rem
stringlengths
0
322k
add
stringlengths
0
2.05M
context
stringlengths
8
228k
_, _, data = self.resource.get_json() self._name = data['db_name']
if ddoc is not None: _, _, data = self.resource('_design', ddoc, '_info').get_json() else: _, _, data = self.resource.get_json() self._name = data['db_name']
def info(self): """Return information about the database as a dictionary.
if not streamed and method in ('GET', 'HEAD') and 'etag' in resp.msg:
if not streamed and method == 'GET' and 'etag' in resp.msg:
def _retry(): conn.close() conn.connect() return _try_request(retries - 1)
for ln in data._iterchunks():
chunks = data._iterchunks() for ln in chunks:
def _changes(self, **opts): _, _, data = self.resource.get('_changes', **opts) for ln in data._iterchunks(): # Skip non-JSON lines (heartbeats). if ln[0] != '{': continue # Yield the update up to and inluding the last_seq line. doc = json.decode(ln) yield doc if 'last_seq' in doc: break
if 'last_seq' in doc: break
def _changes(self, **opts): _, _, data = self.resource.get('_changes', **opts) for ln in data._iterchunks(): # Skip non-JSON lines (heartbeats). if ln[0] != '{': continue # Yield the update up to and inluding the last_seq line. doc = json.decode(ln) yield doc if 'last_seq' in doc: break
return reduce(*cmd, rereduce=True)
return reduce(*cmd, **{'rereduce': True})
def rereduce(*cmd): return reduce(*cmd, rereduce=True)
if headers.get('content-type') == 'application/json':
if 'application/json' in headers.get('content-type'):
def delete_json(self, *a, **k): status, headers, data = self.delete(*a, **k) if headers.get('content-type') == 'application/json': data = json.decode(data.read()) return status, headers, data
if headers.get('content-type') == 'application/json':
if 'application/json' in headers.get('content-type'):
def get_json(self, *a, **k): status, headers, data = self.get(*a, **k) if headers.get('content-type') == 'application/json': data = json.decode(data.read()) return status, headers, data
if headers.get('content-type') == 'application/json':
if 'application/json' in headers.get('content-type'):
def post_json(self, *a, **k): status, headers, data = self.post(*a, **k) if headers.get('content-type') == 'application/json': data = json.decode(data.read()) return status, headers, data
if headers.get('content-type') == 'application/json':
if 'application/json' in headers.get('content-type'):
def put_json(self, *a, **k): status, headers, data = self.put(*a, **k) if headers.get('content-type') == 'application/json': data = json.decode(data.read()) return status, headers, data
Takes since, feed, heartbeat and timeout options. The continuous feed mode isn't supported yet, but normal and longpoll should work. """ if 'feed' in opts and opts['feed'] == 'continuous':
Takes since, feed, heartbeat and timeout options. """ if opts.get('feed') == 'continuous':
def changes(self, **opts): """Retrieve a changes feed from the database.
for docid in db:
def dump_db(dburl, username=None, password=None, boundary=None, output=sys.stdout): db = Database(dburl) if username is not None and password is not None: db.resource.http.add_credentials(username, password) envelope = write_multipart(output, boundary=boundary) for docid in db: doc = db.get(docid, attachments=True) p...
}, )
})
def dump_db(dburl, username=None, password=None, boundary=None, output=sys.stdout): db = Database(dburl) if username is not None and password is not None: db.resource.http.add_credentials(username, password) envelope = write_multipart(output, boundary=boundary) for docid in db: doc = db.get(docid, attachments=True) p...
if not options.compact: return for dbname in dbnames: target_server[dbname].compact()
if options.compact: for (sdb, tdb) in databases: print 'compact', tdb target[tdb].compact()
def main(): usage = '%prog [options] <source> <target>' parser = optparse.OptionParser(usage=usage) parser.add_option('--continuous', action='store_true', dest='continuous', help='trigger continuous replication in cochdb') parser.add_option('--compact', action='store_true', dest='compact', help='compact target databas...
resp, data = self.resource.post('_compact', ddoc)
resp, data = self.resource('_compact').post(ddoc)
def compact(self, ddoc=None): """Compact the database or a design document's index.
elif self.scheme == 'https':
elif scheme == 'https':
def _get_connection(self, url): scheme, host = urlsplit(url, 'http', False)[:2] self.lock.acquire() try: conns = self.conns.setdefault((scheme, host), []) if conns: conn = conns.pop(-1) else: if scheme == 'http': cls = HTTPConnection elif self.scheme == 'https': cls = HTTPSConnection else: raise ValueError('%s is not a...
emit(doc.foo, null);
if(doc.foo !== undefined) { emit(doc.foo, null); }
def test_collation(self): values = [ None, False, True, 1, 2, 3.0, 4, 'a', 'A', 'aa', 'b', 'B', 'ba', 'bb', ['a'], ['b'], ['b', 'c'], ['b', 'c', 'a'], ['b', 'd'], ['b', 'd', 'e'], {'a': 1}, {'a': 2}, {'b': 1}, {'b': 2}, {'b': 2, 'c': 2}, ] self.db['0'] = {'bar': 0} for idx, value in enumerate(values): self.db[str(idx +...
