summarize 1.0.0

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 (62) hide show
  1. data/.gitignore +11 -0
  2. data/README.markdown +42 -0
  3. data/Rakefile +49 -0
  4. data/ext/summarize/article.c +119 -0
  5. data/ext/summarize/dic/bg.xml +101 -0
  6. data/ext/summarize/dic/ca.xml +141 -0
  7. data/ext/summarize/dic/cs.xml +161 -0
  8. data/ext/summarize/dic/cy.xml +118 -0
  9. data/ext/summarize/dic/da.xml +129 -0
  10. data/ext/summarize/dic/de.xml +354 -0
  11. data/ext/summarize/dic/el.xml +80 -0
  12. data/ext/summarize/dic/en.xml +606 -0
  13. data/ext/summarize/dic/eo.xml +171 -0
  14. data/ext/summarize/dic/es.xml +369 -0
  15. data/ext/summarize/dic/et.xml +172 -0
  16. data/ext/summarize/dic/eu.xml +77 -0
  17. data/ext/summarize/dic/fi.xml +105 -0
  18. data/ext/summarize/dic/fr.xml +199 -0
  19. data/ext/summarize/dic/ga.xml +124 -0
  20. data/ext/summarize/dic/gl.xml +290 -0
  21. data/ext/summarize/dic/he.xml +334 -0
  22. data/ext/summarize/dic/hu.xml +280 -0
  23. data/ext/summarize/dic/ia.xml +97 -0
  24. data/ext/summarize/dic/id.xml +75 -0
  25. data/ext/summarize/dic/is.xml +201 -0
  26. data/ext/summarize/dic/it.xml +206 -0
  27. data/ext/summarize/dic/lv.xml +77 -0
  28. data/ext/summarize/dic/mi.xml +76 -0
  29. data/ext/summarize/dic/ms.xml +160 -0
  30. data/ext/summarize/dic/mt.xml +73 -0
  31. data/ext/summarize/dic/nl.xml +245 -0
  32. data/ext/summarize/dic/nn.xml +264 -0
  33. data/ext/summarize/dic/pl.xml +92 -0
  34. data/ext/summarize/dic/pt.xml +365 -0
  35. data/ext/summarize/dic/ro.xml +163 -0
  36. data/ext/summarize/dic/ru.xml +150 -0
  37. data/ext/summarize/dic/sv.xml +255 -0
  38. data/ext/summarize/dic/tl.xml +67 -0
  39. data/ext/summarize/dic/tr.xml +65 -0
  40. data/ext/summarize/dic/uk.xml +98 -0
  41. data/ext/summarize/dic/yi.xml +293 -0
  42. data/ext/summarize/dictionary.c +331 -0
  43. data/ext/summarize/extconf.rb +6 -0
  44. data/ext/summarize/grader-tc.c +185 -0
  45. data/ext/summarize/grader-tc.h +64 -0
  46. data/ext/summarize/grader-tf.c +116 -0
  47. data/ext/summarize/grader.c +85 -0
  48. data/ext/summarize/highlighter.c +128 -0
  49. data/ext/summarize/html.c +131 -0
  50. data/ext/summarize/libots.h +158 -0
  51. data/ext/summarize/parser.c +173 -0
  52. data/ext/summarize/relations.c +163 -0
  53. data/ext/summarize/stemmer.c +332 -0
  54. data/ext/summarize/summarize.c +43 -0
  55. data/ext/summarize/summarize.h +12 -0
  56. data/ext/summarize/text.c +98 -0
  57. data/ext/summarize/wordlist.c +220 -0
  58. data/lib/summarize.rb +91 -0
  59. data/lib/summarize/summarize.bundle +0 -0
  60. data/sample_data/jupiter.txt +15 -0
  61. data/summarize.gemspec +21 -0
  62. metadata +140 -0
@@ -0,0 +1,331 @@
1
+ /*
2
+ * dictionary.c
3
+ *
4
+ * Copyright (C) 2003 Nadav Rotem <nadav256@hotmail.com>
5
+ *
6
+ * This program is free software; you can redistribute it and/or modify
7
+ * it under the terms of the GNU General Public License as published by
8
+ * the Free Software Foundation; either version 2 of the License, or
9
+ * (at your option) any later version.
