def set_body(self, events):
body = self.get_body()
events = self.events[:body.start+1] + events + self.events[body.end:]
if stream_to_str(events) == stream_to_str(self.events):
return False
self.set_changed()
self.events = events
return True
def set_body(self, events):
body = self.get_body()
events = self.events[:body.start + 1] + events + self.events[body.end:]
if stream_to_str(events) == stream_to_str(self.events):
return False
self.set_changed()
self.events = events
return True
def stl(document=None, namespace=freeze({}), prefix=None, events=None,
mode='events', skip=(DOCUMENT_TYPE,)):
# Input
encoding = 'utf-8'
if events is None:
events = document.events
# Prefix
if prefix is not None:
stream = set_prefix(events, prefix)
events = list(stream)
elif type(events) in (GeneratorType, XMLParser):
events = list(events)
# Initialize the namespace stacks
stack = NamespaceStack()
stack.append(namespace)
repeat = NamespaceStack()
# Process
stream = process(events, 0, len(events), stack, repeat, encoding, skip)
# Return
if mode == 'events':
return stream
elif mode == 'xml':
return stream_to_str(stream, encoding)
elif mode == 'xhtml':
return stream_to_str_as_xhtml(stream, encoding)
elif mode == 'html':
return stream_to_str_as_html(stream, encoding)
raise ValueError, 'unexpected mode "%s"' % mode
def stl(document=None, namespace=freeze({}), prefix=None, events=None,
mode='events', skip=(DOCUMENT_TYPE,)):
# Input
encoding = 'utf-8'
if events is None:
events = document.events
# Prefix
if prefix is not None:
stream = set_prefix(events, prefix)
events = list(stream)
elif isinstance(events, (GeneratorType, XMLParser)):
events = list(events)
# Initialize the namespace stacks
stack = NamespaceStack()
stack.append(namespace)
repeat = NamespaceStack()
# Process
stream = process(events, 0, len(events), stack, repeat, encoding, skip)
# Return
if mode == 'events':
return stream
elif mode == 'xml':
return stream_to_str(stream, encoding)
elif mode == 'xhtml':
return stream_to_str_as_xhtml(stream, encoding)
elif mode == 'html':
return stream_to_str_as_html(stream, encoding)
raise ValueError, 'unexpected mode "%s"' % mode
def translate(self, catalog, srx_handler=None):
"""Translate the document and reconstruct an odt document.
"""
# Translate
modified_files = {}
for filename in ['content.xml', 'meta.xml', 'styles.xml']:
events = self.get_events(filename)
translation = translate(events, catalog, srx_handler)
modified_files[filename] = stream_to_str(translation)
# Zip
return zip_data(self.data, modified_files)
def translate(self, catalog, srx_handler=None):
"""Translate the document and reconstruct an odt document.
"""
# Translate
modified_files = {}
for filename in ['content.xml', 'meta.xml', 'styles.xml']:
events = self.get_events(filename)
translation = translate(events, catalog, srx_handler)
modified_files[filename] = stream_to_str(translation)
# Zip
return zip_data(self.data, modified_files)
def to_str(text, format, encoding='utf-8'):
if format == 'xml':
events = block_stream(text)
return stream_to_str(events, encoding)
elif format == 'xhtml':
events = to_html_events(text)
return stream_to_str_as_xhtml(events, encoding)
elif format == 'html':
events = to_html_events(text)
return stream_to_str_as_html(events, encoding)
elif format == 'latex':
events = block_stream(text)
return stream_to_str_as_latex(events, encoding)
raise ValueError, "unexpected format '%s'" % format
def to_str(text, format, encoding='utf-8'):
if format == 'xml':
events = block_stream(text)
return stream_to_str(events, encoding)
elif format == 'xhtml':
events = to_html_events(text)
return stream_to_str_as_xhtml(events, encoding)
elif format == 'html':
events = to_html_events(text)
return stream_to_str_as_html(events, encoding)
elif format == 'latex':
events = block_stream(text)
return stream_to_str_as_latex(events, encoding)
raise ValueError, "unexpected format '%s'" % format
def stl(document=None,
namespace=freeze({}),
prefix=None,
events=None,
mode='events',
skip=(DOCUMENT_TYPE, )):
# Input
encoding = 'utf-8'
if events is None:
events = document.events
# Prefix
if prefix is not None:
stream = set_prefix(events, prefix)
events = list(stream)
elif type(events) in (GeneratorType, XMLParser):
events = list(events)
# Initialize the namespace stacks
stack = NamespaceStack()
stack.append(namespace)
repeat = NamespaceStack()