top, context = get_first(node, r'/([ICV]P|V[VA]|VRD|VSB|VCD)/=TOP $ *=SS', with_context=True)
top, context = get_first(node, r'/([ICV]P|V[VA]|VRD|VSB|VCD)/=TOP $ *=SS', with_context=True, left_to_right=True)
def fix_subject_extraction(self, _, n, pred, w=None, reduced=False): debug("%s", reduced) node = n debug("Fixing subject extraction: %s", lrp_repr(node)) if not reduced: self.remove_null_element(node)
yield Derivation.from_header_and_derivation(header, deriv_string)
deriv = Derivation.from_header_and_derivation(header, deriv_string) yield deriv
def __iter__(self): '''Yields an iterator over this document.''' while True: try: header, deriv_string = self.derivs.next(), self.derivs.next() except StopIteration: self.file.close() raise yield Derivation.from_header_and_derivation(header, deriv_string)
elif is_apposition(node): tag(last_kid, 'r') for kid in node: if not kid.tag.startswith('PU'): if kid.tag.endswith('-APP') and not kid.tag.startswith('CP'): tag(kid, 'A') else: tag(kid, 'a')
def label(root): root = preprocess(root) for node in nodes(root): if node.is_leaf(): continue first_kid, first_kid_index = get_nonpunct_kid(node, get_last=False) last_kid, last_kid_index = get_nonpunct_kid(node, get_last=True) for kid in node: if has_modification_tag(kid): tag(kid, 'm') elif kid.tag == 'MSP': tag...
elif is_apposition(node): tag(last_kid, 'r') for kid in node: if not kid.tag.startswith('PU'): if kid.tag.endswith('-APP') and not kid.tag.startswith('CP'): tag(kid, 'A') else: tag(kid, 'a')
def label(root): root = preprocess(root) for node in nodes(root): if node.is_leaf(): continue first_kid, first_kid_index = get_nonpunct_kid(node, get_last=False) last_kid, last_kid_index = get_nonpunct_kid(node, get_last=True) for kid in node: if has_modification_tag(kid): tag(kid, 'm') elif kid.tag == 'MSP': tag...
debug('final_punctuation_stk: %s', final_punctuation_stk) debug('now going to label: %s', pprint(node))
def label_root(node): debug('originally: %s', pprint(node)) final_punctuation_stk = [] # These derivations consist of a leaf PU root: 24:73(4), 25:81(4), 28:52(21) if node.is_leaf(): return node # inconsistent top-level taggings (0:21(4)) lead to a CCG-like absorption analysis elif is_right_punct_absorption(node): ret...
debug('result: %s', pprint(result))
def label_root(node): debug('originally: %s', pprint(node)) final_punctuation_stk = [] # These derivations consist of a leaf PU root: 24:73(4), 25:81(4), 28:52(21) if node.is_leaf(): return node # inconsistent top-level taggings (0:21(4)) lead to a CCG-like absorption analysis elif is_right_punct_absorption(node): ret...
(C(r'(S[dcl]\S[dcl])/S[dcl]'), C(r'((S[dcl]{_}\S[dcl]{Z}{_})/S[dcl]{Y}){_}'))
(C(r'(S[dcl]\S[dcl])/S[dcl]'), C(r'((S[dcl]{_}\S[dcl]{Z}){_}/S[dcl]{Y}){_}'))
def is_np_n(cat): '''Returns whether _cat_ is the category NP/N.''' return cat.left == NP and cat.right == N
or node.kids[0].tag.startswith(cand) for cand in ('PP-PRD', 'QP-PRD', 'LCP-PRD'))) or
or any(node.kids[0].tag.startswith(cand) for cand in ('PP-PRD', 'QP-PRD', 'LCP-PRD')) ) ) or
def label_node(node, inside_np_internal_structure=False, do_shrink=True): if node.is_leaf(): return node elif node.count() == 1: # shrinkage rules (NP < NN shrinks to NN) if (do_shrink and ((inside_np_internal_structure and node.tag.startswith("NP") and has_noun_tag(node.kids[0]) or node.kids[0].tag == "AD") or # (node...
@echo
def fcomp(l, r): if (l.is_leaf() or r.is_leaf() or l.right != r.left or l.direction != FORWARD or l.direction != r.direction): return None return fake_unify(l, r, l.left / r.right)
def label_adjunction(node, inherit_tag=False, without_labelling=False, inside_np_internal_structure=False): kid_tag = strip_tag_if(not inherit_tag, node.tag)
def label_adjunction(node, without_labelling=False, inside_np_internal_structure=False): kid_tag = node.tag
def label_adjunction(node, inherit_tag=False, without_labelling=False, inside_np_internal_structure=False): kid_tag = strip_tag_if(not inherit_tag, node.tag) if not without_labelling: kids = map(lambda node: label_node(node, inside_np_internal_structure=inside_np_internal_structure), node.kids) else: kids = node.kids ...
def label_apposition(node, inherit_tag=False, inside_np_internal_structure=False): kid_tag = strip_tag_if(not inherit_tag, node.tag)
def label_apposition(node, inside_np_internal_structure=False): kid_tag = node.tag
def label_apposition(node, inherit_tag=False, inside_np_internal_structure=False): kid_tag = strip_tag_if(not inherit_tag, node.tag) if node.count() > 2: # Label the first kid before removing it from the node: if we did this the # other way around, then shrinking (which relies on replace_kid) would not # find _node[0]...
label_node(node[0], inside_np_internal_structure=inside_np_internal_structure) first = node.kids.pop(0) return Node(kid_tag, [first, label_node(node)]) return label_adjunction(node, inherit_tag=inherit_tag) def label_np_internal_structure(node, inherit_tag=False):
first = label_node(node[0], inside_np_internal_structure=inside_np_internal_structure) node.kids.pop(0) return Node(kid_tag, [first, label_node(node, inside_np_internal_structure=inside_np_internal_structure)]) return label_adjunction(node) def label_np_internal_structure(node):
def label_apposition(node, inherit_tag=False, inside_np_internal_structure=False): kid_tag = strip_tag_if(not inherit_tag, node.tag) if node.count() > 2: # Label the first kid before removing it from the node: if we did this the # other way around, then shrinking (which relies on replace_kid) would not # find _node[0]...
