summarize 1.0.1 → 1.0.2

data/README.markdown

@@ -12,7 +12,7 @@
     cd summarize
     rake build
     gem build summarize.gemspec
-    gem install summarize-1.0.gem
+    gem install summarize-1.0.1.gem
 
 ## Usage
 
@@ -27,7 +27,7 @@ Or use the String method
 By default it uses an English dictionary for summarizing but forty languages are supported. Pass in the valid ISO 639 language code to use one. A ratio (default is 25%) can also be passed in.
 
     # Parse an article using Portuguese stemming rules with a ratio of 50%
-    "text to summarize".summarize(:language => 'pt', :ratio => 50)
+    "texto para sumariar".summarize(:language => 'pt', :ratio => 50)
 
 You can also use custom stemming rules
 

data/ext/summarize/extconf.rb

@@ -1,6 +1,10 @@
 require 'mkmf'
 
-$CFLAGS = ENV["CFLAGS"].to_s + " " + `pkg-config --cflags glib-2.0 libxml-2.0`.chomp
-$LDFLAGS = ENV["LDFLAGS"].to_s  + " " + `pkg-config --libs glib-2.0 libxml-2.0`.chomp
+%w(glib-2.0 libxml-2.0).each do |lib|
+  pkg_config lib
+end
+
+# $CFLAGS = ENV["CFLAGS"].to_s + " " + `pkg-config --cflags glib-2.0 libxml-2.0`.chomp
+# $LDFLAGS = ENV["LDFLAGS"].to_s  + " " + `pkg-config --libs glib-2.0 libxml-2.0`.chomp
 
 create_makefile('summarize/summarize')

data/ext/summarize/grader-tf.c

@@ -102,11 +102,14 @@ N = (total-number-of-sentences)
 f = n/N
 */
 
+/*
+ssoper: unused and causing warning messages
+
 double
 ots_calc_idf (const int term_count,const int doc_word_count)
 {
 return -log(doc_word_count/term_count);
-}
+}*/
 
 double
 ots_calc_tf (const int term_count,const int doc_word_count)

data/ext/summarize/highlighter.c

@@ -31,7 +31,7 @@ static int
 ots_highlight_max_line (OtsArticle * Doc)
 {
   GList *li;
-  int max = 0;
+  long int max = 0;
   for (li = (GList *) Doc->lines; li != NULL; li = li->next)
     {
       if (0 == (((OtsSentence *) li->data)->selected))	/* if not selected , count me in */

data/ext/summarize/libots.h

@@ -149,7 +149,7 @@ GList* ots_text_stem_list(const unsigned char *text,const unsigned char *lang_co
 
 
 /*Gives a score on the relations between two lists of topics; simmilar to the inner product*/
-int ots_topic_list_score(const GList *topic_list1,const GList *topic_list2);
+int ots_topic_list_score(GList *topic_list1, GList *topic_list2);
 
 G_END_DECLS
 

data/ext/summarize/parser.c

@@ -29,8 +29,7 @@
 int
 ots_match_post (const char *aWord,const char *post)
 {
-  int i, wlen, plen;
-  
+  long int i, wlen, plen;
 
   wlen = strlen (aWord);
   plen = strlen (post);
@@ -118,7 +117,7 @@ ots_parse_stream(const unsigned char *utf8, size_t len, OtsArticle * Doc)	/*pars
   OtsSentence *tmpLine = ots_append_line (Doc);
   OtsStemRule * rule=Doc->stem;
   gunichar uc;
-  int index = 0;
+  size_t index = 0;
   char *s = (char *) utf8;
   GString *word_buffer = g_string_new (NULL);
 

data/ext/summarize/relations.c

@@ -132,8 +132,8 @@ return topics;
 
 /*Gives a score on the relations between two lists of topics; simmilar to the inner product*/
 int ots_topic_list_score(
-const GList *topic_list1,
-const GList *topic_list2)
+GList *topic_list1,
+GList *topic_list2)
 {
 	int count=0;
 	GList *tmplist1;

data/ext/summarize/stemmer.c

@@ -70,7 +70,7 @@ if (rule != NULL)
 static void
 ots_stem_break (unsigned const char *comp,unsigned char *part_a,unsigned char *part_b)	/*given already alocated part_a and b */
 {				/*example "red|blue" */
-  int i, j, clen;
+  long int i, j, clen;
   i = 0;
   j = 0;
   
