summaryrefslogtreecommitdiff
path: root/extools/sentence_pair.h
diff options
context:
space:
mode:
authortrevor.cohn <trevor.cohn@ec762483-ff6d-05da-a07a-a48fb63a330f>2010-07-15 00:34:58 +0000
committertrevor.cohn <trevor.cohn@ec762483-ff6d-05da-a07a-a48fb63a330f>2010-07-15 00:34:58 +0000
commit3ca6d4b4bbbd0401ab25e7731ce84afb118635d8 (patch)
tree6ce51c1d8fef8ebbd0649b946dd983950295cc25 /extools/sentence_pair.h
parent1350b8e8e465acc9d4d8d43d807cc6093e8f37b9 (diff)
Massacred the pipeline to support source language phrases and contexts.
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@255 ec762483-ff6d-05da-a07a-a48fb63a330f
Diffstat (limited to 'extools/sentence_pair.h')
-rw-r--r--extools/sentence_pair.h8
1 files changed, 5 insertions, 3 deletions
diff --git a/extools/sentence_pair.h b/extools/sentence_pair.h
index d78be359..b5a7ca93 100644
--- a/extools/sentence_pair.h
+++ b/extools/sentence_pair.h
@@ -1,8 +1,10 @@
#ifndef _SENTENCE_PAIR_H_
#define _SENTENCE_PAIR_H_
+#include <map>
#include <utility>
#include <vector>
+#include <boost/tuple/tuple.hpp>
#include "wordid.h"
#include "array2d.h"
@@ -22,12 +24,12 @@ struct AnnotatedParallelSentence {
std::vector<std::vector<std::pair<short, short> > > aligns_by_fword;
// span type information
- Array2D<std::vector<WordID> > span_types; // span_types(i,j) is the list of category
- // types for a span (i,j) in the TARGET language.
+ std::map< boost::tuple<short,short,short,short>, std::vector<WordID> > span_types;
+ // span_types(i,j,k,l) is the list of category span (i,j) in source and (k,l) in the target language.
int f_len, e_len;
- static int ReadAlignmentPoint(const char* buf, int start, int end, bool permit_col, short* a, short* b);
+ static int ReadAlignmentPoint(const char* buf, int start, int end, bool permit_col, short* a, short* b, short* c, short* d);
private:
void Reset();