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
@@ -1,809 +1,809 @@
1
- # Natural Language Toolkit: Shift-Reduce 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 shift-reduce parser.
10
-
11
- The shift-reduce parser maintains a stack, which records the structure
12
- of the portion of the text that has been parsed. The stack is
13
- initially empty. Its contents are shown on the left side of the main
14
- canvas.
15
-
16
- On the right side of the main canvas is the remaining text. This is
17
- the portion of the text which has not yet been considered by the
18
- parser.
19
-
20
- The parser builds up a tree structure for the text using two
21
- operations:
22
-
23
- - "shift" moves the first token from the remaining text to the top
24
- of the stack. In the demo, the top of the stack is its right-hand
25
- side.
26
- - "reduce" uses a grammar production to combine the rightmost stack
27
- elements into a single tree token.
28
-
29
- You can control the parser's operation by using the "shift" and
30
- "reduce" buttons; or you can use the "step" button to let the parser
31
- automatically decide which operation to apply. The parser uses the
32
- following rules to decide which operation to apply:
33
-
34
- - Only shift if no reductions are available.
35
- - If multiple reductions are available, then apply the reduction
36
- whose CFG production is listed earliest in the grammar.
37
-
38
- The "reduce" button applies the reduction whose CFG production is
39
- listed earliest in the grammar. There are two ways to manually choose
40
- which reduction to apply:
41
-
42
- - Click on a CFG production from the list of available reductions,
43
- on the left side of the main window. The reduction based on that
44
- production will be applied to the top of the stack.
45
- - Click on one of the stack elements. A popup window will appear,
46
- containing all available reductions. Select one, and it will be
47
- applied to the top of the stack.
48
-
49
- Note that reductions can only be applied to the top of the stack.
50
-
51
- Keyboard Shortcuts::
52
- [Space]\t Perform the next shift or reduce operation
53
- [s]\t Perform a shift operation
54
- [r]\t Perform a reduction operation
55
- [Ctrl-z]\t Undo most recent operation
56
- [Delete]\t Reset the parser
57
- [g]\t Show/hide available production list
58
- [Ctrl-a]\t Toggle animations
59
- [h]\t Help
60
- [Ctrl-p]\t Print
61
- [q]\t Quit
62
- """
63
-
64
- """
65
- Possible future improvements:
66
- - button/window to change and/or select text. Just pop up a window
67
- with an entry, and let them modify the text; and then retokenize
68
- it? Maybe give a warning if it contains tokens whose types are
69
- not in the grammar.
70
- - button/window to change and/or select grammar. Select from
71
- several alternative grammars? Or actually change the grammar? If
72
- the later, then I'd want to define nltk.draw.cfg, which would be
73
- responsible for that.
74
- """
75
-
76
- import nltk.compat
77
- import tkinter.font
78
- from tkinter import (IntVar, Listbox, Button, Frame, Label, Menu,
79
- Scrollbar, Tk)
80
-
81
- from nltk.tree import Tree
82
- from nltk.parse import SteppingShiftReduceParser
83
- from nltk.util import in_idle
84
- from nltk.draw.util import CanvasFrame, EntryDialog, ShowText, TextWidget
85
- from nltk.draw import CFGEditor, TreeSegmentWidget, tree_to_treesegment
86
-
87
- class ShiftReduceApp(object):
88
- """
89
- A graphical tool for exploring the shift-reduce parser. The tool
90
- displays the parser's stack and the remaining text, and allows the
91
- user to control the parser's operation. In particular, the user
92
- can shift tokens onto the stack, and can perform reductions on the
93
- top elements of the stack. A "step" button simply steps through
94
- the parsing process, performing the operations that
95
- ``nltk.parse.ShiftReduceParser`` would use.
96
- """
97
- def __init__(self, grammar, sent, trace=0):
98
- self._sent = sent
99
- self._parser = SteppingShiftReduceParser(grammar, trace)
100
-
101
- # Set up the main window.
102
- self._top = Tk()
103
- self._top.title('Shift Reduce Parser Application')
104
-
105
- # Animations. animating_lock is a lock to prevent the demo
106
- # from performing new operations while it's animating.
107
- self._animating_lock = 0
108
- self._animate = IntVar(self._top)
109
- self._animate.set(10) # = medium
110
-
111
- # The user can hide the grammar.
112
- self._show_grammar = IntVar(self._top)
113
- self._show_grammar.set(1)
114
-
115
- # Initialize fonts.
116
- self._init_fonts(self._top)
117
-
118
- # Set up key bindings.
119
- self._init_bindings()
120
-
121
- # Create the basic frames.
122
- self._init_menubar(self._top)
123
- self._init_buttons(self._top)
124
- self._init_feedback(self._top)
125
- self._init_grammar(self._top)
126
- self._init_canvas(self._top)
127
-
128
- # A popup menu for reducing.
129
- self._reduce_menu = Menu(self._canvas, tearoff=0)
130
-
131
- # Reset the demo, and set the feedback frame to empty.
132
- self.reset()
133
- self._lastoper1['text'] = ''
134
-
135
- #########################################
136
- ## Initialization Helpers
137
- #########################################
138
-
139
- def _init_fonts(self, root):
140
- # See: <http://www.astro.washington.edu/owen/ROTKFolklore.html>
141
- self._sysfont = tkinter.font.Font(font=Button()["font"])
142
- root.option_add("*Font", self._sysfont)
143
-
144
- # TWhat's our font size (default=same as sysfont)
145
- self._size = IntVar(root)
146
- self._size.set(self._sysfont.cget('size'))
147
-
148
- self._boldfont = tkinter.font.Font(family='helvetica', weight='bold',
149
- size=self._size.get())
150
- self._font = tkinter.font.Font(family='helvetica',
151
- size=self._size.get())
152
-
153
- def _init_grammar(self, parent):
154
- # Grammar view.
155
- self._prodframe = listframe = Frame(parent)
156
- self._prodframe.pack(fill='both', side='left', padx=2)
157
- self._prodlist_label = Label(self._prodframe,
158
- font=self._boldfont,
159
- text='Available Reductions')
160
- self._prodlist_label.pack()
161
- self._prodlist = Listbox(self._prodframe, selectmode='single',
162
- relief='groove', background='white',
163
- foreground='#909090',
164
- font=self._font,
165
- selectforeground='#004040',
166
- selectbackground='#c0f0c0')
167
-
168
- self._prodlist.pack(side='right', fill='both', expand=1)
169
-
170
- self._productions = list(self._parser.grammar().productions())
171
- for production in self._productions:
172
- self._prodlist.insert('end', (' %s' % production))
173
- self._prodlist.config(height=min(len(self._productions), 25))
174
-
175
- # Add a scrollbar if there are more than 25 productions.
176
- if 1:#len(self._productions) > 25:
177
- listscroll = Scrollbar(self._prodframe,
178
- orient='vertical')
179
- self._prodlist.config(yscrollcommand = listscroll.set)
180
- listscroll.config(command=self._prodlist.yview)
181
- listscroll.pack(side='left', fill='y')
182
-
183
- # If they select a production, apply it.
184
- self._prodlist.bind('<<ListboxSelect>>', self._prodlist_select)
185
-
186
- # When they hover over a production, highlight it.
187
- self._hover = -1
188
- self._prodlist.bind('<Motion>', self._highlight_hover)
189
- self._prodlist.bind('<Leave>', self._clear_hover)
190
-
191
- def _init_bindings(self):
192
- # Quit
193
- self._top.bind('<Control-q>', self.destroy)
194
- self._top.bind('<Control-x>', self.destroy)
195
- self._top.bind('<Alt-q>', self.destroy)
196
- self._top.bind('<Alt-x>', self.destroy)
197
-
198
- # Ops (step, shift, reduce, undo)
199
- self._top.bind('<space>', self.step)
200
- self._top.bind('<s>', self.shift)
201
- self._top.bind('<Alt-s>', self.shift)
202
- self._top.bind('<Control-s>', self.shift)
203
- self._top.bind('<r>', self.reduce)
204
- self._top.bind('<Alt-r>', self.reduce)
205
- self._top.bind('<Control-r>', self.reduce)
206
- self._top.bind('<Delete>', self.reset)
207
- self._top.bind('<u>', self.undo)
208
- self._top.bind('<Alt-u>', self.undo)
209
- self._top.bind('<Control-u>', self.undo)
210
- self._top.bind('<Control-z>', self.undo)
211
- self._top.bind('<BackSpace>', self.undo)
212
-
213
- # Misc
214
- self._top.bind('<Control-p>', self.postscript)
215
- self._top.bind('<Control-h>', self.help)
216
- self._top.bind('<F1>', self.help)
217
- self._top.bind('<Control-g>', self.edit_grammar)
218
- self._top.bind('<Control-t>', self.edit_sentence)
219
-
220
- # Animation speed control
221
- self._top.bind('-', lambda e,a=self._animate:a.set(20))
222
- self._top.bind('=', lambda e,a=self._animate:a.set(10))
223
- self._top.bind('+', lambda e,a=self._animate:a.set(4))
224
-
225
- def _init_buttons(self, parent):
226
- # Set up the frames.
