minor changes

cat/versions_available/tmrepository_enhr.cfg

@@ -0,0 +1,8 @@
+dir@#@tmrepository_enhr
+concordia_host@#@concordia.vm.wmi.amu.edu.pl
+concordia_port@#@8800
+tmid@#@1
+desc@#@Welcome to Concordia. The system finds the longest matches of the pattern sentence in its translation memory. This translation memory is over 1M sentences from the TMrepository system (http://concordia.vm.wmi.amu.edu.pl/tmrepository). Please enter a Croatian sentence in the field below and press Enter (or use the search button). You can test the system on predefined samples, simply use the link "show/hide samples" and apply one of the sample sentences. After the search, click on the highlighted fragments to see their context.
+enjoy@#@Enjoy your work with the system!
+prompt@#@Enter search pattern (English sentence):
+suggestion@#@BiHs Komsic resigns from his party

cat/versions_enabled/stocznia_enpl.cfg

@@ -1 +0,0 @@
-../versions_available/stocznia_enpl.cfg

cat/versions_enabled/stocznia_plen.cfg

@@ -1 +0,0 @@
-../versions_available/stocznia_plen.cfg

cat/versions_enabled/tmrepository_enhr.cfg

@@ -0,0 +1 @@
+../versions_available/tmrepository_enhr.cfg

mgiza-aligner/corpus-compilator/Makefile

@@ -1,6 +1,6 @@
 SRC_LANG=pl
 TRG_LANG=en
-CORPUS_NAME=europarl_sample
+CORPUS_NAME=opus
 DICTIONARY_NAME=classyf_popular_medicine
 SEPARATOR=@\#@
 CORPUS_CHUNK_SIZE=100000

mgiza-aligner/corpus-compilator/filter.sh

@@ -0,0 +1,12 @@
+#!/bin/sh
+
+DICTIONARY_NAME=classyf_popular_medicine
+CORPUS_NAME=opus
+
+make clean-filtering
+make dictionaries/$DICTIONARY_NAME.lem
+
+./get_corpus_lines.py dictionaries/$DICTIONARY_NAME.lem corpora/$CORPUS_NAME/report.txt > corpora/$CORPUS_NAME/corpus_lines.txt
+./compile.py corpora/$CORPUS_NAME/src_clean.txt corpora/$CORPUS_NAME/trg_clean.txt corpora/$CORPUS_NAME/corpus_lines.txt corpora/$CORPUS_NAME/src_filtered.txt corpora/$CORPUS_NAME/trg_filtered.txt
+
+

mgiza-aligner/corpus-compilator/setup_solr.sh

@@ -1,6 +1,6 @@
 #!/bin/sh
 
-SOLR_HOME=/home/rafalj/programs/solr-6.0.0
+SOLR_HOME=/home/rjawor/programs/solr-5.5.4
 
 $SOLR_HOME/bin/solr restart
 $SOLR_HOME/bin/solr create -c corpus_compiler

tests/addLemmatizedTM.sh

@@ -1,7 +1,7 @@
 #!/bin/sh
 
-CORPUS_NAME="europarl_sample"
+CORPUS_NAME="tmrepository_enhr"
-SRC_LANG_ID=1
+SRC_LANG_ID=2
-TRG_LANG_ID=2
+TRG_LANG_ID=6
 
 ./addAlignedLemmatizedTM.py $CORPUS_NAME ../mgiza-aligner/corpora/$CORPUS_NAME/src_final.txt $SRC_LANG_ID ../mgiza-aligner/corpora/$CORPUS_NAME/trg_final.txt $TRG_LANG_ID ../mgiza-aligner/corpora/$CORPUS_NAME/aligned_final.txt

tests/generateIndex.py

@@ -0,0 +1,28 @@
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
+
+import unittest
+import json
+import urllib2
+import sys
+import host
+import time
+
+address = 'http://'+host.concordia_host
+if len(host.concordia_port) > 0:
+    address += ':'+host.concordia_port
+
+
+
+print "Generating index..."
+start = time.time()
+data = {
+    'operation': 'refreshIndex',
+    'tmId' : 1
+}
+req = urllib2.Request(address)
+req.add_header('Content-Type', 'application/json')
+urllib2.urlopen(req, json.dumps(data)).read()
+
+end = time.time()
+print "Index regeneration complete. The operation took %.4f s" % (end - start)