nltk 3.2__zip → 3.3__zip

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (406) hide show
  1. {nltk-3.2 → nltk-3.3}/INSTALL.txt +6 -6
  2. {nltk-3.2 → nltk-3.3}/LICENSE.txt +13 -13
  3. {nltk-3.2 → nltk-3.3}/MANIFEST.in +6 -6
  4. {nltk-3.2/nltk.egg-info → nltk-3.3}/PKG-INFO +13 -5
  5. nltk-3.2/README.txt → nltk-3.3/README.md +49 -48
  6. nltk-3.3/nltk/VERSION +1 -0
  7. {nltk-3.2 → nltk-3.3}/nltk/__init__.py +185 -185
  8. {nltk-3.2 → nltk-3.3}/nltk/app/__init__.py +52 -53
  9. {nltk-3.2 → nltk-3.3}/nltk/app/chartparser_app.py +2276 -2276
  10. {nltk-3.2 → nltk-3.3}/nltk/app/chunkparser_app.py +1262 -1264
  11. {nltk-3.2 → nltk-3.3}/nltk/app/collocations_app.py +347 -350
  12. {nltk-3.2 → nltk-3.3}/nltk/app/concordance_app.py +568 -570
  13. {nltk-3.2 → nltk-3.3}/nltk/app/nemo_app.py +156 -155
  14. {nltk-3.2 → nltk-3.3}/nltk/app/rdparser_app.py +892 -894
  15. {nltk-3.2 → nltk-3.3}/nltk/app/srparser_app.py +809 -809
  16. {nltk-3.2 → nltk-3.3}/nltk/app/wordfreq_app.py +32 -32
  17. {nltk-3.2 → nltk-3.3}/nltk/app/wordnet_app.py +968 -968
  18. {nltk-3.2 → nltk-3.3}/nltk/book.py +91 -90
  19. {nltk-3.2 → nltk-3.3}/nltk/ccg/__init__.py +22 -22
  20. {nltk-3.2 → nltk-3.3}/nltk/ccg/api.py +359 -335
  21. {nltk-3.2 → nltk-3.3}/nltk/ccg/chart.py +394 -361
  22. {nltk-3.2 → nltk-3.3}/nltk/ccg/combinator.py +343 -312
  23. {nltk-3.2 → nltk-3.3}/nltk/ccg/lexicon.py +328 -290
  24. nltk-3.3/nltk/ccg/logic.py +46 -0
  25. {nltk-3.2 → nltk-3.3}/nltk/chat/__init__.py +49 -49
  26. {nltk-3.2 → nltk-3.3}/nltk/chat/eliza.py +244 -244
  27. {nltk-3.2 → nltk-3.3}/nltk/chat/iesha.py +140 -140
  28. {nltk-3.2 → nltk-3.3}/nltk/chat/rude.py +92 -92
  29. {nltk-3.2 → nltk-3.3}/nltk/chat/suntsu.py +117 -117
  30. {nltk-3.2 → nltk-3.3}/nltk/chat/util.py +122 -120
  31. {nltk-3.2 → nltk-3.3}/nltk/chat/zen.py +282 -282
  32. {nltk-3.2 → nltk-3.3}/nltk/chunk/__init__.py +190 -190
  33. {nltk-3.2 → nltk-3.3}/nltk/chunk/api.py +51 -51
  34. {nltk-3.2 → nltk-3.3}/nltk/chunk/named_entity.py +331 -331
  35. {nltk-3.2 → nltk-3.3}/nltk/chunk/regexp.py +1390 -1384
  36. {nltk-3.2 → nltk-3.3}/nltk/chunk/util.py +600 -600
  37. {nltk-3.2 → nltk-3.3}/nltk/classify/__init__.py +98 -98
  38. {nltk-3.2 → nltk-3.3}/nltk/classify/api.py +193 -193
  39. {nltk-3.2 → nltk-3.3}/nltk/classify/decisiontree.py +295 -295
  40. {nltk-3.2 → nltk-3.3}/nltk/classify/maxent.py +1510 -1492
  41. {nltk-3.2 → nltk-3.3}/nltk/classify/megam.py +179 -178
  42. {nltk-3.2 → nltk-3.3}/nltk/classify/naivebayes.py +245 -242
  43. {nltk-3.2 → nltk-3.3}/nltk/classify/positivenaivebayes.py +170 -170
  44. {nltk-3.2 → nltk-3.3}/nltk/classify/rte_classify.py +160 -191
  45. {nltk-3.2 → nltk-3.3}/nltk/classify/scikitlearn.py +153 -151
  46. {nltk-3.2 → nltk-3.3}/nltk/classify/senna.py +183 -181
  47. {nltk-3.2 → nltk-3.3}/nltk/classify/svm.py +15 -15
  48. {nltk-3.2 → nltk-3.3}/nltk/classify/tadm.py +113 -112
  49. {nltk-3.2 → nltk-3.3}/nltk/classify/textcat.py +193 -193
  50. {nltk-3.2 → nltk-3.3}/nltk/classify/util.py +324 -312
  51. {nltk-3.2 → nltk-3.3}/nltk/classify/weka.py +346 -345
  52. {nltk-3.2 → nltk-3.3}/nltk/cluster/__init__.py +86 -86
  53. {nltk-3.2 → nltk-3.3}/nltk/cluster/api.py +74 -70
  54. {nltk-3.2 → nltk-3.3}/nltk/cluster/em.py +247 -247
  55. {nltk-3.2 → nltk-3.3}/nltk/cluster/gaac.py +168 -168
  56. {nltk-3.2 → nltk-3.3}/nltk/cluster/kmeans.py +221 -221
  57. {nltk-3.2 → nltk-3.3}/nltk/cluster/util.py +304 -289
  58. nltk-3.3/nltk/collections.py +640 -0
  59. {nltk-3.2 → nltk-3.3}/nltk/collocations.py +388 -388
  60. nltk-3.3/nltk/compat.py +365 -0
  61. {nltk-3.2 → nltk-3.3}/nltk/corpus/__init__.py +341 -320
  62. {nltk-3.2 → nltk-3.3}/nltk/corpus/europarl_raw.py +44 -44
  63. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/__init__.py +147 -145
  64. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/aligned.py +115 -114
  65. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/api.py +448 -446
  66. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/bnc.py +252 -252
  67. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/bracket_parse.py +227 -227
  68. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/categorized_sents.py +179 -178
  69. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/chasen.py +140 -139
  70. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/childes.py +522 -494
  71. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/chunked.py +212 -212
  72. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/cmudict.py +97 -95
  73. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/comparative_sents.py +280 -278
  74. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/conll.py +528 -521
  75. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/crubadan.py +116 -116
  76. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/dependency.py +101 -101
  77. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/framenet.py +3059 -2093
  78. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/ieer.py +112 -111
  79. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/indian.py +85 -86
  80. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/ipipan.py +332 -331
  81. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/knbc.py +187 -187
  82. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/lin.py +156 -156
  83. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/mte.py +311 -310
  84. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/nkjp.py +429 -428
  85. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/nombank.py +421 -421
  86. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/nps_chat.py +73 -73
  87. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/opinion_lexicon.py +115 -115
  88. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/panlex_lite.py +165 -165
  89. nltk-3.3/nltk/corpus/reader/pl196x.py +292 -0
  90. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/plaintext.py +232 -233
  91. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/ppattach.py +96 -95
  92. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/propbank.py +479 -481
  93. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/pros_cons.py +128 -126
  94. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/reviews.py +330 -328
  95. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/rte.py +144 -146
  96. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/semcor.py +256 -256
  97. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/senseval.py +203 -201
  98. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/sentiwordnet.py +137 -137
  99. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/sinica_treebank.py +75 -75
  100. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/string_category.py +60 -59
  101. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/switchboard.py +119 -119
  102. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/tagged.py +295 -294
  103. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/timit.py +452 -450
  104. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/toolbox.py +68 -68
  105. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/twitter.py +155 -157
  106. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/udhr.py +80 -80
  107. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/util.py +804 -800
  108. nltk-3.3/nltk/corpus/reader/verbnet.py +595 -0
  109. nltk-3.3/nltk/corpus/reader/wordlist.py +134 -0
  110. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/wordnet.py +2108 -2036
  111. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/xmldocs.py +392 -387
  112. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/ycoe.py +243 -242
  113. {nltk-3.2 → nltk-3.3}/nltk/corpus/util.py +144 -127
  114. {nltk-3.2 → nltk-3.3}/nltk/data.py +1487 -1451
  115. {nltk-3.2 → nltk-3.3}/nltk/decorators.py +219 -215
  116. {nltk-3.2 → nltk-3.3}/nltk/downloader.py +2278 -2274
  117. {nltk-3.2 → nltk-3.3}/nltk/draw/__init__.py +27 -27
  118. {nltk-3.2 → nltk-3.3}/nltk/draw/cfg.py +774 -777
  119. {nltk-3.2 → nltk-3.3}/nltk/draw/dispersion.py +58 -58
  120. {nltk-3.2 → nltk-3.3}/nltk/draw/table.py +1098 -1098
  121. {nltk-3.2 → nltk-3.3}/nltk/draw/tree.py +963 -966
  122. {nltk-3.2 → nltk-3.3}/nltk/draw/util.py +2356 -2356
  123. {nltk-3.2 → nltk-3.3}/nltk/featstruct.py +2617 -2500
  124. {nltk-3.2 → nltk-3.3}/nltk/grammar.py +1559 -1516
  125. {nltk-3.2 → nltk-3.3}/nltk/help.py +56 -56
  126. {nltk-3.2 → nltk-3.3}/nltk/inference/__init__.py +20 -20
  127. {nltk-3.2 → nltk-3.3}/nltk/inference/api.py +597 -589
  128. {nltk-3.2 → nltk-3.3}/nltk/inference/discourse.py +612 -608
  129. {nltk-3.2 → nltk-3.3}/nltk/inference/mace.py +311 -311
  130. {nltk-3.2 → nltk-3.3}/nltk/inference/nonmonotonic.py +509 -509
  131. {nltk-3.2 → nltk-3.3}/nltk/inference/prover9.py +431 -431
  132. {nltk-3.2 → nltk-3.3}/nltk/inference/resolution.py +687 -687
  133. {nltk-3.2 → nltk-3.3}/nltk/inference/tableau.py +607 -607
  134. {nltk-3.2 → nltk-3.3}/nltk/internals.py +982 -982
  135. {nltk-3.2 → nltk-3.3}/nltk/jsontags.py +63 -63
  136. {nltk-3.2 → nltk-3.3}/nltk/lazyimport.py +142 -142
  137. {nltk-3.2 → nltk-3.3}/nltk/metrics/__init__.py +30 -29
  138. {nltk-3.2 → nltk-3.3}/nltk/metrics/agreement.py +421 -401
  139. nltk-3.3/nltk/metrics/aline.py +607 -0
  140. {nltk-3.2 → nltk-3.3}/nltk/metrics/association.py +414 -410
  141. {nltk-3.2 → nltk-3.3}/nltk/metrics/confusionmatrix.py +206 -206
  142. {nltk-3.2 → nltk-3.3}/nltk/metrics/distance.py +207 -202
  143. {nltk-3.2 → nltk-3.3}/nltk/metrics/paice.py +381 -381
  144. {nltk-3.2 → nltk-3.3}/nltk/metrics/scores.py +228 -228
  145. {nltk-3.2 → nltk-3.3}/nltk/metrics/segmentation.py +231 -232
  146. {nltk-3.2 → nltk-3.3}/nltk/metrics/spearman.py +69 -69
  147. {nltk-3.2 → nltk-3.3}/nltk/misc/__init__.py +11 -11
  148. {nltk-3.2 → nltk-3.3}/nltk/misc/babelfish.py +10 -10
  149. {nltk-3.2 → nltk-3.3}/nltk/misc/chomsky.py +135 -133
  150. {nltk-3.2 → nltk-3.3}/nltk/misc/minimalset.py +83 -83
  151. {nltk-3.2 → nltk-3.3}/nltk/misc/sort.py +157 -157
  152. {nltk-3.2 → nltk-3.3}/nltk/misc/wordfinder.py +129 -129
  153. {nltk-3.2 → nltk-3.3}/nltk/parse/__init__.py +81 -80
  154. {nltk-3.2 → nltk-3.3}/nltk/parse/api.py +66 -66
  155. {nltk-3.2 → nltk-3.3}/nltk/parse/bllip.py +282 -282
  156. {nltk-3.2 → nltk-3.3}/nltk/parse/chart.py +1682 -1681
  157. nltk-3.3/nltk/parse/corenlp.py +780 -0
  158. {nltk-3.2 → nltk-3.3}/nltk/parse/dependencygraph.py +760 -758
  159. {nltk-3.2 → nltk-3.3}/nltk/parse/earleychart.py +452 -451
  160. {nltk-3.2 → nltk-3.3}/nltk/parse/evaluate.py +129 -129
  161. {nltk-3.2 → nltk-3.3}/nltk/parse/featurechart.py +580 -578