kid_tag = strip_tag_if(not inherit_tag, node.tag)
kid_tag = node.tag
def label_np_internal_structure(node, inherit_tag=False): if (node.kids[-1].tag.endswith(':&') # prevent movement when we have an NP with only two children NN ETC and node.count() > 2): etc = node.kids.pop() kid_tag = strip_tag_if(not inherit_tag, node.tag) old_tag = node.tag node.tag = kid_tag return Node(old_tag, ...
return Node(old_tag, [ label_coordination(node, inside_np_internal_structure), etc ])
return Node(old_tag, [ label_coordination(node, inside_np_internal_structure=inside_np_internal_structure), etc ])
def _label_coordination(node, inside_np_internal_structure=False): if (node.kids[-1].tag.endswith(':&') # prevent movement when we have an NP with only two children NN ETC and node.count() > 2): etc = node.kids.pop() kid_tag = base_tag(node.tag, strip_cptb_tag=False) old_tag = node.tag node.tag = kid_tag return Node...
def label_head_initial(node, inherit_tag=False): kid_tag = strip_tag_if(not inherit_tag, node.tag)
def label_head_initial(node): kid_tag = node.tag
def label_head_initial(node, inherit_tag=False): kid_tag = strip_tag_if(not inherit_tag, node.tag) kids = map(label_node, node.kids)[::-1] first_kid, second_kid = kids.pop(), kids.pop() cur = Node(kid_tag, [first_kid, second_kid]) while kids: kid = kids.pop() cur = Node(node.tag, [cur, kid]) cur.tag = node.tag retu...
def label_predication(node, inherit_tag=False):
def label_predication(node):
def label_predication(node, inherit_tag=False): kids = map(label_node, node.kids)
kid_tag = strip_tag_if(not inherit_tag, node.tag)
kid_tag = node.tag
def label_predication(node, inherit_tag=False): kids = map(label_node, node.kids)
trees = parse_tree(sys.stdin.read())
trees = parse_tree(sys.stdin.read(), AugmentedPennParser)
def base_pprint(node, level=0, sep=' ', newline='\n', reduced_leaves=False): out = [] if level == 0: out.append('(') else: out.append( sep * level ) if node.is_leaf(): if reduced_leaves: out.append(node_repr(node)) else: out.append("(%s)" % node_repr(node)) else: # special case for nodes with all-leaf children if no...
elif is_right_adjunction(node): return label_right_adjunction(node)
def label(node, inside_np=False): ''' Labels the descendants of _node_ and returns _node_. ''' if node.category is None: node.category = ptb_to_cat(node) # if this matches the IP root with a *PRO* trace under it, then # we shouldn't map IP -> S, but rather IP -> S\NP if node.tag.startswith('NT'): # map NT -> NP, not N...
else:
if not new_parent_category:
def fix_categories_starting_from(self, node, until):
if cur == SfS and l == N or l == NP: return "nongap_topicalisation"
if cur == SfS and (l == N or l == NP): return "nongap_topicalisation"
def try_unary_rules(l, r, cur): '''Determines if [l r -> cur] matches any unary rules.''' if l == SbNP: for cand_cat, rule in { SfS: "lex_typechange", NP: "lex_typechange", SbNPfSbNP: "lex_typechange", SbS: "lex_typechange", SfSfNP: "np_topicalisation", }.iteritems(): if cur == cand_cat: return rule if config.cn_rule...
if rooted_in_Sdcl(l) and l == r and r == cur: return "vcd_compound"
if rooted_in_Sdcl(l) and l.equal_respecting_features(r) and r == cur: return "vcd_compound"
def try_binary_rules(l, r, cur): if not config.cn_rules: return False # rhs NP needs to be feature-less, otherwise NP NP[conj] gets interpreted # as apposition if r.equal_respecting_features(NP): if cur == NP: if l == NP: return 'np_np_apposition' # NP NP -> NP elif l == S: return 's_np_apposition' # S NP...
return label_adjunction(Node(node.tag, kids), inside_np_internal_structure=inside_np_internal_structure, without_labelling=True)
return reshape_for_coordination(Node(node.tag, kids), inside_np_internal_structure=inside_np_internal_structure) def get_kid(kids, node_tag, seen_cc): print "1: %s" % kids pu = kids.pop() print "2: %s" % kids if seen_cc and len(kids) > 0: xp = kids.pop() print "3: %s" % kids xp_ = Node(xp.tag, [xp, pu]) return...
def label_nonconjunctions(kid): if kid.tag not in ('CC', 'PU'): return label_node(kid, inside_np_internal_structure=inside_np_internal_structure) else: return kid
result = get_first(node, r'*=S $ /DE[CG]/=REL', with_context=True, left_to_right=True)
result = get_first(node, r'/DEC/=REL $ *=S', with_context=True, left_to_right=True) if result is None: result = get_first(node, r'/DEG/=REL $ *=S', with_context=True, left_to_right=True)
def relabel_relativiser(self, node): # Relabel the relativiser category (NP/NP)\S to (NP/NP)\(S|NP) # There's a mis-annotated DEG for DEC in 21:2(1) inter alia result = get_first(node, r'*=S $ /DE[CG]/=REL', with_context=True, left_to_right=True) if result is not None: _, context = result s, relativiser = context['S']...
