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
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453 | class TestErrorMarkup(unittest.TestCase):
"""Test errormarkup functions."""
@staticmethod
def assert_xml_equal(got, want):
"""Check if two stringified xml snippets are equal"""
got = etree.tostring(got, encoding="unicode")
want = etree.tostring(want, encoding="unicode")
checker = doctestcompare.LXMLOutputChecker()
if not checker.check_output(want, got, 0):
message = checker.output_difference(doctest.Example("", want), got, 0)
raise AssertionError(message)
@parameterized.expand(
[
("{a}${b}", (3, 7)),
("{a}${b} {a}${c} ", (11, 15)),
("{a}${b} {{a}${b}}${d} ", (17, 21)),
]
)
def test_regex(self, markup, span):
"""Testing the most important regex."""
# match = errormarkup.LAST_CORRECTION_REGEX.search(markup)
# self.assertEqual(match.span(), span)
match = errormarkup.LAST_CORRECTION_REGEX.search(markup)
self.assertEqual(match.span(), span)
@parameterized.expand(
[
(
"errorlang_infinity",
"<p>{molekylærbiologimi}∞{kal,bio}</p>",
"<p><errorlang>molekylærbiologimi<correct>kal,bio</correct></errorlang></p>",
),
(
"error_parser_errorlang_infinity_with_new_lines",
"<p>\n\n\n\n{molekylærbiologimi}∞{kal,bio}\n\n\n\n</p>",
"<p><errorlang>molekylærbiologimi<correct>kal,bio</correct></errorlang></p>",
),
(
"quote_char",
"<p>{”sjievnnijis”}${conc,vnn-vnnj|sjievnnjis}</p>",
'<p><errorort>”sjievnnijis”<correct errorinfo="conc,vnn-vnnj">sjievnnjis</correct></errorort></p>',
),
(
"error_parser_errorort1",
"<p>{jne.}${adv,typo|jna.}</p>",
'<p><errorort>jne.<correct errorinfo="adv,typo">jna.</correct></errorort></p>',
),
(
"error_parser_error_morphsyn1",
"<p>{Nieiddat leat nuorra}£{a,spred,nompl,nomsg,agr|Nieiddat leat nuorat}</p>",
"<p><errormorphsyn>Nieiddat leat nuorra<correct "
'errorinfo="a,spred,nompl,nomsg,agr">Nieiddat leat nuorat'
"</correct></errormorphsyn></p>",
),
(
"error_parser_with_two_simple_errors",
"<p>gitta {Nordkjosbotn'ii}${Nordkjosbotnii} (mii lea ge "
"{nordkjosbotn}${Nordkjosbotn} sámegillii? Muhtin, veahket mu!) "
"gos</p>",
"<p>gitta "
"<errorort>Nordkjosbotn'ii<correct>Nordkjosbotnii</correct></errorort> "
"(mii lea ge "
"<errorort>nordkjosbotn<correct>Nordkjosbotn</correct></errorort> "
"sámegillii? Muhtin, veahket mu!) gos</p>",
),
(
"only_text_in_element",
"<p>Muittán doložiid</p>",
"<p>Muittán doložiid</p>",
),
(
"paragraph_character",
"<p>Vuodoláhkaj §110a</p>",
"<p>Vuodoláhkaj §110a</p>",
),
(
"errorort1",
"<p>{jne.}${adv,typo|jna.}</p>",
'<p><errorort>jne.<correct errorinfo="adv,typo">jna.</correct></errorort></p>',
),
(
"errorort2",
"<p>{daesn'}${daesnie}</p>",
"<p><errorort>daesn'<correct>daesnie</correct></errorort></p>",
),
(
"input_contains_slash",
"<p>{magistter/}${loan,vowlat,e-a|magisttar}</p>",
'<p><errorort>magistter/<correct errorinfo="loan,vowlat,e-a">'
"magisttar</correct></errorort></p>",
),
(
"error_correct1",
"<p>{1]}§{Ij}</p>",
"<p><error>1]<correct>Ij</correct></error></p>",
),
(
"error_correct2",
"<p>{væ]keles}§{væjkeles}</p>",