  162. {nltk-3.2 → nltk-3.3}/nltk/parse/generate.py +87 -76
  163. {nltk-3.2 → nltk-3.3}/nltk/parse/malt.py +356 -356
  164. {nltk-3.2 → nltk-3.3}/nltk/parse/nonprojectivedependencyparser.py +783 -783
  165. {nltk-3.2 → nltk-3.3}/nltk/parse/pchart.py +522 -522
  166. {nltk-3.2 → nltk-3.3}/nltk/parse/projectivedependencyparser.py +579 -578
  167. {nltk-3.2 → nltk-3.3}/nltk/parse/recursivedescent.py +655 -656
  168. {nltk-3.2 → nltk-3.3}/nltk/parse/shiftreduce.py +458 -459
  169. {nltk-3.2 → nltk-3.3}/nltk/parse/stanford.py +431 -406
  170. {nltk-3.2 → nltk-3.3}/nltk/parse/transitionparser.py +774 -772
  171. {nltk-3.2 → nltk-3.3}/nltk/parse/util.py +231 -231
  172. {nltk-3.2 → nltk-3.3}/nltk/parse/viterbi.py +401 -401
  173. {nltk-3.2 → nltk-3.3}/nltk/probability.py +2410 -2355
  174. {nltk-3.2 → nltk-3.3}/nltk/sem/__init__.py +61 -61
  175. {nltk-3.2 → nltk-3.3}/nltk/sem/boxer.py +1261 -1261
  176. {nltk-3.2 → nltk-3.3}/nltk/sem/chat80.py +782 -782
  177. {nltk-3.2 → nltk-3.3}/nltk/sem/cooper_storage.py +118 -118
  178. {nltk-3.2 → nltk-3.3}/nltk/sem/drt.py +1258 -1257
  179. {nltk-3.2 → nltk-3.3}/nltk/sem/drt_glue_demo.py +485 -489
  180. {nltk-3.2 → nltk-3.3}/nltk/sem/evaluate.py +790 -784
  181. {nltk-3.2 → nltk-3.3}/nltk/sem/glue.py +696 -694
  182. {nltk-3.2 → nltk-3.3}/nltk/sem/hole.py +386 -384
  183. {nltk-3.2 → nltk-3.3}/nltk/sem/lfg.py +210 -208
  184. {nltk-3.2 → nltk-3.3}/nltk/sem/linearlogic.py +451 -449
  185. {nltk-3.2 → nltk-3.3}/nltk/sem/logic.py +1910 -1910
  186. {nltk-3.2 → nltk-3.3}/nltk/sem/relextract.py +476 -483
  187. {nltk-3.2 → nltk-3.3}/nltk/sem/skolemize.py +101 -101
  188. {nltk-3.2 → nltk-3.3}/nltk/sem/util.py +249 -249
  189. {nltk-3.2 → nltk-3.3}/nltk/sentiment/__init__.py +14 -14
  190. {nltk-3.2 → nltk-3.3}/nltk/sentiment/sentiment_analyzer.py +229 -229
  191. {nltk-3.2 → nltk-3.3}/nltk/sentiment/util.py +762 -752
  192. {nltk-3.2 → nltk-3.3}/nltk/sentiment/vader.py +443 -445
  193. {nltk-3.2 → nltk-3.3}/nltk/stem/__init__.py +31 -33
  194. {nltk-3.2 → nltk-3.3}/nltk/stem/api.py +28 -25
  195. nltk-3.3/nltk/stem/arlstem.py +369 -0
  196. {nltk-3.2 → nltk-3.3}/nltk/stem/isri.py +345 -345
  197. {nltk-3.2 → nltk-3.3}/nltk/stem/lancaster.py +337 -310
  198. nltk-3.3/nltk/stem/porter.py +710 -0
  199. {nltk-3.2 → nltk-3.3}/nltk/stem/regexp.py +61 -61
  200. {nltk-3.2 → nltk-3.3}/nltk/stem/rslp.py +142 -142
  201. {nltk-3.2 → nltk-3.3}/nltk/stem/snowball.py +4236 -3712
  202. {nltk-3.2 → nltk-3.3}/nltk/stem/util.py +22 -12
  203. {nltk-3.2 → nltk-3.3}/nltk/stem/wordnet.py +51 -51
  204. {nltk-3.2 → nltk-3.3}/nltk/tag/__init__.py +152 -127
  205. {nltk-3.2 → nltk-3.3}/nltk/tag/api.py +86 -81
  206. {nltk-3.2 → nltk-3.3}/nltk/tag/brill.py +424 -426
  207. {nltk-3.2 → nltk-3.3}/nltk/tag/brill_trainer.py +608 -608
  208. {nltk-3.2 → nltk-3.3}/nltk/tag/crf.py +204 -200
  209. {nltk-3.2 → nltk-3.3}/nltk/tag/hmm.py +1277 -1277
  210. {nltk-3.2 → nltk-3.3}/nltk/tag/hunpos.py +132 -132
  211. {nltk-3.2 → nltk-3.3}/nltk/tag/mapping.py +101 -101
  212. {nltk-3.2 → nltk-3.3}/nltk/tag/perceptron.py +329 -314
  213. {nltk-3.2 → nltk-3.3}/nltk/tag/senna.py +147 -145
  214. {nltk-3.2 → nltk-3.3}/nltk/tag/sequential.py +740 -736
  215. {nltk-3.2 → nltk-3.3}/nltk/tag/stanford.py +219 -196
  216. {nltk-3.2 → nltk-3.3}/nltk/tag/tnt.py +595 -604
  217. {nltk-3.2 → nltk-3.3}/nltk/tag/util.py +72 -72
  218. {nltk-3.2 → nltk-3.3}/nltk/tbl/__init__.py +28 -28
  219. {nltk-3.2 → nltk-3.3}/nltk/tbl/api.py +1 -1
  220. {nltk-3.2 → nltk-3.3}/nltk/tbl/demo.py +366 -366
  221. {nltk-3.2 → nltk-3.3}/nltk/tbl/erroranalysis.py +39 -39
  222. {nltk-3.2 → nltk-3.3}/nltk/tbl/feature.py +259 -261
  223. {nltk-3.2 → nltk-3.3}/nltk/tbl/rule.py +321 -321
  224. {nltk-3.2 → nltk-3.3}/nltk/tbl/template.py +303 -308
  225. {nltk-3.2 → nltk-3.3}/nltk/test/__init__.py +18 -18
  226. {nltk-3.2 → nltk-3.3}/nltk/test/all.py +24 -23
  227. {nltk-3.2 → nltk-3.3}/nltk/test/bleu.doctest +14 -14
  228. {nltk-3.2 → nltk-3.3}/nltk/test/bnc.doctest +60 -60
  229. {nltk-3.2 → nltk-3.3}/nltk/test/ccg.doctest +376 -395
  230. nltk-3.3/nltk/test/ccg_semantics.doctest +553 -0
  231. {nltk-3.2 → nltk-3.3}/nltk/test/chat80.doctest +234 -234
  232. {nltk-3.2 → nltk-3.3}/nltk/test/childes.doctest +184 -184
  233. {nltk-3.2 → nltk-3.3}/nltk/test/childes_fixt.py +14 -13
  234. {nltk-3.2 → nltk-3.3}/nltk/test/chunk.doctest +373 -373
  235. {nltk-3.2 → nltk-3.3}/nltk/test/classify.doctest +185 -185
  236. {nltk-3.2 → nltk-3.3}/nltk/test/classify_fixt.py +11 -10
  237. nltk-3.3/nltk/test/collections.doctest +20 -0
  238. {nltk-3.2 → nltk-3.3}/nltk/test/collocations.doctest +276 -276
  239. {nltk-3.2 → nltk-3.3}/nltk/test/compat.doctest +134 -138
  240. {nltk-3.2 → nltk-3.3}/nltk/test/compat_fixt.py +9 -8
  241. nltk-3.3/nltk/test/concordance.doctest +55 -0
  242. {nltk-3.2 → nltk-3.3}/nltk/test/corpus.doctest +2200 -2212
  243. {nltk-3.2 → nltk-3.3}/nltk/test/corpus_fixt.py +3 -3
  244. {nltk-3.2 → nltk-3.3}/nltk/test/crubadan.doctest +65 -65
  245. {nltk-3.2 → nltk-3.3}/nltk/test/data.doctest +379 -379
  246. {nltk-3.2 → nltk-3.3}/nltk/test/dependency.doctest +241 -241
  247. {nltk-3.2 → nltk-3.3}/nltk/test/discourse.doctest +546 -546
  248. {nltk-3.2 → nltk-3.3}/nltk/test/discourse_fixt.py +14 -13
  249. {nltk-3.2 → nltk-3.3}/nltk/test/doctest_nose_plugin.py +159 -156
  250. {nltk-3.2 → nltk-3.3}/nltk/test/drt.doctest +517 -517
  251. {nltk-3.2 → nltk-3.3}/nltk/test/featgram.doctest +607 -607
  252. {nltk-3.2 → nltk-3.3}/nltk/test/featstruct.doctest +1229 -1230
  253. {nltk-3.2 → nltk-3.3}/nltk/test/framenet.doctest +288 -239
  254. {nltk-3.2 → nltk-3.3}/nltk/test/generate.doctest +67 -67
  255. {nltk-3.2 → nltk-3.3}/nltk/test/gensim.doctest +140 -141
  256. {nltk-3.2 → nltk-3.3}/nltk/test/gensim_fixt.py +10 -9
  257. {nltk-3.2 → nltk-3.3}/nltk/test/gluesemantics.doctest +384 -386
  258. {nltk-3.2 → nltk-3.3}/nltk/test/gluesemantics_malt.doctest +68 -68
  259. {nltk-3.2 → nltk-3.3}/nltk/test/gluesemantics_malt_fixt.py +12 -11
  260. {nltk-3.2 → nltk-3.3}/nltk/test/grammar.doctest +48 -48
  261. {nltk-3.2 → nltk-3.3}/nltk/test/grammartestsuites.doctest +109 -109
  262. {nltk-3.2 → nltk-3.3}/nltk/test/index.doctest +100 -100
  263. {nltk-3.2 → nltk-3.3}/nltk/test/inference.doctest +534 -534
  264. {nltk-3.2 → nltk-3.3}/nltk/test/inference_fixt.py +12 -11
  265. {nltk-3.2 → nltk-3.3}/nltk/test/internals.doctest +140 -140
  266. {nltk-3.2 → nltk-3.3}/nltk/test/japanese.doctest +48 -48
  267. {nltk-3.2 → nltk-3.3}/nltk/test/logic.doctest +1098 -1098
  268. {nltk-3.2 → nltk-3.3}/nltk/test/metrics.doctest +270 -270
  269. {nltk-3.2 → nltk-3.3}/nltk/test/misc.doctest +118 -118
  270. {nltk-3.2 → nltk-3.3}/nltk/test/nonmonotonic.doctest +286 -286
  271. {nltk-3.2 → nltk-3.3}/nltk/test/nonmonotonic_fixt.py +12 -11
  272. {nltk-3.2 → nltk-3.3}/nltk/test/paice.doctest +35 -35
  273. {nltk-3.2 → nltk-3.3}/nltk/test/parse.doctest +884 -884
  274. {nltk-3.2 → nltk-3.3}/nltk/test/portuguese_en.doctest +565 -565
  275. {nltk-3.2 → nltk-3.3}/nltk/test/portuguese_en_fixt.py +16 -16
  276. {nltk-3.2 → nltk-3.3}/nltk/test/probability.doctest +304 -304
  277. {nltk-3.2 → nltk-3.3}/nltk/test/probability_fixt.py +13 -12
  278. {nltk-3.2 → nltk-3.3}/nltk/test/propbank.doctest +176 -176
  279. {nltk-3.2 → nltk-3.3}/nltk/test/relextract.doctest +263 -263
  280. {nltk-3.2 → nltk-3.3}/nltk/test/resolution.doctest +221 -221
  281. {nltk-3.2 → nltk-3.3}/nltk/test/runtests.py +86 -64
  282. {nltk-3.2 → nltk-3.3}/nltk/test/segmentation_fixt.py +10 -10
  283. {nltk-3.2 → nltk-3.3}/nltk/test/semantics.doctest +665 -665
  284. {nltk-3.2 → nltk-3.3}/nltk/test/semantics_fixt.py +7 -7
  285. {nltk-3.2 → nltk-3.3}/nltk/test/sentiment.doctest +233 -233
  286. {nltk-3.2 → nltk-3.3}/nltk/test/sentiwordnet.doctest +41 -41
  287. {nltk-3.2 → nltk-3.3}/nltk/test/simple.doctest +85 -85
  288. {nltk-3.2 → nltk-3.3}/nltk/test/stem.doctest +78 -78
  289. {nltk-3.2 → nltk-3.3}/nltk/test/tag.doctest +33 -33
  290. {nltk-3.2 → nltk-3.3}/nltk/test/tokenize.doctest +273 -166
  291. {nltk-3.2 → nltk-3.3}/nltk/test/toolbox.doctest +307 -307
  292. {nltk-3.2 → nltk-3.3}/nltk/test/translate.doctest +242 -242
  293. {nltk-3.2 → nltk-3.3}/nltk/test/translate_fixt.py +3 -3
  294. {nltk-3.2 → nltk-3.3}/nltk/test/tree.doctest +1101 -1101
  295. {nltk-3.2 → nltk-3.3}/nltk/test/treeprettyprinter.doctest +127 -127
  296. {nltk-3.2 → nltk-3.3}/nltk/test/treetransforms.doctest +156 -156
  297. nltk-3.3/nltk/test/unit/test_2x_compat.py +63 -0
  298. nltk-3.3/nltk/test/unit/test_aline.py +52 -0
  299. nltk-3.3/nltk/test/unit/test_chunk.py +56 -0
  300. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_classify.py +55 -53
  301. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_collocations.py +88 -79
  302. nltk-3.3/nltk/test/unit/test_concordance.py +98 -0
  303. nltk-3.3/nltk/test/unit/test_corenlp.py +1000 -0
  304. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_corpora.py +200 -183
  305. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_corpus_views.py +45 -45
  306. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_hmm.py +88 -88
  307. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_json2csv_corpus.py +188 -187
  308. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_naivebayes.py +26 -26
  309. nltk-3.3/nltk/test/unit/test_rte_classify.py +86 -0
  310. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_seekable_unicode_stream_reader.py +124 -124
  311. nltk-3.3/nltk/test/unit/test_senna.py +74 -0
  312. nltk-3.3/nltk/test/unit/test_stem.py +121 -0
  313. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_tag.py +20 -20
  314. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_tgrep.py +628 -626
  315. nltk-3.3/nltk/test/unit/test_tokenize.py +168 -0
  316. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_twitter_auth.py +185 -185
  317. nltk-3.3/nltk/test/unit/test_wordnet.py +134 -0
  318. nltk-3.3/nltk/test/unit/translate/test_bleu.py +244 -0
  319. nltk-3.3/nltk/test/unit/translate/test_gdfa.py +57 -0
  320. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/test_ibm1.py +75 -75
  321. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/test_ibm2.py +87 -87
  322. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/test_ibm3.py +105 -105
  323. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/test_ibm4.py +126 -126
  324. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/test_ibm5.py +166 -166
  325. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/test_ibm_model.py +270 -270
  326. nltk-3.3/nltk/test/unit/translate/test_nist.py +37 -0
  327. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/test_stack_decoder.py +300 -300
