From 851e389dffdd6996ea32d70defb8906de80b9edc Mon Sep 17 00:00:00 2001
From: Chris Dyer <redpony@gmail.com>
Date: Mon, 14 Dec 2009 20:35:11 -0500
Subject: few small fixes of alignment tools, add new orthographic similarity
 feature for word aligner, final naming of directories, libraries in cdec

---
 decoder/tdict.cc | 49 +++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 49 insertions(+)
 create mode 100644 decoder/tdict.cc

(limited to 'decoder/tdict.cc')

diff --git a/decoder/tdict.cc b/decoder/tdict.cc
new file mode 100644
index 00000000..c00d20b8
--- /dev/null
+++ b/decoder/tdict.cc
@@ -0,0 +1,49 @@
+#include "Ngram.h"
+#include "dict.h"
+#include "tdict.h"
+#include "Vocab.h"
+
+using namespace std;
+
+Vocab* TD::dict_ = new Vocab;
+
+static const string empty;
+static const string space = " ";
+
+WordID TD::Convert(const std::string& s) {
+  return dict_->addWord((VocabString)s.c_str());
+}
+
+const char* TD::Convert(const WordID& w) {
+  return dict_->getWord((VocabIndex)w);
+}
+
+void TD::GetWordIDs(const std::vector<std::string>& strings, std::vector<WordID>* ids) {
+  ids->clear();
+  for (vector<string>::const_iterator i = strings.begin(); i != strings.end(); ++i)
+    ids->push_back(TD::Convert(*i));
+}
+
+std::string TD::GetString(const std::vector<WordID>& str) {
+  string res;
+  for (vector<WordID>::const_iterator i = str.begin(); i != str.end(); ++i)
+    res += (i == str.begin() ? empty : space) + TD::Convert(*i);
+  return res;
+}
+
+void TD::ConvertSentence(const std::string& sent, std::vector<WordID>* ids) {
+  string s = sent;
+  int last = 0;
+  ids->clear();
+  for (int i=0; i < s.size(); ++i)
+    if (s[i] == 32 || s[i] == '\t') {
+      s[i]=0;
+      if (last != i) {
+        ids->push_back(Convert(&s[last]));
+      }
+      last = i + 1;
+    }
+  if (last != s.size())
+    ids->push_back(Convert(&s[last]));
+}
+
-- 
cgit v1.2.3