# Process
stream = process(events, 0, len(events), stack, repeat, encoding, skip)
# Return
try:
if mode == 'events':
return stream
elif mode == 'xml':
return stream_to_str(stream, encoding)
elif mode == 'xhtml':
return stream_to_str_as_xhtml(stream, encoding)
elif mode == 'html':
return stream_to_str_as_html(stream, encoding)
except STLError, e:
error = 'Error in generation of {0}\n'.format(mode)
if document:
error += 'Template {0}\n'.format(document.key)
raise STLError(error + e.message)
def stl(document=None, namespace=freeze({}), prefix=None, events=None,
mode='events', skip=(DOCUMENT_TYPE,)):
# Input
encoding = 'utf-8'
if events is None:
events = document.events
# Prefix
if prefix is not None:
stream = set_prefix(events, prefix)
events = list(stream)
elif type(events) in (GeneratorType, XMLParser):
events = list(events)
# Initialize the namespace stacks
stack = NamespaceStack()
stack.append(namespace)
repeat = NamespaceStack()
# Process
stream = process(events, 0, len(events), stack, repeat, encoding, skip)
# Return
try:
if mode == 'events':
return stream
elif mode == 'xml':
return stream_to_str(stream, encoding)
elif mode == 'xhtml':
return stream_to_str_as_xhtml(stream, encoding)
elif mode == 'html':
return stream_to_str_as_html(stream, encoding)
except STLError, e:
error = 'Error in generation of {0}\n'.format(mode)
if document:
error += 'Template {0}\n'.format(document.key)
raise STLError(error + e.message)
def GET(self, resource, context):
# Content-Type
context.set_content_type('text/opml')
context.set_content_disposition('attachment', 'rss-agregator.opml')
stream = resource.to_opml_stream(context)
return stream_to_str(stream)
def translate(self, catalog, srx_handler=None):
stream = translate(self.events, catalog, srx_handler)
return stream_to_str(stream)
def to_str(self, encoding='UTF-8'):
return stream_to_str(self.events, encoding)
def greek(self):
"""Anonymize the ODF file.
"""
# Verify PIL is installed
if PILImage is None:
err = 'The greeking feature requires the Python Imaging Library'
raise ImportError, err
folder = lfs.open(get_abspath('.'))
err = 'Unexpected "%s" file will be omitted from the greeked document'
modified_files = {}
for filename in self.get_contents():
extension = splitext(filename)[1]
startswith = filename.startswith
# Files to keep as they are
# TODO the manifest.xml file should be properly updated
keep = ['mimetype', 'settings.xml', 'META-INF/manifest.xml']
if filename in keep:
pass
# Content, metadata and style
elif filename in ['content.xml', 'meta.xml', 'styles.xml']:
events = self.get_events(filename)
translation = translate(events, GreekCatalog)
modified_files[filename] = stream_to_str(translation)
# Thumbnails
elif startswith('Thumbnails'):
if extension == '.pdf':
modified_files[filename] = folder.open('thumb.pdf').read()
elif extension == '.png':
modified_files[filename] = folder.open('thumb.png').read()
else:
# Unexpected (TODO use the logging system)
modified_files[filename] = None
print err % filename
# SVM files (usually they are in the Pictures folder)
elif extension == '.svm':
modified_files[filename] = folder.open('square.svm').read()
# Pictures
elif startswith('Pictures') or startswith('media'):
# Try with PIL
file = self.get_file(filename)
file = StringIO(file)
image = PILImage.open(file)
format = image.format
image = image.convert('RGB')
image.filename = filename
draw = PILImageDraw.Draw(image)
# Make a cross
h, l = image.size
draw.rectangle((0, 0, h-1, l-1), fill="grey", outline="black")
draw.line((0, 0, h-1, l-1), fill="black")
draw.line((0, l-1, h-1, 0), fill="black")
# Save
data = StringIO()
image.save(data, format)
modified_files[filename] = data.getvalue()
# Unexpected (TODO use the logging system)
else:
modified_files[filename] = None
print err % filename
return zip_data(self.data, modified_files)
def test_simple(self):
data = 'Hello ${name}'
stream = substitute(data, self.stack, self.repeat)
# Assert
out = stream_to_str(stream)
self.assertEqual(out, 'Hello Toto')
def test_doctype(self):
data = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 3.2 Final//EN">'
stream = HTMLParser(data)
self.assertEqual(stream_to_str(stream), data)
def stream_to_html(stream, encoding='UTF-8', map=stream_to_html_map):
return stream_to_str(stream, encoding=encoding, map=map)
def greek(self):
"""Anonymize the ODF file.