  328. {nltk-3.2 → nltk-3.3}/nltk/test/unit/utils.py +41 -41
  329. {nltk-3.2 → nltk-3.3}/nltk/test/util.doctest +48 -48
  330. {nltk-3.2 → nltk-3.3}/nltk/test/wordnet.doctest +595 -581
  331. {nltk-3.2 → nltk-3.3}/nltk/test/wordnet_fixt.py +7 -6
  332. {nltk-3.2 → nltk-3.3}/nltk/test/wordnet_lch.doctest +53 -53
  333. {nltk-3.2 → nltk-3.3}/nltk/test/wsd.doctest +68 -68
  334. {nltk-3.2 → nltk-3.3}/nltk/text.py +663 -610
  335. {nltk-3.2 → nltk-3.3}/nltk/tgrep.py +938 -937
  336. {nltk-3.2 → nltk-3.3}/nltk/tokenize/__init__.py +130 -107
  337. {nltk-3.2 → nltk-3.3}/nltk/tokenize/api.py +77 -78
  338. {nltk-3.2 → nltk-3.3}/nltk/tokenize/casual.py +343 -342
  339. {nltk-3.2 → nltk-3.3}/nltk/tokenize/mwe.py +120 -120
  340. nltk-3.3/nltk/tokenize/nist.py +173 -0
  341. {nltk-3.2 → nltk-3.3}/nltk/tokenize/punkt.py +1620 -1601
  342. {nltk-3.2 → nltk-3.3}/nltk/tokenize/regexp.py +208 -202
  343. nltk-3.3/nltk/tokenize/repp.py +151 -0
  344. {nltk-3.2 → nltk-3.3}/nltk/tokenize/sexpr.py +140 -142
  345. {nltk-3.2 → nltk-3.3}/nltk/tokenize/simple.py +138 -136
  346. {nltk-3.2 → nltk-3.3}/nltk/tokenize/stanford.py +117 -113
  347. nltk-3.3/nltk/tokenize/stanford_segmenter.py +246 -0
  348. {nltk-3.2 → nltk-3.3}/nltk/tokenize/texttiling.py +458 -460
  349. nltk-3.3/nltk/tokenize/toktok.py +155 -0
  350. nltk-3.3/nltk/tokenize/treebank.py +363 -0
  351. nltk-3.3/nltk/tokenize/util.py +261 -0
  352. {nltk-3.2 → nltk-3.3}/nltk/toolbox.py +499 -497
  353. {nltk-3.2 → nltk-3.3}/nltk/translate/__init__.py +24 -24
  354. {nltk-3.2 → nltk-3.3}/nltk/translate/api.py +331 -321
  355. {nltk-3.2 → nltk-3.3}/nltk/translate/bleu_score.py +615 -556
  356. nltk-3.3/nltk/translate/chrf_score.py +137 -0
  357. {nltk-3.2 → nltk-3.3}/nltk/translate/gale_church.py +248 -235
  358. {nltk-3.2 → nltk-3.3}/nltk/translate/gdfa.py +135 -131
  359. nltk-3.3/nltk/translate/gleu_score.py +193 -0
  360. {nltk-3.2 → nltk-3.3}/nltk/translate/ibm1.py +250 -241
  361. {nltk-3.2 → nltk-3.3}/nltk/translate/ibm2.py +308 -308
  362. {nltk-3.2 → nltk-3.3}/nltk/translate/ibm3.py +337 -337
  363. {nltk-3.2 → nltk-3.3}/nltk/translate/ibm4.py +467 -467
  364. {nltk-3.2 → nltk-3.3}/nltk/translate/ibm5.py +639 -639
  365. {nltk-3.2 → nltk-3.3}/nltk/translate/ibm_model.py +536 -533
  366. {nltk-3.2 → nltk-3.3}/nltk/translate/metrics.py +40 -40
  367. nltk-3.3/nltk/translate/nist_score.py +178 -0
  368. {nltk-3.2 → nltk-3.3}/nltk/translate/phrase_based.py +175 -175
  369. {nltk-3.2 → nltk-3.3}/nltk/translate/ribes_score.py +325 -325
  370. {nltk-3.2 → nltk-3.3}/nltk/translate/stack_decoder.py +499 -495
  371. {nltk-3.2 → nltk-3.3}/nltk/tree.py +1605 -1603
  372. {nltk-3.2 → nltk-3.3}/nltk/treeprettyprinter.py +564 -564
  373. {nltk-3.2 → nltk-3.3}/nltk/treetransforms.py +309 -309
  374. {nltk-3.2 → nltk-3.3}/nltk/twitter/__init__.py +28 -28
  375. {nltk-3.2 → nltk-3.3}/nltk/twitter/api.py +141 -136
  376. {nltk-3.2 → nltk-3.3}/nltk/twitter/common.py +264 -258
  377. {nltk-3.2 → nltk-3.3}/nltk/twitter/twitter_demo.py +288 -288
  378. {nltk-3.2 → nltk-3.3}/nltk/twitter/twitterclient.py +530 -530
  379. {nltk-3.2 → nltk-3.3}/nltk/twitter/util.py +147 -146
  380. nltk-3.3/nltk/util.py +751 -0
  381. {nltk-3.2 → nltk-3.3}/nltk/wsd.py +53 -53
  382. {nltk-3.2 → nltk-3.3/nltk.egg-info}/PKG-INFO +13 -5
  383. {nltk-3.2 → nltk-3.3}/nltk.egg-info/SOURCES.txt +26 -3
  384. nltk-3.3/nltk.egg-info/requires.txt +31 -0
  385. {nltk-3.2 → nltk-3.3}/setup.cfg +0 -1
  386. {nltk-3.2 → nltk-3.3}/setup.py +102 -75
  387. nltk-3.2/nltk/VERSION +0 -1
  388. nltk-3.2/nltk/compat.py +0 -680
  389. nltk-3.2/nltk/corpus/reader/pl196x.py +0 -278
  390. nltk-3.2/nltk/corpus/reader/verbnet.py +0 -408
  391. nltk-3.2/nltk/corpus/reader/wordlist.py +0 -37
  392. nltk-3.2/nltk/sentiment/vader_lexicon.txt +0 -7517
  393. nltk-3.2/nltk/six.py +0 -868
  394. nltk-3.2/nltk/stem/porter.py +0 -692
  395. nltk-3.2/nltk/test/unit/test_2x_compat.py +0 -26
  396. nltk-3.2/nltk/test/unit/test_stem.py +0 -36
  397. nltk-3.2/nltk/test/unit/test_tokenize.py +0 -23
  398. nltk-3.2/nltk/test/unit/translate/test_bleu.py +0 -84
  399. nltk-3.2/nltk/tokenize/treebank.py +0 -124
  400. nltk-3.2/nltk/tokenize/util.py +0 -90
  401. nltk-3.2/nltk/util.py +0 -1364
  402. {nltk-3.2 → nltk-3.3}/nltk/test/unit/__init__.py +0 -0
  403. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/__init__.py +0 -0
  404. {nltk-3.2 → nltk-3.3}/nltk.egg-info/dependency_links.txt +0 -0
  405. {nltk-3.2 → nltk-3.3}/nltk.egg-info/not-zip-safe +0 -0
  406. {nltk-3.2 → nltk-3.3}/nltk.egg-info/top_level.txt +0 -0
@@ -1,894 +1,892 @@
1
- # Natural Language Toolkit: Recursive Descent Parser Application
2
- #
3
- # Copyright (C) 2001-2015 NLTK Project
4
- # Author: Edward Loper <edloper@gmail.com>
5
- # URL: <http://nltk.org/>
6
- # For license information, see LICENSE.TXT
7
-
8
- """
9
- A graphical tool for exploring the recursive descent parser.
10
-
11
- The recursive descent parser maintains a tree, which records the
12
- structure of the portion of the text that has been parsed. It uses
13
- CFG productions to expand the fringe of the tree, and matches its
14
- leaves against the text. Initially, the tree contains the start
15
- symbol ("S"). It is shown in the main canvas, to the right of the
16
- list of available expansions.
17
-
18
- The parser builds up a tree structure for the text using three
19
- operations:
20
-
21
- - "expand" uses a CFG production to add children to a node on the
22
- fringe of the tree.
23
- - "match" compares a leaf in the tree to a text token.
24
- - "backtrack" returns the tree to its state before the most recent
25
- expand or match operation.
26
-
27
- The parser maintains a list of tree locations called a "frontier" to
28
- remember which nodes have not yet been expanded and which leaves have
29
- not yet been matched against the text. The leftmost frontier node is
30
- shown in green, and the other frontier nodes are shown in blue. The
31
- parser always performs expand and match operations on the leftmost
32
- element of the frontier.
33
-
34
- You can control the parser's operation by using the "expand," "match,"
35
- and "backtrack" buttons; or you can use the "step" button to let the
36
- parser automatically decide which operation to apply. The parser uses
37
- the following rules to decide which operation to apply:
38
-
39
- - If the leftmost frontier element is a token, try matching it.
40
- - If the leftmost frontier element is a node, try expanding it with
41
- the first untried expansion.
42
- - Otherwise, backtrack.
43
-
44
- The "expand" button applies the untried expansion whose CFG production
45
- is listed earliest in the grammar. To manually choose which expansion
46
- to apply, click on a CFG production from the list of available
47
- expansions, on the left side of the main window.
48
-
49
- The "autostep" button will let the parser continue applying
50
- applications to the tree until it reaches a complete parse. You can
51
- cancel an autostep in progress at any time by clicking on the
52
- "autostep" button again.
53
-
54
- Keyboard Shortcuts::
55
- [Space]\t Perform the next expand, match, or backtrack operation
56
- [a]\t Step through operations until the next complete parse
57
- [e]\t Perform an expand operation
58
- [m]\t Perform a match operation
59
- [b]\t Perform a backtrack operation
60
- [Delete]\t Reset the parser
61
- [g]\t Show/hide available expansions list
62
- [h]\t Help
63
- [Ctrl-p]\t Print
64
- [q]\t Quit
65
- """
66
- from __future__ import division
67
- import nltk.compat
68
- import tkinter.font
69
- from tkinter import (Listbox, IntVar, Button,
70
- Frame, Label, Menu, Scrollbar, Tk)
71
-
72
- from nltk.tree import Tree
73
- from nltk.util import in_idle
74
- from nltk.parse import SteppingRecursiveDescentParser
75
- from nltk.draw.util import TextWidget, ShowText, CanvasFrame, EntryDialog
76
- from nltk.draw import CFGEditor, TreeSegmentWidget, tree_to_treesegment
77
-
78
- class RecursiveDescentApp(object):
79
- """
80
- A graphical tool for exploring the recursive descent parser. The tool
81
- displays the parser's tree and the remaining text, and allows the
82
- user to control the parser's operation. In particular, the user
83
- can expand subtrees on the frontier, match tokens on the frontier
84
- against the text, and backtrack. A "step" button simply steps
85
- through the parsing process, performing the operations that
86
- ``RecursiveDescentParser`` would use.
87
- """
88
- def __init__(self, grammar, sent, trace=0):
89
- self._sent = sent
90
- self._parser = SteppingRecursiveDescentParser(grammar, trace)
91
-
92
- # Set up the main window.
93
- self._top = Tk()
94
- self._top.title('Recursive Descent Parser Application')
95
-
96
- # Set up key bindings.
97
- self._init_bindings()
98
-
99
- # Initialize the fonts.
100
- self._init_fonts(self._top)
101
-
102
- # Animations. animating_lock is a lock to prevent the demo
103
- # from performing new operations while it's animating.
104
- self._animation_frames = IntVar(self._top)
105
- self._animation_frames.set(5)
106
- self._animating_lock = 0
107
- self._autostep = 0
108
-
109
- # The user can hide the grammar.
110
- self._show_grammar = IntVar(self._top)
111
- self._show_grammar.set(1)
112
-
113
- # Create the basic frames.
114
- self._init_menubar(self._top)
115
- self._init_buttons(self._top)
116
- self._init_feedback(self._top)
117
- self._init_grammar(self._top)
118
- self._init_canvas(self._top)
119
-
120
- # Initialize the parser.
121
- self._parser.initialize(self._sent)
122
-
123
- # Resize callback
124
- self._canvas.bind('<Configure>', self._configure)
125
-
126
- #########################################
127
- ## Initialization Helpers
128
- #########################################
129
-
130
- def _init_fonts(self, root):
131
- # See: <http://www.astro.washington.edu/owen/ROTKFolklore.html>
132
- self._sysfont = tkinter.font.Font(font=Button()["font"])
133
- root.option_add("*Font", self._sysfont)
134
-
135
- # TWhat's our font size (default=same as sysfont)
136
- self._size = IntVar(root)
137
- self._size.set(self._sysfont.cget('size'))
138
-
139
- self._boldfont = tkinter.font.Font(family='helvetica', weight='bold',
140
- size=self._size.get())
141
- self._font = tkinter.font.Font(family='helvetica',
142
- size=self._size.get())
143
- if self._size.get() < 0: big = self._size.get()-2
144
- else: big = self._size.get()+2
145
- self._bigfont = tkinter.font.Font(family='helvetica', weight='bold',
146
- size=big)
147
-
148
- def _init_grammar(self, parent):
149
- # Grammar view.