227
- self._buttonframe = buttonframe = Frame(parent)
228
- buttonframe.pack(fill='none', side='bottom')
229
- Button(buttonframe, text='Step',
230
- background='#90c0d0', foreground='black',
231
- command=self.step,).pack(side='left')
232
- Button(buttonframe, text='Shift', underline=0,
233
- background='#90f090', foreground='black',
234
- command=self.shift).pack(side='left')
235
- Button(buttonframe, text='Reduce', underline=0,
236
- background='#90f090', foreground='black',
237
- command=self.reduce).pack(side='left')
238
- Button(buttonframe, text='Undo', underline=0,
239
- background='#f0a0a0', foreground='black',
240
- command=self.undo).pack(side='left')
241
-
242
- def _init_menubar(self, parent):
243
- menubar = Menu(parent)
244
-
245
- filemenu = Menu(menubar, tearoff=0)
246
- filemenu.add_command(label='Reset Parser', underline=0,
247
- command=self.reset, accelerator='Del')
248
- filemenu.add_command(label='Print to Postscript', underline=0,
249
- command=self.postscript, accelerator='Ctrl-p')
250
- filemenu.add_command(label='Exit', underline=1,
251
- command=self.destroy, accelerator='Ctrl-x')
252
- menubar.add_cascade(label='File', underline=0, menu=filemenu)
253
-
254
- editmenu = Menu(menubar, tearoff=0)
255
- editmenu.add_command(label='Edit Grammar', underline=5,
256
- command=self.edit_grammar,
257
- accelerator='Ctrl-g')
258
- editmenu.add_command(label='Edit Text', underline=5,
259
- command=self.edit_sentence,
260
- accelerator='Ctrl-t')
261
- menubar.add_cascade(label='Edit', underline=0, menu=editmenu)
262
-
263
- rulemenu = Menu(menubar, tearoff=0)
264
- rulemenu.add_command(label='Step', underline=1,
265
- command=self.step, accelerator='Space')
266
- rulemenu.add_separator()
267
- rulemenu.add_command(label='Shift', underline=0,
268
- command=self.shift, accelerator='Ctrl-s')
269
- rulemenu.add_command(label='Reduce', underline=0,
270
- command=self.reduce, accelerator='Ctrl-r')
271
- rulemenu.add_separator()
272
- rulemenu.add_command(label='Undo', underline=0,
273
- command=self.undo, accelerator='Ctrl-u')
274
- menubar.add_cascade(label='Apply', underline=0, menu=rulemenu)
275
-
276
- viewmenu = Menu(menubar, tearoff=0)
277
- viewmenu.add_checkbutton(label="Show Grammar", underline=0,
278
- variable=self._show_grammar,
279
- command=self._toggle_grammar)
280
- viewmenu.add_separator()
281
- viewmenu.add_radiobutton(label='Tiny', variable=self._size,
282
- underline=0, value=10, command=self.resize)
283
- viewmenu.add_radiobutton(label='Small', variable=self._size,
284
- underline=0, value=12, command=self.resize)
285
- viewmenu.add_radiobutton(label='Medium', variable=self._size,
286
- underline=0, value=14, command=self.resize)
287
- viewmenu.add_radiobutton(label='Large', variable=self._size,
288
- underline=0, value=18, command=self.resize)
289
- viewmenu.add_radiobutton(label='Huge', variable=self._size,
290
- underline=0, value=24, command=self.resize)
291
- menubar.add_cascade(label='View', underline=0, menu=viewmenu)
292
-
293
- animatemenu = Menu(menubar, tearoff=0)
294
- animatemenu.add_radiobutton(label="No Animation", underline=0,
295
- variable=self._animate, value=0)
296
- animatemenu.add_radiobutton(label="Slow Animation", underline=0,
297
- variable=self._animate, value=20,
298
- accelerator='-')
299
- animatemenu.add_radiobutton(label="Normal Animation", underline=0,
300
- variable=self._animate, value=10,
301
- accelerator='=')
302
- animatemenu.add_radiobutton(label="Fast Animation", underline=0,
303
- variable=self._animate, value=4,
304
- accelerator='+')
305
- menubar.add_cascade(label="Animate", underline=1, menu=animatemenu)
306
-
307
-
308
- helpmenu = Menu(menubar, tearoff=0)
309
- helpmenu.add_command(label='About', underline=0,
310
- command=self.about)
311
- helpmenu.add_command(label='Instructions', underline=0,
312
- command=self.help, accelerator='F1')
313
- menubar.add_cascade(label='Help', underline=0, menu=helpmenu)
314
-
315
- parent.config(menu=menubar)
316
-
317
- def _init_feedback(self, parent):
318
- self._feedbackframe = feedbackframe = Frame(parent)
319
- feedbackframe.pack(fill='x', side='bottom', padx=3, pady=3)
320
- self._lastoper_label = Label(feedbackframe, text='Last Operation:',
321
- font=self._font)
322
- self._lastoper_label.pack(side='left')
323
- lastoperframe = Frame(feedbackframe, relief='sunken', border=1)
324
- lastoperframe.pack(fill='x', side='right', expand=1, padx=5)
325
- self._lastoper1 = Label(lastoperframe, foreground='#007070',
326
- background='#f0f0f0', font=self._font)
327
- self._lastoper2 = Label(lastoperframe, anchor='w', width=30,
328
- foreground='#004040', background='#f0f0f0',
329
- font=self._font)
330
- self._lastoper1.pack(side='left')
331
- self._lastoper2.pack(side='left', fill='x', expand=1)
332
-
333
- def _init_canvas(self, parent):
334
- self._cframe = CanvasFrame(parent, background='white',
335
- width=525, closeenough=10,
336
- border=2, relief='sunken')
337
- self._cframe.pack(expand=1, fill='both', side='top', pady=2)
338
- canvas = self._canvas = self._cframe.canvas()
339
-
340
- self._stackwidgets = []
341
- self._rtextwidgets = []
342
- self._titlebar = canvas.create_rectangle(0,0,0,0, fill='#c0f0f0',
343
- outline='black')
344
- self._exprline = canvas.create_line(0,0,0,0, dash='.')
345
- self._stacktop = canvas.create_line(0,0,0,0, fill='#408080')
346
- size = self._size.get()+4
347
- self._stacklabel = TextWidget(canvas, 'Stack', color='#004040',
348
- font=self._boldfont)
349
- self._rtextlabel = TextWidget(canvas, 'Remaining Text',
350
- color='#004040', font=self._boldfont)
351
- self._cframe.add_widget(self._stacklabel)
352
- self._cframe.add_widget(self._rtextlabel)
353
-
354
- #########################################
355
- ## Main draw procedure
356
- #########################################
357
-
358
- def _redraw(self):
359
- scrollregion = self._canvas['scrollregion'].split()
360
- (cx1, cy1, cx2, cy2) = [int(c) for c in scrollregion]
361
-
362
- # Delete the old stack & rtext widgets.
363
- for stackwidget in self._stackwidgets:
364
- self._cframe.destroy_widget(stackwidget)
365
- self._stackwidgets = []
366
- for rtextwidget in self._rtextwidgets:
367
- self._cframe.destroy_widget(rtextwidget)
368
- self._rtextwidgets = []
369
-
370
- # Position the titlebar & exprline
371
- (x1, y1, x2, y2) = self._stacklabel.bbox()
372
- y = y2-y1+10
373
- self._canvas.coords(self._titlebar, -5000, 0, 5000, y-4)
374
- self._canvas.coords(self._exprline, 0, y*2-10, 5000, y*2-10)
375
-
376
- # Position the titlebar labels..
377
- (x1, y1, x2, y2) = self._stacklabel.bbox()
378
- self._stacklabel.move(5-x1, 3-y1)
379
- (x1, y1, x2, y2) = self._rtextlabel.bbox()
380
- self._rtextlabel.move(cx2-x2-5, 3-y1)
381
-
382
- # Draw the stack.