"<p><error>væ]keles<correct>væjkeles</correct></error></p>",
),
(
"error_correct3",
"<p>{smávi-}§{smávit-}</p>",
"<p><error>smávi-<correct>smávit-</correct></error></p>",
),
(
"error_correct4",
"<p>{CD:t}§{CD:at}</p>",
"<p><error>CD:t<correct>CD:at</correct></error></p>",
),
(
"error_correct5",
"<p>{DNB-feaskáris}§{DnB-feaskáris}</p>",
"<p><error>DNB-feaskáris<correct>DnB-feaskáris</correct></error></p>",
),
(
"error_correct6",
"<p>{boade}§{boađe}</p>",
"<p><error>boade<correct>boađe</correct></error></p>",
),
(
"error_correct7",
"<p>{2005’as}§{2005:s}</p>",
"<p><error>2005’as<correct>2005:s</correct></error></p>",
),
(
"error_correct8",
"<p>{NSRii}§{NSR:i}</p>",
"<p><error>NSRii<correct>NSR:i</correct></error></p>",
),
(
"error_correct9",
"<p>{Nordkjosbotn'ii}§{Nordkjosbotnii}</p>",
"<p><error>Nordkjosbotn'ii<correct>Nordkjosbotnii</correct>"
"</error></p>",
),
(
"errorort3",
"<p>{nourra}${a,meta|nuorra}</p>",
'<p><errorort>nourra<correct errorinfo="a,meta">nuorra</correct></errorort></p>',
),
(
"error_morphsyn1",
"<p>{Nieiddat leat nuorra}£{a,spred,nompl,nomsg,agr|Nieiddat leat nuorat}</p>",
"<p><errormorphsyn>Nieiddat leat nuorra"
'<correct errorinfo="a,spred,nompl,nomsg,agr">Nieiddat leat '
"nuorat</correct></errormorphsyn></p>",
),
(
"error_syn1",
"<p>{riŋgen nieidda lusa}¥{x,pph|riŋgen niidii}</p>",
'<p><errorsyn>riŋgen nieidda lusa<correct errorinfo="x,pph">'
"riŋgen niidii</correct></errorsyn></p>",
),
(
"error_syn2",
"<p>{ovtta}¥{num,redun| }</p>",
'<p><errorsyn>ovtta<correct errorinfo="num,redun"></correct>'
"</errorsyn></p>",
),
(
"error_lex1",
"<p>{dábálaš}€{adv,adj,der|dábálaččat}</p>",
'<p><errorlex>dábálaš<correct errorinfo="adv,adj,der">dábálaččat'
"</correct></errorlex></p>",
),
(
"error_ortreal1",
"<p>{ráhččamušaid}¢{noun,mix|rahčamušaid}</p>",
'<p><errorortreal>ráhččamušaid<correct errorinfo="noun,mix">'
"rahčamušaid</correct></errorortreal></p>",
),
(
"error_ortreal2",
"<p>gitta {Nordkjosbotn'ii}${Nordkjosbotnii} (mii lea ge "
"{nordkjosbotn}${Nordkjosbotn} sámegillii? Muhtin, veahket mu!) "
"gos</p>",
"<p>gitta <errorort>Nordkjosbotn'ii<correct>Nordkjosbotnii"
"</correct></errorort> (mii lea ge <errorort>nordkjosbotn"
"<correct>Nordkjosbotn</correct></errorort> sámegillii? "
"Muhtin, veahket mu!) gos</p>",
),
(
"error_morphsyn2",
"<p>Čáppa muohtaskulptuvrraid ráhkadeapmi VSM olggobealde lei "
"maiddái ovttasbargu gaskal {skuvla ohppiid}£"
"{noun,attr,gensg,nomsg,case|skuvlla ohppiid}"
"ja VSM.</p>",
"<p>Čáppa muohtaskulptuvrraid ráhkadeapmi VSM olggobealde lei "
"maiddái ovttasbargu gaskal <errormorphsyn>skuvla ohppiid"
'<correct errorinfo="noun,attr,gensg,nomsg,case">skuvlla ohppiid'
"</correct></errormorphsyn> ja VSM.</p>",
),
(
"errorort4",
"<p>- ruksesruonáčalmmehisvuohta lea sullii "
"{8%:as}${acr,suf|8%:s}</p>",
"<p>- ruksesruonáčalmmehisvuohta lea sullii <errorort>8%:as"
'<correct errorinfo="acr,suf">8%:s</correct></errorort></p>',
),
(
"error_ortreal3",
"<p>( {nissonin}¢{noun,suf|nissoniin} dušše {0.6 %:s}£{0.6 %} )</p>",
'<p>( <errorortreal>nissonin<correct errorinfo="noun,suf">'
