def process_conversion(files=None, urls=None, pipeline_id=None, pipeline_type="pipelines", auto_pipeline_id=None, storage_type_name=docvert_storage.storage_type.memory_based, converter=converter_type.python_streaming_to_libreoffice):
if files is None and urls is None:
raise docvert_exception.needs_files_or_urls()
if pipeline_id is None:
raise docvert_exception.unrecognised_pipeline("Unknown pipeline '%s'" % pipeline_id)
storage = docvert_storage.get_storage(storage_type_name)
for filename, data in files.iteritems():
doc_type = document_type.detect_document_type(data)
if doc_type != document_type.types.oasis_open_document:
data = generate_open_document(data, converter)
document_xml = opendocument.extract_useful_open_document_files(data, storage, filename)
process_pipeline(document_xml, pipeline_id, pipeline_type, auto_pipeline_id, storage, filename)
return storage
def process_conversion(
files=None,
urls=None,
pipeline_id=None,
pipeline_type="pipelines",
auto_pipeline_id=None,
storage_type_name=docvert_storage.storage_type.memory_based,
converter=converter_type.python_streaming_to_libreoffice):
if files is None and urls is None:
raise docvert_exception.needs_files_or_urls()
if pipeline_id is None:
raise docvert_exception.unrecognised_pipeline("Unknown pipeline '%s'" %
pipeline_id)
storage = docvert_storage.get_storage(storage_type_name)
for filename, data in files.iteritems():
doc_type = document_type.detect_document_type(data)
if doc_type != document_type.types.oasis_open_document:
data = generate_open_document(data, converter)
document_xml = opendocument.extract_useful_open_document_files(
data, storage, filename)
process_pipeline(document_xml, pipeline_id, pipeline_type,
auto_pipeline_id, storage, filename)
return storage
elif doc_type != document_type.types.oasis_open_document:
try:
data = generate_open_document(data, converter)
doc_type = document_type.types.oasis_open_document
except Exception, e:
if not suppress_errors:
raise e
storage.add("%s/index.txt" % filename, str(e))
if doc_type == document_type.types.oasis_open_document:
if pipeline_id == "open document": #reserved term, for when people want the Open Document file back directly. Don't bother loading pipeline.
storage.add("%s/index.odt" % filename, data)
thumbnail = opendocument.extract_thumbnail(data)
if thumbnail:
storage.add("%s/thumbnail.png" % filename, thumbnail)
else:
document_xml = opendocument.extract_useful_open_document_files(data, storage, filename)
storage.add("%s/opendocument.xml" % filename, document_xml)
process_pipeline(document_xml, pipeline_id, pipeline_type, auto_pipeline_id, storage, filename)
storage.remove("%s/opendocument.xml" % filename)
return storage
def process_pipeline(initial_pipeline_value, pipeline_id, pipeline_type, auto_pipeline_id, storage, storage_prefix=None):
pipeline_definition = docvert_pipeline.get_pipeline_definition(pipeline_type, pipeline_id, auto_pipeline_id)
pipeline = docvert_pipeline.pipeline_processor(storage, pipeline_definition['stages'], pipeline_definition['pipeline_directory'], storage_prefix)
return pipeline.start(initial_pipeline_value)
def generate_open_document(data, converter=converter_type.python_streaming_to_libreoffice):
if converter == converter_type.python_streaming_to_libreoffice:
return docvert_libreoffice.get_client().convert_by_stream(data, docvert_libreoffice.LIBREOFFICE_OPEN_DOCUMENT)
raise docvert_exception.unrecognised_converter("Unknown converter '%s'" % converter)