150
- self._prodframe = listframe = Frame(parent)
151
- self._prodframe.pack(fill='both', side='left', padx=2)
152
- self._prodlist_label = Label(self._prodframe, font=self._boldfont,
153
- text='Available Expansions')
154
- self._prodlist_label.pack()
155
- self._prodlist = Listbox(self._prodframe, selectmode='single',
156
- relief='groove', background='white',
157
- foreground='#909090', font=self._font,
158
- selectforeground='#004040',
159
- selectbackground='#c0f0c0')
160
-
161
- self._prodlist.pack(side='right', fill='both', expand=1)
162
-
163
- self._productions = list(self._parser.grammar().productions())
164
- for production in self._productions:
165
- self._prodlist.insert('end', (' %s' % production))
166
- self._prodlist.config(height=min(len(self._productions), 25))
167
-
168
- # Add a scrollbar if there are more than 25 productions.
169
- if len(self._productions) > 25:
170
- listscroll = Scrollbar(self._prodframe,
171
- orient='vertical')
172
- self._prodlist.config(yscrollcommand = listscroll.set)
173
- listscroll.config(command=self._prodlist.yview)
174
- listscroll.pack(side='left', fill='y')
175
-
176
- # If they select a production, apply it.
177
- self._prodlist.bind('<<ListboxSelect>>', self._prodlist_select)
178
-
179
- def _init_bindings(self):
180
- # Key bindings are a good thing.
181
- self._top.bind('<Control-q>', self.destroy)
182
- self._top.bind('<Control-x>', self.destroy)
183
- self._top.bind('<Escape>', self.destroy)
184
- self._top.bind('e', self.expand)
185
- #self._top.bind('<Alt-e>', self.expand)
186
- #self._top.bind('<Control-e>', self.expand)
187
- self._top.bind('m', self.match)
188
- self._top.bind('<Alt-m>', self.match)
189
- self._top.bind('<Control-m>', self.match)
190
- self._top.bind('b', self.backtrack)
191
- self._top.bind('<Alt-b>', self.backtrack)
192
- self._top.bind('<Control-b>', self.backtrack)
193
- self._top.bind('<Control-z>', self.backtrack)
194
- self._top.bind('<BackSpace>', self.backtrack)
195
- self._top.bind('a', self.autostep)
196
- #self._top.bind('<Control-a>', self.autostep)
197
- self._top.bind('<Control-space>', self.autostep)
198
- self._top.bind('<Control-c>', self.cancel_autostep)
199
- self._top.bind('<space>', self.step)
200
- self._top.bind('<Delete>', self.reset)
201
- self._top.bind('<Control-p>', self.postscript)
202
- #self._top.bind('<h>', self.help)
203
- #self._top.bind('<Alt-h>', self.help)
204
- self._top.bind('<Control-h>', self.help)
205
- self._top.bind('<F1>', self.help)
206
- #self._top.bind('<g>', self.toggle_grammar)
207
- #self._top.bind('<Alt-g>', self.toggle_grammar)
208
- #self._top.bind('<Control-g>', self.toggle_grammar)
209
- self._top.bind('<Control-g>', self.edit_grammar)
210
- self._top.bind('<Control-t>', self.edit_sentence)
211
-
212
- def _init_buttons(self, parent):
213
- # Set up the frames.
214
- self._buttonframe = buttonframe = Frame(parent)
215
- buttonframe.pack(fill='none', side='bottom', padx=3, pady=2)
216
- Button(buttonframe, text='Step',
217
- background='#90c0d0', foreground='black',
218
- command=self.step,).pack(side='left')
219
- Button(buttonframe, text='Autostep',
220
- background='#90c0d0', foreground='black',
221
- command=self.autostep,).pack(side='left')
222
- Button(buttonframe, text='Expand', underline=0,
223
- background='#90f090', foreground='black',
224
- command=self.expand).pack(side='left')
225
- Button(buttonframe, text='Match', underline=0,
226
- background='#90f090', foreground='black',
227
- command=self.match).pack(side='left')
228
- Button(buttonframe, text='Backtrack', underline=0,
229
- background='#f0a0a0', foreground='black',
230
- command=self.backtrack).pack(side='left')
231
- # Replace autostep...
232
- # self._autostep_button = Button(buttonframe, text='Autostep',
233
- # underline=0, command=self.autostep)
234
- # self._autostep_button.pack(side='left')
235
-
236
- def _configure(self, event):
237
- self._autostep = 0
238
- (x1, y1, x2, y2) = self._cframe.scrollregion()
239
- y2 = event.height - 6
240
- self._canvas['scrollregion'] = '%d %d %d %d' % (x1,y1,x2,y2)
241
- self._redraw()
242
-
243
- def _init_feedback(self, parent):
244
- self._feedbackframe = feedbackframe = Frame(parent)
245
- feedbackframe.pack(fill='x', side='bottom', padx=3, pady=3)
246
- self._lastoper_label = Label(feedbackframe, text='Last Operation:',
247
- font=self._font)
248
- self._lastoper_label.pack(side='left')
249
- lastoperframe = Frame(feedbackframe, relief='sunken', border=1)
250
- lastoperframe.pack(fill='x', side='right', expand=1, padx=5)
251
- self._lastoper1 = Label(lastoperframe, foreground='#007070',
252
- background='#f0f0f0', font=self._font)
253
- self._lastoper2 = Label(lastoperframe, anchor='w', width=30,
254
- foreground='#004040', background='#f0f0f0',
255
- font=self._font)
256
- self._lastoper1.pack(side='left')
257
- self._lastoper2.pack(side='left', fill='x', expand=1)
258
-
259
- def _init_canvas(self, parent):
260
- self._cframe = CanvasFrame(parent, background='white',
261
- #width=525, height=250,
262
- closeenough=10,
263
- border=2, relief='sunken')
264
- self._cframe.pack(expand=1, fill='both', side='top', pady=2)
265
- canvas = self._canvas = self._cframe.canvas()
266
-
267
- # Initially, there's no tree or text
268
- self._tree = None
269
- self._textwidgets = []
270
- self._textline = None
271
-
272
- def _init_menubar(self, parent):
273
- menubar = Menu(parent)
274
-
275
- filemenu = Menu(menubar, tearoff=0)
276
- filemenu.add_command(label='Reset Parser', underline=0,
277
- command=self.reset, accelerator='Del')
278
- filemenu.add_command(label='Print to Postscript', underline=0,
279
- command=self.postscript, accelerator='Ctrl-p')
280
- filemenu.add_command(label='Exit', underline=1,
281
- command=self.destroy, accelerator='Ctrl-x')
282
- menubar.add_cascade(label='File', underline=0, menu=filemenu)
283
-
284
- editmenu = Menu(menubar, tearoff=0)
285
- editmenu.add_command(label='Edit Grammar', underline=5,
286
- command=self.edit_grammar,
287
- accelerator='Ctrl-g')
288
- editmenu.add_command(label='Edit Text', underline=5,
289
- command=self.edit_sentence,
290
- accelerator='Ctrl-t')
291
- menubar.add_cascade(label='Edit', underline=0, menu=editmenu)
292
-
293
- rulemenu = Menu(menubar, tearoff=0)
294
- rulemenu.add_command(label='Step', underline=1,
295
- command=self.step, accelerator='Space')
296
- rulemenu.add_separator()
297
- rulemenu.add_command(label='Match', underline=0,
298
- command=self.match, accelerator='Ctrl-m')
299
- rulemenu.add_command(label='Expand', underline=0,
300
- command=self.expand, accelerator='Ctrl-e')
301
- rulemenu.add_separator()
302
- rulemenu.add_command(label='Backtrack', underline=0,
303
- command=self.backtrack, accelerator='Ctrl-b')
304
- menubar.add_cascade(label='Apply', underline=0, menu=rulemenu)
305
-
306
- viewmenu = Menu(menubar, tearoff=0)
307
- viewmenu.add_checkbutton(label="Show Grammar", underline=0,
308
- variable=self._show_grammar,
309
- command=self._toggle_grammar)
310
- viewmenu.add_separator()
311
- viewmenu.add_radiobutton(label='Tiny', variable=self._size,
312
- underline=0, value=10, command=self.resize)
313
- viewmenu.add_radiobutton(label='Small', variable=self._size,
314
- underline=0, value=12, command=self.resize)
315
- viewmenu.add_radiobutton(label='Medium', variable=self._size,
316
- underline=0, value=14, command=self.resize)
317
- viewmenu.add_radiobutton(label='Large', variable=self._size,
318
- underline=0, value=18, command=self.resize)
319
- viewmenu.add_radiobutton(label='Huge', variable=self._size,
320
- underline=0, value=24, command=self.resize)
321
- menubar.add_cascade(label='View', underline=0, menu=viewmenu)
322
-
323
- animatemenu = Menu(menubar, tearoff=0)
324
- animatemenu.add_radiobutton(label="No Animation", underline=0,
325
- variable=self._animation_frames,
326
- value=0)
327
- animatemenu.add_radiobutton(label="Slow Animation", underline=0,
328
- variable=self._animation_frames,
329
- value=10, accelerator='-')
330
- animatemenu.add_radiobutton(label="Normal Animation", underline=0,
331
- variable=self._animation_frames,
332
- value=5, accelerator='=')
333
- animatemenu.add_radiobutton(label="Fast Animation", underline=0,
334
- variable=self._animation_frames,
335
- value=2, accelerator='+')
336
- menubar.add_cascade(label="Animate", underline=1, menu=animatemenu)
337
-
338
-
339
- helpmenu = Menu(menubar, tearoff=0)
340
- helpmenu.add_command(label='About', underline=0,
341
- command=self.about)
342
- helpmenu.add_command(label='Instructions', underline=0,
343
- command=self.help, accelerator='F1')
344
- menubar.add_cascade(label='Help', underline=0, menu=helpmenu)
345
-
346
- parent.config(menu=menubar)
347
-
348
- #########################################
349
- ## Helper
350
- #########################################
351
-
352
- def _get(self, widget, treeloc):
353
- for i in treeloc: widget = widget.subtrees()[i]
354
- if isinstance(widget, TreeSegmentWidget):
355
- widget = widget.label()
356
- return widget
357
-
358
- #########################################
359
- ## Main draw procedure
360
- #########################################
361
-
362
- def _redraw(self):
363
- canvas = self._canvas
364
-
365
- # Delete the old tree, widgets, etc.
366
- if self._tree is not None:
367
- self._cframe.destroy_widget(self._tree)
368
- for twidget in self._textwidgets:
369
- self._cframe.destroy_widget(twidget)
370
- if self._textline is not None:
371
- self._canvas.delete(self._textline)
372
-
373
- # Draw the tree.
374
- helv = ('helvetica', -self._size.get())
375
- bold = ('helvetica', -self._size.get(), 'bold')
376
- attribs = {'tree_color': '#000000', 'tree_width': 2,
377
- 'node_font': bold, 'leaf_font': helv,}
378
- tree = self._parser.tree()
379
- self._tree = tree_to_treesegment(canvas, tree, **attribs)
380
- self._cframe.add_widget(self._tree, 30, 5)
381
-
382
- # Draw the text.
383
- helv = ('helvetica', -self._size.get())
384
- bottom = y = self._cframe.scrollregion()[3]
385
- self._textwidgets = [TextWidget(canvas, word, font=self._font)
386
- for word in self._sent]
387
- for twidget in self._textwidgets:
388
- self._cframe.add_widget(twidget, 0, 0)
389
- twidget.move(0, bottom-twidget.bbox()[3]-5)
390
- y = min(y, twidget.bbox()[1])
391
-
392
- # Draw a line over the text, to separate it from the tree.
393
- self._textline = canvas.create_line(-5000, y-5, 5000, y-5, dash='.')
394
-
395
- # Highlight appropriate nodes.
396
- self._highlight_nodes()
397
- self._highlight_prodlist()
398
-
399
- # Make sure the text lines up.
400
- self._position_text()
401
-
402
-
403
- def _redraw_quick(self):
404
- # This should be more-or-less sufficient after an animation.
405
- self._highlight_nodes()
406
- self._highlight_prodlist()
407
- self._position_text()
408
-
409
- def _highlight_nodes(self):
410
- # Highlight the list of nodes to be checked.
411
- bold = ('helvetica', -self._size.get(), 'bold')
412
- for treeloc in self._parser.frontier()[:1]:
413
- self._get(self._tree, treeloc)['color'] = '#20a050'
414
- self._get(self._tree, treeloc)['font'] = bold
415
- for treeloc in self._parser.frontier()[1:]:
416
- self._get(self._tree, treeloc)['color'] = '#008080'
417
-
418
- def _highlight_prodlist(self):
419
- # Highlight the productions that can be expanded.
420
- # Boy, too bad tkinter doesn't implement Listbox.itemconfig;
421
- # that would be pretty useful here.
422
- self._prodlist.delete(0, 'end')
423
- expandable = self._parser.expandable_productions()
424
- untried = self._parser.untried_expandable_productions()
425
- productions = self._productions
426
- for index in range(len(productions)):
427
- if productions[index] in expandable:
428
- if productions[index] in untried:
429
- self._prodlist.insert(index, ' %s' % productions[index])
430
- else:
431
- self._prodlist.insert(index, ' %s (TRIED)' %
432
- productions[index])
433
- self._prodlist.selection_set(index)
434
- else:
435
- self._prodlist.insert(index, ' %s' % productions[index])
436
-
437
- def _position_text(self):
438
- # Line up the text widgets that are matched against the tree
439
- numwords = len(self._sent)
440
- num_matched = numwords - len(self._parser.remaining_text())
441
- leaves = self._tree_leaves()[:num_matched]
442
- xmax = self._tree.bbox()[0]
443
- for i in range(0, len(leaves)):
444
- widget = self._textwidgets[i]
445
- leaf = leaves[i]
446
- widget['color'] = '#006040'
447
- leaf['color'] = '#006040'
448
- widget.move(leaf.bbox()[0] - widget.bbox()[0], 0)
449
- xmax = widget.bbox()[2] + 10
450
-
451
- # Line up the text widgets that are not matched against the tree.