elif any(kid.lex in ("β€œ", "γ€Œ") for kid in leaf_kids(node)) and any(kid.lex in ("”", "」") for kid in leaf_kids(node)): lqu = first_index_such_that(lambda kid: kid.is_leaf() and kid.lex in ("β€œ", "γ€Œ"), node) rqu = first_index_such_that(lambda kid: kid.is_leaf() and kid.lex in ("”", "」"), node) if rqu != node.count()-1: qu...
def preprocess(root): # IP < PP PU -> PP < PP PU (20:58(1)) if root.count() == 2 and root[1].tag == 'PU' and root[0].tag.startswith('PP'): root.tag = root[0].tag for node in nodes(root): if node.is_leaf(): continue first_kid, first_kid_index = get_nonpunct_kid(node, get_last=False) last_kid, last_kid_index = get_no...
elif is_etc(node): return label_head_final(node)
def label(node, inside_np=False): ''' Labels the descendants of _node_ and returns _node_. ''' if node.category is None: node.category = ptb_to_cat(node) # if this matches the IP root with a *PRO* trace under it, then # we shouldn't map IP -> S, but rather IP -> S\NP if has_noun_tag(node): node.category = N else: node...
elif is_etc(node): return label_head_final(node)
def label(node, inside_np=False): ''' Labels the descendants of _node_ and returns _node_. ''' if node.category is None: node.category = ptb_to_cat(node) # if this matches the IP root with a *PRO* trace under it, then # we shouldn't map IP -> S, but rather IP -> S\NP if has_noun_tag(node): node.category = N else: node...
elif is_PRO_trace(node): new_node = shrink_left(node, node.parent) ret = label(new_node) return ret
def label(node, inside_np=False): ''' Labels the descendants of _node_ and returns _node_. ''' if node.category is None: node.category = ptb_to_cat(node) # if this matches the IP root with a *PRO* trace under it, then # we shouldn't map IP -> S, but rather IP -> S\NP if has_noun_tag(node): node.category = N else: node...
def label_root(node): node.category = ptb_to_cat(node, is_root=True) return label(node)
def label_root(root): root.category = ptb_to_cat(root, is_root=True) for node in nodes(root): if is_PRO_trace(node): new_node = shrink_left(node, node.parent) new_node.tag = base_tag(new_node.tag, strip_cptb_tag=False) inherit_tag(new_node, node) return label(root)
def label_root(node): node.category = ptb_to_cat(node, is_root=True) return label(node)
@echo
def rename_category_while_labelling_with(label_function, node, substitute, when=None): if when and (not when(node.category)): return label_function(node) old_category = node.category node.category = substitute ret = label_function(node) node.category = old_category return ret
@echo
def label_predication(node): node.kids[0] = label(node[0]) node[1].category = node.category | node[0].category node.kids[1] = label(node[1]) return node
@echo
def label_left_absorption(node): node[0].category = ptb_to_cat(node[0]) node[1].category = node.category node.kids[1] = label(node[1]) return node
@echo
def label_right_absorption(node): node[1].category = ptb_to_cat(node[1]) node[0].category = node.category node.kids[0] = label(node[0]) return node
@echo
def label_adjunction(node): node[1].category = ptb_to_cat(node[1], return_none_when_unmatched=True) or node.category node.kids[1] = label(node[1]) node[0].category = featureless(node.category) / featureless(node[1].category) node.kids[0] = label(node[0]) return node
@echo
def label_np_structure(node): node[0].category = node.category / node.category node.kids[0] = label(node[0]) if node.count() > 1: node[1].category = node.category node.kids[1] = label(node[1]) return node
@echo
def label_right_adjunction(node): node[0].category = node.category node.kids[0] = label(node[0]) no_features = featureless(node.category) node[1].category = no_features | no_features node.kids[1] = label(node[1]) return node
@echo
def label_head_final(node): node.kids[0] = label(node[0]) node[1].category = node.category | node[0].category node.kids[1] = label(node[1]) return node
@echo
def label_head_initial(node): node.kids[1] = label(node[1]) node[0].category = node.category / node[1].category node.kids[0] = label(node[0]) return node
@echo
def label_coordination(node, inside_np=False, ucp=False): node[0].category = node.category node.kids[0] = label(node[0], inside_np) node[1].category = ptb_to_cat(node[1]) if ucp else node.category # Label then apply [conj], so that kid categories don't inherit the feature node.kids[1] = label(node[1]) node[1].category...
@echo
def make_atomic_category(atom): return AtomicCategory(atom)
@echo
def np_modifier_tag_to_cat(ptb_tag): ptb_tag = base_tag(ptb_tag) return copy(NPModifierMap.get(ptb_tag, None))
@echo
def is_PRO_trace(node): return (node.count() >= 2 and node[0].count() == 1 and node[0][0].tag == "-NONE-" and node[0][0].lex == "*PRO*")
if node.is_leaf(): return "%s %s" % (node.tag, node.lex)
if hasattr(node, 'category') and node.category is not None: if node.is_leaf(): return "%s {%s} %s" % (node.tag, node.category, node.lex) else: return "%s {%s}" % (node.tag, node.category)
def default_node_repr(node): if node.is_leaf(): return "%s %s" % (node.tag, node.lex) else: return "%s" % node.tag
return "%s" % node.tag def aug_node_repr(node): if node.is_leaf(): return "%s {%s} %s" % (node.tag, node.category, node.lex) else: return "%s {%s}" % (node.tag, node.category)
if node.is_leaf(): return "%s %s" % (node.tag, node.lex) else: return "%s" % node.tag
def default_node_repr(node): if node.is_leaf(): return "%s %s" % (node.tag, node.lex) else: return "%s" % node.tag
if config.aug_pprint: pprint = pprint_with(aug_node_repr) else: pprint = pprint_with(default_node_repr)
pprint = pprint_with(default_node_repr)
def base_pprint(node, level=0, sep=' ', newline='\n', reduced_leaves=False): out = [] if level == 0: out.append('(') else: out.append( sep * level ) if node.is_leaf(): if reduced_leaves: out.append(node_repr(node)) else: out.append("(%s)" % node_repr(node)) else: # special case for nodes with all-leaf children if no...