383
- stackx = 5
384
- for tok in self._parser.stack():
385
- if isinstance(tok, Tree):
386
- attribs = {'tree_color': '#4080a0', 'tree_width': 2,
387
- 'node_font': self._boldfont,
388
- 'node_color': '#006060',
389
- 'leaf_color': '#006060', 'leaf_font':self._font}
390
- widget = tree_to_treesegment(self._canvas, tok,
391
- **attribs)
392
- widget.label()['color'] = '#000000'
393
- else:
394
- widget = TextWidget(self._canvas, tok,
395
- color='#000000', font=self._font)
396
- widget.bind_click(self._popup_reduce)
397
- self._stackwidgets.append(widget)
398
- self._cframe.add_widget(widget, stackx, y)
399
- stackx = widget.bbox()[2] + 10
400
-
401
- # Draw the remaining text.
402
- rtextwidth = 0
403
- for tok in self._parser.remaining_text():
404
- widget = TextWidget(self._canvas, tok,
405
- color='#000000', font=self._font)
406
- self._rtextwidgets.append(widget)
407
- self._cframe.add_widget(widget, rtextwidth, y)
408
- rtextwidth = widget.bbox()[2] + 4
409
-
410
- # Allow enough room to shift the next token (for animations)
411
- if len(self._rtextwidgets) > 0:
412
- stackx += self._rtextwidgets[0].width()
413
-
414
- # Move the remaining text to the correct location (keep it
415
- # right-justified, when possible); and move the remaining text
416
- # label, if necessary.
417
- stackx = max(stackx, self._stacklabel.width()+25)
418
- rlabelwidth = self._rtextlabel.width()+10
419
- if stackx >= cx2-max(rtextwidth, rlabelwidth):
420
- cx2 = stackx + max(rtextwidth, rlabelwidth)
421
- for rtextwidget in self._rtextwidgets:
422
- rtextwidget.move(4+cx2-rtextwidth, 0)
423
- self._rtextlabel.move(cx2-self._rtextlabel.bbox()[2]-5, 0)
424
-
425
- midx = (stackx + cx2-max(rtextwidth, rlabelwidth))/2
426
- self._canvas.coords(self._stacktop, midx, 0, midx, 5000)
427
- (x1, y1, x2, y2) = self._stacklabel.bbox()
428
-
429
- # Set up binding to allow them to shift a token by dragging it.
430
- if len(self._rtextwidgets) > 0:
431
- def drag_shift(widget, midx=midx, self=self):
432
- if widget.bbox()[0] < midx: self.shift()
433
- else: self._redraw()
434
- self._rtextwidgets[0].bind_drag(drag_shift)
435
- self._rtextwidgets[0].bind_click(self.shift)
436
-
437
- # Draw the stack top.
438
- self._highlight_productions()
439
-
440
- def _draw_stack_top(self, widget):
441
- # hack..
442
- midx = widget.bbox()[2]+50
443
- self._canvas.coords(self._stacktop, midx, 0, midx, 5000)
444
-
445
- def _highlight_productions(self):
446
- # Highlight the productions that can be reduced.
447
- self._prodlist.selection_clear(0, 'end')
448
- for prod in self._parser.reducible_productions():
449
- index = self._productions.index(prod)
450
- self._prodlist.selection_set(index)
451
-
452
- #########################################
453
- ## Button Callbacks
454
- #########################################
455
-
456
- def destroy(self, *e):
457
- if self._top is None: return
458
- self._top.destroy()
459
- self._top = None
460
-
461
- def reset(self, *e):
462
- self._parser.initialize(self._sent)
463
- self._lastoper1['text'] = 'Reset App'
464
- self._lastoper2['text'] = ''
465
- self._redraw()
466
-
467
- def step(self, *e):
468
- if self.reduce(): return True
469
- elif self.shift(): return True
470
- else:
471
- if list(self._parser.parses()):
472
- self._lastoper1['text'] = 'Finished:'
473
- self._lastoper2['text'] = 'Success'
474
- else:
475
- self._lastoper1['text'] = 'Finished:'
476
- self._lastoper2['text'] = 'Failure'
477
-
478
- def shift(self, *e):
479
- if self._animating_lock: return
480
- if self._parser.shift():
481
- tok = self._parser.stack()[-1]
482
- self._lastoper1['text'] = 'Shift:'
483
- self._lastoper2['text'] = '%r' % tok
484
- if self._animate.get():
485
- self._animate_shift()
486
- else:
487
- self._redraw()
488
- return True
489
- return False
490
-
491
- def reduce(self, *e):
492
- if self._animating_lock: return
493
- production = self._parser.reduce()
494
- if production:
495
- self._lastoper1['text'] = 'Reduce:'
496
- self._lastoper2['text'] = '%s' % production
497
- if self._animate.get():
498
- self._animate_reduce()
499
- else:
500
- self._redraw()
501
- return production
502
-
503
- def undo(self, *e):
504
- if self._animating_lock: return
505
- if self._parser.undo():
506
- self._redraw()
507
-
508
- def postscript(self, *e):
509
- self._cframe.print_to_file()
510
-
511
- def mainloop(self, *args, **kwargs):
512
- """
513
- Enter the Tkinter mainloop. This function must be called if
514
- this demo is created from a non-interactive program (e.g.
515
- from a secript); otherwise, the demo will close as soon as
516
- the script completes.
517
- """
518
- if in_idle(): return
519
- self._top.mainloop(*args, **kwargs)
520
-
521
- #########################################
522
- ## Menubar callbacks
523
- #########################################
524
-
525
- def resize(self, size=None):
526
- if size is not None: self._size.set(size)
527
- size = self._size.get()
528
- self._font.configure(size=-(abs(size)))
529
- self._boldfont.configure(size=-(abs(size)))
530
- self._sysfont.configure(size=-(abs(size)))
531
-
532
- #self._stacklabel['font'] = ('helvetica', -size-4, 'bold')
533
- #self._rtextlabel['font'] = ('helvetica', -size-4, 'bold')
534
- #self._lastoper_label['font'] = ('helvetica', -size)
535
- #self._lastoper1['font'] = ('helvetica', -size)
536
- #self._lastoper2['font'] = ('helvetica', -size)
537
- #self._prodlist['font'] = ('helvetica', -size)
538
- #self._prodlist_label['font'] = ('helvetica', -size-2, 'bold')
539
- self._redraw()
540
-
541
- def help(self, *e):
542
- # The default font's not very legible; try using 'fixed' instead.
543
- try:
544
- ShowText(self._top, 'Help: Shift-Reduce Parser Application',
545
- (__doc__ or '').strip(), width=75, font='fixed')
546
- except:
547
- ShowText(self._top, 'Help: Shift-Reduce Parser Application',
548
- (__doc__ or '').strip(), width=75)
549
-
550
- def about(self, *e):
551
- ABOUT = ("NLTK Shift-Reduce Parser Application\n"+
552
- "Written by Edward Loper")
553
- TITLE = 'About: Shift-Reduce Parser Application'
554
- try:
555
- from tkinter.messagebox import Message
556
- Message(message=ABOUT, title=TITLE).show()
557
- except:
558
- ShowText(self._top, TITLE, ABOUT)
559
-
560
- def edit_grammar(self, *e):
561
- CFGEditor(self._top, self._parser.grammar(), self.set_grammar)
562
-
563
- def set_grammar(self, grammar):
564
- self._parser.set_grammar(grammar)
565
- self._productions = list(grammar.productions())
566
- self._prodlist.delete(0, 'end')
567
- for production in self._productions:
568
- self._prodlist.insert('end', (' %s' % production))
569
-
570
- def edit_sentence(self, *e):
571
- sentence = " ".join(self._sent)
572
- title = 'Edit Text'
573
- instr = 'Enter a new sentence to parse.'
574
- EntryDialog(self._top, sentence, instr, self.set_sentence, title)
575
-
576
- def set_sentence(self, sent):
577
- self._sent = sent.split() #[XX] use tagged?
578
- self.reset()
579
-
580
- #########################################
581
- ## Reduce Production Selection
582
- #########################################
583
-
584
- def _toggle_grammar(self, *e):
585
- if self._show_grammar.get():
586
- self._prodframe.pack(fill='both', side='left', padx=2,
587
- after=self._feedbackframe)
588
- self._lastoper1['text'] = 'Show Grammar'
589
- else:
590
- self._prodframe.pack_forget()
591
- self._lastoper1['text'] = 'Hide Grammar'
592
- self._lastoper2['text'] = ''
593
-
594
- def _prodlist_select(self, event):
595
- selection = self._prodlist.curselection()
596
- if len(selection) != 1: return
597
- index = int(selection[0])
598
- production = self._parser.reduce(self._productions[index])
599
- if production:
600
- self._lastoper1['text'] = 'Reduce:'
601
- self._lastoper2['text'] = '%s' % production
602
- if self._animate.get():
603
- self._animate_reduce()
604
- else:
605
- self._redraw()
606
- else:
607
- # Reset the production selections.