452
- for i in range(len(leaves), numwords):
453
- widget = self._textwidgets[i]
454
- widget['color'] = '#a0a0a0'
455
- widget.move(xmax - widget.bbox()[0], 0)
456
- xmax = widget.bbox()[2] + 10
457
-
458
- # If we have a complete parse, make everything green :)
459
- if self._parser.currently_complete():
460
- for twidget in self._textwidgets:
461
- twidget['color'] = '#00a000'
462
-
463
- # Move the matched leaves down to the text.
464
- for i in range(0, len(leaves)):
465
- widget = self._textwidgets[i]
466
- leaf = leaves[i]
467
- dy = widget.bbox()[1] - leaf.bbox()[3] - 10.0
468
- dy = max(dy, leaf.parent().label().bbox()[3] - leaf.bbox()[3] + 10)
469
- leaf.move(0, dy)
470
-
471
- def _tree_leaves(self, tree=None):
472
- if tree is None: tree = self._tree
473
- if isinstance(tree, TreeSegmentWidget):
474
- leaves = []
475
- for child in tree.subtrees(): leaves += self._tree_leaves(child)
476
- return leaves
477
- else:
478
- return [tree]
479
-
480
- #########################################
481
- ## Button Callbacks
482
- #########################################
483
-
484
- def destroy(self, *e):
485
- self._autostep = 0
486
- if self._top is None: return
487
- self._top.destroy()
488
- self._top = None
489
-
490
- def reset(self, *e):
491
- self._autostep = 0
492
- self._parser.initialize(self._sent)
493
- self._lastoper1['text'] = 'Reset Application'
494
- self._lastoper2['text'] = ''
495
- self._redraw()
496
-
497
- def autostep(self, *e):
498
- if self._animation_frames.get() == 0:
499
- self._animation_frames.set(2)
500
- if self._autostep:
501
- self._autostep = 0
502
- else:
503
- self._autostep = 1
504
- self._step()
505
-
506
- def cancel_autostep(self, *e):
507
- #self._autostep_button['text'] = 'Autostep'
508
- self._autostep = 0
509
-
510
- # Make sure to stop auto-stepping if we get any user input.
511
- def step(self, *e): self._autostep = 0; self._step()
512
- def match(self, *e): self._autostep = 0; self._match()
513
- def expand(self, *e): self._autostep = 0; self._expand()
514
- def backtrack(self, *e): self._autostep = 0; self._backtrack()
515
-
516
- def _step(self):
517
- if self._animating_lock: return
518
-
519
- # Try expanding, matching, and backtracking (in that order)
520
- if self._expand(): pass
521
- elif self._parser.untried_match() and self._match(): pass
522
- elif self._backtrack(): pass
523
- else:
524
- self._lastoper1['text'] = 'Finished'
525
- self._lastoper2['text'] = ''
526
- self._autostep = 0
527
-
528
- # Check if we just completed a parse.
529
- if self._parser.currently_complete():
530
- self._autostep = 0
531
- self._lastoper2['text'] += ' [COMPLETE PARSE]'
532
-
533
- def _expand(self, *e):
534
- if self._animating_lock: return
535
- old_frontier = self._parser.frontier()
536
- rv = self._parser.expand()
537
- if rv is not None:
538
- self._lastoper1['text'] = 'Expand:'
539
- self._lastoper2['text'] = rv
540
- self._prodlist.selection_clear(0, 'end')
541
- index = self._productions.index(rv)
542
- self._prodlist.selection_set(index)
543
- self._animate_expand(old_frontier[0])
544
- return True
545
- else:
546
- self._lastoper1['text'] = 'Expand:'
547
- self._lastoper2['text'] = '(all expansions tried)'
548
- return False
549
-
550
- def _match(self, *e):
551
- if self._animating_lock: return
552
- old_frontier = self._parser.frontier()
553
- rv = self._parser.match()
554
- if rv is not None:
555
- self._lastoper1['text'] = 'Match:'
556
- self._lastoper2['text'] = rv
557
- self._animate_match(old_frontier[0])
558
- return True
559
- else:
560
- self._lastoper1['text'] = 'Match:'
561
- self._lastoper2['text'] = '(failed)'
562
- return False
563
-
564
- def _backtrack(self, *e):
565
- if self._animating_lock: return
566
- if self._parser.backtrack():
567
- elt = self._parser.tree()
568
- for i in self._parser.frontier()[0]:
569
- elt = elt[i]
570
- self._lastoper1['text'] = 'Backtrack'
571
- self._lastoper2['text'] = ''
572
- if isinstance(elt, Tree):
573
- self._animate_backtrack(self._parser.frontier()[0])
574
- else:
575
- self._animate_match_backtrack(self._parser.frontier()[0])
576
- return True
577
- else:
578
- self._autostep = 0
579
- self._lastoper1['text'] = 'Finished'
580
- self._lastoper2['text'] = ''
581
- return False
582
-
583
- def about(self, *e):
584
- ABOUT = ("NLTK Recursive Descent Parser Application\n"+
585
- "Written by Edward Loper")
586
- TITLE = 'About: Recursive Descent Parser Application'
587
- try:
588
- from tkinter.messagebox import Message
589
- Message(message=ABOUT, title=TITLE).show()
590
- except:
591
- ShowText(self._top, TITLE, ABOUT)
592
-
593
- def help(self, *e):
594
- self._autostep = 0
595
- # The default font's not very legible; try using 'fixed' instead.
596
- try:
597
- ShowText(self._top, 'Help: Recursive Descent Parser Application',
598
- (__doc__ or '').strip(), width=75, font='fixed')
599
- except:
600
- ShowText(self._top, 'Help: Recursive Descent Parser Application',
601
- (__doc__ or '').strip(), width=75)
602
-
603
- def postscript(self, *e):
604
- self._autostep = 0
605
- self._cframe.print_to_file()
606
-
607
- def mainloop(self, *args, **kwargs):
608
- """
609
- Enter the Tkinter mainloop. This function must be called if
610
- this demo is created from a non-interactive program (e.g.
611
- from a secript); otherwise, the demo will close as soon as
612
- the script completes.
613
- """
614
- if in_idle(): return
615
- self._top.mainloop(*args, **kwargs)
616
-
617
- def resize(self, size=None):
618
- if size is not None: self._size.set(size)
619
- size = self._size.get()
620
- self._font.configure(size=-(abs(size)))
621
- self._boldfont.configure(size=-(abs(size)))
622
- self._sysfont.configure(size=-(abs(size)))
623
- self._bigfont.configure(size=-(abs(size+2)))
624
- self._redraw()
625
-
626
- #########################################
627
- ## Expand Production Selection
628
- #########################################
629
-
630
- def _toggle_grammar(self, *e):
631
- if self._show_grammar.get():
632
- self._prodframe.pack(fill='both', side='left', padx=2,
633
- after=self._feedbackframe)
634
- self._lastoper1['text'] = 'Show Grammar'
635
- else:
636
- self._prodframe.pack_forget()
637
- self._lastoper1['text'] = 'Hide Grammar'
638
- self._lastoper2['text'] = ''
639
-
640
- # def toggle_grammar(self, *e):
641
- # self._show_grammar = not self._show_grammar
642
- # if self._show_grammar:
643
- # self._prodframe.pack(fill='both', expand='y', side='left',
644
- # after=self._feedbackframe)
645
- # self._lastoper1['text'] = 'Show Grammar'
646
- # else:
647
- # self._prodframe.pack_forget()
648
- # self._lastoper1['text'] = 'Hide Grammar'
649
- # self._lastoper2['text'] = ''
650
-
651
- def _prodlist_select(self, event):
652
- selection = self._prodlist.curselection()
653
- if len(selection) != 1: return
654
- index = int(selection[0])
655
- old_frontier = self._parser.frontier()
656
- production = self._parser.expand(self._productions[index])
657
-
658
- if production:
659
- self._lastoper1['text'] = 'Expand:'
660
- self._lastoper2['text'] = production
661
- self._prodlist.selection_clear(0, 'end')
662
- self._prodlist.selection_set(index)
663
- self._animate_expand(old_frontier[0])
664
- else:
665
- # Reset the production selections.
666
- self._prodlist.selection_clear(0, 'end')
667
- for prod in self._parser.expandable_productions():
668
- index = self._productions.index(prod)
669
- self._prodlist.selection_set(index)
670
-
671
- #########################################
672
- ## Animation
673
- #########################################
674
-
675
- def _animate_expand(self, treeloc):
676
- oldwidget = self._get(self._tree, treeloc)
677
- oldtree = oldwidget.parent()
678
- top = not isinstance(oldtree.parent(), TreeSegmentWidget)
679
-
680
- tree = self._parser.tree()
681
- for i in treeloc:
682
- tree = tree[i]
683
-
684
- widget = tree_to_treesegment(self._canvas, tree,
685
- node_font=self._boldfont,
686
- leaf_color='white',
687
- tree_width=2, tree_color='white',
688
- node_color='white',
689
- leaf_font=self._font)
690
- widget.label()['color'] = '#20a050'
691
-
692
- (oldx, oldy) = oldtree.label().bbox()[:2]
693
- (newx, newy) = widget.label().bbox()[:2]
694
- widget.move(oldx-newx, oldy-newy)
695
-
696
- if top:
697
- self._cframe.add_widget(widget, 0, 5)
698
- widget.move(30-widget.label().bbox()[0], 0)
699
- self._tree = widget
700
- else:
701
- oldtree.parent().replace_child(oldtree, widget)
702
-
703
- # Move the children over so they don't overlap.
704
- # Line the children up in a strange way.
705
- if widget.subtrees():
706
- dx = (oldx + widget.label().width()/2 -
707
- widget.subtrees()[0].bbox()[0]/2 -
708
- widget.subtrees()[0].bbox()[2]/2)
709
- for subtree in widget.subtrees(): subtree.move(dx, 0)
710
-
711
- self._makeroom(widget)
712
-
713
- if top:
714
- self._cframe.destroy_widget(oldtree)
715
- else:
716
- oldtree.destroy()
717
-
718
- colors = ['gray%d' % (10*int(10*x/self._animation_frames.get()))
719
- for x in range(self._animation_frames.get(),0,-1)]
720
-
721
- # Move the text string down, if necessary.
722
- dy = widget.bbox()[3] + 30 - self._canvas.coords(self._textline)[1]
723
- if dy > 0:
724
- for twidget in self._textwidgets: twidget.move(0, dy)
725
- self._canvas.move(self._textline, 0, dy)
726
-
727
- self._animate_expand_frame(widget, colors)
728
-
729
- def _makeroom(self, treeseg):
730
- """
731
- Make sure that no sibling tree bbox's overlap.
732
- """
733
- parent = treeseg.parent()
734
- if not isinstance(parent, TreeSegmentWidget): return
735
-
736
- index = parent.subtrees().index(treeseg)
737
-
738
- # Handle siblings to the right
739
- rsiblings = parent.subtrees()[index+1:]
740
- if rsiblings:
741
- dx = treeseg.bbox()[2] - rsiblings[0].bbox()[0] + 10
742
- for sibling in rsiblings: sibling.move(dx, 0)
743
-
744
- # Handle siblings to the left
745
- if index > 0:
746
- lsibling = parent.subtrees()[index-1]
747
- dx = max(0, lsibling.bbox()[2] - treeseg.bbox()[0] + 10)
748
- treeseg.move(dx, 0)
749
-
750
- # Keep working up the tree.
751
- self._makeroom(parent)
752
-
753
- def _animate_expand_frame(self, widget, colors):
754
- if len(colors) > 0:
755
- self._animating_lock = 1
756
- widget['color'] = colors[0]
757
- for subtree in widget.subtrees():
758
- if isinstance(subtree, TreeSegmentWidget):
759
- subtree.label()['color'] = colors[0]
760
- else:
761
- subtree['color'] = colors[0]
762
- self._top.after(50, self._animate_expand_frame,
763
- widget, colors[1:])
764
- else:
765
- widget['color'] = 'black'
766
- for subtree in widget.subtrees():
767
- if isinstance(subtree, TreeSegmentWidget):
768
- subtree.label()['color'] = 'black'
769
- else:
770
- subtree['color'] = 'black'
771
- self._redraw_quick()
772
- widget.label()['color'] = 'black'
773
- self._animating_lock = 0
774
- if self._autostep: self._step()
775
-
776
- def _animate_backtrack(self, treeloc):
777
- # Flash red first, if we're animating.
