Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly
Sign up
Appearance settings

Commit 3e25850

Browse files
Extend Romance block to support Spanish
1 parent 5a7dd36 commit 3e25850

File tree

1 file changed

+72
-13
lines changed

1 file changed

+72
-13
lines changed

‎udapi/block/msf/romance/romance.py

Lines changed: 72 additions & 13 deletions
Original file line numberDiff line numberDiff line change
@@ -2,6 +2,9 @@
22
import udapi.block.msf.phrase
33
from enum import Enum
44

5+
AUXES_HAVE = ['ter', 'haber']
6+
AUXES_BE = ['estar']
7+
58
class Aspect(str, Enum):
69
IMP = 'Imp'
710
IMPPROG = 'ImpProg'
@@ -41,7 +44,7 @@ def process_node(self, node):
4144
self.process_copulas(node,cop,auxes,refl,expl)
4245
return
4346

44-
if node.upos == 'VERB':
47+
if node.upos == 'VERB':#TODO maybe add 'or node.feats['VerbForm'] == 'Part'?
4548
auxes = [x for x in node.children if x.udeprel == 'aux']
4649
aux_pass = [x for x in node.children if x.deprel == 'aux:pass']
4750
auxes_without_pass = [x for x in node.children if x.udeprel == 'aux' and x.deprel != 'aux:pass']
@@ -66,29 +69,48 @@ def process_node(self, node):
6669
phrase_ords = [node.ord] + [r.ord for r in refl]
6770
phrase_ords.sort()
6871

72+
# Portuguese
6973
# presente -> PhraseTense=Pres, PhraseAspect=''
7074
# Futuro do presente -> PhraseTense=Fut, PhraseAspect=''
75+
76+
# Spanish
77+
# presente -> PhraseTense=Pres, PhraseAspect=''
78+
# futuro simple -> PhraseTense=Fut, PhraseAspect=''
7179
aspect = ''
7280
tense = node.feats['Tense']
7381

7482
if node.feats['Mood'] == 'Ind':
75-
83+
84+
# Portuguese
7685
# pretérito imperfeito -> PhraseTense=Past, PhraseAspect=Imp
86+
87+
# Spanish
88+
# pretérito imperfecto -> PhraseTense=Past, PhraseAspect=Imp
7789
if node.feats['Tense'] == 'Imp':
7890
tense=Tense.PAST.value
7991
aspect=Aspect.IMP.value
8092

93+
# Portuguese
8194
# pretérito perfeito -> PhraseTense=Past, PhraseAspect=Perf
95+
96+
# Spanish
97+
# pretérito perfecto -> PhraseTense=Past, PhraseAspect=Perf
8298
if node.feats['Tense'] == 'Past':
8399
aspect=Aspect.PERF.value
84100

101+
# Portuguese
85102
# pretérito mais que perfeito simples -> PhraseTense=Past, PhraseAspect=Pqp
86103
if node.feats['Tense'] == 'Pqp':
87104
tense=Tense.PAST.value
88105
aspect=Aspect.PQP.value
89106

107+
# Portuguese
90108
# subjunctive presente -> PhraseTense=Pres, PhraseAspect=''
91109
# subjunctive futuro -> PhraseTense=Fut, PhraseAspect=''
110+
111+
# Spanish
112+
# subjunctive presente -> PhraseTense=Pres, PhraseAspect=''
113+
# subjunctive futuro -> PhraseTense=Fut, PhraseAspect='' TODO not annotated in treebanks?
92114
if node.feats['Mood'] == 'Sub':
93115

94116
if node.feats['Tense'] == 'Past':
@@ -99,7 +121,11 @@ def process_node(self, node):
99121
tense=Tense.PAST.value
100122
aspect=Aspect.IMP.value
101123

124+
# Portuguese
102125
# Futuro do pretérito (cnd) -> PhraseTense=Pres, PhraseAspect='', PhraseMood=Cnd
126+
127+
# Spanish
128+
# pospretérito (cnd) -> PhraseTense=Pres, PhraseAspect='', PhraseMood=Cnd
103129
if node.feats['Mood'] == 'Cnd':
104130
aspect=''
105131
tense=Tense.PRES.value
@@ -146,7 +172,6 @@ def process_node(self, node):
146172
else:
147173
self.process_periphrastic_verb_forms(aux_pass[0], auxes_without_pass, refl, auxes, node)
148174

149-
150175
def process_periphrastic_verb_forms(self, node, auxes, refl, all_auxes, head_node):
151176
"""
152177
Parameters
@@ -166,19 +191,24 @@ def process_periphrastic_verb_forms(self, node, auxes, refl, all_auxes, head_nod
166191

