Disabled new implementation in favor of the old one - all that's needed is a c implementation of apply delta data

Author: Byron

_delta_apply.c

@@ -450,6 +450,10 @@ void DCV_replace_one_by_many(const DeltaChunkVector* from, DeltaChunkVector* to,
 
 	// If we are somewhere in the middle, we have to make some space
 	if (DCV_last(to) != at) {
+		// IMPORTANT: This memmove kills the performance in case of large deltas
+		// Causing everything to slow down enormously. Its logical, as the memory
+		// gets shifted each time we insert nodes, for each chunk, for ever smaller
+		// chunks depending on the deltas
 		memmove((void*)(at+from->size), (void*)(at+1), (size_t)((DCV_end(to) - (at+1)) * sizeof(DeltaChunk)));
 	}
 

stream.py

@@ -325,7 +325,7 @@ def __init__(self, stream_list):
 		self._dstreams = tuple(stream_list[:-1])
 		self._br = 0
 
-	def _set_cache_(self, attr):
+	def _set_cache_too_slow(self, attr):
 		# the direct algorithm is fastest and most direct if there is only one 
 		# delta. Also, the extra overhead might not be worth it for items smaller
 		# than X - definitely the case in python, every function call costs 
@@ -360,7 +360,7 @@ def _set_cache_(self, attr):
 
 		self._mm_target.seek(0)
 
-	def _set_cache_brute_(self, attr):
+	def _set_cache_(self, attr):
 		"""If we are here, we apply the actual deltas"""
 
 		buffer_info_list = list()