608
- self._prodlist.selection_clear(0, 'end')
609
- for prod in self._parser.reducible_productions():
610
- index = self._productions.index(prod)
611
- self._prodlist.selection_set(index)
612
-
613
- def _popup_reduce(self, widget):
614
- # Remove old commands.
615
- productions = self._parser.reducible_productions()
616
- if len(productions) == 0: return
617
-
618
- self._reduce_menu.delete(0, 'end')
619
- for production in productions:
620
- self._reduce_menu.add_command(label=str(production),
621
- command=self.reduce)
622
- self._reduce_menu.post(self._canvas.winfo_pointerx(),
623
- self._canvas.winfo_pointery())
624
-
625
- #########################################
626
- ## Animations
627
- #########################################
628
-
629
- def _animate_shift(self):
630
- # What widget are we shifting?
631
- widget = self._rtextwidgets[0]
632
-
633
- # Where are we shifting from & to?
634
- right = widget.bbox()[0]
635
- if len(self._stackwidgets) == 0: left = 5
636
- else: left = self._stackwidgets[-1].bbox()[2]+10
637
-
638
- # Start animating.
639
- dt = self._animate.get()
640
- dx = (left-right)*1.0/dt
641
- self._animate_shift_frame(dt, widget, dx)
642
-
643
- def _animate_shift_frame(self, frame, widget, dx):
644
- if frame > 0:
645
- self._animating_lock = 1
646
- widget.move(dx, 0)
647
- self._top.after(10, self._animate_shift_frame,
648
- frame-1, widget, dx)
649
- else:
650
- # but: stacktop??
651
-
652
- # Shift the widget to the stack.
653
- del self._rtextwidgets[0]
654
- self._stackwidgets.append(widget)
655
- self._animating_lock = 0
656
-
657
- # Display the available productions.
658
- self._draw_stack_top(widget)
659
- self._highlight_productions()
660
-
661
- def _animate_reduce(self):
662
- # What widgets are we shifting?
663
- numwidgets = len(self._parser.stack()[-1]) # number of children
664
- widgets = self._stackwidgets[-numwidgets:]
665
-
666
- # How far are we moving?
667
- if isinstance(widgets[0], TreeSegmentWidget):
668
- ydist = 15 + widgets[0].label().height()
669
- else:
670
- ydist = 15 + widgets[0].height()
671
-
672
- # Start animating.
673
- dt = self._animate.get()
674
- dy = ydist*2.0/dt
675
- self._animate_reduce_frame(dt/2, widgets, dy)
676
-
677
- def _animate_reduce_frame(self, frame, widgets, dy):
678
- if frame > 0:
679
- self._animating_lock = 1
680
- for widget in widgets: widget.move(0, dy)
681
- self._top.after(10, self._animate_reduce_frame,
682
- frame-1, widgets, dy)
683
- else:
684
- del self._stackwidgets[-len(widgets):]
685
- for widget in widgets:
686
- self._cframe.remove_widget(widget)
687
- tok = self._parser.stack()[-1]
688
- if not isinstance(tok, Tree): raise ValueError()
689
- label = TextWidget(self._canvas, str(tok.label()), color='#006060',
690
- font=self._boldfont)
691
- widget = TreeSegmentWidget(self._canvas, label, widgets,
692
- width=2)
693
- (x1, y1, x2, y2) = self._stacklabel.bbox()
694
- y = y2-y1+10
695
- if not self._stackwidgets: x = 5
696
- else: x = self._stackwidgets[-1].bbox()[2] + 10
697
- self._cframe.add_widget(widget, x, y)
698
- self._stackwidgets.append(widget)
699
-
700
- # Display the available productions.
701
- self._draw_stack_top(widget)
702
- self._highlight_productions()
703
-
704
- # # Delete the old widgets..
705
- # del self._stackwidgets[-len(widgets):]
706
- # for widget in widgets:
707
- # self._cframe.destroy_widget(widget)
708
- #
709
- # # Make a new one.
710
- # tok = self._parser.stack()[-1]
711
- # if isinstance(tok, Tree):
712
- # attribs = {'tree_color': '#4080a0', 'tree_width': 2,
713
- # 'node_font': bold, 'node_color': '#006060',
714
- # 'leaf_color': '#006060', 'leaf_font':self._font}
715
- # widget = tree_to_treesegment(self._canvas, tok.type(),
716
- # **attribs)
717
- # widget.node()['color'] = '#000000'
718
- # else:
719
- # widget = TextWidget(self._canvas, tok.type(),
720
- # color='#000000', font=self._font)
721
- # widget.bind_click(self._popup_reduce)
722
- # (x1, y1, x2, y2) = self._stacklabel.bbox()
723
- # y = y2-y1+10
724
- # if not self._stackwidgets: x = 5
725
- # else: x = self._stackwidgets[-1].bbox()[2] + 10
726
- # self._cframe.add_widget(widget, x, y)
727
- # self._stackwidgets.append(widget)
728
-
729
- #self._redraw()
730
- self._animating_lock = 0
731
-
732
- #########################################
733
- ## Hovering.
734
- #########################################
735
-
736
- def _highlight_hover(self, event):
737
- # What production are we hovering over?
738
- index = self._prodlist.nearest(event.y)
739
- if self._hover == index: return
740
-
741
- # Clear any previous hover highlighting.
742
- self._clear_hover()
743
-
744
- # If the production corresponds to an available reduction,
745
- # highlight the stack.
746
- selection = [int(s) for s in self._prodlist.curselection()]
747
- if index in selection:
748
- rhslen = len(self._productions[index].rhs())
749
- for stackwidget in self._stackwidgets[-rhslen:]:
750
- if isinstance(stackwidget, TreeSegmentWidget):
751
- stackwidget.label()['color'] = '#00a000'
752
- else:
753
- stackwidget['color'] = '#00a000'
754
-
755
- # Remember what production we're hovering over.
756
- self._hover = index
757
-
758
- def _clear_hover(self, *event):
759
- # Clear any previous hover highlighting.
760
- if self._hover == -1: return
761
- self._hover = -1
762
- for stackwidget in self._stackwidgets:
763
- if isinstance(stackwidget, TreeSegmentWidget):
764
- stackwidget.label()['color'] = 'black'
765
- else:
766
- stackwidget['color'] = 'black'
767
-
768
-
769
- def app():
770
- """
771
- Create a shift reduce parser app, using a simple grammar and
772
- text.
773
- """
774
-
775
- from nltk.grammar import Nonterminal, Production, CFG
776
- nonterminals = 'S VP NP PP P N Name V Det'
777
- (S, VP, NP, PP, P, N, Name, V, Det) = [Nonterminal(s)
778
- for s in nonterminals.split()]
779
-
780
- productions = (
781
- # Syntactic Productions
782
- Production(S, [NP, VP]),
783
- Production(NP, [Det, N]),
784
- Production(NP, [NP, PP]),
785
- Production(VP, [VP, PP]),
786
- Production(VP, [V, NP, PP]),
787
- Production(VP, [V, NP]),
788
- Production(PP, [P, NP]),
789
-
790
- # Lexical Productions
791
- Production(NP, ['I']), Production(Det, ['the']),
792
- Production(Det, ['a']), Production(N, ['man']),
793
- Production(V, ['saw']), Production(P, ['in']),
794
- Production(P, ['with']), Production(N, ['park']),
795
- Production(N, ['dog']), Production(N, ['statue']),
796
- Production(Det, ['my']),
797
- )
798
-
799
- grammar = CFG(S, productions)
800
-
801
- # tokenize the sentence
802
- sent = 'my dog saw a man in the park with a statue'.split()
803
-
804
- ShiftReduceApp(grammar, sent).mainloop()
805
-
806
- if __name__ == '__main__':
807
- app()
808
-
809
- __all__ = ['app']
1
+ # Natural Language Toolkit: Shift-Reduce 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 shift-reduce parser.
10
+
11
+ The shift-reduce parser maintains a stack, which records the structure
12
+ of the portion of the text that has been parsed. The stack is
13
+ initially empty. Its contents are shown on the left side of the main
14
+ canvas.
15
+
16
+ On the right side of the main canvas is the remaining text. This is
17
+ the portion of the text which has not yet been considered by the
18
+ parser.
19
+
20
+ The parser builds up a tree structure for the text using two
21
+ operations:
22
+
23
+ - "shift" moves the first token from the remaining text to the top
24
+ of the stack. In the demo, the top of the stack is its right-hand
25
+ side.
26
+ - "reduce" uses a grammar production to combine the rightmost stack
27
+ elements into a single tree token.