"nissoniin</correct></errorortreal> dušše <errormorphsyn>0.6 %:s"
"<correct>0.6 %</correct></errormorphsyn> )</p>",
),
(
"errorort5",
"<p>(haploida) ja {njiŋŋalas}${noun,á|njiŋŋálas} {ságahuvvon}"
"${verb,a|sagahuvvon} manneseallas (diploida)</p>",
'<p>(haploida) ja <errorort>njiŋŋalas<correct errorinfo="noun,á">'
"njiŋŋálas</correct></errorort> <errorort>ságahuvvon"
'<correct errorinfo="verb,a">sagahuvvon</correct></errorort> '
"manneseallas (diploida)</p>",
),
(
"errorort6",
"<p>(gii oahpaha) {giinu}${x,notcmp|gii nu} manai {intiánalávlagat}"
"${loan,conc|indiánalávlagat} {guovža-klána}${noun,cmp|guovžaklána} "
"olbmuid</p>",
'<p>(gii oahpaha) <errorort>giinu<correct errorinfo="x,notcmp">'
"gii nu</correct></errorort> manai <errorort>intiánalávlagat"
'<correct errorinfo="loan,conc">indiánalávlagat</correct>'
'</errorort> <errorort>guovža-klána<correct errorinfo="noun,cmp">'
"guovžaklána</correct></errorort> olbmuid</p>",
),
(
"error_format",
"<p>{{A B}‰{notspace|A B} C}‰{notspace|A B C}</p>",
"<p>"
"<errorformat>"
"<errorformat>"
'A B<correct errorinfo="notspace">A B</correct></errorformat>'
' C<correct errorinfo="notspace">A B C</correct></errorformat>'
"</p>",
),
(
"preserve_space_at_end_of_sentence",
"<p>buvttadeaddji Anstein {Mikkelsens}${typo|Mikkelsen} lea "
"ráhkadan. </p>",
'<p>buvttadeaddji Anstein <errorort>Mikkelsens<correct errorinfo="typo">Mikkelsen</correct></errorort> lea ráhkadan. </p>',
),
(
"place_error_elements_before_old_element1",
"<p>buvttadeaddji Anstein {Mikkelsens}${typo|Mikkelsen} lea "
"ráhkadan. {bálkkášumi}${vowlat,á-a|bálkkašumi} miessemánu. <span "
'type="quote" xml:lang="eng">«Best Shorts Competition»</span></p>',
"<p>buvttadeaddji Anstein <errorort>Mikkelsens"
'<correct errorinfo="typo">Mikkelsen</correct></errorort> lea ráhkadan. '
'<errorort>bálkkášumi<correct errorinfo="vowlat,á-a">bálkkašumi</correct></errorort>'
' miessemánu. <span type="quote" xml:lang="eng">«Best Shorts '
"Competition»</span></p>",
),
(
"place_error_elements_before_old_element2",
"<p>{Mikkelsens}${typo|Mikkelsen} lea ráhkadan. "
"{bálkkášumi}${vowlat,á-a|bálkkašumi} miessemánu. "
'<span type="quote" xml:lang="eng">'
"«Best Shorts Competition»</span></p>",
'<p><errorort>Mikkelsens<correct errorinfo="typo">Mikkelsen</correct></errorort> lea ráhkadan. <errorort>bálkkášumi<correct errorinfo="vowlat,á-a">bálkkašumi</correct></errorort> miessemánu. <span type="quote" xml:lang="eng">«Best Shorts Competition»</span></p>',
),
(
"place_error_element_after_old_element",
'<p>I 1864 ga han ut boka <span type="quote" xml:lang="swe">'
'"Fornuftigt Madstel"</span>. {Asbjørsen}${prop,typo|Asbjørnsen} '
"døde 5. januar 1885, nesten 73 år gammel.</p>",
'<p>I 1864 ga han ut boka <span type="quote" xml:lang="swe">'
'"Fornuftigt Madstel"</span>. <errorort>Asbjørsen'
'<correct errorinfo="prop,typo">Asbjørnsen</correct></errorort> '
"døde 5. januar 1885, nesten 73 år gammel.</p>",
),
(
"place_error_element_before_and_after_old_element",
"<p>buvttadeaddji Anstein {Mikkelsens}${typo|Mikkelsen} lea "
