Add language to detect_all output

chardet/__init__.py

@@ -24,12 +24,15 @@
 __all__ = ['UniversalDetector', 'detect', 'detect_all', '__version__', 'VERSION']
 
 
-def detect(byte_str):
+def detect(byte_str, chunk_size=100000):
     """
     Detect the encoding of the given byte string.
 
     :param byte_str:     The byte sequence to examine.
     :type byte_str:      ``bytes`` or ``bytearray``
+    :param chunk_size:   Number of bytes to pass to feed into underlying
+                         UniversalDetector at a time.
+    :type chunk_size:    ``int``
     """
     if not isinstance(byte_str, bytearray):
         if not isinstance(byte_str, bytes):
@@ -38,7 +41,10 @@ def detect(byte_str):
         else:
             byte_str = bytearray(byte_str)
     detector = UniversalDetector()
-    detector.feed(byte_str)
+    for i in range(len(byte_str) // chunk_size + 1):
+        detector.feed(byte_str[i * chunk_size: (i + 1) * chunk_size])
+        if detector.done:
+            break
     return detector.close()
 
 
@@ -55,7 +61,6 @@ def detect_all(byte_str):
                             '{}'.format(type(byte_str)))
         else:
             byte_str = bytearray(byte_str)
-
     detector = UniversalDetector()
     detector.feed(byte_str)
     detector.close()
@@ -74,7 +79,8 @@ def detect_all(byte_str):
                                                             charset_name)
                 results.append({
                     'encoding': charset_name,
-                    'confidence': prober.get_confidence()
+                    'confidence': prober.get_confidence(),
+                    'language': prober.language,
                 })
         if len(results) > 0:
             return sorted(results, key=lambda result: -result['confidence'])