node[0].category = node.category
node[0].category = ptb_to_cat(node[0]) if ucp else node.category
def label_coordination(node, inside_np=False, ucp=False): node[0].category = node.category node.kids[0] = label(node[0], inside_np) node[1].category = ptb_to_cat(node[1]) if ucp else node.category # Label then apply [conj], so that kid categories don't inherit the feature node.kids[1] = label(node[1]) node[1].category...
(node.tag.startswith('ADVP') and node[0].tag == 'CC') or
(node.tag.startswith('ADVP') and node[0].tag in ('CC', 'PN')) or
def _label_node(node, inside_np_internal_structure=False, do_shrink=True): if node.is_leaf(): return node elif node.count() == 1: # shrinkage rules (NP < NN shrinks to NN) if (do_shrink and ((inside_np_internal_structure and ((node.tag.startswith("NP") and (not node.tag.endswith(':A')) and has_noun_tag(node[0])) or no...
(node.tag.startswith('NP') and node[0].tag.startswith('NT')) or
(node.tag.startswith('NP') and (node[0].tag.startswith('NT') or node[0].tag.startswith('DT'))) or
def _label_node(node, inside_np_internal_structure=False, do_shrink=True): if node.is_leaf(): return node elif node.count() == 1: # shrinkage rules (NP < NN shrinks to NN) if (do_shrink and ((inside_np_internal_structure and ((node.tag.startswith("NP") and (not node.tag.endswith(':A')) and has_noun_tag(node[0])) or no...
(node.tag.startswith('DP') and node[0].tag.startswith('NN')) or
(node.tag.startswith('DP') and (node[0].tag.startswith('NN') or node[0].tag.startswith('PN'))) or
def _label_node(node, inside_np_internal_structure=False, do_shrink=True): if node.is_leaf(): return node elif node.count() == 1: # shrinkage rules (NP < NN shrinks to NN) if (do_shrink and ((inside_np_internal_structure and ((node.tag.startswith("NP") and (not node.tag.endswith(':A')) and has_noun_tag(node[0])) or no...
def label_coordination(node): def _label_coordination(node, inside_np_internal_structure=False): if (node.kids[-1].tag.endswith(':&') and node.count() > 2): etc = node.kids.pop() kid_tag = base_tag(tag, strip_cptb_tag=False) old_tag = node.tag node.tag = kid_tag return Node(old_tag, [ label_coordination(node, insid...
def _label_coordination(node, inside_np_internal_structure=False): if (node.kids[-1].tag.endswith(':&') and node.count() > 2): etc = node.kids.pop() kid_tag = base_tag(tag, strip_cptb_tag=False) old_tag = node.tag node.tag = kid_tag return Node(old_tag, [ label_coordination(node, inside_np_internal_structure), etc ...
def label_coordination(node): def _label_coordination(node, inside_np_internal_structure=False): if (node.kids[-1].tag.endswith(':&') # prevent movement when we have an NP with only two children NN ETC and node.count() > 2): etc = node.kids.pop() kid_tag = base_tag(tag, strip_cptb_tag=False) old_tag = node.tag node.t...
top, context = get_first(node, r'/([ICV]P|V[VA]|VRD|VSB|VCD)/=TOP $ *=SS', with_context=True, left_to_right=True) ss = context.ss
result = get_first(node, r'{ /([ICV]P|V[VA]|VRD|VSB|VCD)/=TOP $ *=SS } ! > /([ICV]P|V[VA]|VRD|VSB|VCD)/', with_context=True, left_to_right=True) if not result: debug('Could not find verbal category; did not create null relativiser.') return
def fix_subject_extraction(self, _, n, pred, w=None, reduced=False): debug("%s", reduced) node = n debug("Fixing subject extraction: %s", lrp_repr(node)) if not reduced: self.remove_null_element(node)
debug("Creating null relativiser unary category: %s", ss.category/ss.category) replace_kid(top.parent, top, Node(ss.category/ss.category, "NN", [top]))
top, context = result SS = context.ss.category debug("Creating null relativiser unary category: %s", SS/SS) replace_kid(top.parent, top, Node(SS/SS, "NN", [top]))
def fix_subject_extraction(self, _, n, pred, w=None, reduced=False): debug("%s", reduced) node = n debug("Fixing subject extraction: %s", lrp_repr(node)) if not reduced: self.remove_null_element(node)
node[0].category = node.category / node.category
node[0].category = featureless(node.category) / featureless(node.category)
def label_np_structure(node): node[0].category = node.category / node.category node.kids[0] = label(node[0]) if node.count() > 1: node[1].category = node.category node.kids[1] = label(node[1]) return node
(r'/(IP|CP-CND|VP)/=P < {/-TPC-\d+:t$/a=T $ /(IP|CP-CND)/=S }', self.fix_topicalisation_with_gap),
(r'/(IP|CP-CND)/=P < {/-TPC-\d+:t$/a=T $ /(IP|CP-CND)/=S }', self.fix_topicalisation_with_gap),
def pattern(self): return list(( # must come before object extraction (r'*=TOP $ /-SBJ-d+/a=N < { * < /LB/=BEI } << { /NP-(?:TPC|OBJ)/ < ^/\*/ $ /V[PV]|VRD|VSB|VCD/=PRED }', self.fix_reduced_long_bei_gap), (r'*=TOP < { * < /LB/=BEI } << { /NP-(?:TPC|OBJ)/ < ^/\*/ $ /V[PV]|VRD|VSB|VCD/=PRED }', self.fix_r...