'ráhkadan. <span type="quote" xml:lang="eng">«Best Shorts '
"Competition»</span> {bálkkášumi}${vowlat,á-a|bálkkašumi} "
"miessemánu.</p>",
"<p>buvttadeaddji Anstein <errorort>Mikkelsens"
'<correct errorinfo="typo">Mikkelsen</correct></errorort> lea '
'ráhkadan. <span type="quote" xml:lang="eng">«Best Shorts '
"Competition»</span> <errorort>bálkkášumi"
'<correct errorinfo="vowlat,á-a">bálkkašumi</correct></errorort> '
"miessemánu.</p>",
),
(
"markup3Levels",
"<p>buvttadeaddji Anstein <errorort>Mikkelsens<correct "
'errorinfo="typo">Mikkelsen</correct></errorort> lea ráhkadan. '
'<span type="quote" xml:lang="eng">«Best Shorts Competition»'
'</span> <errorort>bálkkášumi<correct errorinfo="vowlat,á-a">'
"bálkkašumi</correct></errorort> miessemánu. <em>buvttadeaddji "
"Anstein {Mikkelsens}${typo|Mikkelsen} lea ráhkadan. <span "
'type="quote" xml:lang="eng">«Best Shorts Competition»</span> '
"{bálkkášumi}${vowlat,á-a|bálkkašumi} miessemánu.</em></p>",
"<p>buvttadeaddji Anstein <errorort>Mikkelsens<correct "
'errorinfo="typo">Mikkelsen</correct></errorort> lea ráhkadan. '
'<span type="quote" xml:lang="eng">«Best Shorts Competition»'
'</span> <errorort>bálkkášumi<correct errorinfo="vowlat,á-a">'
"bálkkašumi</correct></errorort> miessemánu. <em>buvttadeaddji "
'Anstein <errorort>Mikkelsens<correct errorinfo="typo">Mikkelsen'
'</correct></errorort> lea ráhkadan. <span type="quote" '
'xml:lang="eng">«Best Shorts Competition»</span> <errorort>'
'bálkkášumi<correct errorinfo="vowlat,á-a">bálkkašumi</correct>'
"</errorort> miessemánu.</em></p>",
),
(
"inline multiple corrections",
"<p>{leimme}£{leimmet///leat}</p>",
"<p><errormorphsyn>leimme"
"<correct>leimmet</correct>"
"<correct>leat</correct>"
"</errormorphsyn></p>",
),
]
)
def test_convert_to_errormarkupxml(self, name, in_string, want_string):
"""Test plain errormarkup."""
in_elem = etree.fromstring(in_string)
want = etree.fromstring(want_string)
errormarkup.convert_to_errormarkupxml(in_elem)
self.assert_xml_equal(in_elem, want)
# Nested markup
@parameterized.expand(
[
(
"<p>{{šaddai}${verb,conc|šattai} ollu áššit}£{verb,fin,pl3prs,sg3prs,tense|šadde ollu áššit}</p>",
'<p><errormorphsyn><errorort>šaddai<correct errorinfo="verb,conc">'
"šattai</correct></errorort> ollu áššit<correct "
'errorinfo="verb,fin,pl3prs,sg3prs,tense">šadde ollu áššit'
"</correct></errormorphsyn></p>",
),
(
"<p>{guokte {ganddat}§{n,á|gánddat}}£{n,nump,gensg,nompl,case|guokte gándda}</p>",
'<p><errormorphsyn>guokte <error>ganddat<correct errorinfo="n,á">'
"gánddat</correct></error><correct "
'errorinfo="n,nump,gensg,nompl,case">guokte gándda</correct>'
"</errormorphsyn></p>",
),
(
"<p>{Nieiddat leat {nourra}${adj,meta|nuorra}}"
"£{adj,spred,nompl,nomsg,agr|Nieiddat leat nuorat}</p>",
"<p><errormorphsyn>Nieiddat leat <errorort>nourra<correct "
'errorinfo="adj,meta">nuorra</correct></errorort>'
'<correct errorinfo="adj,spred,nompl,nomsg,agr">Nieiddat leat '
"nuorat</correct></errormorphsyn></p>",
),
(
"<p>{leat {okta máná}£{n,spred,nomsg,gensg,case|okta mánná}}"
"£{v,v,sg3prs,pl3prs,agr|lea okta mánná}</p>",
