def parse_string(self, text, tarsqidoc):
"""Parse the TTK string and put the contents in the appropriate parts of the
SourceDoc."""
self.lif = LIF(json_string=text)
tarsqidoc.sourcedoc = SourceDoc()
tarsqidoc.sourcedoc.text = self.lif.text.value
tarsqidoc.sourcedoc.lif = self.lif
def parse_string(self, text):
"""Parses a text string and returns a SourceDoc. Simply dumps the full
string into the text variable of the SourceDoc."""
sourcedoc = SourceDoc(None)
# TODO: do we need to ensure the text is unicode?
sourcedoc.text = text
return TarsqiDocument(sourcedoc, {})
def _parse(self, tarsqidoc):
self.sourcedoc = SourceDoc(None)
self.tarsqidoc = tarsqidoc
self.tarsqidoc.sourcedoc = self.sourcedoc
self.sourcedoc.text = self.topnodes['text'].firstChild.data
self._add_source_tags()
self._add_tarsqi_tags()
self._add_comments()
self._add_metadata()
def parse_string(self, text, tarsqidoc):
"""Parses a text string and returns a SourceDoc. Uses the ParseFile routine of
the expat parser, where all the handlers are set up to fill in the text
and tags in SourceDoc."""
self.sourcedoc = SourceDoc(None)
# TODO: do we need to make sure that text is unicode?
self.parser.Parse(text)
self.sourcedoc.finish()
tarsqidoc.sourcedoc = self.sourcedoc
def parse_file(self, filename, tarsqidoc):
"""Parses filename and returns a SourceDoc. Uses the ParseFile routine
of the expat parser, where all the handlers are set up to fill in the
text and tags in SourceDoc."""
self.sourcedoc = SourceDoc(filename)
# TODO: should this be codecs.open() for non-ascii?
self.parser.ParseFile(open(filename))
self.sourcedoc.finish()
tarsqidoc.sourcedoc = self.sourcedoc
def parse_file(self, filename):
"""Parse the TTK file and put the contents in the appropriate parts of
the SourceDoc."""
self._load_dom(filename)
self.sourcedoc = SourceDoc(filename)
self.tarsqidoc = TarsqiDocument(self.sourcedoc, {})
self.sourcedoc.text = self.topnodes['text'].firstChild.data
self._add_source_tags()
self._add_tarsqi_tags()
self._add_comments()
self._add_metadata()
return self.tarsqidoc
def parse_file(self, filename, tarsqidoc):
"""Parse the TTK file and put the contents in the appropriate parts of
the SourceDoc."""
if self.is_container(filename):
self.container = Container(json_file=filename)
self.lif = self.container.payload
else:
self.container = None
self.lif = LIF(json_file=filename)
tarsqidoc.sourcedoc = SourceDoc(filename)
tarsqidoc.sourcedoc.text = self.lif.text.value
tarsqidoc.sourcedoc.lif = self.lif
tarsqidoc.sourcedoc.lif_container = self.container
def parse_file(self, filename, tarsqidoc):
"""Parses filename and returns a SourceDoc. Uses the ParseFile routine
of the expat parser, where all the handlers are set up to fill in the
text and tags in SourceDoc."""
self.sourcedoc = SourceDoc(filename)
# TODO: should this be codecs.open() for non-ascii?
# self.parser.ParseFile(open(filename))
# NOTE: actually, the above line needed to replaced with the following
# while preparing to port code to Python3.
content = codecs.open(filename).read()
self.parser.Parse(content)
self.sourcedoc.finish()
tarsqidoc.sourcedoc = self.sourcedoc
def parse_file(self, filename):
"""Parses filename and returns a SourceDoc. Simply dumps the full file
content into the text variable of the SourceDoc."""
sourcedoc = SourceDoc(filename)
sourcedoc.text = codecs.open(filename, encoding='utf8').read()
return TarsqiDocument(sourcedoc, {})
def parse_string(self, text, tarsqidoc):
"""Parses a text string and returns a SourceDoc. Simply dumps the full
string into the text variable of the SourceDoc."""
tarsqidoc.sourcedoc = SourceDoc(None)
# TODO: do we need to ensure the text is unicode?
tarsqidoc.sourcedoc.text = text