Merge branch 'master' of ssh://github.com/pks/nlp_scripts

1 files changed, 18 insertions, 0 deletions

diff --git a/langid-polyglot b/langid-polyglot
new file mode 100755
index 0000000..0b0b20c
--- /dev/null
+++ b/langid-polyglot
@@ -0,0 +1,18 @@
+#!/usr/bin/env python
+
+import polyglot
+from polyglot.detect import Detector
+import fileinput
+
+for line in fileinput.input():
+    try:
+        for lang in Detector(line).languages:
+            if lang.confidence > 80.0:
+                print(lang.confidence)
+            else:
+                print("??")
+            break
+    except polyglot.detect.base.UnknownLanguage:
+        print("??")
+        pass
+