@@ -106,7 +106,7 @@ ots_stem_break (unsigned const char *comp,unsigned char *part_a,unsigned char *p
 static unsigned char *
 ots_stem_remove_pre (unsigned const char *aWord,unsigned const char *pre,unsigned const char *new)
 {
-  int i, plen, wlen, nlen;
+  long int i, plen, wlen, nlen;
   unsigned char *new_str = NULL;
 
   if (aWord==NULL) return NULL;
@@ -135,7 +135,7 @@ ots_stem_remove_pre (unsigned const char *aWord,unsigned const char *pre,unsigne
 static unsigned char *
 ots_stem_remove_post (unsigned const char *aWord,unsigned const char *post,unsigned const char *new)
 {
-  unsigned int i, wlen, plen, nlen;
+  unsigned long int i, wlen, plen, nlen;
   unsigned char *new_str = NULL;
 
   if ((NULL==aWord)||(NULL==post)||(NULL==new)) return NULL;

data/ext/summarize/summarize.c

@@ -9,15 +9,14 @@
 #include "libots.h"
 #include "summarize.h"
 
-const char *OTS_ERROR_BAD_DICT = "Cannot load dictionary file";
 
 void Init_summarize() {
  VALUE rb_mOts = rb_define_module("Summarize");
  rb_define_module_function(rb_mOts, "summarize", summarize, 3);
 }
 
-static VALUE summarize(const VALUE self, const VALUE rb_str, const VALUE rb_dict_file, const VALUE rb_ratio) {
-  int length = RSTRING_LEN(rb_str);
+static VALUE summarize(const VALUE self, volatile VALUE rb_str, volatile VALUE rb_dict_file, const VALUE rb_ratio) {
+  long int length = RSTRING_LEN(rb_str);
   char *text = StringValuePtr(rb_str);
   char *dictionary_file = StringValuePtr(rb_dict_file);
   int ratio = NUM2INT(rb_ratio);
@@ -27,7 +26,7 @@ static VALUE summarize(const VALUE self, const VALUE rb_str, const VALUE rb_dict
 
   if (!ots_load_xml_dictionary(doc, dictionary_file)) {
     ots_free_article(doc);
-    rb_raise(rb_eRuntimeError, OTS_ERROR_BAD_DICT);
+    rb_raise(rb_eRuntimeError, "Cannot load dictionary file");
     return Qnil;
   }
 

data/ext/summarize/text.c

@@ -57,7 +57,7 @@ static void
 ots_print_line (FILE * stream, const OtsSentence * aLine)
 {
   unsigned char *utf8_txt;
-  size_t len;
+  size_t len = 0;
   utf8_txt = ots_get_line_text (aLine, TRUE, &len);
   fwrite (utf8_txt, 1, len, stream);
   g_free (utf8_txt);
@@ -69,7 +69,7 @@ ots_get_doc_text (const OtsArticle * Doc, size_t * out_len)
   GList *li;
   GString *text;
   unsigned char *utf8_data;
-  size_t line_len;
+  size_t line_len = 0;
 
   text = g_string_new (NULL);
 

data/lib/summarize.rb

@@ -10,7 +10,7 @@ class Hash #:nodoc:
 end unless {}.respond_to? 'symbolize_keys'
 
 module Summarize
-  VERSION = "1.0.1"
+  VERSION = "1.0.2"
 
   LANGUAGES = [
     'bg', # Bulgarian
@@ -84,6 +84,9 @@ class String
   # language::
   #   An ISO 639-1 language code. See Summarize::LANGUAGES for the supported list.
   #
+  # dictionary::
+  #   A path to a custom stemming XML file
+  #
   # == Returns:
   # A string summary
   #
@@ -105,6 +108,9 @@ class File
   # language::
   #   An ISO 639-1 language code. See Summarize::LANGUAGES for the supported list.
   #
+  # dictionary::
+  #   A path to a custom stemming XML file
+  #
   # == Returns:
   # A string summary
   #

metadata

@@ -5,8 +5,8 @@ version: !ruby/object:Gem::Version
   segments: 
   - 1
   - 0
-  - 1
-  version: 1.0.1
+  - 2
+  version: 1.0.2
 platform: ruby
 authors: 
 - Sean Soper