778
- if self._animation_frames.get() == 0: colors = []
779
- else: colors = ['#a00000', '#000000', '#a00000']
780
- colors += ['gray%d' % (10*int(10*x/(self._animation_frames.get())))
781
- for x in range(1, self._animation_frames.get()+1)]
782
-
783
- widgets = [self._get(self._tree, treeloc).parent()]
784
- for subtree in widgets[0].subtrees():
785
- if isinstance(subtree, TreeSegmentWidget):
786
- widgets.append(subtree.label())
787
- else:
788
- widgets.append(subtree)
789
-
790
- self._animate_backtrack_frame(widgets, colors)
791
-
792
- def _animate_backtrack_frame(self, widgets, colors):
793
- if len(colors) > 0:
794
- self._animating_lock = 1
795
- for widget in widgets: widget['color'] = colors[0]
796
- self._top.after(50, self._animate_backtrack_frame,
797
- widgets, colors[1:])
798
- else:
799
- for widget in widgets[0].subtrees():
800
- widgets[0].remove_child(widget)
801
- widget.destroy()
802
- self._redraw_quick()
803
- self._animating_lock = 0
804
- if self._autostep: self._step()
805
-
806
- def _animate_match_backtrack(self, treeloc):
807
- widget = self._get(self._tree, treeloc)
808
- node = widget.parent().label()
809
- dy = ((node.bbox()[3] - widget.bbox()[1] + 14) /
810
- max(1, self._animation_frames.get()))
811
- self._animate_match_backtrack_frame(self._animation_frames.get(),
812
- widget, dy)
813
-
814
- def _animate_match(self, treeloc):
815
- widget = self._get(self._tree, treeloc)
816
-
817
- dy = ((self._textwidgets[0].bbox()[1] - widget.bbox()[3] - 10.0) /
818
- max(1, self._animation_frames.get()))
819
- self._animate_match_frame(self._animation_frames.get(), widget, dy)
820
-
821
- def _animate_match_frame(self, frame, widget, dy):
822
- if frame > 0:
823
- self._animating_lock = 1
824
- widget.move(0, dy)
825
- self._top.after(10, self._animate_match_frame,
826
- frame-1, widget, dy)
827
- else:
828
- widget['color'] = '#006040'
829
- self._redraw_quick()
830
- self._animating_lock = 0
831
- if self._autostep: self._step()
832
-
833
- def _animate_match_backtrack_frame(self, frame, widget, dy):
834
- if frame > 0:
835
- self._animating_lock = 1
836
- widget.move(0, dy)
837
- self._top.after(10, self._animate_match_backtrack_frame,
838
- frame-1, widget, dy)
839
- else:
840
- widget.parent().remove_child(widget)
841
- widget.destroy()
842
- self._animating_lock = 0
843
- if self._autostep: self._step()
844
-
845
- def edit_grammar(self, *e):
846
- CFGEditor(self._top, self._parser.grammar(), self.set_grammar)
847
-
848
- def set_grammar(self, grammar):
849
- self._parser.set_grammar(grammar)
850
- self._productions = list(grammar.productions())
851
- self._prodlist.delete(0, 'end')
852
- for production in self._productions:
853
- self._prodlist.insert('end', (' %s' % production))
854
-
855
- def edit_sentence(self, *e):
856
- sentence = " ".join(self._sent)
857
- title = 'Edit Text'
858
- instr = 'Enter a new sentence to parse.'
859
- EntryDialog(self._top, sentence, instr, self.set_sentence, title)
860
-
861
- def set_sentence(self, sentence):
862
- self._sent = sentence.split() #[XX] use tagged?
863
- self.reset()
864
-
865
- def app():
866
- """
867
- Create a recursive descent parser demo, using a simple grammar and
868
- text.
869
- """
870
- from nltk.grammar import CFG
871
- grammar = CFG.fromstring("""
872
- # Grammatical productions.
873
- S -> NP VP
874
- NP -> Det N PP | Det N
875
- VP -> V NP PP | V NP | V
876
- PP -> P NP
877
- # Lexical productions.
878
- NP -> 'I'
879
- Det -> 'the' | 'a'
880
- N -> 'man' | 'park' | 'dog' | 'telescope'
881
- V -> 'ate' | 'saw'
882
- P -> 'in' | 'under' | 'with'
883
- """)
884
-
885
- sent = 'the dog saw a man in the park'.split()
886
-
887
- RecursiveDescentApp(grammar, sent).mainloop()
888
-
889
- if __name__ == '__main__':
890
- app()
891
-
892
- __all__ = ['app']
893
-
894
-
1
+ # Natural Language Toolkit: Recursive Descent Parser Application
2
+ #
3
+ # Copyright (C) 2001-2018 NLTK Project
4
+ # Author: Edward Loper <edloper@gmail.com>
5
+ # URL: <http://nltk.org/>
6
+ # For license information, see LICENSE.TXT
7
+
8
+ """
9
+ A graphical tool for exploring the recursive descent parser.
10
+
11
+ The recursive descent parser maintains a tree, which records the
12
+ structure of the portion of the text that has been parsed. It uses
13
+ CFG productions to expand the fringe of the tree, and matches its
14
+ leaves against the text. Initially, the tree contains the start
15
+ symbol ("S"). It is shown in the main canvas, to the right of the
16
+ list of available expansions.
17
+
18
+ The parser builds up a tree structure for the text using three
19
+ operations:
20
+
21
+ - "expand" uses a CFG production to add children to a node on the
22
+ fringe of the tree.
23
+ - "match" compares a leaf in the tree to a text token.
24
+ - "backtrack" returns the tree to its state before the most recent
25
+ expand or match operation.
26
+
27
+ The parser maintains a list of tree locations called a "frontier" to
28
+ remember which nodes have not yet been expanded and which leaves have
29
+ not yet been matched against the text. The leftmost frontier node is
30
+ shown in green, and the other frontier nodes are shown in blue. The
31
+ parser always performs expand and match operations on the leftmost
32
+ element of the frontier.
33
+
34
+ You can control the parser's operation by using the "expand," "match,"
35
+ and "backtrack" buttons; or you can use the "step" button to let the
36
+ parser automatically decide which operation to apply. The parser uses
37
+ the following rules to decide which operation to apply:
38
+
39
+ - If the leftmost frontier element is a token, try matching it.
40
+ - If the leftmost frontier element is a node, try expanding it with
41
+ the first untried expansion.
42
+ - Otherwise, backtrack.
43
+
44
+ The "expand" button applies the untried expansion whose CFG production
45
+ is listed earliest in the grammar. To manually choose which expansion
46
+ to apply, click on a CFG production from the list of available
47
+ expansions, on the left side of the main window.
48
+
49
+ The "autostep" button will let the parser continue applying
50
+ applications to the tree until it reaches a complete parse. You can
51
+ cancel an autostep in progress at any time by clicking on the
52
+ "autostep" button again.
53
+
54
+ Keyboard Shortcuts::
55
+ [Space]\t Perform the next expand, match, or backtrack operation
56
+ [a]\t Step through operations until the next complete parse
57
+ [e]\t Perform an expand operation
58
+ [m]\t Perform a match operation
59
+ [b]\t Perform a backtrack operation
60
+ [Delete]\t Reset the parser
61
+ [g]\t Show/hide available expansions list
62
+ [h]\t Help
63
+ [Ctrl-p]\t Print
64
+ [q]\t Quit
65
+ """
66
+ from __future__ import division
67
+
68
+ from six.moves.tkinter_font import Font
69
+ from six.moves.tkinter import (Listbox, IntVar, Button, Frame, Label, Menu,
70
+ Scrollbar, Tk)
71
+
72
+ from nltk.tree import Tree
73
+ from nltk.util import in_idle
74
+ from nltk.parse import SteppingRecursiveDescentParser
75
+ from nltk.draw.util import TextWidget, ShowText, CanvasFrame, EntryDialog
76
+ from nltk.draw import CFGEditor, TreeSegmentWidget, tree_to_treesegment
77
+
78
+ class RecursiveDescentApp(object):
79
+ """
80
+ A graphical tool for exploring the recursive descent parser. The tool
81
+ displays the parser's tree and the remaining text, and allows the
82
+ user to control the parser's operation. In particular, the user
83
+ can expand subtrees on the frontier, match tokens on the frontier
84
+ against the text, and backtrack. A "step" button simply steps
85
+ through the parsing process, performing the operations that
86
+ ``RecursiveDescentParser`` would use.
87
+ """
88
+ def __init__(self, grammar, sent, trace=0):
89
+ self._sent = sent
90
+ self._parser = SteppingRecursiveDescentParser(grammar, trace)
91
+
92
+ # Set up the main window.
93
+ self._top = Tk()
94
+ self._top.title('Recursive Descent Parser Application')
95
+
96
+ # Set up key bindings.
97
+ self._init_bindings()
98
+
99
+ # Initialize the fonts.
100
+ self._init_fonts(self._top)
101
+
102
+ # Animations. animating_lock is a lock to prevent the demo
103
+ # from performing new operations while it's animating.
104
+ self._animation_frames = IntVar(self._top)
105
+ self._animation_frames.set(5)
106
+ self._animating_lock = 0
107
+ self._autostep = 0
108
+
109
+ # The user can hide the grammar.
110
+ self._show_grammar = IntVar(self._top)
111
+ self._show_grammar.set(1)
112
+
113
+ # Create the basic frames.
114
+ self._init_menubar(self._top)
115
+ self._init_buttons(self._top)
116
+ self._init_feedback(self._top)
117
+ self._init_grammar(self._top)
118
+ self._init_canvas(self._top)
119
+
120
+ # Initialize the parser.
121
+ self._parser.initialize(self._sent)
122
+
123
+ # Resize callback
124
+ self._canvas.bind('<Configure>', self._configure)
125
+
126
+ #########################################
127
+ ## Initialization Helpers
128
+ #########################################
129
+
130
+ def _init_fonts(self, root):
131
+ # See: <http://www.astro.washington.edu/owen/ROTKFolklore.html>
132
+ self._sysfont = Font(font=Button()["font"])
133
+ root.option_add("*Font", self._sysfont)
134
+
135
+ # TWhat's our font size (default=same as sysfont)
136
+ self._size = IntVar(root)
137
+ self._size.set(self._sysfont.cget('size'))
138
+
139
+ self._boldfont = Font(family='helvetica', weight='bold',
140
+ size=self._size.get())
141
+ self._font = Font(family='helvetica',
142
+ size=self._size.get())
143
+ if self._size.get() < 0: big = self._size.get()-2
144
+ else: big = self._size.get()+2
145
+ self._bigfont = Font(family='helvetica', weight='bold',
146
+ size=big)
147
+
148
+ def _init_grammar(self, parent):
149
+ # Grammar view.
150
+ self._prodframe = listframe = Frame(parent)
151
+ self._prodframe.pack(fill='both', side='left', padx=2)
152
+ self._prodlist_label = Label(self._prodframe, font=self._boldfont,
153
+ text='Available Expansions')
154
+ self._prodlist_label.pack()
155
+ self._prodlist = Listbox(self._prodframe, selectmode='single',
156
+ relief='groove', background='white',
157
+ foreground='#909090', font=self._font,
158
+ selectforeground='#004040',
159
+ selectbackground='#c0f0c0')
160
+
161
+ self._prodlist.pack(side='right', fill='both', expand=1)
162
+
163
+ self._productions = list(self._parser.grammar().productions())
164
+ for production in self._productions:
165
+ self._prodlist.insert('end', (' %s' % production))
166
+ self._prodlist.config(height=min(len(self._productions), 25))
167
+
168
+ # Add a scrollbar if there are more than 25 productions.
169
+ if len(self._productions) > 25:
170
+ listscroll = Scrollbar(self._prodframe,
171
+ orient='vertical')
172
+ self._prodlist.config(yscrollcommand = listscroll.set)
173
+ listscroll.config(command=self._prodlist.yview)
174
+ listscroll.pack(side='left', fill='y')
175
+
176
+ # If they select a production, apply it.
177
+ self._prodlist.bind('<<ListboxSelect>>', self._prodlist_select)
178
+
179
+ def _init_bindings(self):
180
+ # Key bindings are a good thing.
181
+ self._top.bind('<Control-q>', self.destroy)
182
+ self._top.bind('<Control-x>', self.destroy)
183
+ self._top.bind('<Escape>', self.destroy)
184
+ self._top.bind('e', self.expand)
185
+ #self._top.bind('<Alt-e>', self.expand)
186
+ #self._top.bind('<Control-e>', self.expand)
187
+ self._top.bind('m', self.match)
188
+ self._top.bind('<Alt-m>', self.match)
189
+ self._top.bind('<Control-m>', self.match)
190
+ self._top.bind('b', self.backtrack)
191
+ self._top.bind('<Alt-b>', self.backtrack)
192
+ self._top.bind('<Control-b>', self.backtrack)
193
+ self._top.bind('<Control-z>', self.backtrack)
194
+ self._top.bind('<BackSpace>', self.backtrack)
195
+ self._top.bind('a', self.autostep)
196
+ #self._top.bind('<Control-a>', self.autostep)
197
+ self._top.bind('<Control-space>', self.autostep)
198
+ self._top.bind('<Control-c>', self.cancel_autostep)
199
+ self._top.bind('<space>', self.step)
200
+ self._top.bind('<Delete>', self.reset)
201
+ self._top.bind('<Control-p>', self.postscript)
202
+ #self._top.bind('<h>', self.help)
203
+ #self._top.bind('<Alt-h>', self.help)
204
+ self._top.bind('<Control-h>', self.help)
205
+ self._top.bind('<F1>', self.help)
206
+ #self._top.bind('<g>', self.toggle_grammar)
207
+ #self._top.bind('<Alt-g>', self.toggle_grammar)
208
+ #self._top.bind('<Control-g>', self.toggle_grammar)
209
+ self._top.bind('<Control-g>', self.edit_grammar)
210
+ self._top.bind('<Control-t>', self.edit_sentence)
211
+
212
+ def _init_buttons(self, parent):
213
+ # Set up the frames.
214
+ self._buttonframe = buttonframe = Frame(parent)
215
+ buttonframe.pack(fill='none', side='bottom', padx=3, pady=2)
216
+ Button(buttonframe, text='Step',
217
+ background='#90c0d0', foreground='black',
218
+ command=self.step,).pack(side='left')
219
+ Button(buttonframe, text='Autostep',
220
+ background='#90c0d0', foreground='black',
221
+ command=self.autostep,).pack(side='left')
222
+ Button(buttonframe, text='Expand', underline=0,
223
+ background='#90f090', foreground='black',
224
+ command=self.expand).pack(side='left')
225
+ Button(buttonframe, text='Match', underline=0,
226
+ background='#90f090', foreground='black',
227
+ command=self.match).pack(side='left')
228
+ Button(buttonframe, text='Backtrack', underline=0,
229
+ background='#f0a0a0', foreground='black',
230
+ command=self.backtrack).pack(side='left')
231
+ # Replace autostep...
