file("test.docx")
text = converter.to_text("", blob, "application/msword")
def test_pdf_to_text(self):
blob = self.read_file("onepage.pdf")
text = converter.to_text("", blob, "application/pdf")