167192
if len(auxes) == 1:
168193
# Cnd
169-
if ((auxes[0].lemma =='ter' and node.feats['VerbForm'] == 'Part') or (auxes[0].lemma =='estar' and node.feats['VerbForm'] == 'Ger')) and auxes[0].feats['Mood'] == 'Cnd':
194+
if ((auxes[0].lemma inAUXES_HAVE and node.feats['VerbForm'] == 'Part') or (auxes[0].lemma inAUXES_BE and node.feats['VerbForm'] == 'Ger')) and auxes[0].feats['Mood'] == 'Cnd':
170195
phrase_ords = [head_node.ord] + [x.ord for x in all_auxes] + [r.ord for r in refl] + [r.ord for r in refl]
171196
phrase_ords.sort()
172197

198+
# Portuguese
173199
# aux estar cond + gerund -> PhraseTense=Pres, PhraseAspect=Prog, PhraseMood=Cnd
174200
if auxes[0].lemma == 'estar':
175201
tense=Tense.PRES.value
176202
aspect=Aspect.PROG.value
177203

178-
# Futuro do pretérito composto -> PhraseTense=Past, PhraseAspect=Perf, PhraseMood=Cnd
204+
# Portuguese
205+
# Futuro do pretérito composto -> PhraseTense=Past, PhraseAspect='', PhraseMood=Cnd
206+
207+
# Spanish
208+
# Antepospretérito -> PhraseTense=Past, PhraseAspect='', PhraseMood=Cnd
179209
else:
180210
tense=Tense.PAST.value
181-
aspect=Aspect.PERF.value
211+
aspect=''
182212

183213
self.write_node_info(head_node,
184214
tense=tense,
@@ -193,26 +223,30 @@ def process_periphrastic_verb_forms(self, node, auxes, refl, all_auxes, head_nod
193223
return
194224

195225
# Auxiliary 'estar' followed by a gerund
196-
if auxes[0].lemma =='estar' and node.feats['VerbForm'] == 'Ger':
226+
if auxes[0].lemma inAUXES_BE and node.feats['VerbForm'] == 'Ger':
197227
phrase_ords = [head_node.ord] + [x.ord for x in all_auxes] + [r.ord for r in refl]
198228
phrase_ords.sort()
199229

230+
# Portuguese + Spanish
200231
# pretérito imperfeito (aux estar) -> PhraseTense=Past, PhraseAspect=ImpProg
201232
# subjunctive pretérito imperfeito (aux estar) -> PhraseTense=Past, PhraseAspect=ImpProg, PhraseMood=Sub
202233
if auxes[0].feats['Tense'] == 'Imp':
203234
tense=Tense.PAST.value
204235
aspect=Aspect.IMPPROG.value
205236

237+
# Portuguese + Spanish
206238
# pretérito perfeito (aux estar) -> PhraseTense=Past, PhraseAspect=PerfProg
207239
elif auxes[0].feats['Tense'] == 'Past':
208240
tense=Tense.PAST.value
209241
aspect=Aspect.PERFPROG.value
210242

243+
# Portuguese + Spanish
211244
# conditional (aux estar) -> PhraseTense=Pres, PhraseAspect=Prog, PhraseMood=Cnd
212245
elif auxes[0].feats['Mood'] == 'Cnd':
213246
tense=Tense.PRES.value
214247
aspect=Aspect.PROG.value
215248

249+
# Portuguese + Spanish
216250
# presente (aux estar) -> PhraseTense=Pres, PhraseAspect=Prog
217251
# futuro do presente (aux estar) -> PhraseTense=Fut, PhraseAspect=Prog
218252
# subjunctive presente (aux estar) -> PhraseTense=Pres, PhraseAspect=Prog, PhraseMood=Sub
@@ -232,22 +266,38 @@ def process_periphrastic_verb_forms(self, node, auxes, refl, all_auxes, head_nod
232266
expl=expl,
233267
ords=phrase_ords)
234268

235-
# Auxiliary 'ter' followed by a participle
236-
if auxes[0].lemma =='ter' and node.feats['VerbForm'] == 'Part':
269+
# Auxiliary 'ter' / 'haber' followed by a participle
270+
if auxes[0].lemma inAUXES_HAVE and node.feats['VerbForm'] == 'Part':
237271
phrase_ords = [head_node.ord] + [x.ord for x in all_auxes] + [r.ord for r in refl]
238272
phrase_ords.sort()
239273

274+
# Portuguese
240275
# futuro do presente composto (aux ter) -> PhraseTense=Fut, PhraseAspect=Perf
276+
277+
# Spanish
278+
# Futuro compuesto antefuturo -> PhraseTense=Fut, PhraseAspect=Perf
241279
aspect=Aspect.PERF.value
242280
tense=auxes[0].feats['Tense']
243281

