1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
|
#!/usr/bin/env python
#-*- coding:utf-8 -*-
#Frog Wrapper with XML input and FoLiA output support
from __future__ import print_function, unicode_literals, division, absolute_import
import getopt
import lxml.etree
import sys
import os
import codecs
if __name__ == "__main__":
sys.path.append(sys.path[0] + '/../..')
os.environ['PYTHONPATH'] = sys.path[0] + '/../..'
import pynlpl.formats.folia as folia
from pynlpl.clients.frogclient import FrogClient
def legacyout(i, word,lemma,morph,pos):
if word:
out = str(i + 1) + "\t" + word + "\t" + lemma + "\t" + morph + "\t" + pos
print(out.encode('utf-8'))
else:
print()
def usage():
print >>sys.stderr,"frogwrapper.py [options]"
print >>sys.stderr,"------------------------------------------------------"
print >>sys.stderr,"Input file:"
print >>sys.stderr,"\t--txt=[file] Plaintext input"
print >>sys.stderr,"\t--xml=[file] XML Input"
print >>sys.stderr,"\t--folia=[file] FoLiA XML Input"
print >>sys.stderr,"Frog settings:"
print >>sys.stderr,"\t-p [port] Port the Frog server is running on"
print >>sys.stderr,"Output type:"
print >>sys.stderr,"\t--id=[ID] ID for outputted FoLiA XML Document"
print >>sys.stderr,"\t--legacy Use legacy columned output instead of FoLiA"
print >>sys.stderr,"\t-o Write output to input file (only works for --folia)"
print >>sys.stderr,"XML Input:"
print >>sys.stderr,"\t--selectsen=[expr] Use xpath expression to select sentences"
print >>sys.stderr,"\t--selectpar=[expr] Use xpath expression to select paragraphs"
print >>sys.stderr,"\t--idattrib=[attrb] Copy ID from this attribute"
print >>sys.stderr,"Text Input:"
print >>sys.stderr,"\t-N No structure"
print >>sys.stderr,"\t-S One sentence per line (strict)"
print >>sys.stderr,"\t-P One paragraph per line"
print >>sys.stderr,"\t-I Value in first column (tab seperated) is ID!"
print >>sys.stderr,"\t-E [encoding] Encoding of input file (default: utf-8)"
try:
opts, files = getopt.getopt(sys.argv[1:], "hSPINEp:o", ["txt=","xml=", "folia=","id=",'legacy','tok','selectsen=','selectpar=','idattrib='])
except getopt.GetoptError as err:
# print help information and exit:
print(str(err))
usage()
sys.exit(1)
textfile = xmlfile = foliafile = None
foliaid = 'UNTITLED'
legacy = None
tok = False
idinfirstcolumn = False
encoding = 'utf-8'
mode='s'
xpathselect = ''
idattrib=''
port = None
save = False
for o, a in opts:
if o == "-h":
usage()
sys.exit(0)
elif o == "-I":
idinfirstcolumn = True
elif o == "-S":
mode = 's'
elif o == "-P":
mode = 'p'
elif o == "-p":
port = int(a)
elif o == "-N":
mode = 'n'
elif o == "-E":
encoding = a
elif o == "--selectsen":
mode='s'
xpathselect = a
elif o == "--selectpar":
mode='p'
xpathselect = a
elif o == "--idattrib":
idattrib = a
elif o == "--txt":
textfile = a
elif o == "--xml":
xmlfile = a
elif o == "--folia":
foliafile = a
elif o == "--id":
foliaid = a #ID
elif o == "-o":
save = True
elif o == "--legacy":
legacy = True
elif o == "--tok":
tok = True
else:
print >>sys.stderr, "ERROR: Unknown option:",o
sys.exit(1)
if not port:
print >> sys.stderr,"ERROR: No port specified to connect to Frog server"
sys.exit(2)
elif (not textfile and not xmlfile and not foliafile):
print >> sys.stderr,"ERROR: Specify a file with either --txt, --xml or --folia"
sys.exit(2)
elif xmlfile and not xpathselect:
print >> sys.stderr,"ERROR: You need to specify --selectsen or --selectpar when using --xml"
sys.exit(2)
frogclient = FrogClient('localhost',port)
idmap = []
data = []
if textfile:
f = codecs.open(textfile, 'r', encoding)
for line in f.readlines():
if idinfirstcolumn:
id, line = line.split('\t',1)
idmap.append(id.strip())
else:
idmap.append(None)
data.append(line.strip())
f.close()
if xmlfile:
xmldoc = lxml.etree.parse(xmlfile)
for node in xmldoc.xpath(xpathselect):
if idattrib:
if idattrib in node.attrib:
idmap.append(node.attrib[idattrib])
else:
print >>sys.stderr,"WARNING: Attribute " + idattrib + " not found on node!"