10
+ *
11
+ * This program is distributed in the hope that it will be useful,
12
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
13
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14
+ * GNU Library General Public License for more details.
15
+ *
16
+ * You should have received a copy of the GNU General Public License
17
+ * along with this program; if not, write to the Free Software
18
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
19
+ */
20
+
21
+ #include <stdio.h>
22
+ #include <stdlib.h>
23
+ #include <string.h>
24
+
25
+ #include "libots.h"
26
+ #include "grader-tc.h"
27
+
28
+ #include <libxml/xmlmemory.h>
29
+ #include <libxml/parser.h>
30
+
31
+
32
+ /* loads the xml dictionary file to memory*/
33
+
34
+ gboolean
35
+ ots_load_xml_dictionary (OtsArticle * Doc,unsigned const char *name)
36
+ {
37
+
38
+ xmlDocPtr doc=NULL;
39
+ xmlNodePtr head=NULL;
40
+ xmlNodePtr stem=NULL;
41
+ xmlNodePtr pre=NULL;
42
+ xmlNodePtr post=NULL;
43
+ xmlNodePtr syno=NULL; /* synonyms */
44
+ xmlNodePtr manual=NULL; /* manual */
45
+ xmlNodePtr step1_pre=NULL; /* step1 */
46
+ xmlNodePtr step1_post=NULL; /* step1 */
47
+
48
+ xmlNodePtr parse=NULL; /* parser rules */
49
+ xmlNodePtr pbreak=NULL;
50
+ xmlNodePtr pdbreak=NULL;
51
+
52
+ xmlNodePtr tc_words=NULL; /* term count dictionary */
53
+ xmlNodePtr tf_words=NULL; /* term frequency dictionary */
54
+
55
+
56
+ OtsStemRule * rule=Doc->stem;
57
+
58
+ char *local_dict_name;
59
+
60
+ local_dict_name = g_strdup_printf ("%s.xml", name);
61
+
62
+
63
+ if (g_file_test(local_dict_name,G_FILE_TEST_EXISTS))
64
+ doc = xmlParseFile (local_dict_name); /* it warns to the screen so we cant use it; enable for web services only */
65
+ if (doc == NULL) return (FALSE);
66
+
67
+ head = xmlDocGetRootElement (doc);
68
+ if (head == NULL)
69
+ {
70
+ fprintf (stderr, "empty document\n");
71
+ xmlFreeDoc (doc);
72
+ return (FALSE);
73
+ }
74
+
75
+ if (xmlStrcmp (head->name, (const xmlChar *) "dictionary"))
76
+ {
77
+ fprintf (stderr, "%s", head->name);
78
+ xmlFreeDoc (doc);
79
+ return (FALSE);
80
+ }
81
+
82
+ if (head != NULL)
83
+ stem = head->xmlChildrenNode;
84
+ while ((stem != NULL)
85
+ && (xmlStrcmp (stem->name, (const xmlChar *) "stemmer")))
86
+ {
87
+ stem = stem->next;
88
+ }
89
+
90
+ if (head != NULL)
91
+ parse = head->xmlChildrenNode;
92
+ while ((parse != NULL)
93
+ && (xmlStrcmp (parse->name, (const xmlChar *) "parser")))
94
+ {
95
+ parse = parse->next;
96
+ }
97
+
98
+ if (head != NULL)
99
+ tc_words = head->xmlChildrenNode;
100
+ while ((tc_words != NULL)
101
+ && (xmlStrcmp (tc_words->name, (const xmlChar *) "grader-tc")))
102
+ {
103
+ tc_words = tc_words->next;
