def loads_xml(bytestring, encoding=None, cls=ListDmrs, convert_legacy_prontype=True, **kwargs):
"""
Currently processes "<dmrs>...</dmrs>"
To be updated for "<dmrslist>...</dmrslist>"...
Expects a bytestring; to load from a string instead, specify encoding
Produces a ListDmrs by default; for a different type, specify cls
"""
if encoding:
bytestring = bytestring.encode(encoding)
xml = ET.XML(bytestring)
dmrs = cls(**kwargs)
dmrs.cfrom = int(xml.get('cfrom')) if 'cfrom' in xml.attrib else None
dmrs.cto = int(xml.get('cto')) if 'cto' in xml.attrib else None
dmrs.surface = xml.get('surface')
dmrs.ident = int(xml.get('ident')) if 'ident' in xml.attrib else None
# top may be set as a graph attribute or as a link (see below)
top_id = int(xml.get('top')) if 'top' in xml.attrib else None
index_id = int(xml.get('index')) if 'index' in xml.attrib else None
for elem in xml:
if elem.tag == 'node':
node = Node.from_xml(elem, convert_legacy_prontype)
dmrs.add_node(node)
elif elem.tag == 'link':
link = Link.from_xml(elem)
if link.start == 0:
# this would overwrite any graph-level top attribute
# (see above), but let's assume we won't encounter
# both in the same graph
top_id = link.end
else:
dmrs.add_link(link)
else:
raise PydmrsValueError(elem.tag)
if top_id:
dmrs.top = dmrs[top_id]
if index_id:
dmrs.index = dmrs[index_id]
return dmrs
def loads_xml(bytestring,
encoding=None,
cls=ListDmrs,
convert_legacy_prontype=True,
**kwargs):
"""
Currently processes "<dmrs>...</dmrs>"
To be updated for "<dmrslist>...</dmrslist>"...
Expects a bytestring; to load from a string instead, specify encoding
Produces a ListDmrs by default; for a different type, specify cls
"""
if encoding:
bytestring = bytestring.encode(encoding)
xml = ET.XML(bytestring)
dmrs = cls(**kwargs)
dmrs.cfrom = int(xml.get('cfrom')) if 'cfrom' in xml.attrib else None
dmrs.cto = int(xml.get('cto')) if 'cto' in xml.attrib else None
dmrs.surface = xml.get('surface')
dmrs.ident = int(xml.get('ident')) if 'ident' in xml.attrib else None
index_id = int(xml.get('index')) if 'index' in xml.attrib else None
top_id = None
for elem in xml:
if elem.tag == 'node':
node = Node.from_xml(elem, convert_legacy_prontype)
dmrs.add_node(node)
elif elem.tag == 'link':
link = Link.from_xml(elem)
if link.start == 0:
top_id = link.end
else:
dmrs.add_link(link)
else:
raise PydmrsValueError(elem.tag)
if top_id:
dmrs.top = dmrs[top_id]
if index_id:
dmrs.index = dmrs[index_id]
return dmrs