idmap.append(None)
else:
idmap.append(None)
data.append(node.text)
if foliafile:
foliadoc = folia.Document(file=foliafile)
if not foliadoc.declared(folia.AnnotationType.TOKEN):
foliadoc.declare(folia.AnnotationType.TOKEN, set='http://ilk.uvt.nl/folia/sets/ucto-nl.foliaset', annotator='Frog',annotatortype=folia.AnnotatorType.AUTO)
if not foliadoc.declared(folia.AnnotationType.POS):
foliadoc.declare(folia.AnnotationType.POS, set='http://ilk.uvt.nl/folia/sets/cgn-legacy.foliaset', annotator='Frog',annotatortype=folia.AnnotatorType.AUTO)
if not foliadoc.declared(folia.AnnotationType.LEMMA):
foliadoc.declare(folia.AnnotationType.LEMMA, set='http://ilk.uvt.nl/folia/sets/mblem-nl.foliaset', annotator='Frog',annotatortype=folia.AnnotatorType.AUTO)
foliadoc.language('nld')
text = foliadoc.data[-1]
for p in foliadoc.paragraphs():
found_s = False
for s in p.sentences():
found_w = False
for w in s.words():
found_w = True
found_s = True
if found_w:
#pass tokenised sentence
words = s.words()
response = frogclient.process(" ".join([unicode(w) for w in words]))
for i, (word, lemma, morph, pos) in enumerate(response):
if legacy: legacyout(i,word,lemma,morph,pos)
if unicode(words[i]) == word:
if lemma:
words[i].append( folia.LemmaAnnotation(foliadoc, cls=lemma) )
if pos:
words[i].append( folia.PosAnnotation(foliadoc, cls=pos) )
else:
print >>sys.stderr,"WARNING: Out of sync after calling Frog! ", i, word
else:
#pass untokenised sentence
try:
sentext = s.text()
except folia.NoSuchText:
continue
response = frogclient.process(sentext)
for i, (word, lemma, morph, pos) in enumerate(response):
if legacy: legacyout(i,word,lemma,morph,pos)
if word:
w = folia.Word(foliadoc, text=word, generate_id_in=s)
if lemma:
w.append( folia.LemmaAnnotation(foliadoc, cls=lemma) )
if pos:
w.append( folia.PosAnnotation(foliadoc, cls=pos) )
s.append(w)
if not found_s:
#pass paragraph
try:
partext = p.text()
except folia.NoSuchText:
continue
s = folia.Sentence(foliadoc, generate_id_in=p)
response = frogclient.process(partext)
for i, (word, lemma, morph, pos) in enumerate(response):
if (not word or i == len(response) - 1) and len(s) > 0:
#gap or end of response: terminate sentence
p.append(s)
s = folia.Sentence(foliadoc, generate_id_in=p)
elif word:
w = folia.Word(foliadoc, text=word, generate_id_in=s)
if lemma:
w.append( folia.LemmaAnnotation(foliadoc, cls=lemma) )
if pos:
w.append( folia.PosAnnotation(foliadoc, cls=pos) )
s.append(w)
else:
foliadoc = folia.Document(id=foliaid)
foliadoc.declare(folia.AnnotationType.TOKEN, set='http://ilk.uvt.nl/folia/sets/ucto-nl.foliaset', annotator='Frog',annotatortype=folia.AnnotatorType.AUTO)
foliadoc.declare(folia.AnnotationType.POS, set='http://ilk.uvt.nl/folia/sets/cgn-legacy.foliaset', annotator='Frog',annotatortype=folia.AnnotatorType.AUTO)
foliadoc.declare(folia.AnnotationType.LEMMA, set='http://ilk.uvt.nl/folia/sets/mblem-nl.foliaset', annotator='Frog',annotatortype=folia.AnnotatorType.AUTO)
foliadoc.language('nld')
text = folia.Text(foliadoc, id=foliadoc.id + '.text.1')
foliadoc.append(text)
curid = None
for (fragment, id) in zip(data,idmap):
if mode == 's' or mode == 'n':
if id:
s = folia.Sentence(foliadoc, id=id)
else:
s = folia.Sentence(foliadoc, generate_id_in=text)
elif mode == 'p':
if id:
p = folia.Paragraph(foliadoc, id=id)
else:
p = folia.Paragraph(foliadoc, generate_id_in=text)
s = folia.Sentence(foliadoc, generate_id_in=p)
curid = s.id
response = frogclient.process(fragment)
for i, (word, lemma, morph, pos) in enumerate(response):
if legacy:
legacyout(i,word,lemma,morph,pos)
continue
if word:
w = folia.Word(foliadoc, text=word, generate_id_in=s)
if lemma:
w.append( folia.LemmaAnnotation(foliadoc, cls=lemma) )
if pos:
w.append( folia.PosAnnotation(foliadoc, cls=pos) )
s.append(w)
if (not word or i == len(response) - 1) and len(s) > 0:
#gap or end of response: terminate sentence
if mode == 'p':
p.append(s)
if (i == len(response) - 1):
text.append(p)
elif mode == 'n' or (mode == 's' and i == len(response) - 1):
text.append(s)
elif mode == 's':
continue
if i < len(response) - 1: #not done yet?
#create new sentence
if mode == 'p':
s = folia.Sentence(foliadoc, generate_id_in=p)
elif mode == 'n' and id:
#no id for this unforeseen sentence, make something up
s = folia.Sentence(foliadoc, id=curid+'.X')
print("WARNING: Sentence found that was not in original",file=sys.stderr)
if not legacy:
print(foliadoc.xmlstring())
if save and foliafile:
foliadoc.save()
|