"""
# Verify PIL is installed
if PILImage is None:
err = 'The greeking feature requires the Python Imaging Library'
raise ImportError, err
folder = lfs.open(get_abspath('.'))
err = 'Unexpected "%s" file will be omitted from the greeked document'
modified_files = {}
for filename in self.get_contents():
extension = splitext(filename)[1]
startswith = filename.startswith
# Files to keep as they are
# TODO the manifest.xml file should be properly updated
keep = ['mimetype', 'settings.xml', 'META-INF/manifest.xml']
if filename in keep:
pass
# Content, metadata and style
elif filename in ['content.xml', 'meta.xml', 'styles.xml']:
events = self.get_events(filename)
translation = translate(events, GreekCatalog)
modified_files[filename] = stream_to_str(translation)
# Thumbnails
elif startswith('Thumbnails'):
if extension == '.pdf':
modified_files[filename] = folder.open('thumb.pdf').read()
elif extension == '.png':
modified_files[filename] = folder.open('thumb.png').read()
else:
# Unexpected (TODO use the logging system)
modified_files[filename] = None
print err % filename
# SVM files (usually they are in the Pictures folder)
elif extension == '.svm':
modified_files[filename] = folder.open('square.svm').read()
# Pictures
elif startswith('Pictures') or startswith('media'):
# Try with PIL
file = self.get_file(filename)
file = StringIO(file)
image = PILImage.open(file)
format = image.format
image = image.convert('RGB')
image.filename = filename
draw = PILImageDraw.Draw(image)
# Make a cross
h, l = image.size
draw.rectangle((0, 0, h - 1, l - 1),
fill="grey",
outline="black")
draw.line((0, 0, h - 1, l - 1), fill="black")
draw.line((0, l - 1, h - 1, 0), fill="black")
# Save
data = StringIO()
image.save(data, format)
modified_files[filename] = data.getvalue()
# Unexpected (TODO use the logging system)
else:
modified_files[filename] = None
print err % filename
return zip_data(self.data, modified_files)
def stream_to_str_as_xhtml(stream, encoding='UTF-8'):
content_type = 'application/xhtml+xml; charset=%s' % encoding
stream = set_content_type(stream, content_type)
return stream_to_str(stream, encoding)
def translate(self, catalog, srx_handler=None):
stream = translate(self.events, catalog, srx_handler)
return stream_to_str(stream)
def _get_translatable_blocks(events):
# Default value
encoding = 'utf-8'
# To identify the begin/end format
id = 0
id_stack = []
context_stack = [None]
stream = None
message = Message()
skip_level = 0
for event in events:
type, value, line = event
# Set the good encoding
if type == XML_DECL:
encoding = value[1]
# And now, we catch only the good events
elif type == START_ELEMENT:
if skip_level > 0:
skip_level += 1
if stream:
stream.append(event)
continue
else:
tag_uri, tag_name, attributes = value
schema = get_element_schema(tag_uri, tag_name)
# Context management
if schema.context is not None:
context_stack.append(schema.context)
# Skip content ?
if schema.skip_content:
skip_level = 1
if id_stack:
stream = [event]
continue
# Is inline ?
elif schema.is_inline:
id += 1
id_stack.append(id)
start_format = _make_start_format(tag_uri, tag_name,
attributes, encoding)
message.append_start_format(start_format, id, line)
continue
elif id_stack:
skip_level = 1
stream = [event]
continue
elif type == END_ELEMENT:
if skip_level > 0:
skip_level -= 1
if stream:
stream.append(event)
if skip_level == 0:
id += 1
aux = stream_to_str(stream, encoding)
aux = unicode(aux, encoding)
aux = [(aux, False, context_stack[-1])]
message.append_start_format(aux, id, line)
message.append_end_format([], id, line)
stream = None
continue
else:
tag_uri, tag_name = value[:2]
schema = get_element_schema(tag_uri, tag_name)
# Context management
if schema.context is not None:
context_stack.pop()
# Is inline ?
if schema.is_inline:
message.append_end_format([(get_end_tag(value), False,
None)], id_stack.pop(), line)
continue
elif type == TEXT:
# Not empty ?
if stream:
stream.append(event)
continue
elif skip_level == 0 and (value.strip() != '' or message):
value = XMLContent.encode(value)
value = unicode(value, encoding)
message.append_text(value, line, context_stack[-1])
continue
elif type == COMMENT:
if stream:
stream.append(event)
continue
elif message:
id += 1
if isinstance(value, str):
value = unicode(value, encoding)
value = u'<!--%s-->' % value
message.append_start_format([(value, False, None)], id, line)
message.append_end_format([], id, line)
continue
# Not a good event => break + send the event
if message:
yield MESSAGE, message, message.get_line()
message = Message()
yield event
# Send the last message!
if message:
yield MESSAGE, message, message.get_line()
def to_str(self, encoding='UTF-8'):
return stream_to_str(self.events, encoding)
def test_simple(self):
data = 'Hello ${name}'
stream = substitute(data, self.stack, self.repeat)
# Assert
out = stream_to_str(stream)
self.assertEqual(out, 'Hello Toto')
def test_doctype(self):
data = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 3.2 Final//EN">'
stream = HTMLParser(data)
self.assertEqual(stream_to_str(stream), data)