104
+ }
105
+
106
+
107
+ if (head != NULL)
108
+ tf_words = head->xmlChildrenNode;
109
+ while ((tf_words != NULL)
110
+ && (xmlStrcmp (tf_words->name, (const xmlChar *) "grader-tf")))
111
+ {
112
+ tf_words = tf_words->next;
113
+ }
114
+
115
+
116
+
117
+ if (stem != NULL)
118
+ pre = stem->xmlChildrenNode;
119
+ while ((pre != NULL) && (xmlStrcmp (pre->name, (const xmlChar *) "pre")))
120
+ {
121
+ pre = pre->next;
122
+ }
123
+
124
+ if (stem != NULL)
125
+ post = stem->xmlChildrenNode;
126
+ while ((post != NULL) && (xmlStrcmp (post->name, (const xmlChar *) "post")))
127
+ {
128
+ post = post->next;
129
+ }
130
+
131
+
132
+ if (stem != NULL)
133
+ syno = stem->xmlChildrenNode;
134
+ while ((syno != NULL)
135
+ && (xmlStrcmp (syno->name, (const xmlChar *) "synonyms")))
136
+ {
137
+ syno = syno->next;
138
+ }
139
+
140
+ if (stem != NULL)
141
+ manual = stem->xmlChildrenNode;
142
+ while ((manual != NULL)
143
+ && (xmlStrcmp (manual->name, (const xmlChar *) "manual")))
144
+ {
145
+ manual = manual->next;
146
+ }
147
+
148
+
149
+ if (stem != NULL)
150
+ step1_pre = stem->xmlChildrenNode;
151
+ while ((step1_pre != NULL)
152
+ && (xmlStrcmp (step1_pre->name, (const xmlChar *) "step1_pre")))
153
+ {
154
+ step1_pre = step1_pre->next;
155
+ }
156
+
157
+
158
+
159
+ if (stem != NULL)
160
+ step1_post = stem->xmlChildrenNode;
161
+ while ((step1_post != NULL)
162
+ && (xmlStrcmp (step1_post->name, (const xmlChar *) "step1_post")))
163
+ {
164
+ step1_post = step1_post->next;
165
+ }
166
+
167
+
168
+ if (pre != NULL)
169
+ pre = pre->xmlChildrenNode; /*point to first word */
170
+ while (pre != NULL)
171
+ {
172
+ if (0 == xmlStrcmp (pre->name, (const xmlChar *) "rule"))
173
+ rule->RemovePre =
174
+ g_list_append (rule->RemovePre,
175
+ (xmlNodeListGetString
176
+ (doc, pre->xmlChildrenNode, 1)));
177
+ pre = pre->next;
178
+ }
179
+
180
+
181
+ if (post != NULL)
182
+ post = post->xmlChildrenNode;
183
+ while (post != NULL)
184
+ {
185
+ if (0 == xmlStrcmp (post->name, (const xmlChar *) "rule"))
186
+ rule->RemovePost =
187
+ g_list_append (rule->RemovePost,
188
+ (xmlNodeListGetString
189
+ (doc, post->xmlChildrenNode, 1)));
190
+ post = post->next;
191
+ }
192
+
193
+ if (syno != NULL)
194
+ syno = syno->xmlChildrenNode;
195
+ while (syno != NULL)
196
+ {
197
+ if (0 == xmlStrcmp (syno->name, (const xmlChar *) "rule"))
198
+ rule->synonyms =
199
+ g_list_append (rule->synonyms,
200
+ (xmlNodeListGetString
201
+ (doc, syno->xmlChildrenNode, 1)));
202
+ syno = syno->next;
203
+ }
204
+
205
+ if (manual != NULL)
206
+ manual = manual->xmlChildrenNode;
207
+ while (manual != NULL)
208
+ {
209
+ if (0 == xmlStrcmp (manual->name, (const xmlChar *) "rule"))
