From 5530575ae0ad939e17f08d6bd49978acea388ab7 Mon Sep 17 00:00:00 2001
From: Paul Baltescu <pauldb89@gmail.com>
Date: Mon, 28 Jan 2013 11:56:31 +0000
Subject: Initial working commit.

---
 extractor/grammar_extractor.cc | 45 ++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 45 insertions(+)
 create mode 100644 extractor/grammar_extractor.cc

(limited to 'extractor/grammar_extractor.cc')

diff --git a/extractor/grammar_extractor.cc b/extractor/grammar_extractor.cc
new file mode 100644
index 00000000..3014c2e9
--- /dev/null
+++ b/extractor/grammar_extractor.cc
@@ -0,0 +1,45 @@
+#include "grammar_extractor.h"
+
+#include <iterator>
+#include <sstream>
+#include <vector>
+
+using namespace std;
+
+vector<string> Tokenize(const string& sentence) {
+  vector<string> result;
+  result.push_back("<s>");
+
+  istringstream buffer(sentence);
+  copy(istream_iterator<string>(buffer),
+       istream_iterator<string>(),
+       back_inserter(result));
+
+  result.push_back("</s>");
+  return result;
+}
+
+GrammarExtractor::GrammarExtractor(
+    shared_ptr<SuffixArray> source_suffix_array,
+    shared_ptr<DataArray> target_data_array,
+    const Alignment& alignment, const Precomputation& precomputation,
+    int min_gap_size, int max_rule_span, int max_nonterminals,
+    int max_rule_symbols, bool use_baeza_yates) :
+    vocabulary(make_shared<Vocabulary>()),
+    rule_factory(source_suffix_array, target_data_array, alignment,
+        vocabulary, precomputation, min_gap_size, max_rule_span,
+        max_nonterminals, max_rule_symbols, use_baeza_yates) {}
+
+void GrammarExtractor::GetGrammar(const string& sentence) {
+  vector<string> words = Tokenize(sentence);
+  vector<int> word_ids = AnnotateWords(words);
+  rule_factory.GetGrammar(word_ids);
+}
+
+vector<int> GrammarExtractor::AnnotateWords(const vector<string>& words) {
+  vector<int> result;
+  for (string word: words) {
+    result.push_back(vocabulary->GetTerminalIndex(word));
+  }
+  return result;
+}
-- 
cgit v1.2.3