for node, ctx in find_all(top, r'/VP/=VP < /NP/=NP < /(QP|V[PV])/=QP', with_context=True):
for node, ctx in find_all(top, r'/VP/=VP <1 /NP/=NP <2 /(QP|V[PV])/=QP', with_context=True):
def clusterfix(self, top, pp, p, s, t): debug("Fixing argument cluster coordination: %s", pprint(top)) debug('T: %s', t) # 1. Shrink the verb (node T) self.fix_object_gap(pp, p, t, s) # 2. Reattach the verb above the TOP node new_node = Node(top.category, 'TAG', top.kids) top.kids = [t, new_node] # (Reattaching parent ...
node.kids[0] = node[0][0]
node.kids = node[0].kids
def label(root): for node in nodes(root): if node.is_leaf(): continue first_kid, first_kid_index = get_nonpunct_kid(node, get_last=False) last_kid, last_kid_index = get_nonpunct_kid(node, get_last=True) # CPTB/Chinese-specific fixes # --------------------------- # PP(P CP NP) in derivations like 5:11(3) should be P...
ModifierTags = frozenset(("TPC", "LOC", "EXT", "ADV", "DIR", "IO", "LGS", "MNR", "PN", "PRP", "TMP", "TTL")) ModifierTagsRegex = "(?:" + "|".join(ModifierTags) + ")"
def is_rooted_in(subcat, cat, respecting_features=False): cur = cat while not cur.is_leaf() and cur.left: cur = cur.left return cur.equal_respecting_features(subcat) if respecting_features else cur == subcat
(r'^/\*T\*/ > { /[NPQ]P(?:-(?:TPC|LOC|EXT|ADV|DIR|IO|LGS|MNR|PN|PRP|TMP|TTL))?(?!-\d+)/=K >> { /[ICV]P/ $ {/WH[NP]P(-\d+)?/ > { /CP/=PRED > *=N } } } }', self.fix_nongap_extraction),
(r'''^/\*T\*/ > { /[NPQ]P(?:-%(tags)s)?(?!-\d+)/=K >> { /[ICV]P/ $ {/WH[NP]P(-\d+)?/ > { /CP/=PRED > *=N } } } }''' % { 'tags': ModifierTagsRegex }, self.fix_nongap_extraction),
def pattern(self): return list(( # /VP/ < { /VP:c/ < /V[PVEC]|VRD|VSB|VCD/ < /NP/=NP < /QP/=QP } < { /VP:c/ < /NP/ < /QP/ ! < /V[PVEC]|VRD|VSB|VCD/ } # must come before object extraction (r'*=TOP $ /-SBJ-d+/a=N < { * < /LB/=BEI } << { /NP-(?:TPC|OBJ)/ < ^/\*/ $ /V[PV]|VRD|VSB|VCD/=PRED }', self.fix_reduced_long_bei_gap...
inherit_tag(ctx['S'], ctx['P']) self.fix_object_gap(ctx['PP'], ctx['P'], ctx['T'], ctx['S']) self.fix_categories_starting_from(ctx['S'], g)
inherit_tag(ctx.s, ctx.p) self.fix_object_gap(ctx.pp, ctx.p, ctx.t, ctx.s) self.fix_categories_starting_from(ctx.s, g)
def fix_rnr(self, rnr, g): debug("Fixing RNR: %s", pprint(g)) index = get_trace_index_from_tag(rnr.lex) # -i debug("index: %s", index) expr = r'*=PP < { *=P < { *=T < ^/\*RNR\*%s/ $ *=S } }' % index for node, ctx in find_all(g, expr, with_context=True): inherit_tag(ctx['S'], ctx['P']) self.fix_object_gap(ctx['PP'], ctx...
argument = ctx['T'] self.fix_object_gap(ctx['PP'], ctx['P'], ctx['T'], ctx['S'])
argument = ctx.t self.fix_object_gap(ctx.pp, ctx.p, ctx.t, ctx.s)
def fix_rnr(self, rnr, g): debug("Fixing RNR: %s", pprint(g)) index = get_trace_index_from_tag(rnr.lex) # -i debug("index: %s", index) expr = r'*=PP < { *=P < { *=T < ^/\*RNR\*%s/ $ *=S } }' % index for node, ctx in find_all(g, expr, with_context=True): inherit_tag(ctx['S'], ctx['P']) self.fix_object_gap(ctx['PP'], ctx...
debug('PP: %s, P: %s, T: %s, S: %s', *map(lrp_repr, (ctx['PP'],ctx['P'],ctx['T'],ctx['S'])))
debug('PP: %s, P: %s, T: %s, S: %s', *map(lrp_repr, (ctx.pp, ctx.p, ctx.t, ctx.s)))
def fix_rnr(self, rnr, g): debug("Fixing RNR: %s", pprint(g)) index = get_trace_index_from_tag(rnr.lex) # -i debug("index: %s", index) expr = r'*=PP < { *=P < { *=T < ^/\*RNR\*%s/ $ *=S } }' % index for node, ctx in find_all(g, expr, with_context=True): inherit_tag(ctx['S'], ctx['P']) self.fix_object_gap(ctx['PP'], ctx...