210
+ rule->manual =
211
+ g_list_append (rule->manual,
212
+ (xmlNodeListGetString
213
+ (doc, manual->xmlChildrenNode, 1)));
214
+ manual = manual->next;
215
+ }
216
+
217
+
218
+
219
+
220
+ if (step1_pre != NULL)
221
+ step1_pre = step1_pre->xmlChildrenNode;
222
+ while (step1_pre != NULL)
223
+ {
224
+ if (0 == xmlStrcmp (step1_pre->name, (const xmlChar *) "rule"))
225
+ rule->step1_pre =
226
+ g_list_append (rule->step1_pre,
227
+ (xmlNodeListGetString
228
+ (doc, step1_pre->xmlChildrenNode, 1)));
229
+ step1_pre = step1_pre->next;
230
+ }
231
+
232
+
233
+
234
+ if (step1_post != NULL)
235
+ step1_post = step1_post->xmlChildrenNode;
236
+ while (step1_post != NULL)
237
+ {
238
+ if (0 == xmlStrcmp (step1_post->name, (const xmlChar *) "rule"))
239
+ rule->step1_post =
240
+ g_list_append (rule->step1_post,
241
+ (xmlNodeListGetString
242
+ (doc, step1_post->xmlChildrenNode, 1)));
243
+ step1_post = step1_post->next;
244
+ }
245
+
246
+ if (parse != NULL)
247
+ pbreak = parse->xmlChildrenNode;
248
+ while ((pbreak != NULL) && (xmlStrcmp (pbreak->name, (const xmlChar *) "linebreak")))
249
+ {
250
+ pbreak = pbreak->next;
251
+ }
252
+
253
+
254
+
255
+ if (parse != NULL)
256
+ pdbreak = parse->xmlChildrenNode;
257
+ while ((pdbreak != NULL) && (xmlStrcmp (pdbreak->name, (const xmlChar *) "linedontbreak")))
258
+ {
259
+ pdbreak = pdbreak->next;
260
+ }
261
+
262
+
263
+ /*Parser break*/
264
+ if (pbreak != NULL)
265
+ pbreak = pbreak->xmlChildrenNode;
266
+ while (pbreak != NULL)
267
+ {
268
+ if (0 == xmlStrcmp (pbreak->name, (const xmlChar *) "rule"))
269
+ rule->ParserBreak =
270
+ g_list_append (rule->ParserBreak,
271
+ (xmlNodeListGetString
272
+ (doc, pbreak->xmlChildrenNode, 1)));
273
+ pbreak = pbreak->next;
274
+ }
275
+
276
+ /*Parser Don't break*/
277
+ if (pdbreak != NULL)
278
+ pdbreak = pdbreak->xmlChildrenNode;
279
+ while (pdbreak != NULL)
280
+ {
281
+ if (0 == xmlStrcmp (pdbreak->name, (const xmlChar *) "rule"))
282
+ rule->ParserDontBreak =
283
+ g_list_append (rule->ParserDontBreak,
284
+ (xmlNodeListGetString
285
+ (doc, pdbreak->xmlChildrenNode, 1)));
286
+ pdbreak = pdbreak->next;
287
+ }
288
+
289
+ /*Term Count load dict*/
290
+
291
+ if (tc_words != NULL)
292
+ tc_words = tc_words->xmlChildrenNode;
293
+ while (tc_words != NULL)
294
+ {
295
+ if (0 == xmlStrcmp (tc_words->name, (const xmlChar *) "word"))
296
+ {
297
+ xmlChar *key;
298
+ key=xmlNodeListGetString(doc, tc_words->xmlChildrenNode,1);
299
+ Doc->dict = g_list_append (Doc->dict,(gpointer)ots_new_wordEntery(key));
300
+ xmlFree(key);
301
+ }
302
+ tc_words = tc_words->next;
303
+ }
304
+
305
+
306