232
+ # self._autostep_button = Button(buttonframe, text='Autostep',
233
+ # underline=0, command=self.autostep)
234
+ # self._autostep_button.pack(side='left')
235
+
236
+ def _configure(self, event):
237
+ self._autostep = 0
238
+ (x1, y1, x2, y2) = self._cframe.scrollregion()
239
+ y2 = event.height - 6
240
+ self._canvas['scrollregion'] = '%d %d %d %d' % (x1,y1,x2,y2)
241
+ self._redraw()
242
+
243
+ def _init_feedback(self, parent):
244
+ self._feedbackframe = feedbackframe = Frame(parent)
245
+ feedbackframe.pack(fill='x', side='bottom', padx=3, pady=3)
246
+ self._lastoper_label = Label(feedbackframe, text='Last Operation:',
247
+ font=self._font)
248
+ self._lastoper_label.pack(side='left')
249
+ lastoperframe = Frame(feedbackframe, relief='sunken', border=1)
250
+ lastoperframe.pack(fill='x', side='right', expand=1, padx=5)
251
+ self._lastoper1 = Label(lastoperframe, foreground='#007070',
252
+ background='#f0f0f0', font=self._font)
253
+ self._lastoper2 = Label(lastoperframe, anchor='w', width=30,
254
+ foreground='#004040', background='#f0f0f0',
255
+ font=self._font)
256
+ self._lastoper1.pack(side='left')
257
+ self._lastoper2.pack(side='left', fill='x', expand=1)
258
+
259
+ def _init_canvas(self, parent):
260
+ self._cframe = CanvasFrame(parent, background='white',
261
+ #width=525, height=250,
262
+ closeenough=10,
263
+ border=2, relief='sunken')
264
+ self._cframe.pack(expand=1, fill='both', side='top', pady=2)
265
+ canvas = self._canvas = self._cframe.canvas()
266
+
267
+ # Initially, there's no tree or text
268
+ self._tree = None
269
+ self._textwidgets = []
270
+ self._textline = None
271
+
272
+ def _init_menubar(self, parent):
273
+ menubar = Menu(parent)
274
+
275
+ filemenu = Menu(menubar, tearoff=0)
276
+ filemenu.add_command(label='Reset Parser', underline=0,
277
+ command=self.reset, accelerator='Del')
278
+ filemenu.add_command(label='Print to Postscript', underline=0,
279
+ command=self.postscript, accelerator='Ctrl-p')
280
+ filemenu.add_command(label='Exit', underline=1,
281
+ command=self.destroy, accelerator='Ctrl-x')
282
+ menubar.add_cascade(label='File', underline=0, menu=filemenu)
283
+
284
+ editmenu = Menu(menubar, tearoff=0)
285
+ editmenu.add_command(label='Edit Grammar', underline=5,
286
+ command=self.edit_grammar,
287
+ accelerator='Ctrl-g')
288
+ editmenu.add_command(label='Edit Text', underline=5,
289
+ command=self.edit_sentence,
290
+ accelerator='Ctrl-t')
291
+ menubar.add_cascade(label='Edit', underline=0, menu=editmenu)
292
+
293
+ rulemenu = Menu(menubar, tearoff=0)
294
+ rulemenu.add_command(label='Step', underline=1,
295
+ command=self.step, accelerator='Space')
296
+ rulemenu.add_separator()
297
+ rulemenu.add_command(label='Match', underline=0,
298
+ command=self.match, accelerator='Ctrl-m')
299
+ rulemenu.add_command(label='Expand', underline=0,
300
+ command=self.expand, accelerator='Ctrl-e')
301
+ rulemenu.add_separator()
302
+ rulemenu.add_command(label='Backtrack', underline=0,
303
+ command=self.backtrack, accelerator='Ctrl-b')
304
+ menubar.add_cascade(label='Apply', underline=0, menu=rulemenu)
305
+
306
+ viewmenu = Menu(menubar, tearoff=0)
307
+ viewmenu.add_checkbutton(label="Show Grammar", underline=0,
308
+ variable=self._show_grammar,
309
+ command=self._toggle_grammar)
310
+ viewmenu.add_separator()
311
+ viewmenu.add_radiobutton(label='Tiny', variable=self._size,
312
+ underline=0, value=10, command=self.resize)
313
+ viewmenu.add_radiobutton(label='Small', variable=self._size,
314
+ underline=0, value=12, command=self.resize)
315
+ viewmenu.add_radiobutton(label='Medium', variable=self._size,
316
+ underline=0, value=14, command=self.resize)
317
+ viewmenu.add_radiobutton(label='Large', variable=self._size,
318
+ underline=0, value=18, command=self.resize)
319
+ viewmenu.add_radiobutton(label='Huge', variable=self._size,
320
+ underline=0, value=24, command=self.resize)
321
+ menubar.add_cascade(label='View', underline=0, menu=viewmenu)
322
+
323
+ animatemenu = Menu(menubar, tearoff=0)
324
+ animatemenu.add_radiobutton(label="No Animation", underline=0,
325
+ variable=self._animation_frames,
326
+ value=0)
327
+ animatemenu.add_radiobutton(label="Slow Animation", underline=0,
328
+ variable=self._animation_frames,
329
+ value=10, accelerator='-')
330
+ animatemenu.add_radiobutton(label="Normal Animation", underline=0,
331
+ variable=self._animation_frames,
332
+ value=5, accelerator='=')
333
+ animatemenu.add_radiobutton(label="Fast Animation", underline=0,
334
+ variable=self._animation_frames,
335
+ value=2, accelerator='+')
336
+ menubar.add_cascade(label="Animate", underline=1, menu=animatemenu)
337
+
338
+
339
+ helpmenu = Menu(menubar, tearoff=0)
340
+ helpmenu.add_command(label='About', underline=0,
341
+ command=self.about)
342
+ helpmenu.add_command(label='Instructions', underline=0,
343
+ command=self.help, accelerator='F1')
344
+ menubar.add_cascade(label='Help', underline=0, menu=helpmenu)
345
+
346
+ parent.config(menu=menubar)
347
+
348
+ #########################################
349
+ ## Helper
350
+ #########################################
351
+
352
+ def _get(self, widget, treeloc):
353
+ for i in treeloc: widget = widget.subtrees()[i]
354
+ if isinstance(widget, TreeSegmentWidget):
355
+ widget = widget.label()
356
+ return widget
357
+
358
+ #########################################
359
+ ## Main draw procedure
360
+ #########################################
361
+
362
+ def _redraw(self):
363
+ canvas = self._canvas
364
+
365
+ # Delete the old tree, widgets, etc.
366
+ if self._tree is not None:
367
+ self._cframe.destroy_widget(self._tree)
368
+ for twidget in self._textwidgets:
369
+ self._cframe.destroy_widget(twidget)
370
+ if self._textline is not None:
371
+ self._canvas.delete(self._textline)
372
+
373
+ # Draw the tree.
374
+ helv = ('helvetica', -self._size.get())
375
+ bold = ('helvetica', -self._size.get(), 'bold')
376
+ attribs = {'tree_color': '#000000', 'tree_width': 2,
377
+ 'node_font': bold, 'leaf_font': helv,}
378
+ tree = self._parser.tree()
379
+ self._tree = tree_to_treesegment(canvas, tree, **attribs)
380
+ self._cframe.add_widget(self._tree, 30, 5)
381
+
382
+ # Draw the text.
383
+ helv = ('helvetica', -self._size.get())
384
+ bottom = y = self._cframe.scrollregion()[3]
385
+ self._textwidgets = [TextWidget(canvas, word, font=self._font)
386
+ for word in self._sent]
387
+ for twidget in self._textwidgets:
388
+ self._cframe.add_widget(twidget, 0, 0)
389
+ twidget.move(0, bottom-twidget.bbox()[3]-5)
390
+ y = min(y, twidget.bbox()[1])
391
+
392
+ # Draw a line over the text, to separate it from the tree.
393
+ self._textline = canvas.create_line(-5000, y-5, 5000, y-5, dash='.')
394
+
395
+ # Highlight appropriate nodes.
396
+ self._highlight_nodes()
397
+ self._highlight_prodlist()
398
+
399
+ # Make sure the text lines up.
400
+ self._position_text()
401
+
402
+
403
+ def _redraw_quick(self):
404
+ # This should be more-or-less sufficient after an animation.
405
+ self._highlight_nodes()
406
+ self._highlight_prodlist()
407
+ self._position_text()
408
+
409
+ def _highlight_nodes(self):
410
+ # Highlight the list of nodes to be checked.
411
+ bold = ('helvetica', -self._size.get(), 'bold')
412
+ for treeloc in self._parser.frontier()[:1]:
413
+ self._get(self._tree, treeloc)['color'] = '#20a050'
414
+ self._get(self._tree, treeloc)['font'] = bold
415
+ for treeloc in self._parser.frontier()[1:]:
416
+ self._get(self._tree, treeloc)['color'] = '#008080'
417
+
418
+ def _highlight_prodlist(self):
419
+ # Highlight the productions that can be expanded.
420
+ # Boy, too bad tkinter doesn't implement Listbox.itemconfig;
421
+ # that would be pretty useful here.
422
+ self._prodlist.delete(0, 'end')
423
+ expandable = self._parser.expandable_productions()
424
+ untried = self._parser.untried_expandable_productions()
425
+ productions = self._productions
426
+ for index in range(len(productions)):
427
+ if productions[index] in expandable:
428
+ if productions[index] in untried:
429
+ self._prodlist.insert(index, ' %s' % productions[index])
430
+ else:
431
+ self._prodlist.insert(index, ' %s (TRIED)' %
432
+ productions[index])
433
+ self._prodlist.selection_set(index)
434
+ else:
435
+ self._prodlist.insert(index, ' %s' % productions[index])
436
+
437
+ def _position_text(self):
438
+ # Line up the text widgets that are matched against the tree
439
+ numwords = len(self._sent)
440
+ num_matched = numwords - len(self._parser.remaining_text())
441
+ leaves = self._tree_leaves()[:num_matched]
442
+ xmax = self._tree.bbox()[0]
443
+ for i in range(0, len(leaves)):
444
+ widget = self._textwidgets[i]
445
+ leaf = leaves[i]
446
+ widget['color'] = '#006040'
447
+ leaf['color'] = '#006040'
448
+ widget.move(leaf.bbox()[0] - widget.bbox()[0], 0)
449
+ xmax = widget.bbox()[2] + 10
450
+
451
+ # Line up the text widgets that are not matched against the tree.
452
+ for i in range(len(leaves), numwords):
453
+ widget = self._textwidgets[i]
454
+ widget['color'] = '#a0a0a0'
455
+ widget.move(xmax - widget.bbox()[0], 0)
456
+ xmax = widget.bbox()[2] + 10
457
+
458
+ # If we have a complete parse, make everything green :)
459
+ if self._parser.currently_complete():
460
+ for twidget in self._textwidgets:
461
+ twidget['color'] = '#00a000'
462
+
463
+ # Move the matched leaves down to the text.
464
+ for i in range(0, len(leaves)):
465
+ widget = self._textwidgets[i]
466
+ leaf = leaves[i]
467
+ dy = widget.bbox()[1] - leaf.bbox()[3] - 10.0
468
+ dy = max(dy, leaf.parent().label().bbox()[3] - leaf.bbox()[3] + 10)
469
+ leaf.move(0, dy)
470
+
471
+ def _tree_leaves(self, tree=None):
472
+ if tree is None: tree = self._tree
473
+ if isinstance(tree, TreeSegmentWidget):
474
+ leaves = []
475
+ for child in tree.subtrees(): leaves += self._tree_leaves(child)
476
+ return leaves
477
+ else:
478
+ return [tree]
479
+
480
+ #########################################
481
+ ## Button Callbacks
482
+ #########################################
483
+
484
+ def destroy(self, *e):
485
+ self._autostep = 0
486
+ if self._top is None: return
487
+ self._top.destroy()
488
+ self._top = None
489
+
490
+ def reset(self, *e):
491
+ self._autostep = 0
492
+ self._parser.initialize(self._sent)
493
+ self._lastoper1['text'] = 'Reset Application'
494
+ self._lastoper2['text'] = ''
495
+ self._redraw()
496
+
497
+ def autostep(self, *e):
498
+ if self._animation_frames.get() == 0:
499
+ self._animation_frames.set(2)
500
+ if self._autostep:
501
+ self._autostep = 0
502
+ else:
503
+ self._autostep = 1
504
+ self._step()
505
+
506
+ def cancel_autostep(self, *e):
507
+ #self._autostep_button['text'] = 'Autostep'
508
+ self._autostep = 0
509
+
510
+ # Make sure to stop auto-stepping if we get any user input.
511
+ def step(self, *e): self._autostep = 0; self._step()
512
+ def match(self, *e): self._autostep = 0; self._match()
513
+ def expand(self, *e): self._autostep = 0; self._expand()
514
+ def backtrack(self, *e): self._autostep = 0; self._backtrack()
515
+
516
+ def _step(self):
517
+ if self._animating_lock: return
518
+
519
+ # Try expanding, matching, and backtracking (in that order)
520
+ if self._expand(): pass
521
+ elif self._parser.untried_match() and self._match(): pass
522
+ elif self._backtrack(): pass
523
+ else:
524
+ self._lastoper1['text'] = 'Finished'
525
+ self._lastoper2['text'] = ''
526
+ self._autostep = 0
527
+
528
+ # Check if we just completed a parse.