new_g.category = ctx['S'].category.left
new_g.category = ctx.s.category.left
def fix_rnr(self, rnr, g): debug("Fixing RNR: %s", pprint(g)) index = get_trace_index_from_tag(rnr.lex) # -i debug("index: %s", index) expr = r'*=PP < { *=P < { *=T < ^/\*RNR\*%s/ $ *=S } }' % index for node, ctx in find_all(g, expr, with_context=True): inherit_tag(ctx['S'], ctx['P']) self.fix_object_gap(ctx['PP'], ctx...
pp, context = get_first(node, r'*=PP < { *=P < { /WH[NP]P/=T $ *=S } }', with_context=True) p, t, s = context['P'], context['T'], context['S']
pp, ctx = get_first(node, r'*=PP < { *=P < { /WH[NP]P/=T $ *=S } }', with_context=True) p, t, s = ctx.p, ctx.t, ctx.s
def remove_null_element(self, node): # Remove the null element WHNP and its trace -NONE- '*OP*' and shrink tree pp, context = get_first(node, r'*=PP < { *=P < { /WH[NP]P/=T $ *=S } }', with_context=True) p, t, s = context['P'], context['T'], context['S']
s, relativiser = context['S'], context['REL']
s, relativiser = context.s, context.rel
def relabel_relativiser(self, node): # Relabel the relativiser category (NP/NP)\S to (NP/NP)\(S|NP) result = get_first(node, r'*=S $ /(DEC|SP)/=REL', with_context=True, left_to_right=True)
for trace_NP, context in find_all(node, expr, with_context=True): pp, p, t, s = (context[n] for n in "PP P T S".split())
for trace_NP, ctx in find_all(node, expr, with_context=True): pp, p, t, s = ctx.pp, ctx.p, ctx.t, ctx.s
def fix_subject_extraction(self, _, n, pred, w=None, reduced=False): debug("%s", reduced) node = n debug("Fixing subject extraction: %s", lrp_repr(node)) if not reduced: self.remove_null_element(node)
ss = context["SS"]
ss = context.ss
def fix_subject_extraction(self, _, n, pred, w=None, reduced=False): debug("%s", reduced) node = n debug("Fixing subject extraction: %s", lrp_repr(node)) if not reduced: self.remove_null_element(node)
expr = r'*=PP < { *=P < { /[NPQ]P(?:-(?:TPC|LOC|EXT|ADV|DIR|IO|LGS|MNR|PN|PRP|TMP|TTL))?%s/=T << ^/\*T\*/ $ *=S } }' % index
expr = (r'*=PP < { *=P < { /[NPQ]P(?:-%(tags)s)?%(index)s/=T << ^/\*T\*/ $ *=S } }' % { 'tags': ModifierTagsRegex, 'index': index })
def fix_nongap_extraction(self, _, n, pred, k): node = n debug("Fixing nongap extraction: %s", pprint(node)) debug("k %s", pprint(k)) self.remove_null_element(node)
for trace_NP, context in find_all(node, expr, with_context=True): pp, p, t, s = (context[n] for n in "PP P T S".split())
for trace_NP, ctx in find_all(node, expr, with_context=True): pp, p, t, s = ctx.pp, ctx.p, ctx.t, ctx.s
def fix_nongap_extraction(self, _, n, pred, k): node = n debug("Fixing nongap extraction: %s", pprint(node)) debug("k %s", pprint(k)) self.remove_null_element(node)
ss = context["SS"]
ss = context.ss
def fix_nongap_extraction(self, _, n, pred, k): node = n debug("Fixing nongap extraction: %s", pprint(node)) debug("k %s", pprint(k)) self.remove_null_element(node)
for trace_NP, context in find_all(node, expr, with_context=True): top, pp, p, t, s = (context[n] for n in "TOP PP P T S".split())
for trace_NP, ctx in find_all(node, expr, with_context=True): top, pp, p, t, s = ctx.top, ctx.pp, ctx.p, ctx.t, ctx.s
def fix_object_extraction(self, _, n, pred, w=None, reduced=False): node = n debug("Fixing object extraction: %s", lrp_repr(node)) if not reduced: self.remove_null_element(node) if w: index = get_trace_index_from_tag(w.tag) else: index = '' expr = r'/IP/=TOP << { *=PP < { *=P < { /NP-OBJ/=T << ^/\*T\*%s/ $ *=S } } }'...
