aboutsummaryrefslogtreecommitdiffstats
path: root/lowercasing_test/src/tests/lowercasing/casingvariants_fastlib.cpp
diff options
context:
space:
mode:
authorJon Bratseth <bratseth@yahoo-inc.com>2016-06-15 23:09:44 +0200
committerJon Bratseth <bratseth@yahoo-inc.com>2016-06-15 23:09:44 +0200
commit72231250ed81e10d66bfe70701e64fa5fe50f712 (patch)
tree2728bba1131a6f6e5bdf95afec7d7ff9358dac50 /lowercasing_test/src/tests/lowercasing/casingvariants_fastlib.cpp
Publish
Diffstat (limited to 'lowercasing_test/src/tests/lowercasing/casingvariants_fastlib.cpp')
-rw-r--r--lowercasing_test/src/tests/lowercasing/casingvariants_fastlib.cpp41
1 files changed, 41 insertions, 0 deletions
diff --git a/lowercasing_test/src/tests/lowercasing/casingvariants_fastlib.cpp b/lowercasing_test/src/tests/lowercasing/casingvariants_fastlib.cpp
new file mode 100644
index 00000000000..d7478c4755a
--- /dev/null
+++ b/lowercasing_test/src/tests/lowercasing/casingvariants_fastlib.cpp
@@ -0,0 +1,41 @@
+// Copyright 2016 Yahoo Inc. Licensed under the terms of the Apache 2.0 license. See LICENSE in the project root.
+#include <vespa/fastos/fastos.h>
+#include <vespa/fastlib/text/normwordfolder.h>
+#include <fstream>
+#include <iostream>
+
+ucs4_t
+getUCS4Char(const char *src)
+{
+ return Fast_UnicodeUtil::GetUTF8Char(src);
+}
+
+int
+main(int argc, char ** argv)
+{
+ assert(argc == 3);
+ (void) argc;
+ std::ifstream input(argv[1]);
+ std::ifstream ref(argv[2]);
+ Fast_NormalizeWordFolder wordFolder;
+ char inputBuf[128];
+ char refBuf[128];
+ char lowerBuf[128];
+ while (input.good()) {
+ input.getline(inputBuf, 128);
+ ref.getline(refBuf, 128);
+ ucs4_t inputChar = getUCS4Char(inputBuf);
+ ucs4_t refChar = getUCS4Char(refBuf);
+ ucs4_t lowerChar = wordFolder.ToFold(inputChar);
+ Fast_UnicodeUtil::utf8ncopy(lowerBuf, &lowerChar, 128, 1);
+ if (refChar != lowerChar) {
+ printf("input(%s,%u,0x%X), lower(%s,%u,0x%X), ref(%s,%u,0x%X) \n",
+ inputBuf, inputChar, inputChar,
+ lowerBuf, lowerChar, lowerChar,
+ refBuf, refChar, refChar);
+ }
+ }
+ input.close();
+ return 0;
+}
+