"<p><errormorphsyn>leat <errormorphsyn>okta máná<correct "
'errorinfo="n,spred,nomsg,gensg,case">okta mánná</correct>'
'</errormorphsyn><correct errorinfo="v,v,sg3prs,pl3prs,agr">'
"lea okta mánná</correct></errormorphsyn></p>",
),
(
"<p>heaitit {dáhkaluddame}${verb,a|dahkaluddame} ahte sis "
"{máhkaš}¢{adv,á|mahkáš} livččii {makkarge}${adv,á|makkárge} "
"politihkka, muhto rahpasit baicca muitalivčče {{makkar}"
"${interr,á|makkár} soga}€{man soga} sii {ovddasttit}"
"${verb,conc|ovddastit}.</p>",
'<p>heaitit <errorort>dáhkaluddame<correct errorinfo="verb,a">'
"dahkaluddame</correct></errorort> ahte sis <errorortreal>"
'máhkaš<correct errorinfo="adv,á">mahkáš</correct></errorortreal>'
'livččii <errorort>makkarge<correct errorinfo="adv,á">makkárge'
"</correct></errorort> politihkka, muhto rahpasit baicca "
"muitalivčče <errorlex><errorort>makkar<correct "
'errorinfo="interr,á">makkár</correct></errorort> soga<correct>'
"man soga</correct></errorlex> sii <errorort>ovddasttit"
'<correct errorinfo="verb,conc">ovddastit</correct></errorort>.</p>',
),
(
"<p>{{Bearpmahat}${noun,svow|Bearpmehat} "
"{earuha}€{verb,v,w|sirre}}£{verb,fin,pl3prs,sg3prs,agr|Bearpmehat"
" sirrejit} uskki ja loaiddu.</p>",
"<p><errormorphsyn><errorort>Bearpmahat<correct "
'errorinfo="noun,svow">Bearpmehat</correct></errorort><errorlex>'
'earuha<correct errorinfo="verb,v,w">sirre</correct></errorlex>'
'<correct errorinfo="verb,fin,pl3prs,sg3prs,agr">Bearpmehat '
"sirrejit</correct></errormorphsyn> uskki ja loaiddu.</p>",
),
(
"<p>Mirja ja Line leaba {{gulahallan olbmožat}"
"¢{noun,cmp|gulahallanolbmožat}}€{gulahallanolbmot}</p>",
"<p>Mirja ja Line leaba <errorlex><errorortreal>gulahallan "
'olbmožat<correct errorinfo="noun,cmp">gulahallanolbmožat'
"</correct></errorortreal><correct>gulahallanolbmot</correct>"
"</errorlex></p>",
),
(
"<p>{Ovddit geasis}£{noun,advl,gensg,locsg,case|Ovddit geasi} "
"{{{čoaggen}${verb,mono|čoggen} ollu jokŋat}"
"£{noun,obj,genpl,nompl,case|čoggen ollu joŋaid} ja sarridat}"
"£{noun,obj,genpl,nompl,case|čoggen ollu joŋaid ja sarridiid}</p>",
"<p><errormorphsyn>Ovddit geasis<correct "
'errorinfo="noun,advl,gensg,locsg,case">Ovddit geasi</correct>'
"</errormorphsyn><errormorphsyn><errormorphsyn><errorort>čoaggen"
'<correct errorinfo="verb,mono">čoggen</correct></errorort> '
'ollu jokŋat<correct errorinfo="noun,obj,genpl,nompl,case">'
"čoggen ollu joŋaid</correct></errormorphsyn> ja sarridat"
'<correct errorinfo="noun,obj,genpl,nompl,case">čoggen ollu '
"joŋaid ja sarridiid</correct></errormorphsyn></p>",
),
(
"<p>Bruk {{epoxi}${noun,cons|epoksy} lim}¢{noun,mix|epoksylim} "
"med god kvalitet.</p>",
"<p>Bruk <errorortreal><errorort>epoxi"
'<correct errorinfo="noun,cons">epoksy</correct></errorort> lim'
'<correct errorinfo="noun,mix">epoksylim</correct></errorortreal> '
"med god kvalitet.</p>",
),
]
)
def test_nested_markup(self, in_string, want_string):
in_elem = etree.fromstring(in_string)
want = etree.fromstring(want_string)
errormarkup.convert_to_errormarkupxml(in_elem)
self.assert_xml_equal(in_elem, want)
|