28
+
29
+ You can control the parser's operation by using the "shift" and
30
+ "reduce" buttons; or you can use the "step" button to let the parser
31
+ automatically decide which operation to apply. The parser uses the
32
+ following rules to decide which operation to apply:
33
+
34
+ - Only shift if no reductions are available.
35
+ - If multiple reductions are available, then apply the reduction
36
+ whose CFG production is listed earliest in the grammar.
37
+
38
+ The "reduce" button applies the reduction whose CFG production is
39
+ listed earliest in the grammar. There are two ways to manually choose
40
+ which reduction to apply:
41
+
42
+ - Click on a CFG production from the list of available reductions,
43
+ on the left side of the main window. The reduction based on that
44
+ production will be applied to the top of the stack.
45
+ - Click on one of the stack elements. A popup window will appear,
46
+ containing all available reductions. Select one, and it will be
47
+ applied to the top of the stack.
48
+
49
+ Note that reductions can only be applied to the top of the stack.
50
+
51
+ Keyboard Shortcuts::
52
+ [Space]\t Perform the next shift or reduce operation
53
+ [s]\t Perform a shift operation
54
+ [r]\t Perform a reduction operation
55
+ [Ctrl-z]\t Undo most recent operation
56
+ [Delete]\t Reset the parser
57
+ [g]\t Show/hide available production list
58
+ [Ctrl-a]\t Toggle animations
59
+ [h]\t Help
60
+ [Ctrl-p]\t Print
61
+ [q]\t Quit
62
+
63
+ """
64
+
65
+ from six.moves.tkinter_font import Font
66
+ from six.moves.tkinter import (IntVar, Listbox, Button, Frame, Label, Menu,
67
+ Scrollbar, Tk)
68
+
69
+ from nltk.tree import Tree
70
+ from nltk.parse import SteppingShiftReduceParser
71
+ from nltk.util import in_idle
72
+ from nltk.draw.util import CanvasFrame, EntryDialog, ShowText, TextWidget
73
+ from nltk.draw import CFGEditor, TreeSegmentWidget, tree_to_treesegment
74
+
75
+ """
76
+ Possible future improvements:
77
+ - button/window to change and/or select text. Just pop up a window
78
+ with an entry, and let them modify the text; and then retokenize
79
+ it? Maybe give a warning if it contains tokens whose types are
80
+ not in the grammar.
81
+ - button/window to change and/or select grammar. Select from
82
+ several alternative grammars? Or actually change the grammar? If
83
+ the later, then I'd want to define nltk.draw.cfg, which would be
84
+ responsible for that.
85
+ """
86
+
87
+ class ShiftReduceApp(object):
88
+ """
89
+ A graphical tool for exploring the shift-reduce parser. The tool
90
+ displays the parser's stack and the remaining text, and allows the
91
+ user to control the parser's operation. In particular, the user
92
+ can shift tokens onto the stack, and can perform reductions on the
93
+ top elements of the stack. A "step" button simply steps through
94
+ the parsing process, performing the operations that
95
+ ``nltk.parse.ShiftReduceParser`` would use.
96
+ """
97
+ def __init__(self, grammar, sent, trace=0):
98
+ self._sent = sent
99
+ self._parser = SteppingShiftReduceParser(grammar, trace)
100
+
101
+ # Set up the main window.
102
+ self._top = Tk()
103
+ self._top.title('Shift Reduce Parser Application')
104
+
105
+ # Animations. animating_lock is a lock to prevent the demo
106
+ # from performing new operations while it's animating.
107
+ self._animating_lock = 0
108
+ self._animate = IntVar(self._top)
109
+ self._animate.set(10) # = medium
110
+
111
+ # The user can hide the grammar.
112
+ self._show_grammar = IntVar(self._top)
113
+ self._show_grammar.set(1)
114
+
115
+ # Initialize fonts.
116
+ self._init_fonts(self._top)
117
+
118
+ # Set up key bindings.
119
+ self._init_bindings()
120
+
121
+ # Create the basic frames.
122
+ self._init_menubar(self._top)
123
+ self._init_buttons(self._top)
124
+ self._init_feedback(self._top)
125
+ self._init_grammar(self._top)
126
+ self._init_canvas(self._top)
127
+
128
+ # A popup menu for reducing.
129
+ self._reduce_menu = Menu(self._canvas, tearoff=0)
130
+
131
+ # Reset the demo, and set the feedback frame to empty.
132
+ self.reset()
133
+ self._lastoper1['text'] = ''
134
+
135
+ #########################################
136
+ ## Initialization Helpers
137
+ #########################################
138
+
139
+ def _init_fonts(self, root):
140
+ # See: <http://www.astro.washington.edu/owen/ROTKFolklore.html>
141
+ self._sysfont = Font(font=Button()["font"])
142
+ root.option_add("*Font", self._sysfont)
143
+
144
+ # TWhat's our font size (default=same as sysfont)
145
+ self._size = IntVar(root)
146
+ self._size.set(self._sysfont.cget('size'))
147
+
148
+ self._boldfont = Font(family='helvetica', weight='bold',
149
+ size=self._size.get())
150
+ self._font = Font(family='helvetica',
151
+ size=self._size.get())
152
+
153
+ def _init_grammar(self, parent):
154
+ # Grammar view.
155
+ self._prodframe = listframe = Frame(parent)
156
+ self._prodframe.pack(fill='both', side='left', padx=2)
157
+ self._prodlist_label = Label(self._prodframe,
158
+ font=self._boldfont,
159
+ text='Available Reductions')
160
+ self._prodlist_label.pack()
161
+ self._prodlist = Listbox(self._prodframe, selectmode='single',
162
+ relief='groove', background='white',
163
+ foreground='#909090',
164
+ font=self._font,
165
+ selectforeground='#004040',
166
+ selectbackground='#c0f0c0')
167
+
168
+ self._prodlist.pack(side='right', fill='both', expand=1)
169
+
170
+ self._productions = list(self._parser.grammar().productions())
171
+ for production in self._productions:
172
+ self._prodlist.insert('end', (' %s' % production))
173
+ self._prodlist.config(height=min(len(self._productions), 25))
174
+
175
+ # Add a scrollbar if there are more than 25 productions.
176
+ if 1:#len(self._productions) > 25:
177
+ listscroll = Scrollbar(self._prodframe,
178
+ orient='vertical')
179
+ self._prodlist.config(yscrollcommand = listscroll.set)
180
+ listscroll.config(command=self._prodlist.yview)
181
+ listscroll.pack(side='left', fill='y')
182
+
183
+ # If they select a production, apply it.
184
+ self._prodlist.bind('<<ListboxSelect>>', self._prodlist_select)
185
+
186
+ # When they hover over a production, highlight it.
187
+ self._hover = -1
188
+ self._prodlist.bind('<Motion>', self._highlight_hover)
189
+ self._prodlist.bind('<Leave>', self._clear_hover)
190
+
191
+ def _init_bindings(self):
192
+ # Quit
193
+ self._top.bind('<Control-q>', self.destroy)
194
+ self._top.bind('<Control-x>', self.destroy)
195
+ self._top.bind('<Alt-q>', self.destroy)
196
+ self._top.bind('<Alt-x>', self.destroy)
197
+
198
+ # Ops (step, shift, reduce, undo)
199
+ self._top.bind('<space>', self.step)
200
+ self._top.bind('<s>', self.shift)
201
+ self._top.bind('<Alt-s>', self.shift)
202
+ self._top.bind('<Control-s>', self.shift)
203
+ self._top.bind('<r>', self.reduce)
204
+ self._top.bind('<Alt-r>', self.reduce)
205
+ self._top.bind('<Control-r>', self.reduce)
206
+ self._top.bind('<Delete>', self.reset)
207
+ self._top.bind('<u>', self.undo)
208
+ self._top.bind('<Alt-u>', self.undo)
209
+ self._top.bind('<Control-u>', self.undo)
210
+ self._top.bind('<Control-z>', self.undo)
211
+ self._top.bind('<BackSpace>', self.undo)
212
+
213
+ # Misc
214
+ self._top.bind('<Control-p>', self.postscript)
215
+ self._top.bind('<Control-h>', self.help)
216
+ self._top.bind('<F1>', self.help)
217
+ self._top.bind('<Control-g>', self.edit_grammar)
218
+ self._top.bind('<Control-t>', self.edit_sentence)
219
+
220
+ # Animation speed control
221
+ self._top.bind('-', lambda e,a=self._animate:a.set(20))
222
+ self._top.bind('=', lambda e,a=self._animate:a.set(10))
223
+ self._top.bind('+', lambda e,a=self._animate:a.set(4))
224
+
225
+ def _init_buttons(self, parent):
226
+ # Set up the frames.