+ /*Term Frequency load dict*/
307
+
308
+ if (tf_words != NULL)
309
+ tf_words = tf_words->xmlChildrenNode;
310
+ while (tf_words != NULL)
311
+ {
312
+ if (0 == xmlStrcmp (tf_words->name, (const xmlChar *) "word"))
313
+ {
314
+ xmlChar *key;
315
+ xmlChar *idf_key;
316
+ key=xmlNodeListGetString(doc, tf_words->xmlChildrenNode,1);
317
+
318
+ idf_key=xmlGetProp(tf_words,"idf");
319
+ Doc->tf_terms = g_list_append (Doc->tf_terms,ots_new_OtsWordTF(key,atof(idf_key)));
320
+ xmlFree(key);
321
+ xmlFree(idf_key);
322
+ }
323
+ tf_words = tf_words->next;
324
+ }
325
+
326
+
327
+ xmlFreeDoc(doc);
328
+ xmlCleanupParser ();
329
+ g_free(local_dict_name);
330
+ return (TRUE);
331
+ }
@@ -0,0 +1,6 @@
1
+ require 'mkmf'
2
+
3
+ $CFLAGS = ENV["CFLAGS"].to_s + " " + `pkg-config --cflags glib-2.0 libxml-2.0`.chomp
4
+ $LDFLAGS = ENV["LDFLAGS"].to_s + " " + `pkg-config --libs glib-2.0 libxml-2.0`.chomp
5
+
6
+ create_makefile('summarize/summarize')
@@ -0,0 +1,185 @@
1
+ /*
2
+ * grader-tc.c
3
+ *
4
+ * Copyright (C) 2003 Nadav Rotem <nadav256@hotmail.com>
5
+ *
6
+ * This program is free software; you can redistribute it and/or modify
7
+ * it under the terms of the GNU General Public License as published by
8
+ * the Free Software Foundation; either version 2 of the License, or
9
+ * (at your option) any later version.
10
+ *
11
+ * This program is distributed in the hope that it will be useful,
12
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
13
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14
+ * GNU Library General Public License for more details.
15
+ *
16
+ * You should have received a copy of the GNU General Public License
17
+ * along with this program; if not, write to the Free Software
18
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
19
+ */
20
+
21
+ #include <stdio.h>
22
+ #include <stdlib.h>
23
+ #include <string.h>
24
+ #include "libots.h"
25
+
26
+
27
+ #include "grader-tc.h"
28
+
29
+
30
+ /*Grader - Term count algorithm*/
31
+ /*This is non-normelized term frequency algorithm without using inverse document frequency database */
32
+
33
+ #define NUM_KEY_WORDS 100 /* use first n key words only */
34
+
35
+ int
36
+ ots_get_article_word_count (const OtsArticle * Doc)
37
+ {
38
+ GList *li;
39
+ int articleWC;
40
+ articleWC = 0;
41
+
42
+ if (Doc==NULL) return 0;
43
+
44
+ for (li = (GList *) Doc->lines; li != NULL; li = li->next)
45
+ {
46
+ articleWC += ((OtsSentence *) li->data)->wc;
47
+ }
48
+
49
+ return articleWC;
50
+ }
51
+
52
+
53
+ /*take this line and add each word to the "wordStat" list
54
+ * this list will hold all of the words in the article and the number
55
+ * of times they appeared in the article.