282+
# Portuguese
244283
# pretérito perfeito composto (aux ter) -> PhraseTense=PastPres, PhraseAspect=Perf
245284
# subjonctive pretérito perfeito composto (aux ter) -> PhraseTense=PastPres, PhraseAspect=Perf, PhraseMood=Sub
285+
246286
if auxes[0].feats['Tense'] == 'Pres':
247-
tense=Tense.PASTPRES.value
248287

249-
# pretérito mais que perfeito composto (aux ter/haver) -> PhraseTense=Past, PhraseAspect=Pqp
288+
# Spanish
289+
# Pretérito perfecto compuesto ante presente -> PhraseTense=Past, PhraseAspect=Perf
290+
if auxes[0].lemma == 'haber' and auxes[0].feats['Mood'] != 'Sub':
291+
tense = Tense.PAST.value
292+
else:
293+
tense=Tense.PASTPRES.value
294+
295+
# Portuguese
296+
# pretérito mais que perfeito composto (aux ter) -> PhraseTense=Past, PhraseAspect=Pqp
250297
# subjonctive pretérito mais-que-perfeito composto (aux ter) -> PhraseTense=Past, PhraseAspect=Pqp, PhraseMood=Sub
298+
299+
# Spanish
300+
# pretérito pluscuamperfecto -> PhraseTense=Past, PhraseAspect=Pqp
251301
elif auxes[0].feats['Tense'] in ['Imp', 'Past']: # TODO prej neni v Past, jenom Imp
252302
tense=Tense.PAST.value
253303
aspect=Aspect.PQP.value
@@ -263,6 +313,8 @@ def process_periphrastic_verb_forms(self, node, auxes, refl, all_auxes, head_nod
263313
expl=expl,
264314
ords=phrase_ords)
265315

316+
# Portuguese
317+
# pretérito mais que perfeito composto (aux haver) -> PhraseTense=Past, PhraseAspect=Perf
266318
if auxes[0].lemma == 'haver' and auxes[0].feats['Tense'] == 'Imp' and node.feats['VerbForm'] == 'Part':
267319
phrase_ords = [head_node.ord] + [x.ord for x in all_auxes] + [r.ord for r in refl]
268320
phrase_ords.sort()
@@ -303,6 +355,7 @@ def process_periphrastic_verb_forms(self, node, auxes, refl, all_auxes, head_nod
303355

304356

305357
# auxiliary 'ir' followed by infinitive
358+
# TODO solve these verb forms for Spanish (VERB 'ir' + ADP 'a' + infinitive)
306359
if auxes[0].lemma == 'ir' and node.feats['VerbForm'] == 'Inf':
307360
phrase_ords = [head_node.ord] + [x.ord for x in all_auxes] + [r.ord for r in refl]
308361
phrase_ords.sort()
@@ -368,7 +421,11 @@ def process_periphrastic_verb_forms(self, node, auxes, refl, all_auxes, head_nod
368421
ords=phrase_ords)
369422

370423
elif len(auxes) == 2:
424+
# Portuguese
371425
# auxiliry 'ir' followed by auxiliary 'estar' in infinitive and a gerund
426+
427+
# TODO Spanish
428+
# VERB 'ir' + ADP 'a' + AUX 'estar'.Inf + gerund
372429
if auxes[0].lemma == 'ir' and auxes[1].lemma == 'estar' and node.feats['VerbForm'] == 'Ger':
373430
phrase_ords = [head_node.ord] + [x.ord for x in all_auxes] + [r.ord for r in refl]
374431
phrase_ords.sort()
@@ -431,8 +488,8 @@ def process_periphrastic_verb_forms(self, node, auxes, refl, all_auxes, head_nod
431488

432489

433490

434-
# Cnd (only ter), Sub and Past,Pres,Fut tenses: 2 auxes - ter + estar
435-
if auxes[0].lemma in ['ter', 'haver'] and auxes[1].lemma == 'estar' and node.feats['VerbForm'] == 'Ger':
491+
# Cnd (only ter/haber), Sub and Past,Pres,Fut tenses: 2 auxes - ter/haber + estar
492+
if auxes[0].lemma in AUXES_HAVE and auxes[1].lemma == 'estar' and node.feats['VerbForm'] == 'Ger':
436493
phrase_ords = [head_node.ord] + [x.ord for x in all_auxes] + [r.ord for r in refl]
437494
phrase_ords.sort()
438495

@@ -473,6 +530,8 @@ def process_periphrastic_verb_forms(self, node, auxes, refl, all_auxes, head_nod
473530
return
474531

475532
def process_copulas(self, node, cop, auxes, refl, expl):
533+
534+
aspect = ''
476535

477536
if not auxes:
478537
tense = cop[0].feats['Tense']

0 commit comments

Comments
(0)

AltStyle によって変換されたページ (->オリジナル) /