227
+ self._buttonframe = buttonframe = Frame(parent)
228
+ buttonframe.pack(fill='none', side='bottom')
229
+ Button(buttonframe, text='Step',
230
+ background='#90c0d0', foreground='black',
231
+ command=self.step,).pack(side='left')
232
+ Button(buttonframe, text='Shift', underline=0,
233
+ background='#90f090', foreground='black',
234
+ command=self.shift).pack(side='left')
235
+ Button(buttonframe, text='Reduce', underline=0,
236
+ background='#90f090', foreground='black',
237
+ command=self.reduce).pack(side='left')
238
+ Button(buttonframe, text='Undo', underline=0,
239
+ background='#f0a0a0', foreground='black',
240
+ command=self.undo).pack(side='left')
241
+
242
+ def _init_menubar(self, parent):
243
+ menubar = Menu(parent)
244
+
245
+ filemenu = Menu(menubar, tearoff=0)
246
+ filemenu.add_command(label='Reset Parser', underline=0,
247
+ command=self.reset, accelerator='Del')
248
+ filemenu.add_command(label='Print to Postscript', underline=0,
249
+ command=self.postscript, accelerator='Ctrl-p')
250
+ filemenu.add_command(label='Exit', underline=1,
251
+ command=self.destroy, accelerator='Ctrl-x')
252
+ menubar.add_cascade(label='File', underline=0, menu=filemenu)
253
+
254
+ editmenu = Menu(menubar, tearoff=0)
255
+ editmenu.add_command(label='Edit Grammar', underline=5,
256
+ command=self.edit_grammar,
257
+ accelerator='Ctrl-g')
258
+ editmenu.add_command(label='Edit Text', underline=5,
259
+ command=self.edit_sentence,
260
+ accelerator='Ctrl-t')
261
+ menubar.add_cascade(label='Edit', underline=0, menu=editmenu)
262
+
263
+ rulemenu = Menu(menubar, tearoff=0)
264
+ rulemenu.add_command(label='Step', underline=1,
265
+ command=self.step, accelerator='Space')
266
+ rulemenu.add_separator()
267
+ rulemenu.add_command(label='Shift', underline=0,
268
+ command=self.shift, accelerator='Ctrl-s')
269
+ rulemenu.add_command(label='Reduce', underline=0,
270
+ command=self.reduce, accelerator='Ctrl-r')
271
+ rulemenu.add_separator()
272
+ rulemenu.add_command(label='Undo', underline=0,
273
+ command=self.undo, accelerator='Ctrl-u')
274
+ menubar.add_cascade(label='Apply', underline=0, menu=rulemenu)
275
+
276
+ viewmenu = Menu(menubar, tearoff=0)
277
+ viewmenu.add_checkbutton(label="Show Grammar", underline=0,
278
+ variable=self._show_grammar,
279
+ command=self._toggle_grammar)
280
+ viewmenu.add_separator()
281
+ viewmenu.add_radiobutton(label='Tiny', variable=self._size,
282
+ underline=0, value=10, command=self.resize)
283
+ viewmenu.add_radiobutton(label='Small', variable=self._size,
284
+ underline=0, value=12, command=self.resize)
285
+ viewmenu.add_radiobutton(label='Medium', variable=self._size,
286
+ underline=0, value=14, command=self.resize)
287
+ viewmenu.add_radiobutton(label='Large', variable=self._size,
288
+ underline=0, value=18, command=self.resize)
289
+ viewmenu.add_radiobutton(label='Huge', variable=self._size,
290
+ underline=0, value=24, command=self.resize)
291
+ menubar.add_cascade(label='View', underline=0, menu=viewmenu)
292
+
293
+ animatemenu = Menu(menubar, tearoff=0)
294
+ animatemenu.add_radiobutton(label="No Animation", underline=0,
295
+ variable=self._animate, value=0)
296
+ animatemenu.add_radiobutton(label="Slow Animation", underline=0,
297
+ variable=self._animate, value=20,
298
+ accelerator='-')
299
+ animatemenu.add_radiobutton(label="Normal Animation", underline=0,
300
+ variable=self._animate, value=10,
301
+ accelerator='=')
302
+ animatemenu.add_radiobutton(label="Fast Animation", underline=0,
303
+ variable=self._animate, value=4,
304
+ accelerator='+')
305
+ menubar.add_cascade(label="Animate", underline=1, menu=animatemenu)
306
+
307
+
308
+ helpmenu = Menu(menubar, tearoff=0)
309
+ helpmenu.add_command(label='About', underline=0,
310
+ command=self.about)
311
+ helpmenu.add_command(label='Instructions', underline=0,
312
+ command=self.help, accelerator='F1')
313
+ menubar.add_cascade(label='Help', underline=0, menu=helpmenu)
314
+
315
+ parent.config(menu=menubar)
316
+
317
+ def _init_feedback(self, parent):
318
+ self._feedbackframe = feedbackframe = Frame(parent)
319
+ feedbackframe.pack(fill='x', side='bottom', padx=3, pady=3)
320
+ self._lastoper_label = Label(feedbackframe, text='Last Operation:',
321
+ font=self._font)
322
+ self._lastoper_label.pack(side='left')
323
+ lastoperframe = Frame(feedbackframe, relief='sunken', border=1)
324
+ lastoperframe.pack(fill='x', side='right', expand=1, padx=5)
325
+ self._lastoper1 = Label(lastoperframe, foreground='#007070',
326
+ background='#f0f0f0', font=self._font)
327
+ self._lastoper2 = Label(lastoperframe, anchor='w', width=30,
328
+ foreground='#004040', background='#f0f0f0',
329
+ font=self._font)
330
+ self._lastoper1.pack(side='left')
331
+ self._lastoper2.pack(side='left', fill='x', expand=1)
332
+
333
+ def _init_canvas(self, parent):
334
+ self._cframe = CanvasFrame(parent, background='white',
335
+ width=525, closeenough=10,
336
+ border=2, relief='sunken')
337
+ self._cframe.pack(expand=1, fill='both', side='top', pady=2)
338
+ canvas = self._canvas = self._cframe.canvas()
339
+
340
+ self._stackwidgets = []
341
+ self._rtextwidgets = []
342
+ self._titlebar = canvas.create_rectangle(0,0,0,0, fill='#c0f0f0',
343
+ outline='black')
344
+ self._exprline = canvas.create_line(0,0,0,0, dash='.')
345
+ self._stacktop = canvas.create_line(0,0,0,0, fill='#408080')
346
+ size = self._size.get()+4
347
+ self._stacklabel = TextWidget(canvas, 'Stack', color='#004040',
348
+ font=self._boldfont)
349
+ self._rtextlabel = TextWidget(canvas, 'Remaining Text',
350
+ color='#004040', font=self._boldfont)
351
+ self._cframe.add_widget(self._stacklabel)
352
+ self._cframe.add_widget(self._rtextlabel)
353
+
354
+ #########################################
355
+ ## Main draw procedure
356
+ #########################################
357
+
358
+ def _redraw(self):
359
+ scrollregion = self._canvas['scrollregion'].split()
360
+ (cx1, cy1, cx2, cy2) = [int(c) for c in scrollregion]
361
+
362
+ # Delete the old stack & rtext widgets.
363
+ for stackwidget in self._stackwidgets:
364
+ self._cframe.destroy_widget(stackwidget)
365
+ self._stackwidgets = []
366
+ for rtextwidget in self._rtextwidgets:
367
+ self._cframe.destroy_widget(rtextwidget)
368
+ self._rtextwidgets = []
369
+
370
+ # Position the titlebar & exprline
371
+ (x1, y1, x2, y2) = self._stacklabel.bbox()
372
+ y = y2-y1+10
373
+ self._canvas.coords(self._titlebar, -5000, 0, 5000, y-4)
374
+ self._canvas.coords(self._exprline, 0, y*2-10, 5000, y*2-10)
375
+
376
+ # Position the titlebar labels..
377
+ (x1, y1, x2, y2) = self._stacklabel.bbox()
378
+ self._stacklabel.move(5-x1, 3-y1)
379
+ (x1, y1, x2, y2) = self._rtextlabel.bbox()
380
+ self._rtextlabel.move(cx2-x2-5, 3-y1)
381
+
382
+ # Draw the stack.