529
+ if self._parser.currently_complete():
530
+ self._autostep = 0
531
+ self._lastoper2['text'] += ' [COMPLETE PARSE]'
532
+
533
+ def _expand(self, *e):
534
+ if self._animating_lock: return
535
+ old_frontier = self._parser.frontier()
536
+ rv = self._parser.expand()
537
+ if rv is not None:
538
+ self._lastoper1['text'] = 'Expand:'
539
+ self._lastoper2['text'] = rv
540
+ self._prodlist.selection_clear(0, 'end')
541
+ index = self._productions.index(rv)
542
+ self._prodlist.selection_set(index)
543
+ self._animate_expand(old_frontier[0])
544
+ return True
545
+ else:
546
+ self._lastoper1['text'] = 'Expand:'
547
+ self._lastoper2['text'] = '(all expansions tried)'
548
+ return False
549
+
550
+ def _match(self, *e):
551
+ if self._animating_lock: return
552
+ old_frontier = self._parser.frontier()
553
+ rv = self._parser.match()
554
+ if rv is not None:
555
+ self._lastoper1['text'] = 'Match:'
556
+ self._lastoper2['text'] = rv
557
+ self._animate_match(old_frontier[0])
558
+ return True
559
+ else:
560
+ self._lastoper1['text'] = 'Match:'
561
+ self._lastoper2['text'] = '(failed)'
562
+ return False
563
+
564
+ def _backtrack(self, *e):
565
+ if self._animating_lock: return
566
+ if self._parser.backtrack():
567
+ elt = self._parser.tree()
568
+ for i in self._parser.frontier()[0]:
569
+ elt = elt[i]
570
+ self._lastoper1['text'] = 'Backtrack'
571
+ self._lastoper2['text'] = ''
572
+ if isinstance(elt, Tree):
573
+ self._animate_backtrack(self._parser.frontier()[0])
574
+ else:
575
+ self._animate_match_backtrack(self._parser.frontier()[0])
576
+ return True
577
+ else:
578
+ self._autostep = 0
579
+ self._lastoper1['text'] = 'Finished'
580
+ self._lastoper2['text'] = ''
581
+ return False
582
+
583
+ def about(self, *e):
584
+ ABOUT = ("NLTK Recursive Descent Parser Application\n"+
585
+ "Written by Edward Loper")
586
+ TITLE = 'About: Recursive Descent Parser Application'
587
+ try:
588
+ from six.moves.tkinter_messagebox import Message
589
+ Message(message=ABOUT, title=TITLE).show()
590
+ except:
591
+ ShowText(self._top, TITLE, ABOUT)
592
+
593
+ def help(self, *e):
594
+ self._autostep = 0
595
+ # The default font's not very legible; try using 'fixed' instead.
596
+ try:
597
+ ShowText(self._top, 'Help: Recursive Descent Parser Application',
598
+ (__doc__ or '').strip(), width=75, font='fixed')
599
+ except:
600
+ ShowText(self._top, 'Help: Recursive Descent Parser Application',
601
+ (__doc__ or '').strip(), width=75)
602
+
603
+ def postscript(self, *e):
604
+ self._autostep = 0
605
+ self._cframe.print_to_file()
606
+
607
+ def mainloop(self, *args, **kwargs):
608
+ """
609
+ Enter the Tkinter mainloop. This function must be called if
610
+ this demo is created from a non-interactive program (e.g.
611
+ from a secript); otherwise, the demo will close as soon as
612
+ the script completes.
613
+ """
614
+ if in_idle(): return
615
+ self._top.mainloop(*args, **kwargs)
616
+
617
+ def resize(self, size=None):
618
+ if size is not None: self._size.set(size)
619
+ size = self._size.get()
620
+ self._font.configure(size=-(abs(size)))
621
+ self._boldfont.configure(size=-(abs(size)))
622
+ self._sysfont.configure(size=-(abs(size)))
623
+ self._bigfont.configure(size=-(abs(size+2)))
624
+ self._redraw()
625
+
626
+ #########################################
627
+ ## Expand Production Selection
628
+ #########################################
629
+
630
+ def _toggle_grammar(self, *e):
631
+ if self._show_grammar.get():
632
+ self._prodframe.pack(fill='both', side='left', padx=2,
633
+ after=self._feedbackframe)
634
+ self._lastoper1['text'] = 'Show Grammar'
635
+ else:
636
+ self._prodframe.pack_forget()
637
+ self._lastoper1['text'] = 'Hide Grammar'
638
+ self._lastoper2['text'] = ''
639
+
640
+ # def toggle_grammar(self, *e):
641
+ # self._show_grammar = not self._show_grammar
642
+ # if self._show_grammar:
643
+ # self._prodframe.pack(fill='both', expand='y', side='left',
644
+ # after=self._feedbackframe)
645
+ # self._lastoper1['text'] = 'Show Grammar'
646
+ # else:
647
+ # self._prodframe.pack_forget()
648
+ # self._lastoper1['text'] = 'Hide Grammar'
649
+ # self._lastoper2['text'] = ''
650
+
651
+ def _prodlist_select(self, event):
652
+ selection = self._prodlist.curselection()
653
+ if len(selection) != 1: return
654
+ index = int(selection[0])
655
+ old_frontier = self._parser.frontier()
656
+ production = self._parser.expand(self._productions[index])
657
+
658
+ if production:
659
+ self._lastoper1['text'] = 'Expand:'
660
+ self._lastoper2['text'] = production
661
+ self._prodlist.selection_clear(0, 'end')
662
+ self._prodlist.selection_set(index)
663
+ self._animate_expand(old_frontier[0])
664
+ else:
665
+ # Reset the production selections.
666
+ self._prodlist.selection_clear(0, 'end')
667
+ for prod in self._parser.expandable_productions():
668
+ index = self._productions.index(prod)
669
+ self._prodlist.selection_set(index)
670
+
671
+ #########################################
672
+ ## Animation
673
+ #########################################
674
+
675
+ def _animate_expand(self, treeloc):
676
+ oldwidget = self._get(self._tree, treeloc)
677
+ oldtree = oldwidget.parent()
678
+ top = not isinstance(oldtree.parent(), TreeSegmentWidget)
679
+
680
+ tree = self._parser.tree()
681
+ for i in treeloc:
682
+ tree = tree[i]
683
+
684
+ widget = tree_to_treesegment(self._canvas, tree,
685
+ node_font=self._boldfont,
686
+ leaf_color='white',
687
+ tree_width=2, tree_color='white',
688
+ node_color='white',
689
+ leaf_font=self._font)
690
+ widget.label()['color'] = '#20a050'
691
+
692
+ (oldx, oldy) = oldtree.label().bbox()[:2]
693
+ (newx, newy) = widget.label().bbox()[:2]
694
+ widget.move(oldx-newx, oldy-newy)
695
+
696
+ if top:
697
+ self._cframe.add_widget(widget, 0, 5)
698
+ widget.move(30-widget.label().bbox()[0], 0)
699
+ self._tree = widget
700
+ else:
701
+ oldtree.parent().replace_child(oldtree, widget)
702
+
703
+ # Move the children over so they don't overlap.
704
+ # Line the children up in a strange way.
705
+ if widget.subtrees():
706
+ dx = (oldx + widget.label().width()/2 -
707
+ widget.subtrees()[0].bbox()[0]/2 -
708
+ widget.subtrees()[0].bbox()[2]/2)
709
+ for subtree in widget.subtrees(): subtree.move(dx, 0)
710
+
711
+ self._makeroom(widget)
712
+
713
+ if top:
714
+ self._cframe.destroy_widget(oldtree)
715
+ else:
716
+ oldtree.destroy()
717
+
718
+ colors = ['gray%d' % (10*int(10*x/self._animation_frames.get()))
719
+ for x in range(self._animation_frames.get(),0,-1)]
720
+
721
+ # Move the text string down, if necessary.
722
+ dy = widget.bbox()[3] + 30 - self._canvas.coords(self._textline)[1]
723
+ if dy > 0:
724
+ for twidget in self._textwidgets: twidget.move(0, dy)
725
+ self._canvas.move(self._textline, 0, dy)
726
+
727
+ self._animate_expand_frame(widget, colors)
728
+
729
+ def _makeroom(self, treeseg):
730
+ """
731
+ Make sure that no sibling tree bbox's overlap.
732
+ """
733
+ parent = treeseg.parent()
734
+ if not isinstance(parent, TreeSegmentWidget): return
735
+
736
+ index = parent.subtrees().index(treeseg)
737
+
738
+ # Handle siblings to the right
739
+ rsiblings = parent.subtrees()[index+1:]
740
+ if rsiblings:
741
+ dx = treeseg.bbox()[2] - rsiblings[0].bbox()[0] + 10
742
+ for sibling in rsiblings: sibling.move(dx, 0)
743
+
744
+ # Handle siblings to the left
745
+ if index > 0:
746
+ lsibling = parent.subtrees()[index-1]
747
+ dx = max(0, lsibling.bbox()[2] - treeseg.bbox()[0] + 10)
748
+ treeseg.move(dx, 0)
749
+
750
+ # Keep working up the tree.
751
+ self._makeroom(parent)
752
+
753
+ def _animate_expand_frame(self, widget, colors):
754
+ if len(colors) > 0:
755
+ self._animating_lock = 1
756
+ widget['color'] = colors[0]
757
+ for subtree in widget.subtrees():
758
+ if isinstance(subtree, TreeSegmentWidget):
759
+ subtree.label()['color'] = colors[0]
760
+ else:
761
+ subtree['color'] = colors[0]
762
+ self._top.after(50, self._animate_expand_frame,
763
+ widget, colors[1:])
764
+ else:
765
+ widget['color'] = 'black'
766
+ for subtree in widget.subtrees():
767
+ if isinstance(subtree, TreeSegmentWidget):
768
+ subtree.label()['color'] = 'black'
769
+ else:
770
+ subtree['color'] = 'black'
771
+ self._redraw_quick()
772
+ widget.label()['color'] = 'black'
773
+ self._animating_lock = 0
774
+ if self._autostep: self._step()
775
+
776
+ def _animate_backtrack(self, treeloc):
777
+ # Flash red first, if we're animating.
778
+ if self._animation_frames.get() == 0: colors = []
779
+ else: colors = ['#a00000', '#000000', '#a00000']
780
+ colors += ['gray%d' % (10*int(10*x/(self._animation_frames.get())))
781
+ for x in range(1, self._animation_frames.get()+1)]
782
+
783
+ widgets = [self._get(self._tree, treeloc).parent()]
784
+ for subtree in widgets[0].subtrees():
785
+ if isinstance(subtree, TreeSegmentWidget):
786
+ widgets.append(subtree.label())
787
+ else:
788
+ widgets.append(subtree)
789
+
790
+ self._animate_backtrack_frame(widgets, colors)
791
+
792
+ def _animate_backtrack_frame(self, widgets, colors):
793
+ if len(colors) > 0:
794
+ self._animating_lock = 1
795
+ for widget in widgets: widget['color'] = colors[0]
796
+ self._top.after(50, self._animate_backtrack_frame,
797
+ widgets, colors[1:])
798
+ else:
799
+ for widget in widgets[0].subtrees():
800
+ widgets[0].remove_child(widget)
801
+ widget.destroy()
802
+ self._redraw_quick()
803
+ self._animating_lock = 0
804
+ if self._autostep: self._step()
805
+
806
+ def _animate_match_backtrack(self, treeloc):
807
+ widget = self._get(self._tree, treeloc)
808
+ node = widget.parent().label()
809
+ dy = ((node.bbox()[3] - widget.bbox()[1] + 14) /
810
+ max(1, self._animation_frames.get()))
811
+ self._animate_match_backtrack_frame(self._animation_frames.get(),
812
+ widget, dy)
813
+
814
+ def _animate_match(self, treeloc):
815
+ widget = self._get(self._tree, treeloc)
816
+
817
+ dy = ((self._textwidgets[0].bbox()[1] - widget.bbox()[3] - 10.0) /
818
+ max(1, self._animation_frames.get()))
819
+ self._animate_match_frame(self._animation_frames.get(), widget, dy)
820
+
821
+ def _animate_match_frame(self, frame, widget, dy):
822
+ if frame > 0:
823
+ self._animating_lock = 1
824
+ widget.move(0, dy)
825
+ self._top.after(10, self._animate_match_frame,
826
+ frame-1, widget, dy)
827
+ else:
828
+ widget['color'] = '#006040'
829
+ self._redraw_quick()
830
+ self._animating_lock = 0
831
+ if self._autostep: self._step()
832
+
833
+ def _animate_match_backtrack_frame(self, frame, widget, dy):
834
+ if frame > 0:
835
+ self._animating_lock = 1
836
+ widget.move(0, dy)
837
+ self._top.after(10, self._animate_match_backtrack_frame,
838
+ frame-1, widget, dy)
839
+ else:
840
+ widget.parent().remove_child(widget)
841
+ widget.destroy()
842
+ self._animating_lock = 0
843
+ if self._autostep: self._step()
844
+
845
+ def edit_grammar(self, *e):
846
+ CFGEditor(self._top, self._parser.grammar(), self.set_grammar)
847
+
848
+ def set_grammar(self, grammar):
849
+ self._parser.set_grammar(grammar)
850
+ self._productions = list(grammar.productions())
851
+ self._prodlist.delete(0, 'end')
852
+ for production in self._productions:
853
+ self._prodlist.insert('end', (' %s' % production))
854
+
855
+ def edit_sentence(self, *e):
856
+ sentence = " ".join(self._sent)
857
+ title = 'Edit Text'
858
+ instr = 'Enter a new sentence to parse.'
859
+ EntryDialog(self._top, sentence, instr, self.set_sentence, title)
860
+
861
+ def set_sentence(self, sentence):
862
+ self._sent = sentence.split() #[XX] use tagged?
863
+ self.reset()
864
+
865
+ def app():
866
+ """
867
+ Create a recursive descent parser demo, using a simple grammar and
868
+ text.
869
+ """
870
+ from nltk.grammar import CFG
871
+ grammar = CFG.fromstring("""
872
+ # Grammatical productions.
873
+ S -> NP VP
874
+ NP -> Det N PP | Det N
875
+ VP -> V NP PP | V NP | V
876
+ PP -> P NP
877
+ # Lexical productions.
878
+ NP -> 'I'
879
+ Det -> 'the' | 'a'
880
+ N -> 'man' | 'park' | 'dog' | 'telescope'
881
+ V -> 'ate' | 'saw'
882
+ P -> 'in' | 'under' | 'with'
883
+ """)
884
+
885
+ sent = 'the dog saw a man in the park'.split()
886
+
887
+ RecursiveDescentApp(grammar, sent).mainloop()
888
+
889
+ if __name__ == '__main__':
890
+ app()
891
+
892
+ __all__ = ['app']