_, ctx = result; ss = ctx['SS']
_, ctx = result; ss = ctx.ss
def fix_object_extraction(self, _, n, pred, w=None, reduced=False): node = n debug("Fixing object extraction: %s", lrp_repr(node)) if not reduced: self.remove_null_element(node) if w: index = get_trace_index_from_tag(w.tag) else: index = '' expr = r'/IP/=TOP << { *=PP < { *=P < { /NP-OBJ/=T << ^/\*T\*%s/ $ *=S } } }'...
bei, context = get_first(top, r'*=S [ $ /LB/=BEI | $ ^"η”±"=BEI ]', with_context=True) s = context['S'] bei = context['BEI']
bei, ctx = get_first(top, r'*=S [ $ /LB/=BEI | $ ^"η”±"=BEI ]', with_context=True) s, bei = ctx.s, ctx.bei
def relabel_bei_category(self, top, pred): bei, context = get_first(top, r'*=S [ $ /LB/=BEI | $ ^"η”±"=BEI ]', with_context=True) s = context['S'] bei = context['BEI']
ba, context = get_first(top, r'*=S [ $ /BA/=BA ]', with_context=True) s, ba = context['S'], context['BA']
_, ctx = get_first(top, r'*=S [ $ /BA/=BA ]', with_context=True) s, ba = ctx.s, ctx.ba
def relabel_ba_category(self, top, ba): ba, context = get_first(top, r'*=S [ $ /BA/=BA ]', with_context=True) s, ba = context['S'], context['BA']
trace_NP, context = get_first(top, expr, with_context=True) pp, p, t, s = (context[n] for n in "PP P T S".split())
trace_NP, ctx = get_first(top, expr, with_context=True) pp, p, t, s = ctx.pp, ctx.p, ctx.t, ctx.s
def fix_long_bei_gap(self, node, bei, pred, top, n=None, reduced=False): debug("Fixing long bei gap: %s", lrp_repr(node))
for trace_NP, context in find_all(top, r'*=PP < {*=P < { /NP-OBJ/=T < ^/\*-/ $ *=S } }', with_context=True):
for trace_NP, ctx in find_all(top, r'*=PP < {*=P < { /NP-OBJ/=T < ^/\*-/ $ *=S } }', with_context=True):
def fix_ba_object_gap(self, node, top, c, ba): debug("Fixing ba-construction object gap: %s" % lrp_repr(node))
pp, p, t, s = (context[n] for n in "PP P T S".split())
pp, p, t, s = ctx.pp, ctx.p, ctx.t, ctx.s
def fix_ba_object_gap(self, node, top, c, ba): debug("Fixing ba-construction object gap: %s" % lrp_repr(node))
self.fix_categories_starting_from(ctx['S'], until=top)
self.fix_categories_starting_from(ctx.s, until=top)
def fix_topicalisation_with_gap(self, node, p, s, t): debug("Fixing topicalisation with gap:\nnode=%s\ns=%s\nt=%s", lrp_repr(node), pprint(s), pprint(t))
(C(r'(S\S)\(S\S)'), C(r'((S{Y}\S{Z}){Y}\(S{Y}\S{Z}){Y}){_}'))
(C(r'(S\S)\(S\S)'), C(r'((S{Y}\S{Z}){Y}\(S{Y}\S{Z}){Y}){_}')),
def is_np_n(cat): '''Returns whether _cat_ is the category NP/N.''' return cat.left == NP and cat.right == N
if cur == NfN and (l == SbNP or l == SfNP or l == S): return "null_relativiser_typechange"
if cur in (NfN, C('(N/N)/(N/N)')) and (l == SbNP or l == SfNP or l == S): return "null_relativiser_typechange"
def try_unary_rules(l, r, cur): '''Determines if [l r -> cur] matches any unary rules.''' if l == SbNP: for cand_cat, rule in { SfS: "lex_typechange", NP: "lex_typechange", SbNPfSbNP: "lex_typechange", SbS: "lex_typechange", SfSfNP: "np_topicalisation", }.iteritems(): if cur == cand_cat: return rule if config.cn_rule...
if cur.direction == BACKWARD and cur.direction != r.direction:
if cur.direction == BACKWARD and cur.direction == r.direction:
def try_composition(l, r, cur, examine_modes=False): '''Determines if [l r -> cur] matches any composition rules. If _examine_modes_ is true, then the modes of the arguments are checked to see if they permit composition.''' if is_composition(l, r, cur, examine_modes): if l.direction == FORWARD: # Forward harmonic or cr...
if cur.direction == FORWARD and cur.direction != r.direction:
if cur.direction == FORWARD and cur.direction == l.direction:
def try_composition(l, r, cur, examine_modes=False): '''Determines if [l r -> cur] matches any composition rules. If _examine_modes_ is true, then the modes of the arguments are checked to see if they permit composition.''' if is_composition(l, r, cur, examine_modes): if l.direction == FORWARD: # Forward harmonic or cr...
if cur.direction == BACKWARD and cur.direction == r.direction:
if cur.direction == BACKWARD and cur.direction == l.direction:
def try_composition(l, r, cur, examine_modes=False): '''Determines if [l r -> cur] matches any composition rules. If _examine_modes_ is true, then the modes of the arguments are checked to see if they permit composition.''' if is_composition(l, r, cur, examine_modes): if l.direction == FORWARD: # Forward harmonic or cr...
if not reduced: self.remove_null_element(node)
if pred.tag.startswith('NP'): n[0].kids.pop(0) else: if not reduced: self.remove_null_element(node)
def fix_subject_extraction(self, _, n, pred, w=None, reduced=False): debug("%s", reduced) node = n debug("Fixing subject extraction: %s", lrp_repr(node)) if not reduced: self.remove_null_element(node)
if not reduced: self.remove_null_element(node)
if pred.tag.startswith('NP'): n[0].kids.pop(0) else: if not reduced: self.remove_null_element(node)
def fix_object_extraction(self, _, n, pred, w=None, reduced=False): node = n debug("Fixing object extraction: %s", lrp_repr(node)) if not reduced: self.remove_null_element(node) if w: index = get_trace_index_from_tag(w.tag) else: index = '' expr = r'/IP/=TOP << { *=PP < { *=P < { /NP-(OBJ|EXT)/=T << ^/\*T\*%s/ $ *=S ...