383
+ stackx = 5
384
+ for tok in self._parser.stack():
385
+ if isinstance(tok, Tree):
386
+ attribs = {'tree_color': '#4080a0', 'tree_width': 2,
387
+ 'node_font': self._boldfont,
388
+ 'node_color': '#006060',
389
+ 'leaf_color': '#006060', 'leaf_font':self._font}
390
+ widget = tree_to_treesegment(self._canvas, tok,
391
+ **attribs)
392
+ widget.label()['color'] = '#000000'
393
+ else:
394
+ widget = TextWidget(self._canvas, tok,
395
+ color='#000000', font=self._font)
396
+ widget.bind_click(self._popup_reduce)
397
+ self._stackwidgets.append(widget)
398
+ self._cframe.add_widget(widget, stackx, y)
399
+ stackx = widget.bbox()[2] + 10
400
+
401
+ # Draw the remaining text.
402
+ rtextwidth = 0
403
+ for tok in self._parser.remaining_text():
404
+ widget = TextWidget(self._canvas, tok,
405
+ color='#000000', font=self._font)
406
+ self._rtextwidgets.append(widget)
407
+ self._cframe.add_widget(widget, rtextwidth, y)
408
+ rtextwidth = widget.bbox()[2] + 4
409
+
410
+ # Allow enough room to shift the next token (for animations)
411
+ if len(self._rtextwidgets) > 0:
412
+ stackx += self._rtextwidgets[0].width()
413
+
414
+ # Move the remaining text to the correct location (keep it
415
+ # right-justified, when possible); and move the remaining text
416
+ # label, if necessary.
417
+ stackx = max(stackx, self._stacklabel.width()+25)
418
+ rlabelwidth = self._rtextlabel.width()+10
419
+ if stackx >= cx2-max(rtextwidth, rlabelwidth):
420
+ cx2 = stackx + max(rtextwidth, rlabelwidth)
421
+ for rtextwidget in self._rtextwidgets:
422
+ rtextwidget.move(4+cx2-rtextwidth, 0)
423
+ self._rtextlabel.move(cx2-self._rtextlabel.bbox()[2]-5, 0)
424
+
425
+ midx = (stackx + cx2-max(rtextwidth, rlabelwidth))/2
426
+ self._canvas.coords(self._stacktop, midx, 0, midx, 5000)
427
+ (x1, y1, x2, y2) = self._stacklabel.bbox()
428
+
429
+ # Set up binding to allow them to shift a token by dragging it.
430
+ if len(self._rtextwidgets) > 0:
431
+ def drag_shift(widget, midx=midx, self=self):
432
+ if widget.bbox()[0] < midx: self.shift()
433
+ else: self._redraw()
434
+ self._rtextwidgets[0].bind_drag(drag_shift)
435
+ self._rtextwidgets[0].bind_click(self.shift)
436
+
437
+ # Draw the stack top.
438
+ self._highlight_productions()
439
+
440
+ def _draw_stack_top(self, widget):
441
+ # hack..
442
+ midx = widget.bbox()[2]+50
443
+ self._canvas.coords(self._stacktop, midx, 0, midx, 5000)
444
+
445
+ def _highlight_productions(self):
446
+ # Highlight the productions that can be reduced.
447
+ self._prodlist.selection_clear(0, 'end')
448
+ for prod in self._parser.reducible_productions():
449
+ index = self._productions.index(prod)
450
+ self._prodlist.selection_set(index)
451
+
452
+ #########################################
453
+ ## Button Callbacks
454
+ #########################################
455
+
456
+ def destroy(self, *e):
457
+ if self._top is None: return
458
+ self._top.destroy()
459
+ self._top = None
460
+
461
+ def reset(self, *e):
462
+ self._parser.initialize(self._sent)
463
+ self._lastoper1['text'] = 'Reset App'
464
+ self._lastoper2['text'] = ''
465
+ self._redraw()
466
+
467
+ def step(self, *e):
468
+ if self.reduce(): return True
469
+ elif self.shift(): return True
470
+ else:
471
+ if list(self._parser.parses()):
472
+ self._lastoper1['text'] = 'Finished:'
473
+ self._lastoper2['text'] = 'Success'
474
+ else:
475
+ self._lastoper1['text'] = 'Finished:'
476
+ self._lastoper2['text'] = 'Failure'
477
+
478
+ def shift(self, *e):
479
+ if self._animating_lock: return
480
+ if self._parser.shift():
481
+ tok = self._parser.stack()[-1]
482
+ self._lastoper1['text'] = 'Shift:'
483
+ self._lastoper2['text'] = '%r' % tok
484
+ if self._animate.get():
485
+ self._animate_shift()
486
+ else:
487
+ self._redraw()
488
+ return True
489
+ return False
490
+
491
+ def reduce(self, *e):
492
+ if self._animating_lock: return
493
+ production = self._parser.reduce()
494
+ if production:
495
+ self._lastoper1['text'] = 'Reduce:'
496
+ self._lastoper2['text'] = '%s' % production
497
+ if self._animate.get():
498
+ self._animate_reduce()
499
+ else:
500
+ self._redraw()
501
+ return production
502
+
503
+ def undo(self, *e):
504
+ if self._animating_lock: return
505
+ if self._parser.undo():
506
+ self._redraw()
507
+
508
+ def postscript(self, *e):
509
+ self._cframe.print_to_file()
510
+
511
+ def mainloop(self, *args, **kwargs):
512
+ """
513
+ Enter the Tkinter mainloop. This function must be called if
514
+ this demo is created from a non-interactive program (e.g.
515
+ from a secript); otherwise, the demo will close as soon as
516
+ the script completes.
517
+ """
518
+ if in_idle(): return
519
+ self._top.mainloop(*args, **kwargs)
520
+
521
+ #########################################
522
+ ## Menubar callbacks
523
+ #########################################
524
+
525
+ def resize(self, size=None):
526
+ if size is not None: self._size.set(size)
527
+ size = self._size.get()
528
+ self._font.configure(size=-(abs(size)))
529
+ self._boldfont.configure(size=-(abs(size)))
530
+ self._sysfont.configure(size=-(abs(size)))
531
+
532
+ #self._stacklabel['font'] = ('helvetica', -size-4, 'bold')
533
+ #self._rtextlabel['font'] = ('helvetica', -size-4, 'bold')
534
+ #self._lastoper_label['font'] = ('helvetica', -size)
535
+ #self._lastoper1['font'] = ('helvetica', -size)
536
+ #self._lastoper2['font'] = ('helvetica', -size)
537
+ #self._prodlist['font'] = ('helvetica', -size)
538
+ #self._prodlist_label['font'] = ('helvetica', -size-2, 'bold')
539
+ self._redraw()
540
+
541
+ def help(self, *e):
542
+ # The default font's not very legible; try using 'fixed' instead.
543
+ try:
544
+ ShowText(self._top, 'Help: Shift-Reduce Parser Application',
545
+ (__doc__ or '').strip(), width=75, font='fixed')
546
+ except:
547
+ ShowText(self._top, 'Help: Shift-Reduce Parser Application',
548
+ (__doc__ or '').strip(), width=75)
549
+
550
+ def about(self, *e):
551
+ ABOUT = ("NLTK Shift-Reduce Parser Application\n"+
552
+ "Written by Edward Loper")
553
+ TITLE = 'About: Shift-Reduce Parser Application'
554
+ try:
555
+ from six.moves.tkinter_messagebox import Message
556
+ Message(message=ABOUT, title=TITLE).show()
557
+ except:
558
+ ShowText(self._top, TITLE, ABOUT)
559
+
560
+ def edit_grammar(self, *e):
561
+ CFGEditor(self._top, self._parser.grammar(), self.set_grammar)
562
+
563
+ def set_grammar(self, grammar):
564
+ self._parser.set_grammar(grammar)
565
+ self._productions = list(grammar.productions())
566
+ self._prodlist.delete(0, 'end')
567
+ for production in self._productions:
568
+ self._prodlist.insert('end', (' %s' % production))
569
+
570
+ def edit_sentence(self, *e):
571
+ sentence = " ".join(self._sent)
572
+ title = 'Edit Text'
573
+ instr = 'Enter a new sentence to parse.'
574
+ EntryDialog(self._top, sentence, instr, self.set_sentence, title)
575
+
576
+ def set_sentence(self, sent):
577
+ self._sent = sent.split() #[XX] use tagged?
578
+ self.reset()
579
+
580
+ #########################################
581
+ ## Reduce Production Selection
582
+ #########################################
583
+
584
+ def _toggle_grammar(self, *e):
585
+ if self._show_grammar.get():
586
+ self._prodframe.pack(fill='both', side='left', padx=2,
587
+ after=self._feedbackframe)
588
+ self._lastoper1['text'] = 'Show Grammar'
589
+ else:
590
+ self._prodframe.pack_forget()
591
+ self._lastoper1['text'] = 'Hide Grammar'
592
+ self._lastoper2['text'] = ''
593
+
594
+ def _prodlist_select(self, event):
595
+ selection = self._prodlist.curselection()
596
+ if len(selection) != 1: return
597
+ index = int(selection[0])
598
+ production = self._parser.reduce(self._productions[index])
599
+ if production:
600
+ self._lastoper1['text'] = 'Reduce:'
601
+ self._lastoper2['text'] = '%s' % production
602
+ if self._animate.get():
603
+ self._animate_reduce()
604
+ else:
605
+ self._redraw()
606
+ else:
607
+ # Reset the production selections.
