summaryrefslogtreecommitdiff
path: root/sa-extract/sa2cdec.py
diff options
context:
space:
mode:
authorChris Dyer <cdyer@cab.ark.cs.cmu.edu>2012-05-23 16:39:28 -0400
committerChris Dyer <cdyer@cab.ark.cs.cmu.edu>2012-05-23 16:39:28 -0400
commit57339faa0074d35783213b03f921ee7ad13f9481 (patch)
treeec809850793af5171780a8ff88218cbcfac8b112 /sa-extract/sa2cdec.py
parent6cc7919ee64eada4b0c6d551d2a16cd5644a0266 (diff)
parent70e50adc591ca6e3e56a1ced66bdb133eb89807b (diff)
Merge branch 'master' of github.com:redpony/cdec
Diffstat (limited to 'sa-extract/sa2cdec.py')
-rwxr-xr-xsa-extract/sa2cdec.py19
1 files changed, 19 insertions, 0 deletions
diff --git a/sa-extract/sa2cdec.py b/sa-extract/sa2cdec.py
new file mode 100755
index 00000000..55fb19f3
--- /dev/null
+++ b/sa-extract/sa2cdec.py
@@ -0,0 +1,19 @@
+#!/usr/bin/env python
+import sys
+
+featNames = [ line.strip() for line in open(sys.argv[1]) if not line.startswith('#') ]
+
+for line in sys.stdin:
+ try:
+ (lhs, src, tgt, feats, align) = line.strip("\n").split(' ||| ')
+ except:
+ print >>sys.stderr, 'WARNING: No alignments:', line
+ try:
+ (lhs, src, tgt, feats) = line.strip().split(' ||| ')
+ align = ''
+ except:
+ print >>sys.stderr, "ERROR: Malformed line:", line
+ raise
+ featValues = feats.split(' ')
+ namedFeats = ' '.join( name+"="+value for (name, value) in zip(featNames, featValues) )
+ print " ||| ".join( (lhs, src, tgt, namedFeats, align) )