def _dict_post_parse(tokens, loc, string):
acc = tokens.tokens
for a in list(acc):
for tok in list(a):
if not tok:
continue
if is_text(tok):
new_tok = Annotation(tok, a.start, a.end, [])
else:
kv = list(tok)
key = kv[0]
value = kv[1:]
new_tok = Annotation(text(key), tok.start, tok.end, value)
acc.append(new_tok)
return tokens
def extractText(tokens, loc, string):
start, d, end = tokens
content = string[start:end]
annotations = [
Annotation(k, v[0].start, v[-1].end, v) for k, v in d.items()
]
return ParseResults(d.type, start, end, [content] + annotations)
def group_list(tokens):
start, end = tokens.start, tokens.end
# RE-MATCH :(
sub_string = tokens.tokens[0]
found = tokens.type.regex.match(sub_string)
lookup = dict(reversed(p) for p in found.re.groupindex.items())
ann = []
pe = 0
for i, (s, e) in enumerate(found.regs[1:], start=1):
if s == -1:
continue # NOT FOUND
g = sub_string[s:e]
if pe <= s:
ann.append(g)
pe = e
ii = chr(i + ord("0"))
ann.append(Annotation(ii, s + start, e + start, [g]))
n = lookup.get(i)
if n:
ann.append(Annotation(n, s + start, e + start, [g]))
return ParseResults(_plain_group, start, end, ann)
def _dict_post_parse(tokens, loc, string):
acc = tokens.tokens
for a in list(acc):
for tok in list(a):
if not tok:
continue
kv = list(tok)
key = kv[0]
value = kv[1:]
new_tok = Annotation(text(key), value)
acc.append(new_tok)
return tokens