608
+ self._prodlist.selection_clear(0, 'end')
609
+ for prod in self._parser.reducible_productions():
610
+ index = self._productions.index(prod)
611
+ self._prodlist.selection_set(index)
612
+
613
+ def _popup_reduce(self, widget):
614
+ # Remove old commands.
615
+ productions = self._parser.reducible_productions()
616
+ if len(productions) == 0: return
617
+
618
+ self._reduce_menu.delete(0, 'end')
619
+ for production in productions:
620
+ self._reduce_menu.add_command(label=str(production),
621
+ command=self.reduce)
622
+ self._reduce_menu.post(self._canvas.winfo_pointerx(),
623
+ self._canvas.winfo_pointery())
624
+
625
+ #########################################
626
+ ## Animations
627
+ #########################################
628
+
629
+ def _animate_shift(self):
630
+ # What widget are we shifting?
631
+ widget = self._rtextwidgets[0]
632
+
633
+ # Where are we shifting from & to?
634
+ right = widget.bbox()[0]
635
+ if len(self._stackwidgets) == 0: left = 5
636
+ else: left = self._stackwidgets[-1].bbox()[2]+10
637
+
638
+ # Start animating.
639
+ dt = self._animate.get()
640
+ dx = (left-right)*1.0/dt
641
+ self._animate_shift_frame(dt, widget, dx)
642
+
643
+ def _animate_shift_frame(self, frame, widget, dx):
644
+ if frame > 0:
645
+ self._animating_lock = 1
646
+ widget.move(dx, 0)
647
+ self._top.after(10, self._animate_shift_frame,
648
+ frame-1, widget, dx)
649
+ else:
650
+ # but: stacktop??
651
+
652
+ # Shift the widget to the stack.
653
+ del self._rtextwidgets[0]
654
+ self._stackwidgets.append(widget)
655
+ self._animating_lock = 0
656
+
657
+ # Display the available productions.
658
+ self._draw_stack_top(widget)
659
+ self._highlight_productions()
660
+
661
+ def _animate_reduce(self):
662
+ # What widgets are we shifting?
663
+ numwidgets = len(self._parser.stack()[-1]) # number of children
664
+ widgets = self._stackwidgets[-numwidgets:]
665
+
666
+ # How far are we moving?
667
+ if isinstance(widgets[0], TreeSegmentWidget):
668
+ ydist = 15 + widgets[0].label().height()
669
+ else:
670
+ ydist = 15 + widgets[0].height()
671
+
672
+ # Start animating.
673
+ dt = self._animate.get()
674
+ dy = ydist*2.0/dt
675
+ self._animate_reduce_frame(dt/2, widgets, dy)
676
+
677
+ def _animate_reduce_frame(self, frame, widgets, dy):
678
+ if frame > 0:
679
+ self._animating_lock = 1
680
+ for widget in widgets: widget.move(0, dy)
681
+ self._top.after(10, self._animate_reduce_frame,
682
+ frame-1, widgets, dy)
683
+ else:
684
+ del self._stackwidgets[-len(widgets):]
685
+ for widget in widgets:
686
+ self._cframe.remove_widget(widget)
687
+ tok = self._parser.stack()[-1]
688
+ if not isinstance(tok, Tree): raise ValueError()
689
+ label = TextWidget(self._canvas, str(tok.label()), color='#006060',
690
+ font=self._boldfont)
691
+ widget = TreeSegmentWidget(self._canvas, label, widgets,
692
+ width=2)
693
+ (x1, y1, x2, y2) = self._stacklabel.bbox()
694
+ y = y2-y1+10
695
+ if not self._stackwidgets: x = 5
696
+ else: x = self._stackwidgets[-1].bbox()[2] + 10
697
+ self._cframe.add_widget(widget, x, y)
698
+ self._stackwidgets.append(widget)
699
+
700
+ # Display the available productions.
701
+ self._draw_stack_top(widget)
702
+ self._highlight_productions()
703
+
704
+ # # Delete the old widgets..
705
+ # del self._stackwidgets[-len(widgets):]
706
+ # for widget in widgets:
707
+ # self._cframe.destroy_widget(widget)
708
+ #
709
+ # # Make a new one.
710
+ # tok = self._parser.stack()[-1]
711
+ # if isinstance(tok, Tree):
712
+ # attribs = {'tree_color': '#4080a0', 'tree_width': 2,
713
+ # 'node_font': bold, 'node_color': '#006060',
714
+ # 'leaf_color': '#006060', 'leaf_font':self._font}
715
+ # widget = tree_to_treesegment(self._canvas, tok.type(),
716
+ # **attribs)
717
+ # widget.node()['color'] = '#000000'
718
+ # else:
719
+ # widget = TextWidget(self._canvas, tok.type(),
720
+ # color='#000000', font=self._font)
721
+ # widget.bind_click(self._popup_reduce)
722
+ # (x1, y1, x2, y2) = self._stacklabel.bbox()
723
+ # y = y2-y1+10
724
+ # if not self._stackwidgets: x = 5
725
+ # else: x = self._stackwidgets[-1].bbox()[2] + 10
726
+ # self._cframe.add_widget(widget, x, y)
727
+ # self._stackwidgets.append(widget)
728
+
729
+ #self._redraw()
730
+ self._animating_lock = 0
731
+
732
+ #########################################
733
+ ## Hovering.
734
+ #########################################
735
+
736
+ def _highlight_hover(self, event):
737
+ # What production are we hovering over?
738
+ index = self._prodlist.nearest(event.y)
739
+ if self._hover == index: return
740
+
741
+ # Clear any previous hover highlighting.
742
+ self._clear_hover()
743
+
744
+ # If the production corresponds to an available reduction,
745
+ # highlight the stack.
746
+ selection = [int(s) for s in self._prodlist.curselection()]
747
+ if index in selection:
748
+ rhslen = len(self._productions[index].rhs())
749
+ for stackwidget in self._stackwidgets[-rhslen:]:
750
+ if isinstance(stackwidget, TreeSegmentWidget):
751
+ stackwidget.label()['color'] = '#00a000'
752
+ else:
753
+ stackwidget['color'] = '#00a000'
754
+
755
+ # Remember what production we're hovering over.
756
+ self._hover = index
757
+
758
+ def _clear_hover(self, *event):
759
+ # Clear any previous hover highlighting.
760
+ if self._hover == -1: return
761
+ self._hover = -1
762
+ for stackwidget in self._stackwidgets:
763
+ if isinstance(stackwidget, TreeSegmentWidget):
764
+ stackwidget.label()['color'] = 'black'
765
+ else:
766
+ stackwidget['color'] = 'black'
767
+
768
+
769
+ def app():
770
+ """
771
+ Create a shift reduce parser app, using a simple grammar and
772
+ text.
773
+ """
774
+
775
+ from nltk.grammar import Nonterminal, Production, CFG
776
+ nonterminals = 'S VP NP PP P N Name V Det'
777
+ (S, VP, NP, PP, P, N, Name, V, Det) = [Nonterminal(s)
778
+ for s in nonterminals.split()]
779
+
780
+ productions = (
781
+ # Syntactic Productions
782
+ Production(S, [NP, VP]),
783
+ Production(NP, [Det, N]),
784
+ Production(NP, [NP, PP]),
785
+ Production(VP, [VP, PP]),
786
+ Production(VP, [V, NP, PP]),
787
+ Production(VP, [V, NP]),
788
+ Production(PP, [P, NP]),
789
+
790
+ # Lexical Productions
791
+ Production(NP, ['I']), Production(Det, ['the']),
792
+ Production(Det, ['a']), Production(N, ['man']),
793
+ Production(V, ['saw']), Production(P, ['in']),
794
+ Production(P, ['with']), Production(N, ['park']),
795
+ Production(N, ['dog']), Production(N, ['statue']),
796
+ Production(Det, ['my']),
797
+ )
798
+
799
+ grammar = CFG(S, productions)
800
+
801
+ # tokenize the sentence
802
+ sent = 'my dog saw a man in the park with a statue'.split()
803
+
804
+ ShiftReduceApp(grammar, sent).mainloop()
805
+
806
+ if __name__ == '__main__':
807
+ app()
808
+
809
+ __all__ = ['app']