def get_image(self, transformations=None):
cache_filename = 'base_image'
logger.debug('Page cache filename: %s', cache_filename)
cache_file = self.cache_partition.get_file(filename=cache_filename)
if cache_file:
logger.debug('Page cache file "%s" found', cache_filename)
with cache_file.open() as file_object:
converter = ConverterBase.get_converter_class()(
file_object=file_object
)
converter.seek_page(page_number=0)
# This code is also repeated below to allow using a context
# manager with cache_file.open and close it automatically.
# Apply runtime transformations
for transformation in transformations:
converter.transform(transformation=transformation)
return converter.get_page()
else:
logger.debug('Page cache file "%s" not found', cache_filename)
try:
with self.document_version.get_intermediate_file() as file_object:
converter = ConverterBase.get_converter_class()(
file_object=file_object
)
converter.seek_page(page_number=self.page_number - 1)
page_image = converter.get_page()
# Since open "wb+" doesn't create files, create it explicitly
with self.cache_partition.create_file(filename=cache_filename) as file_object:
file_object.write(page_image.getvalue())
# Apply runtime transformations
for transformation in transformations:
converter.transform(transformation=transformation)
return converter.get_page()
except Exception as exception:
# Cleanup in case of error
logger.error(
'Error creating page cache file "%s"; %s',
cache_filename, exception
)
raise
def get_image(self, transformations=None):
cache_filename = self.cache_filename
file_object = None
try:
file_object = open(file=self.get_full_path(), mode='rb')
converter = ConverterBase.get_converter_class()(
file_object=file_object)
page_image = converter.get_page()
# Since open "wb+" doesn't create files, check if the file
# exists, if not then create it
if not self.storage.exists(cache_filename):
self.storage.save(name=cache_filename,
content=ContentFile(content=''))
with self.storage.open(name=cache_filename,
mode='wb+') as file_object:
file_object.write(page_image.getvalue())
except Exception as exception:
# Cleanup in case of error
logger.error('Error creating staging file cache "%s"; %s',
cache_filename, exception)
self.storage.delete(cache_filename)
if file_object:
file_object.close()
raise
for transformation in transformations:
converter.transform(transformation=transformation)
result = converter.get_page()
file_object.close()
return result
def update_page_count(self, save=True):
try:
with self.open() as file_object:
converter = ConverterBase.get_converter_class()(
file_object=file_object, mime_type=self.mimetype
)
detected_pages = converter.get_page_count()
except PageCountError:
# If converter backend doesn't understand the format,
# use 1 as the total page count
pass
else:
DocumentPage = apps.get_model(
app_label='documents', model_name='DocumentPage'
)
with transaction.atomic():
self.pages.all().delete()
for page_number in range(detected_pages):
DocumentPage.objects.create(
document_version=self, page_number=page_number + 1
)
if save:
self.save()
return detected_pages
def get_intermediate_file(self):
cache_filename = 'intermediate_file'
cache_file = self.cache_partition.get_file(filename=cache_filename)
if cache_file:
logger.debug('Intermidiate file found.')
return cache_file.open()
else:
logger.debug('Intermidiate file not found.')
try:
with self.open() as version_file_object:
converter = ConverterBase.get_converter_class()(
file_object=version_file_object
)
with converter.to_pdf() as pdf_file_object:
with self.cache_partition.create_file(filename=cache_filename) as file_object:
shutil.copyfileobj(
fsrc=pdf_file_object, fdst=file_object
)
return self.cache_partition.get_file(filename=cache_filename).open()
except InvalidOfficeFormat:
return self.open()
except Exception as exception:
logger.error(
'Error creating intermediate file "%s"; %s.',
cache_filename, exception
)
cache_file = self.cache_partition.get_file(filename=cache_filename)
if cache_file:
cache_file.delete()
raise
def detect_orientation(self):
with self.document_version.open() as file_object:
converter = ConverterBase.get_converter_class()(
file_object=file_object,
mime_type=self.document_version.mimetype
)
return converter.detect_orientation(
page_number=self.page_number
)
def execute(self, file_object, language=None, transformations=None):
self.language = language
if not transformations:
transformations = []
self.converter = ConverterBase.get_converter_class()(
file_object=file_object)
for transformation in transformations:
self.converter.transform(transformation=transformation)