56
+ */
57
+
58
+ static void
59
+ ots_line_add_wordlist(OtsArticle * Doc,const OtsSentence * aLine)
60
+ {
61
+ GList *li;
62
+ if ((aLine==NULL) ||(NULL==Doc)) { return;}
63
+
64
+ for (li = (GList *) aLine->words; li != NULL; li = li->next) /* for each word in the sentence Do: */
65
+ if (li->data && strlen (li->data)) ots_add_wordstat (Doc, (char *)li->data);
66
+
67
+ return;
68
+ }
69
+
70
+ static void
71
+ ots_create_wordlist(OtsArticle * Doc)
72
+ {
73
+ GList *line;
74
+ if (Doc==NULL) return;
75
+
76
+ for (line = (GList *) Doc->lines; line != NULL; line = line->next)
77
+ {
78
+ OtsSentence * aLine=line->data;
79
+ if (aLine)
80
+ ots_line_add_wordlist(Doc,aLine);
81
+ }
82
+ }
83
+
84
+
85
+
86
+
87
+ static int
88
+ keyVal (const int n) /* Ugly , I know */
89
+ {
90
+ if (n == 1) return 3;
91
+ if (n == 2) return 2;
92
+ if (n == 3) return 2;
93
+ if (n == 4) return 2;
94
+ return 1;
95
+ }
96
+
97
+
98
+ static void
99
+ ots_grade_line (GList *impList, OtsSentence * aLine,
100
+ OtsStemRule * rule)
101
+ {
102
+ GList *li;
103
+ GList *di;
104
+ int n;
105
+ char *tmp_stem;
106
+
107
+ if ((aLine==NULL)||(rule==NULL)||(impList==NULL)) return;
108
+
109
+ for (li = (GList *) aLine->words; li != NULL; li = li->next) /* for each word */
110
+ {
111
+ n = 0;
112
+ tmp_stem = ots_stem_strip ((unsigned char *) li->data, rule);
113
+
114
+ for (di = (GList *) impList;
115
+ ((di != NULL) && (n < NUM_KEY_WORDS)); di = di->next)
116
+ {
117
+ n++;
118
+ if ((NULL!=((OtsWordEntery *) di->data)->stem) && (NULL!=tmp_stem))
119
+ if (0 == strcmp ((((OtsWordEntery *) di->data)->stem), tmp_stem))
120
+ {
121
+ /* debug:
122
+ if (0!=strcmp((((OtsWordEntery *) di->data)->word),li->data))
123
+ printf("[%s][%s] stem[%s]\n",(((OtsWordEntery *) di->data)->word),li->data,tmp);*/
124
+
125
+ aLine->score += (((OtsWordEntery *) di->data)->occ) * keyVal (n);
126
+ }
127
+
128
+ }
129
+
130
+ g_free (tmp_stem);
131
+ }
132
+
133
+ }
134
+
135
+
136
+ void
137
+ ots_create_title_tc(OtsArticle * Doc)
138
+ {
139
+
140
+ char *tmp;
141
+ char *word;
142
+ int i;
143
+ GString *title;
144
+ if (NULL==Doc) return;
145
+
146
+ title=g_string_new(NULL);
147
+
148
+ for (i=0;i<5;i++)
149
+ {
150
+ word = ots_word_in_list(Doc->ImpWords,i);
151
+ if (word) g_string_append(title,word); else break;
152
+ if (i<4) g_string_append(title,",");
153
+ }
154
+
155
+ tmp=title->str;
156
+ if (NULL!=title) g_string_free(title,FALSE);
157
+ Doc->title=tmp;
158
+ }
159
+
160
+
161
+ void
162
+ ots_grade_doc_tc (OtsArticle * Doc)
163
+ {
164
+
165
+ GList *li;
166
+ if (NULL==Doc) return;
167
+ ots_create_wordlist(Doc);
168
+
169
+
170
+ Doc->ImpWords=ots_union_list (Doc->wordStat, Doc->dict); /* subtract from the Article wordlist all the words in the dic file (on , the , is...) */
171
+ Doc->ImpWords=ots_sort_list (Doc->ImpWords); /* sort the list , top 3 is what the article talks about (SARS , virus , cure ... ) */
172
+
173
+ /*to print wordlist: ots_print_wordlist (stdout, Doc->ImpWords);*/
174
+
175
+ if (0 == Doc->lineCount) return;
176
+
177
+ for (li = (GList *) Doc->lines; li != NULL; li = li->next)
178
+ {
179
+ if (li->data)
180
+ ots_grade_line (Doc->ImpWords, (OtsSentence *) li->data, Doc->stem);
181
+ }
182
+
183
+
184
+ ots_create_title_tc(Doc);
185
+ }