def get_pipeline_xml(pipeline_type, pipeline_id, auto_pipeline_id):
path = os.path.join(docvert_root, "pipelines", pipeline_type, pipeline_id,
"pipeline.xml")
if not os.path.exists(path):
raise docvert_exception.unrecognised_pipeline(
"Unknown pipeline_id '%s' (checked %s)" % (pipeline_id, path))
autopipeline_path = None
xml = lxml.etree.parse(path)
if xml.getroot().tag == "autopipeline":
if auto_pipeline_id is None:
raise docvert_exception.unrecognised_auto_pipeline(
"Unknown auto pipeline '%s'" % auto_pipeline_id)
autopipeline_path = os.path.join(docvert_root, "pipelines",
"auto_pipelines", auto_pipeline_id,
"pipeline.xml")
if not os.path.exists(path):
raise docvert_exception.unrecognised_auto_pipeline(
"Unknown auto pipeline '%s'" % auto_pipeline_id)
custom_stages = "".join(map(lxml.etree.tostring, xml.getroot()))
autopipeline = ""
try:
autopipeline_handle = open(autopipeline_path)
except IOError, e:
autopipeline_path_with_default = os.path.join(
docvert_root, "pipelines", "auto_pipelines",
"%s.default" % auto_pipeline_id, "pipeline.xml")
autopipeline_handle = open(autopipeline_path_with_default)
autopipeline = autopipeline_handle.read().replace(
'{{custom-stages}}', custom_stages)
xml = lxml.etree.fromstring(autopipeline)
xml = xml.getroottree()
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 get_pipeline_xml(namespaced_pipeline_id, auto_pipeline_id):
path = os.path.join(docvert_root, "pipelines", namespaced_pipeline_id, "pipeline.xml")
autopath = None
if not os.path.exists(path):
raise docvert_exception.unrecognised_pipeline("Unknown pipeline '%s' (checked %s)" % (namespaced_pipeline_id, path))
xml = lxml.etree.parse(path)
if xml.getroot().tag == "autopipeline":
if auto_pipeline_id is None:
raise docvert_exception.unrecognised_auto_pipeline("Unknown auto pipeline '%s'" % auto_pipeline_id)
raise Exception("Sorry, auto pipelines aren't implemented yet.")
autopath = os.path.join(docvert_root, "pipelines", "autopipeline", auto_pipeline_id, "pipeline.xml")
if not os.path.exists(path):
raise docvert_exception.unrecognised_auto_pipeline("Unknown auto pipeline '%s'" % auto_pipeline_id)
return dict(xml=xml, pipeline_directory=os.path.dirname(path), path=path, autopath=autopath)
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, suppress_errors=False):
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)
def _title(name, files, data):
filename = os.path.basename(name).replace('\\','-').replace('/','-').replace(':','-')
if len(filename) == 0:
filename = "document.odt"
if files.has_key(filename):
if data and hasattr(files[filename], 'read') and files[filename].getvalue() == data:
return filename
unique = 1
potential_filename = filename
while files.has_key(potential_filename):
unique += 1
if filename.count("."):
potential_filename = filename.replace(".", "%i." % unique, 1)
else:
potential_filename = filename + str(unique)
filename = potential_filename
return filename
for filename, data in files.iteritems():
storage.set_friendly_name(filename, filename)
for url in urls:
try:
data = urllib2.urlopen(url, None, http_timeout).read()
doc_type = document_type.detect_document_type(data)
if doc_type == document_type.types.html:
data = html_to_opendocument(data, url)
filename = _title(url, files, data)
storage.set_friendly_name(filename, "%s (%s)" % (filename, url))
files[filename] = StringIO.StringIO(data)
except IOError, e:
filename = _title(url, files, None)
storage.set_friendly_name(filename, "%s (%s)" % (filename, url))
files[filename] = Exception("Download error from %s: %s" % (url, e))
def get_pipeline_xml(pipeline_type, pipeline_id, auto_pipeline_id):
path = os.path.join(docvert_root, "pipelines", pipeline_type, pipeline_id, "pipeline.xml")
if not os.path.exists(path):
raise docvert_exception.unrecognised_pipeline("Unknown pipeline_id '%s' (checked %s)" % (pipeline_id, path))
autopipeline_path = None
xml = lxml.etree.parse(path)
if xml.getroot().tag == "autopipeline":
if auto_pipeline_id is None:
raise docvert_exception.unrecognised_auto_pipeline("Unknown auto pipeline '%s'" % auto_pipeline_id)
autopipeline_path = os.path.join(docvert_root, "pipelines", "auto_pipelines", auto_pipeline_id, "pipeline.xml")
if not os.path.exists(path):
raise docvert_exception.unrecognised_auto_pipeline("Unknown auto pipeline '%s'" % auto_pipeline_id)
custom_stages = "".join(map(lxml.etree.tostring,xml.getroot()))
autopipeline = ""
try:
autopipeline_handle = open(autopipeline_path)
except IOError, e:
autopipeline_path_with_default = os.path.join(docvert_root, "pipelines", "auto_pipelines", "%s.default" % auto_pipeline_id, "pipeline.xml")
autopipeline_handle = open(autopipeline_path_with_default)
autopipeline = autopipeline_handle.read().replace('{{custom-stages}}', custom_stages)
xml = lxml.etree.fromstring(autopipeline)
xml = xml.getroottree()
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
