Add reset() method to EncodingDetector

chatnoir-eu · Mar 6, 2023 · a43df7a · a43df7a
1 parent 310aa25
commit a43df7a
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 0 deletions.
diff --git a/resiliparse/resiliparse/parse/encoding.pxd b/resiliparse/resiliparse/parse/encoding.pxd
@@ -21,6 +21,7 @@ cdef class EncodingDetector:
 
     cpdef void update(self, const string& data)
     cpdef str encoding(self, bint html5_compatible=*)
+    cpdef void reset(self)
 
 cpdef str detect_encoding(bytes data, size_t max_len=*, bint html5_compatible=*, bint from_html_meta=*)
 cpdef str bytes_to_str(bytes data, str encoding=*, str errors=*, fallback_encodings=*, bint strip_bom=*)

diff --git a/resiliparse/resiliparse/parse/encoding.pyx b/resiliparse/resiliparse/parse/encoding.pyx
@@ -163,6 +163,14 @@ cdef class EncodingDetector:
 
         return enc
 
+    cpdef void reset(self):
+        """
+        reset(self)
+        
+        Manually reset the encoding detector state.
+        """
+        uchardet_reset(self.d)
+
     def __dealloc__(self):
         if self.d != NULL:
             uchardet_delete(self.d)